THE LITERATE LEXICON IN NARRATIVE AND EXPOSITORY WRITING:
A DEVELOPMENTAL STUDY OF CHILDREN AND ADOLESCENTS
by
LEI SUN
A DISSERTATION
Presented to the Department of Special Education
and Clinical Sciences
and the Graduate School of the University of Oregon
in partial fulfillment of the requirements
for the degree of
Doctor of Philosophy
September 2008
11
University of Oregon Graduate School
Confirmation' of Approval and Acceptance of Dissertation prepared by:
Lei Sun
Title:
"The Literate Lexicon in Narrative and Expository Writing: A Developmental Study of Children
and Adolescents"
This dissertation has been accepted and approved in partial fulfillment of the requirements for
the degree in the Department of Special Education and Clinical Sciences by:
Marilyn Nippold, Chairperson, Special Education and Clinical Sciences
Roland Good, Member, Special Education and Clinical Sciences
Deborah Olson, Member, Special Education and Clinical Sciences
Nathaniel Teich, Outside Member, English
and Richard Linton, Vice President for Research and Graduate Studies/Dean of the Graduate
School for the University of Oregon.
September 6, 2008
Original approval signatures are on file with the Graduate School and the University of Oregon
Libraries.
© 2008 LEI SUN
111
IV
An Abstract of the Dissertation of
Lei Sun for the degree of Doctor of Philosophy
in the Department of Special Education and Clinical Sciences
to be taken September 2008
Title: THE LITERATE LEXICON IN NARRATIVE AND EXPOSITORY WRITING:
A DEVELOPMENTAL STUDY OF CHILDREN AND ADOLESCENTS
Three types of literate words, including abstract nouns (freedom, challenge), mental
state verbs (assume, explain), and derivatives (relationship, respectful), were examined 1Il
narrative and expository 'writing in typically develuping children and adolescents. It was
predicted that older students would use literate words more frequently than younger
students, and that literate words would occur more frequently in expository writing than in
narrative writing. One hundred and twenty typically developing children and adolescents
including forty 5th graders, forty 8th graders, and forty 11 th graders wrote one narrative
and one expository essay at school. The results showed that genre had a substantial impact
on the use ofliterate words in the writing of school-age children lli'ld adolescents, except
metalinguistic verbs. Moreover, literate words were used significantly more often in
expository than in narrative text, except derived adjectives. Additionally, metalinguistic
verbs occurred more often in narrative writing than expository writing; however, the
difference was not statistically significant.
Age-related increases occurred in the use of abstract nouns, derived nominals,
early/late developing metacognitive verbs and late developing metalinguistic verbs in
narrative writing. Age-related increases also occurred in the use of derived adjectives, and
late developing metacognitive and metalinguistic verbs in expository writing. The present
study adds to the knowledge base concerning the development of literate word use in
narrative and expository writing in typically developing children and adolescents.
v
CURRICULUM VITAE
NAME OF AUTHOR: Lei Sun
PLACE OF BIRTH: Taipei, Taiwan
GRADUATE AND UNDERGRADUATE SCHOOLS ATTENDED:
University of Oregon
Shih Hsin University, Taipei, Taiwan
National Taipei University, Taipei, Taiwan
DEGREES AWARDED:
Doctor of Philosophy in Communication Disorders and Sciences, 2008
University of Oregon.
Master of Science in Communication Disorders and Sciences, 2004
University of Oregon.
Master of Arts in Communication, 2001
Shih Hsin University
Bachelor of Arts in Public Administration and Politics, 1998
National Taipei University
AREAS OF SPECIAL INTEREST:
Child Language Development
School-Age Children Language Development
Literacy
Word Learning Process and Strategy
Literate Lexicon
Word Use in Writing
Morphology Development
Mental Imagery in Literacy
PROFESSIONAL EXPERIENCE:
Graduate Teaching Fellow, Communication Disorders and Sciences Program,
University of Oregon, Fall 2006
VI
VB
Teaching Assistant, Communication Disorders and Sciences Program,
University of Oregon, Fall 2004; 2005, Winter 2005
Research Assistant, Communication Disorders and Sciences Program,
University of Oregon, 2005-2006
PUBLICATIONS:
Nippold, M., & Sun, L. (in press). Knowledge of morphologically complex words: A
developmental study of older children and young adolescents. Language, Speech,
and Hearing Services in School.
V111
ACKNOWLEDGMENTS
Extreme thanks to my advisor, Dr. Marilyn Nippold, for encouraging me, helping
me overcome the challenges and providing assistance in the preparation of my dissertation.
Thank you to my committee: Marilyn Nippold, Ph.D., Roland Good, Ph.D., Deborah
Olson, Ph.D., and Nathaniel Teich, Ph.D. I appreciate their support and valuable input in
assisting me in this process.
Sincere appreciation goes to my best friend/partner in my life, Pei-Fang Hung.
Without her unconditional love, support and patience, I would not have been able to
achieve this goal. I thank her for being there through thick and thin, joy and tears. In
addition, I wish to express sincere appreciation to Tracy Mainsfield, Ph.D. for solving my
millions of questions about coding and the SALT program and his neverending support
for the past few years. I also would like to thank Shelley Watson and Shiree Nguyen.
Being friends with them is a blessing. Thanks for their encouragement and always being
there for me. The days in Eugene and Seattle are unforgettable.
Thanks to my parents for spiritually and financially supporting me for the past
few years, to my little sister for always being on my side and making me laugh, and my
grandma's continuous love. Finally, I thank the students and teachers who participated in
this study who provided me with invaluable information.
IX
TABLE OF CONTENTS
Chapter Page
I. BACKGROUND AND INTRODUCTION . 1
Literate Lexicon and Core Lexicon 3
Word Knowledge Tests 6
Writing as an Important Window........... 7
Later Language Development and Children with Language Difficulties 10
Purpose of the Present Study 13
II. LITERATURE REVIEW 15
Literate Lexicon 15
Narrative and Expository Texts 17
Spoken and Written Modality.... 21
Abstract Nouns 26
Dual Coding Theory 27
Evidence for Dual Coding Theory.......................................... 28
Difference in Defining Concrete and Abstract Nouns :............. 31
Use of Abstract Nouns in Written Language 33
Summary of Abstract Nouns 34
Mental State Verbs 35
Mental State Verbs and Theory of Mind 36
Mental State Terms and Literacy............................................................ 39
Development of Mental State Terms in Preschoolers .. 41
Development of Mental State Terms in School-Age Children and
Adolescents 43
Mental State Terms in Writing 46
Summary of Mental State Verbs 47
Derivational Morphology.... 48
What Is Derivational Morphology? 49
The Process of Derivational Morphology......... 51
Development of Derivational Morphology in School-age Children 53
xChapter Page
Derivational Morphology in Vocabulary Learning and Spelling 57
Derivational Morphology in Writing 59
Summary of Derivational Morphology 62
Research Rationale and Hypotheses of the Current Study 64
III. METHOD . 69
Participants 69
Recruiting and Sampling 69
Demographics 70
Materials 70
Experimental Materials in Writing 70
Instructions for the Narrative Essay................................................. 71
Instructions for the Expository Essay.............................................. 71
Teacher Questionnaire 72
Procedures 73
Coding and SALT Program 74
Transcription and SALT Program 74
Inclusion Criteria for Literate Lexicon 75
Abstract Nouns 75
Derivatives 76
Mental State Verbs 78
Coding for the Literate Lexicon 79
Interrater Reliability in Coding 80
Data Analysis 80
Quantitative Analysis . 81
Content Analysis 82
Teacher Questionnaire Report 83
IV. RESULTS 84
Quantitative Results 84
General Comparison between Age Groups and Gemes 84
Total Number of Words (TW) 84
Xl
Chapter Page
Total Number of Utterances (TTU) 86
Mean Length of Utterance in Words (MLTU) 86
Literate Words Use across Age Groups and Genres 88
Abstract Nouns (ABN) 88
Derived Nominals (DRN).... 89
Derived Adjectives (DRA) 90
Early Developing Metacognitive Verbs (EMCV) 92
Early Developing Metalinguistic Verbs (EMLV)............................. 93
Later Developing Metacognitive Verbs (LMCV)............................. 95
Later Developing Metalinguistic Verbs (LMLV)............ 95
Summary of Quantitative Results of Literate Word Use 96
Content Analysis Results 97
Word Lists from Two Genres across Three Age Groups 97
Abstract Nouns 98
Derived Nominals 99
Derived Adjectives 101
Mental State Verbs 103
Questionnaire Results 104
Frequency 105
Instruction 105
Types of Feedback 106
Word Knowledge and Genre Ratings 106
V. DISCUSSION 109
Analysis of Writing Productivity in Age Groups and Genres 110
Literate Word Use across Three Age Groups 111
Abstract Nouns 111
Derivatives - Derived Nominals and Derived Adjectives 112
Mental State Verbs - Metacognitive and Metalinguistic Verbs 114
Literate Word Use between Narrative and Expository Writing 117
Abstract Nouns 117
Derivatives - Derived Nominals and Derived Adjectives 118
xu
Chapter Page
Mental State Verbs - Metacognitive and Metalinguistic Verbs 119
Vocabulary Instruction in Writing 120
Conclusions 122
Limitations and Future Research.................................................................. 124
Limitations 124
Future Research 125
APPENDICES
A. INSTRUCTION SHEET -- NARRATIVE ESSAy...................................... 128
B. INSTRUCTION SHEET -- EXPOSITORY ESSAy..................................... 129
C. TEACHER QUESTIONNAIRE 130
D. DESCRIPTIVE STATISTIC TABLE OF 10 DEPENDENT VARIABLES-
MEAN, STANDARD DEVIATION AND RANGE...................................... 136
BIBLIOGRAPHy................ 137
Xlll
LIST OF FIGURES
R~re P~
4.1 Percentages of Each Literate Word Use in Two Gemes across Three
Groups 98
XIV
LIST OF TABLES
Table Page
3.1 Coding System for Three Types of Literate Words 76
3.2 Samples of Suffixes Used to Create Nouns and Adjectives 78
4.1 Two-Way, Mixed-Effects Analysis of Variance on TW 85
4.2 Two-Way, Mixed-Effects Analysis of Variance on TTU 87
4.3 Two-Way, Mixed-Effects Analysis of Variance on MLTU 88
4.4 Two-Way, Mixed-Effects Analysis of Variance on ABN 90
4.5 Two-Way, Mixed-Effects Analysis of Variance on DRN 91
4.6 Two-Way, Mixed-Effects Analysis of Variance on DRA 92
4.7 Two-Way, Mixed-Effects Analysis of Variance on EMCV 93
4.8 Two-Way, Mixed-Effects Analysis of Variance on EMLV 94
4.9 Two-Way, Mixed-Effects Analysis of Variance on LMCV 96
4.10 Two-Way, Mixed-Effects Analysis of Variance on LMLV 97
4.11 Examples of Abstract Nouns Used in Writings across Three Age
Groups 99
4.12 Examples of Derived Nominals Used in Writings across Three Age
Groups........................................................................................................ 100
4.13 Examples of Derived Adjectives Used in Writings across Three Age
Groups........................................................................................................ 102
4.14 Later Developing Metacognitive and Metalinguistic Verbs Used in
Writings across Three Age Groups ,. 104
4.15 Summary of Teacher Questionnaire Report 108
1CHAPTER I
BACKGROUND AND INTRODUCTION
The lexicon is central in communication, providing a window of observation into
language acquisition and processing (Clark, 1993). Moreover, the lexicon is an important
element of both spoken and written language development and is a crucial component of
higher cognitive activity (Ravid, 2004). Dockrell and Messer (2004) stated that the
lexicon is important in both comprehension and production. Additionally, a
well-developed receptive vocabulary is a prerequisite for fluent reading, a critical link
between decoding and comprehension (Joshi, 2005). In general, children with a larger
vocabulary tend to continue to expand their word sets faster and to understand texts more
easily than children with a smaller vocabulary. Therefore, lexical development is
important in communication, reading, and academic achievement across the school years
and is a foundation for gaining world knowledge throughout life.
An individual's lexical knowledge refers to the person's knowledge of vocabulary.
Aitchison (2003) referred to the lexicon as "the human word store." Ravid (2004)
claimed that the lexicon is the repository of words that speakers and writers use for
language comprehension and production. In other words, the lexicon is the knowledge
2that a person has about a language including the sounds and use of words, meanings, and
word categorizations (Brent & Cartwright, 2002; Clark, 1993). Therefore, the lexicon
provides a bridge between sounds and meaning, linking phonological properties to
semantic and syntactic information about the words (Marslen-Wilson, 2002). Because the
lexicon is stored and organized in a specific way in memory, mature adults can recognize
and retrieve words effortlessly and quickly. The lexicon is more complex than a mental
dictionary, organizing the vocabulary of a language according to certain principles but
also utilizing rules to generate the meanings of new words.
Based on Nagy and Anderson's (1984) report, children acquire 3000 words per
year between third and ninth grades and an average high school graduate knows about
45,000 words. Bloom (2001) stated that an average 17-year-old English speaker knows
more than 60,000 words and some literate adults even know more than 100,000 words.
Although there is no consistent number defining how many words a typically developing
adolescent knows, it is critical for researchers to understand how children learn so many
words at such an amazing rate. A child's lexicon grows with age in different ways without
direct instruction, such as: 1) quick incidentallearning/fast mapping (Dockrell & Messer,
2004; Oetting, Rice, & Swank, 1995); 2) use of context clues (Cain, Oakhill, & Elbro,
2003; Cain, Oakhill, & Lemmon, 2004; Chaffin, Morris, & Seely, 2001); 3) syntactic
bootstrapping (Oetting, 1999; O'Hara & Johnston, 1997; Rice, Cleave, & Oetting, 2000);
and 4) morphological generalization (Larsen & Nippold, 2007; Nippold, 2007). A child's
vocabulary expands tremendously from the very first year, and continues to grow through
3the school~age years, adolescence, and early adulthood. However, vocabulary acquired in
early childhood is different from that learned in middle and high schools.
Literate Lexicon and Core Lexicon
There are several differences between the lexical items acquired in early
childhood and those learned during the school-age years. Ravid (2004) defined the core
lexicon as basic vocabulary acquired by preschoolers mainly from spoken language, and
the literate lexicon as advanced vocabulary items that are more abstract, less familiar, and
acquired mainly through schooling and education. Ravid (2004) further suggested that
cognitive development and linguistic literacy contributed to advanced lexical
development. Cognitive development allows older children to process, retrieve, and store
lexical items in appropriate contexts, and linguistic literacy provides children plenty of
exposure to complex syntactic structures and advanced lexical items in textbooks,
lectures, and content area readings (Nippold, 1993). Thus, through schooling and
exposure to print, children are able to expand their vocabulary greatly and boost their
literate language development. Vocabulary items that school-age children encounter are
more abstract, complex, longer, with multiple affixes, and are used to reflect internal
states (Nippold, 2007; Ravid, 2004; Westby, 1990). Therefore, use of an advanced
lexicon reflects later language development and cognitive development.
The literate lexicon was examined in spoken narratives of school-age children in
Greenhalgh and Strong's study (2001), specifically conjunctions, adverbs, and mental
state verbs. Nippold, Ward-Lonergan, and Fanning (2005) studied adverbial conjuncts,
4abstract nouns, and metaverbs in the persuasive writing of school-age children,
adolescents, and adults. Bar-Han and Berman (2007) investigated the use of Latinate
portion of the English lexicon, a type of high linguistic register such as conflict and
pressure, in spoken and written narrative and expository production (Bar-Han & Berman,
2007). Additionally, five literate language features (elaborated noun phrases, conjunctions,
adverbs and mental and linguistic verbs) were examined in preschool spoken narrative in
Curenton and Justice's study (2004).
Another study done by Ravid (2004) investigated correct production in Hebrew of
derived nominals, passive constructions, and denominal adjectives in tasks completed by
school-age children. Nouns were further analyzed on ten-point noun scale, concrete to
abstract, in spoken and written narrative and expository produced by school-age children,
adolescents and adults in Ravid's study (Ravid, 2006). The latest study done by Berman
and Nir-Sagiv(2007) studied the use of nouns, Latinate words (high register linguistic),
and polysyllabic words in written narrative and expository in school-age children,
adolescents and adults. A very limited number of studies have looked at how school-age
children use the literate lexicon, and previous studies focused on different aspects of
literate language and the literate lexicon. Given that use of the literate lexicon reflects
later language development, more research into the types and uses of literate lexicon is
needed in order to provide a complete picture of the development of the literate lexicon.
According to previous studies, use of mental state verbs including metacognitive
and metalinguistic verbs is important in literate language development (Greenhalgh &
Strong, 2001; Nippold, 2007; Westby, 1990). Additionally, mental state verbs provide a
5bridge to connect speakers/writers and their audience and a link between theory of mind
and language development (Astington & Jenkins, 1999; Booth, Hall, Robinson, & Kim,
1997; Olson & Astington, 1993). In addition to mental state verbs, most of the complex
and longer words that school-age children encountered were derivatives. Derivatives are
words with affixes, and usually stem from the root words (e.g., "happiness" is derived
from "happy"). Complex derived words become increasingly important throughout the
school years in content area reading, writing, textbooks, and literature (Nippold, 2007).
Children use morphological generation strategies to generate word meanings while
reading, but use of derivatives is a more complex process and requires knowledge of
spelling, syntax, and morphology. Therefore, use of derivatives is an important aspect of
later language development that cannot be overlooked.
Another aspect of later language development that differs from early language
development is abstract thought (Nippold, 2007). Concrete objects and thoughts
dominate early childhood language development; however, the importance of abstract
thinking and reasoning increases through cognitive and linguistic development. Abstract
nouns are one of the representations of abstract thought. An abstract noun refers to states,
events, concepts, feelings, and qualities that have no physical existence and anything that
cannot be perceived through five physical senses: the opposite ofa concrete noun. These
three important types of literate words provide insight into later language and literacy
development. Therefore, it is critical that research not only focuses on the literate lexicon
and language development, but also examines different types of literate words.
6Word Knowledge Tests
The literate lexicon is associated with word knowledge. How children's word
knowledge is measured is critical to educators, clinicians, parents, and researchers.
Dockrell and Messer (2004) claimed that word knowledge should be assessed with
different measures and should consider the quality and quantity of children's vocabulary
knowledge. Westby (1990) suggested using whole language theory to assess language
development and to further evaluate children with language disabilities. In the whole
language model, children are evaluated in different aspects of language such as semantics,
pragmatics, and syntax in four modalities (i.e., reading, writing, speaking, and listening).
This is based on the view that understanding some aspects of a word does not necessarily
indicate understanding of the word's meaning in a more complex context. Thus, in order
to provide a full picture of children's word knowledge, assessment in both
comprehension and production of vocabulary is essential.
The standardized vocabulary tests such as Peabody Picture Vocabulary Test-
Fourth Edition (PPVT-4) (Dunn & Dunn, 2007) and some standardized language tests
containing vocabulary knowledge subtests such as Clinical Evaluation of Language
Fundamentals 3 (CELF-3) (Semel, Wiig & Secord, 1995), and Test of Language
Development-Primary, Third edition (TOLD-P:3) (Newcomer & Hammill, 1997) were
commonly used to detect vocabulary knowledge of the examinee. However, the focuses
of the lexical knowledge tests are often limited to the comprehension and production of
the meaning, vocabulary use, and word association (Laufer, Elder, Hill & Congdon,
2004). Westby (1990) claimed that many language tests assess discrete aspects of
semantic abilities. Laufer, Elder, Hill, and Congdon (2004) pointed out that the best way
to evaluate children's vocabulary size and strength was to measure different dimensions
of word knowledge because word knowledge meant not only understanding the link
between the meaning and the form but also use of the word correctly in free production.
Vocabulary tests have been questioned by researchers (Curtis, 1987; Laufer et aI., 2004),
and most of the tests focused on word or sentence level comprehension and production.
Therefore, it is important that researchers not only assess word knowledge under one
aspect of language but also examine the quality of word knowledge. In addition to
determining how many words a child knows through vocabulary tests, understanding
how children use these words in daily life is essential to both language development and
instruction design.
Writing as an Important Window
Language develops and expands continuously in life. Throughout the school-age
years, writing play an important role in learning and academic success and is viewed as
schooled language competence (Perfetti & McCutchen, 1987). Writing is the most
complex aspect of language and is important during the school-age years while written
language becomes crucial in academic success and knowledge acquisition (Morris &
Crump, 1982). Writers do not simply write the words down, but rather also need to plan,
organize, spell, and compose the content according to the potential audience and geme
requirements. Perfetti and McCutchen (1987) suggested that three components
comprising writing competence including discourse schema knowledge, lexical
7
8knowledge, and syntactic knowledge, and that writers manipulate words and phrases to
meet the discourse requirements by having sufficient lexical and syntactic knowledge.
Therefore, using appropriate words in writing is a critical process for writers to correctly
convey the meaning with confidence and effectiveness.
Written texts were often found to contain low-frequency lexical items and words
referring to abstract entities (Bar-Ilan & Berman, 2007; Chafe & Danielewicz, 1987;
Nippold, 2007; Nippold,Ward-Lonergan & Fanning, 2005). In addition, children used
more complex syntax and diverse vocabulary in written text than spoken text (Bar-Ilan &
Berman, 2007; Scott, 1988). Therefore, writing is a natural and practical way to examine
how children use words in regular settings because children need to have full knowledge
of the word in order to use it appropriately given the context, genre, and audience. Thus,
examining word use in writing provides a window to explore the literate lexicon and
school-age language development.
According to Flower and Hayes' writing model (1984), the writing process
contains three steps (i.e., planning, translating, and reviewing), and is interactive with the
writer's knowledge of the topic, audience, and writing plans. All of these steps must be
coordinated within the limits of working memory (McCutchen, Covill, Hoyne, & Mildes,
1994); however, the translating step was less elaborated in Flower and Hayes' model.
Berninger and Swanson (1994) and McCutchen and colleagues (1994) further explained
what happened during the translating stage. In the translating stage, writers need to
retrieve appropriate words and further generate sentences and paragraphs. If the
translating process does not operate fluently, then the limited working memory available
9for online planning and reviewing is hindered (McCutchen, Covill, Hoyne, & Mildes,
1994). Berninger and Swanson (1994) further divided the translating step into two
components, namely text generation and transcription, where text generation contains the
production of words, sentences, and paragraphs. Therefore, word use is in the process of
text generation and translation and is an important representation of children's word
knowledge and literate language development.
Writers need to consider audience and genre in the writing process; therefore, it is
important to take genre into account to further explore whether writers use vocabulary
differently. Genre is a map or plan for discourse and varies in its functions (Westby,
1994). Scott pointed out that genre is more powerful than modality on the effect of a
text's form (1994), and the distinction between narrative and expository discourse
provides most classification scheme differences (2005). Thus, it is advisable to compare
language samples in these two distinct types of discourse while examining language use
and development. Since the literate lexicon is used more often in formal writing than in
speaking (Ravid, 2004) and by school-age children than by preschoolers, the focus of the
present study was on written language in school-age children and adolescents.
Compared to spoken language development, fewer studies have been devoted to
written language. Moreover, most of these studies focused on cohesion, organization,
syntactic complexity, information density, and semantic density and diversity (e.g.,
Berman & Nir-Sagiv, 2007; Hidi & Hildyard, 1983; McCutchen, 1987; O'Donnell, 1974;
Ravid & Berman, 2006; Ravid & Zilberbuch, 2003; Verhoeven et ai., 2002). Although
some of the studies explored the literate word use in writing, the focus ofliterate word
10
use in these studies was on Latinate lexicon (Berman & Bar-Ilan, 2007; Berman &
Nir-Sagiv, 2007), abstract nouns (Berman & Nir-Sagiv, 2007; Nippold, Ward-Lonergan &
Fanning, 2005; Ravid, 2006) and adverbial conjunctions and metaverbs (Nippold,
Ward-Lonergan & Fanning, 2005). Very few studies looked into different types of literate
words that school-age children and adolescents produced in different writing genres.
However, as stated previously, it is critical to assess language development in multiple
dimensions and across different age spans to fully understand continuous language
development.
Later Language Development and Children with Language Difficulties
Traditionally, older children have received less attention in the study of language
development because later attaimnents in language are subtle (Nippold, 1993; 1995).
However, in studying the literate lexicon and writing, school-age children and
adolescents must be included because the use of advanced and complex language
develops relatively late. Differences in language use between the school-age years and
early childhood show themselves in quality and quantity, resulting from schooling,
exposure to print, and opportunities to write (Nippold, 2007). Older children not only use
advanced vocabulary and compose sentences with more complex syntax but also consider
their potential audience, context, and discourse more maturely. Although preschoolers use
aspects of literate language such as conjunctions and mental state verbs in narratives
(Curenton & Justice, 2004), the use of these types of words may not reflect their true
understanding since linguistic literacy and cognitive maturity are still developing.
11
Therefore, studying school-age children's written language samples is the best way to
explore how children use and understand the literate lexicon.
In particular, little attention has been paid to how school-age children and
adolescents use the literate lexicon in writing. Writing provides a window to examine
how children use words they have learned, and school-age children are perfect candidates
for exploring later language development. For the present study, task of narrative and
expository writing not only supplied research evidence to further understand word use in
different gemes of writing but also was embedded in the school curriculum. Therefore,
this study offered valuable research evidence and a more complete picture of literate
lexicon development during the school-age years.
Studying typically developing children and adolescents provides valuable
information for educators and clinicians to determine when children are at risk. It is
critical to first understand how typically developing children build up and utilize their
language in order to use the information as a foundation to identify children with
language difficulties. The American Speech-Language-Hearing Association (ASHA,
2001) stated that children with spoken language problems commonly have difficulty
learning to read and write, since that challenge involves the same language components
as listening and speaking, such as phonology, morphology, syntax, and semantics. Gillam
and Johnston (1992) examined spoken and written narratives of typically developing
children and children with language learning difficulties. They found out that children
with language learning difficulties produced less complex sentences and made more
grammatically unacceptable sentences, especially in writing. Morris and Crump (1982)
12
found that children with learning disabilities used less variety in their written vocabulary,
although there was no difference in T-unit length between typically developing children
and children with learning disabilities. Children with specific language impairment (SLI)
were compared with chronological age matched (CA) and language age matched (LA)
groups in writing in the Mackie and Dockrell study (2004). Children with SLI produced
shorter stories, had more difficulties with generating text, and made more syntactical
errors in writing compared to the CA and LA marched peers. They claimed that children
with SLI have difficulties with translating the ideas into written language because of the
limited linguistic and cognitive resources.
Windsor, Scott, and Street (2000) examined the use of verb morphology (such as
regular past tense, and third person singular present tense) and noun morphology (such as
articles, regular plural, and possessives) in spoken and written language production for
children with language learning disabilities (LLD). They found that children with LLD
had difficulties with marking verb morphology appropriately in their written and spoken
language. Another study done by McFadden and Gillam (1996) used a holistic scoring
system to evaluate spoken and written narratives generated by children with language
disorders. Their findings indicated that children with language disorders produced
narratives with less complex structure and lower quality compared to their peers. Similar
results were obtained in Scott and Windsor (2000), where genre and modality effects
were more prominent in school-age children with language learning disabilities. These
children performed more poorly in writing and in expository discourse compared to their
peers. Additionally, Lewis and colleagues (1998) claimed that children with speech
13
and/or language disorders were at risk for writing difficulties at school age, and the risk
continued even after their spoken language improved.
According to ASHA (2001), Speech Language Pathologists are required to
provide language assessment and intervention for young children at risk to prevent later
written language problems. However, few studies address writing issues, and no study to
date has focused on word use in writing in children with language difficulties. Lewis and
the colleagues (1998) pointed out that since reading, writing, and spelling shared
common linguistic processes, children with spoken language difficulties were at risk for
writing difficulties. Given that, researchers advocate the importance of evaluating
children's written language skill since development of writing and written language
deficits were less studied and understood (Singer, 1995; Windsor, Scott & Street,
2000).Thus, the present study provided important evidence of how typically developing
children use the literate lexicon in different genres of writing in order to supply
developmental data for educators and clinicians to detect children at risk and further
provide intervention in writing and literate lexicon learning.
Purpose of the Present Study
The purpose of the present study was to provide research evidence and
developmental norms regarding how typically developing school-age children and
adolescents use the literate lexicon in different genres of writing. Word knowledge refers
not only to understanding meaning and form, but also having knowledge about how to
use words appropriately in different contexts, genres, and for varying audiences.
14
Investigating word use in writing provides a window into language development and is a
practical way to understand how literate language develops in typical school-age children
and adolescents. The present study not only integrated the literate lexicon/language and
later language development, but also provided a clear picture of how school-age children
and adolescents continuously develop their literate use of language.
Examining how typically developing children process language offers useful
information to identify children at risk. Since literate language plays an important role
during the school-age years, it is critical to assess children's literate lexicon/language
knowledge to detect any difficulties in word knowledge and language use. Therefore,
developmental study is critical not only to understand language development, but also to
use it as a reference to detect children at risk. Moreover, the present study integrated later
language, literate lexicon development, and writing to serve as a foundation for clinicians
and teachers to identify children with writing difficulties and prevent writing problems.
15
CHAPTER II
LITERATURE REVIEW
Literate Lexicon
The literate lexicon is more than just the set of words learned in early childhood.
Based on Ravid's definition (2004), the core lexicon refers to the basic vocabulary gained
during preschool and mainly derives from spoken language. In comparison, the advanced
or literate lexicon reflects maturation in cognitive and language development and
increases during the school years. The literate lexicon commonly shows itself in scholarly
contexts in high school and college, such as in textbooks, lectures, or technical
presentations (Nippold, 1993). Overall, literate words are longer, more abstract and
complex, and less familiar than those in the core lexicon, often appearing with multiple
affixes which make these words more difficult to learn and retain. Nippold,
Ward-Lonergan, and Fanning (2005) asked 60 school-age children, 60 adolescents, and
60 adults to write an essay on a controversial topic to investigate syntactic, semantic, and
pragmatic development in persuasive writing. They found that the use of three types of
literate words -- adverbial conjuncts, abstract nouns, and metaverbs -- increased from
16
childhood to adulthood. However, although the literate lexicon develops gradually, the
use of these words remains low even in adults.
Oral and literate language both develop (Paul, 2007), but they differ from one
another. The school-age years are important for literate language development since
children at these ages encounter more complex and less concrete advanced words through
exposure to textbooks, lectures, and reference woks such as encyclopedias. Nippold
(1993) claimed that two crucial components of semantic development during adolescence
are the literate lexicon and figurative language. Adolescents not only acquire a larger
vocabulary, but also are able to elaborate upon the meanings of known words and to
understand the connections among words, such as derivatives and synonyms (Nippold,
2007; Paul, 2007). With cognitive and linguistic development, adolescents have the
ability to acquire more advanced words and to expand their lexical inventory through
categorizing words and utilizing word rules. School-age children and adolescents develop
advanced, literate vocabularies through opportunities to read texts with complex syntactic
structures and to write in professional ways.
Literate language development displays some distinct signs. Paul (2007)
categorized six types of words that involve literate language forms, including: 1)
advanced adverbial conjuncts (e.g., moreover, in contrast); 2) adverbs oflikelihood and
magnitude (e.g., possibly, extremely); 3) technical terms related to curricular content (e.g.,
bacteria, proton); 4) metalinguistic and metacognitive verbs (e.g., imply, hypothesis); 5)
words with multiple meanings (e.g., strike, short); and 6) words with multiple
grammatical functions (e.g., hard, sweet). In addition, use of words such as conjunctions,
17
elaborate nouns phrases, metacognitive and metalinguistic verbs, and adverbs, helps
researchers and teachers estimate a child's literate language ability (Greenhalgh & Strong,
2001; Westby, 1999).
Development of an advanced lexicon interacts with syntactic complexity, genres
(e.g., narrative, expository), and modality (e.g., speech, writing) (Ravid, 2004).
Verhoeven and colleagues (2002) claimed that it is a critical process in later language
development to distinguish the difference between modality and genre. Therefore, it is
important to consider genres and modality as factors while comparing spoken and written
language performance (Scott & Windsor, 2000).
Narrative and Expository Texts
The ability to talk and write with different genres is an important factor in
development (Verhoeven, Aparici, Cahana-Amitary, van Hell, Kriz, Viguie-Simon, 2002).
Different genres have a significant impact on how information is presented, varying with
broad ranges of vocabulary, clauses, phrases, sentences, and organization. Therefore,
while talking about language development, different genres need to be taken into account
to represent language performance without bias.
In the Greenhalgh and Strong (2001) study, there was no age-level difference
found when literate language use was examined in spoken narratives in typically
developing children compared to children with language impairments in four different
age groups (from 7 to 10 years old). One of the explanations of failure to find age-level
differences was that only narratives were collected in this study, and literate language is
18
commonly shown in written language. Children use different kinds of vocabulary
depending on contextual factors, the way they represent the content, and the requirements
of different discourse genres. Thus, the frequency of different types of words used may
vary with genre and modality.
A genre is a plan for discourse and text, and different genres may have different
functions (Westby, 1994). Each genre has a predominant discourse form to be learned
through numerous writing experiences and genres influence choices in lexical, phrase,
sentence, and text levels (Scott, 1994). Only two genres were investigated in the present
study, because there is a major distinction between narrative and expository genres
(Ravid & Zilberbuch, 2003). Narratives are about people engaged in events with a
sequential timeline, including personal experiences, fictional narratives/stories, and
film/book/TV program summaries (Hadley, 1998; Scott, 1988). They are chronologically
based and agent-focused (Scott & Windsor, 2000), as in retelling and generating stories.
In contrast, expository texts usually contain factual or technical information such as
cause-effect explanations and procedural directions (Hadley, 1998), having a
non-temporal, logically-based, and argumentative structure (Ragnarsdottir, Aparici,
Cahana-Amitay, van Hell, & Viguie, 2002). Ragnarsdottir and colleagues (2002) claimed
that one major difference between narrative and expository texts is verbal paradigms. For
example, past tense is typically used in narrative, but present tense is characteristic of
expository text.
Narrative skills develop earlier than those used in writing expository text, where
narrative is the earliest extended monologue text produced by children. Narratives receive
19
great interest because they are a universal genre and have a central place in human lives
(Westby, 1994). Moreover, narrative text expresses people's thoughts and plays an
important role in comprehension and production for academic success. In contrast, the
acquisition of expository text structure requires exposure to lectures, textbooks, and
writing and reading experiences in school. Berman and Katzenberger (2004) suggested
that the well-formed expository texts are constructed beginning from high school.
By adolescence, children are able to organize the information hierarchically in an
expository text, including a statement in the introduction followed by supporting
arguments and a conclusion, with a main ideas and claims. Scott and Windsor (2000)
stated that sentential grammatical complexity is affected by genre in children age 9 to 12.
Although children use different linguistic markers and verbal expressions to show
distinctive knowledge between narrative and expository texts around 4th grade (Berman,
2004; Berman & Katzenberger, 2004; Berman & Nir-Sagiv, 2004), a clear distinction,
such as syntactic structures and lexical selection, is not formed until adolescence (Ravid
& Zilberbuch, 2003).
San Jose (1973, cited in Morris & Crump, 1982) pointed out the effect of
discourse genre on average T-unit length in writing as early as 4th grade, and T-unit
length was longer in argumentative discourse than descriptive text. In Scott and Windsor
(2000), they compared spoken and written narrative and expository discourse between
typically developing children and children with learning disabilities. The results indicated
that narrative summaries had longer T-units and a greater total number of words, and that
productivity and fluency were greater in narrative than expository discourse. However,
20
average T-unit length, used to measure grammatical complexity, was longer in expository
summaries than narrative ones. In addition, expository text is usually longer than
narratives. Another study done by Hidi and Hildyard (1983) indicated that narrative text
was produced more easily, and was found to be longer, better formed, and more coherent
than expository text in both 3rd and 5th graders.
The differences between the two gemes have been compared across several
measures in several studies, namely: productivity, grammatical complexity, lexical
diversity, and grammatical errors (Scott & Windsor, 2000); use of ten types of nouns
(Ravid, 2006); text openings (Berman & Katzenberger, 2004); lexical and
morphosyntactic constructions (Berman & Nir-Sagiv, 2004); noun and verb phrases
(Berman, 2004); content and text organization (Langer, 1985; Berman & Nir-Sagiv,
2007); semantic well formedness and cohesion (Hidi & Hildyard, 1983); clause
packaging (Verhoeven, Aparici, Cahana-Arnitary, van Hell, Kriz, Viguie-Simon, 2002);
Latinate and Germanic lexicon use (Bar-Ilan & Berman, 2007).
It is a gradualleaming process to understand how to alter content organization,
select appropriate words, use proper syntactic structures, and target the right audience
according to the specific geme type. Therefore, one of the crucial components oflater
language development is to be familiar with the written language requirements of
different discourse styles (Ravid & Tolchinsky, 2002). Berman and Nir-Sagiv (2004)
confirmed that one of the late-developing language abilities is use of geme-appropriate
context structure and consistent register distinction according to different discourse types.
For example, logical connectors, present tense, and argument statements are used in
21
expository text; on the other hand, past tense, concrete description, and illustration are
seen in narratives.
Berman and Katzenberger (2004) pointed out that expository and narrative texts
reflect different cognitive processes. In addition, expository text develops later than
narrative text because of sociocognitive factors and different types of experiences.
Westby (1994) argued that narrative text requires children to integrate language skill into
a literate language production without assistance; however, the acquisition of expository
text needs exposure to formal schooling. In general, it is easier to describe an experienced
event or to tell a story than it is to make a logical statement or argument about a
controversial issue. Therefore, expository and narrative texts are different not only in
content, lexical use, and syntactic structure, but also in developmental and cognitive
processes.
Spoken and Written Modality
One turning point in child language development is the transition from oral to
written language (Verhoeven et aI, 2002). In order to study the lexicon, it is necessary to
consider modality because lexical measures are sensitive to geme and modality
(Stromqvist, Nordqvist & Wengelin, 2004). Thus, in addition to gemes, modality needs to
be taken into account in order to get a full picture of language development. Modality of
language, describing how information is communicated and delivered, often refers to
spoken and written language (Westby, 1985). Speech and writing are both language
expressions, sharing some features, distinct in others. In general, spoken language tends
22
to be face-to-face (or voice-to-voice), concrete, contextually embedded, rapidly processed,
more spontaneous (i.e., less planned), with less word variation; in contrast, written
language tends to represent a monologue, and is abstract, contextually extracted,
deliberately processed, planned (i.e., editable), structured, and uses a formal vocabulary
register.
These two modalities are both expressive, closely interacting with each other in
the language development process. Common variables used to compare these two
modalities are syntactic structures (i.e., sentences, clauses, phrases), semantic use (e.g.,
lexical density and diversity), cohesion (e.g., organization), number of words, and
connectors used (Scott, 1988). In addition to understanding the similarities and
differences in oral and written language, it is worth examining the process wherein a
child begins to write, and then reaches a level required to talk and write appropriately
according to the context, audience, and purpose.
Kroll (1981) pointed out four phases describing the developmental process of
speaking and writing. Younger children have well-developed oral language but very
minimal written language. During this preparation phase, younger children use oral
language to learn and to provide a basis for written language. Children's written language
lags behind oral language when they begin to write, and they focus more on spelling and
forming letters instead of on structures and formation. When children begin to write, they
write as if they were talking. During this consolidation phase, writing and speaking begin
to integrate.
23
Narrative serves as an important transition from oral language to literate language
and is the first step that children learn to produce a monologue without interactive
dialogue (Westby, 1985). Younger children produce less complex sentences in writing
than in speaking, but the sentences in writing become more complex during the
secondary school years (Scott & Windsor, 2000).Through schooling and experience,
children learn the difference between speaking and writing, and their essays acquire a
format that is different from speech. During adolescence and adulthood, writing and
speaking are used interchangeably, maturely and appropriately according to audience,
purposes, and context.
O'Donnell (1974) indicated that writing tended to be more complex in syntactic
structure and had a greater degree of syntactic density. In addition, T-units, one main
clause with all the subordinate clauses and nonclausal phrases attached to or embedded in
it (Scott, 1988; Nippold, 2007), were longer in writing and more T-units contain
dependent clauses in writing than in speech. Ravid and Berman (2006) studied spoken
and written narrative texts in English and Hebrew in 4th, 7th, and 11 th graders and
graduate students. In their study, oral narratives were longer but less dense in information
compared to written narratives; moreover, informational density was closely related to
modality beginning in 4th grade.
McCutchen (1987) pointed out that spoken texts were longer and more elaborated
than written texts, but writing was more coherent than speaking. Additionally, explicit
ties such as conjunctions and dependent clauses were shown in writing but not in spoken
texts. Similar results were confirmed in Hidi and Hildyard's study (1983). Spoken texts
.....- ----- -_._---
24
were longer than written texts for both narrative and expository gemes, and written texts
tended to be more cohesive than spoken texts. Not surprisingly, a developmental trend
was shown in both quality and quantity of written and oral production.
Chafe and Danielewicz (1987) compared four types of language samples:
conversation, lecture, letters, and academic writing from twenty graduate students and
professors. It is noteworthy that speakers tended to have a limited variety of vocabulary
use and to use more new words and new meanings of old words compared to writers. In
addition, more literate vocabulary was used in academic writing but more colloquial
vocabulary was used in conversation. Therefore, because of the nature of writing, it leads
to richer and more explicit uses of words.
Some researchers expanded the understanding of spoken and written language
relationships from typically developing children to children with language and learning
difficulties. Scott and Windsor (2000) compared productivity, lexical diversity and
grammatical errors of school-age children with language learning disabilities and their
language and age matched peers in different modalities and gemes. The findings revealed
that spoken texts were longer and produced faster than written texts. Moreover, the
grammatical errors were more commonly seen in writing than that in speaking for all
three groups. Combining with gemes, written expository was difficult for children with
language learning disabilities and language matched peers, and there was no grammatical
complexity increase in written expository for these two groups.
Children with language and learning difficulties and typically developing children
were also compared in spoken and written narratives in Gillam and Johnston's (1992)
25
study. Their results showed that spoken narrative was longer but not necessary more
complex than written narrative. In addition, children with language and learning
difficulties produced more ungrammatical sentences in writing; however, the relationship
between spoken and written language was similar to that of matched peers. In the Morris
and Crump (1982) study, four groups of school-age children with learning disabilities and
their age-matched peers were studied on written texts. Although the T-units length
increased with ages, it did not differentiate these two groups of children. However, when
syntactic maturity was taken into account, typically developing children performed better
than children with learning disabilities. In addition, typically developing children used a
greater variety of vocabulary words in writing.
Overall, spoken and written languages are critical in language development,
learning, and communication. These two modalities share similarities, but becoming a
mature writer is not only putting words in print but requiring selection and organization
of ideas and knowing how to produce content that is appropriate, along with discourse
and modality. According to the literature review, spoken text uses less diverse vocabulary
and more frequent words; however, more advanced literate lexicon and a wider variety of
words are seen in written text (Stromqvist, Nordqvist & Wengelin, 2004). Therefore,
Lexical use in spoken text tends to be less formal and less informative compared to
written text (Ravid & Tolchinsky, 2002). Beyond lexical use differences written text often
contains longer sentences, denser information, and more complex syntactic structure.
The literature related to the literate lexicon, geme, and modality has been
reviewed to provide a solid foundation for the present study. Since the main focus of the
26
present study was to investigate three types of literate words used across two different
genres in writing, a literature review of three types of literate words, including abstract
nouns, mental state verbs, and derivatives was essential.
Abstract Nouns
Concrete words such as flower and dinosaur and abstract nouns such asfreedom
and loneliness not only have different functions in language but also are processed
differently in the brain. According to Paivio's dual coding theory (1991), a verbal system
which refers to verbal codes and a nonverbal system which implies mental imagery are
processed differently. The two systems can be activated by each other in either parallel or
integrated ways. In addition to the different processing methods between verbal and
nonverbal, there is a discrepancy between abstract and concrete language processing.
Concrete and abstract words are processed differently in the brain because of activation
of different part of the brain. Moreover, concrete words have greater cognitive processing
advantages over abstract words (West & Holcomb, 2000). Apparently, concrete words
additionally activate regions in the right hemisphere and form nonverbal images, so
concrete language is retained longer, recalled faster, and more easily comprehended due
to the activation of both verbal and nonverbal systems. On the other hand, abstract
language depends heavily on verbal associations for its meaning and is mainly processed
by the verbal system.
27
Dual Coding Theory
It is assumed that concrete nouns are learned through sensory experience while
the meanings of abstract nouns are derived mainly from networks that are mostly made
up of abstract words (Sabsevitz, Meddler, Seidenberg & Binder, 2005). Dual coding
theory proposes that abstract concepts are stored in the language dominate hemisphere,
the left hemisphere, but that concrete concepts are processed in both hemisphere.
However, if dual coding theory is applicable not only to words but also to sentences, then
paragraphs and texts are worth investigating. Sadoski (1995) conducted a study to
investigate whether concreteness effects could be extended beyond the word level such as
text. The concreteness effect was found in paragraphs drawn from naturally occurring
texts. In addition, concrete content was better recalled when concrete and abstract
information was equally familiar to the readers, but when abstract content was more
familiar to readers than concrete content, they were equally recalled. Therefore, in
addition to the concreteness effect, content familiarity had influence on later recall. But,
concrete content was twice as memorable as abstract content when the text was
controlled in readability, comprehensibility, and other context differences. Thus, dual
coding theory explained the concreteness effect to natural text, which is beyond the word
and sentence levels.
In contrast to dual coding theory, the context availability theory views the
concreteness effect from a different standpoint. Context availability theory
(Schwanenflugrl, Akin & Luh, 1992) claims that concrete and abstract information is
integrated into the learner's prior knowledge and is processed differently because of
28
distinct context support, not different brain systems. Schwanenflugrl and colleagues
(1992) argued that when context availability was controlled, subjects recalled abstract
and concrete word without any differences. In addition, when imageability and context
availability were highly correlated, the independent effect of imagery was difficult to
detect. According to context availability theory, concrete information is easier to
comprehend and remember because it is easier to relate to prior knowledge. Context
availability theory claims that it is easier to retrieve a context in which a concrete word
appears such as lamp than to retrieve a context in which an abstract word appears such as
peace (Altarriba, Bauer & Benvenuto, 1999). In other words, concrete words have
stronger links to contextual information and a greater quantity of information available,
in contrast to abstract words (Sabsevitz, Meddler, Seidenberg & Binder, 2005). However,
abstract information is more vague and difficult to process due to the lack of built-in
contextual support (Kounios & Holcomb, 1994).
Evidence for Dual Coding Theory
In order to test how concreteness effects work and if dual coding theory or
context availability theory better explain word processing in the brain, several imaging
studies were done to provide the experimental evidence. Jessen and colleagues (2000)
attempted to provide evidence for dual coding theory through event-related functional
fMRI and to explore which cortical regions are responsible for the additional
concreteness effect. Fourteen German-speaking adults participated in this study. Two
hundred and forty words including 120 concrete and 120 abstract words were used as
stimuli. Subjects had to encode as many words as possible and the word recognition test
29
was administrated later. The test consisted of 80 experimental words and 80 distractor
words, and the subjects were asked to identify previously presented words. The results of
this study showed that higher concreteness resulted in better recognition, so the superior
concreteness effect was achieved. The findings of this study supported dual coding theory,
that concrete words activated in the right hemisphere during decoding. In addition, the
parietal region, which was related to spatial imagery, was activated in some participants;
thus, a spatial associative encoding strategy was used in some participants to memorize
the concrete words.
Sabsevitz and colleagues (2005) conducted an fMRI study to determine if
concreteness effects exist. Twenty-eight adults had to select the word that was most
similar to the meaning of the sample noun. The stimuli, which were composed of
concrete and abstract nouns, were visually presented. The results indicated that abstract
nouns were mainly activated and processed in the left hemisphere while concrete nouns
were processed bilaterally. Therefore, this study again supported dual coding theory with
the evidence of bilateral activation during the processing of concrete nouns. Another
fMRI study done by Noppeney and Price (2004) asked 15 adults to choose the word that
had a similar meaning to the sample word. The words were visually represented within a
triad and drawn from abstract concepts (motive, intent, attempt), color (green,pink, red),
sounds (tone, melody, noise), and hand movements (squeeze, comb, brush). The results
showed that abstract concepts activated a left-lateralized frontal temporal system that is
usually involved in semantic processing. Additionally, abstract concepts and sensory-
based semantics were processed in different neural systems. Noppeney and Price
-------
30
suggested that since abstract concepts could not be represented by real objects, people
might generate a semantic context to obtain meaning.
The imageability on word processing was also investigated through
Event-Related potentials (ERPs). Swaab, Baynes, and Knight (2002) presented 320 word
pairs auditorally to 12 adults. The stimuli were composed of concrete, imageable, and
abstract, non-imageable words, and the word pairs were either related or unrelated in
meaning. For example, bread-butter was highly-imageable and related, but pace-dispute
was low-imageable and unrelated. Participants were asked to decide whether the words in
pairs were related semantically. They pointed out that imageable, concrete words were
stored differently from abstract, non-imageable words and activated both verbal and
image-based store systems. Additionally, the image-based system can be activated during
normal word processing without sentence or context support.
Kellogg, Olive and Piolat (2007) asked 60 college students to write definitions of
10 concrete or abstract nouns and to perform a verbal or visual working memory task at
the same time. They found that the visual WM task showed interference only when
participants wrote the definitions of the concrete nouns but not for the definitions of
abstract nouns. They stated that participants did not need to generate the images while
writing the definitions of abstract nouns; thus, the interference did not show up. In
addition, participants responded more slowly in defining abstract nouns while performing
the verbal WM task but not the visual WM task. Therefore, their findings were consistent
with dual coding theory since the visual WM task only interfered while participants
defined concrete nouns.
31
Another ERP study was done by Nittono, Suehiro and Hori (2002). Twelve
college students were asked to rate the imageability of the words that were visually
presented to them on a five-point scale and to recall as many words as they could later.
The stimuli were composed of high imagery words, low imagery words, and non-words.
The results indicated that high imagery words activated additional imagery-related
networks that were located in the right hemisphere and were not activated by low
imagery words. Thus, the results again supported dual coding theory.
Mellet, Tzourio, Denis and Mazoyer (1998) investigated the activation of cortical
areas through PET when people listened to word definitions. Participants were in three
conditions including generating mental imagery while listening to concrete word
definitions, listening to abstract word definitions, and silence conditions. Eight college
students listened to 15 words and their definitions and then recalled the words they had
just heard. The findings indicated that participants recalled more concrete words than
abstract words after listening to word definitions. They also found that listening to the
abstract word definitions activated the language area more intensively than listening to
the concrete ones.
Difference in Defining Concrete and Abstract Nouns
In addition to imaging evidence, some behavioral evidence revealed that people
process concrete and abstract nouns differently. Sadoski and his colleagues (1997) asked
48 college students to write definitions of 5 concrete and 5 abstract nouns and to rate their
thinking strategies while defining the words. The findings indicated that participants
tended to use the imagery strategy while defining a concrete word, but used the
32
verbal-associate strategy for abstract words. In addition, the definitions of the concrete
nouns were more detailed and informative compared to the abstract words.
McGhee-Bidlack (1991) recruited 120 subjects (ages 10, 14, and 18) to define
eight concrete and eight abstract nouns. Results revealed that concrete and abstract nouns
were defined differently and that concrete nouns were easier to define than abstract nouns.
In addition, all groups defined concrete nouns mostly by class and characteristics;
however, abstract nouns were not well defined by class and characteristics until 18
years-old. The quality of the definitions of both concrete and abstract nouns improved
gradually with age but defining abstract nouns apparently develops later than concrete
nouns. McGhee-Bidlack stated that concrete nouns are related to tangible objects;
therefore, children have better built-in concepts for concrete nouns but not abstract ones
and are able to learn concrete words prior to abstract words.
Another definition study done by Nippold and colleagues (1999) asked four
age-groups (12, 15, 18, and 23) to define 16 abstract nouns. The findings indicated that
the development of the ability to define abstract nouns improved with age into early
adulthood but was still incomplete by late adolescence. Nippold et al (1999) claimed that
the improvement might result from the development of abstract reasoning as a way of
thinking; therefore, early adulthood was a time to refine word knowledge and
metalinguistic ability. Nippold (1999) conducted another study and tested the same
abstract nouns on 16 skilled readers and 16 poor readers, age 15. She pointed out that
skilled readers provided more informative, accurate, and clear definitions than poor
readers.
33
Use ofAbstract Nouns in Written Language
According to the imaging evidence and behavioral evidence reviewed above,
abstract and concrete nouns are processed differently. However, how people use abstract
and concrete nouns in actual language production is less understood especially in written
language. Although spoken and written compositions are different, these two modalities
are both language production and may share similar cognitive mechanisms (Sadoski,
Kealy, Goetz, & Raivio, 1997). Some studies (McGhee-Bidlack, 1991; Nippold, 1999;
Nippold, Hegel, Sohlberg, & Schwarz, 1999) attempted to examine how people
composed definitions for abstract and concrete nouns through written modality, but how
children use abstract nouns in actual writing is unclear and has received less attention.
Sadoski et al (1997) recommended that it is necessary to study longer written
compositions in different genres with sufficient time limits.
Ravid (2006) used ten nouns, including concrete nouns (ball, bike), proper names
(Tiffany, Mike), collection/location (library, class), role nouns (cousin, neighbor), generic
nouns (people, somebody), temporal nouns (week, months), event nouns (game, party),
imaginable abstract nouns (yells, kicks), abstract nouns (control, purpose), and derived
abstract nominals (discussion, pressure) to examine how Hebrew school-age children
utilized nouns while writing and telling expository texts and narrative stories. She found
that all age-groups (4th, 7th, 11th graders and adults) used more abstract and high-level
nouns in expository and written texts than narrative and spoken texts, and a gradual
increase toward adolescence was shown in both category and abstractness. In the use of
abstract nouns within both modalities across all age groups, 12% ofthe words used in
34
expository text were abstract nouns, but abstract nouns only accounted for 7% of the
word use in narratives.
On of the sub-studies done by Berman and Nir-Sagiv (2007) categorized Ravid's
(2006) noun scale into four levels based on the abstractness and imageability. The
participants were asked to tell and write a story regarding "problems between people"
(narrative) and discuss "problems between people" (expository) orally and in written
format. The results revealed that participants tended to use more concrete/imageable
nouns in narrative, but use more abstract and low frequency words in the expository
writing.
Summary afAbstract Nauns
Use of abstract nouns is one of the features in later language development.
Nippold (2007) stated four contrasts between early and later language development
including sources of language learning, metalinguistic competence, abstract thought, and
social perspective-taking; thus, the use of abstract nouns is evidence of thinking
abstractly. Dual coding theory provides a foundation for how and why people process
abstract and concrete nouns differently in the brain, and definition tasks additionally
supply evidence of distinct quality and quantity while defining these two types of nouns.
In both spoken and written discourse, abstract nouns have received little attention
from researchers. However, language development is a continuum and lexical learning is
a gradual rather than all-or-none process. Therefore, in order to fully understand language
development, it is important to study the lexical use from school-age to adulthood and to
detect the differences from early to late language development. Although the changes in
35
quantity might be subtle, changes in quality of language use show differences (Nippold,
2007; Nippold, Mansfield & Billow, 2007; Scott, 1988). Nippold and colleagues (2005)
claimed that the use of abstract nouns increased between childhood and adulthood, and
that the word learning process heavily focused on abstract concepts in adolescents and
adults. Since children's thoughts become increasingly abstract with age, measurement on
how school-age children use abstract nouns is needed. The study of abstract nouns offers
insight to researchers for understanding later language development. This information
also serves as an index for clinicians to evaluate language processes in children with
language difficulties.
Mental State Verbs
In addition to abstract nouns, the use of mental state verbs is sensitive to
developmental growth. During the school years, words are used not only to label familiar
experiences but also to reflect internal states, which are critical in reading comprehension
and communication (Westby, 1990). In both spoken and written language, it is important
to correctly convey people's beliefs and the truth of what is communicated
(Fuste-Herrmann, Silliman, Bahr, Fasnacht, Federico, 2006). Therefore, mental state
terms, including metacognitive and metalinguistic verbs, are worthy of study because
they play important roles in social and conversational understanding and represent
people's internal mental states.
- -- ..__._----
36
Mental State Verbs and Theory ofMind
According to Olson and Astington (1993), metacognition is talking about
cognition and metalinguistics is talking about language, and they are all
metarepresentation. In addition, metacognitive and metalinguistic terms provide language
for and about thinking. When people say, "I think an elephant is a mammal," it represents
language for thinking to reflect on things in the world or to change one's opinions about
things. Mental state terms also are used to reflect our own or other peoples' thoughts; thus,
they are language about thinking (Olson & Astington, 1993). Astington and Olson (1990)
claimed that metacognitive verbs, such as assume, doubt, infer, and remember, and
metalinguistic verbs, such as admit, conclude, confirm, andpredict, facilitate children's
thinking and talking. In addition, use of these terms could help children to recognize
different mental states and speech acts in order to make the meaning clear. Moreover,
mental state verbs not only are used to monitor, interpret, and express internal mental
states of people, but also to provide a possible window into how the mind develops.
Theory of mind, the ability to understand that people have beliefs, desires, and
intentions, is important in the development of pragmatics. Most children produce at least
one mental state verb around two years old (Olson & Astington, 1993; Shatz, Wellman &
Silber, 1983). Basically, two common tasks have been used to examine the development
of theory of mind and false belief. During the first task, one character hid an object in one
place and then left and the second character moved the object while the first character
was gone. The participant was asked where the first character would look for the object
upon returning. The other task is an unexpected-content task where the experimenter used
37
a commonly seen and familiar container such as a candy box, but replaced the candies
with other objects such as crayons. After the child leamed that the box contained
something unexpected, the experimenter closed the box and asked the child what a friend
would think is in the box before he/she looked inside.
When theory of mind comes to language, mental state terms are a salient and
explicit way to represent and detect one's thinking and intemal states. In addition, use of
metacognitive and metalinguistic terms provided the first evidence of children's
development of theory of mind (Olson & Astington, 1993). Booth and colleagues (1997)
stated that cognitive words are an important process to develop awareness of intemal
mind activities and a vehicle to explore one's thought. Therefore, mental state terms are
critical in cognition and language development because these terms provide evidence of
development of theory of mind and false belief. What is the relationship between
language and theory of mind?
Three major views about the relationship between language and theory of mind
were discussed by Astington and Jenkins (1999) and Miller (2004). One of the major
views is that theory of mind develops first and then language builds on it. The second
point of view is that development of theory of mind depends on language competence,
including pragmatics, semantics (Astington & Jenkins, 1999), and syntax (de Villiers &
de Villiers, 2000) (Slade & Ruffman, 2005). Another view claimed that theory of mind
and language are intertwined in development, and depend on the third factor such as
working memory or executive function. The research results did not reach consensus on
38
which factor strongly predicts the development of theory of mind; however, it is clear that
there is a strong relationship between theory of mind and language competence.
Although the views on the relationship between theory of mind and language
were divergent, several studies showed that mental state terms emerge around two to
three years old (Bretherton & Beeghly, 1982; Shatz et aI, 1983). At this time, children
begin to be aware of other peoples' thoughts and minds, although how much children
understand other peoples' thoughts remain in question (Olson & Astington, 1993).
Preschoolers begin to use lower level mental state terms, for example, know, remember,
and guess; however, they do not grasp the complete meanings of these mental state terms
until the early school years (Astington & Olson, 1990). Research showed that children
begin to distinguish different certainty levels of mental state terms from four or five years
old (Johnson & Maratsos, 1977; Johnson & Wellman, 1980; Moore & Davidge, 1989)
and understand that people have different beliefs, and that beliefs change depending on
time, the environment, and circumstances (Olson & Astington, 1993).
Understanding continues to develop well into adolescence for high level mental
state terms such as infer and hypothesize (Schwanenflugel, Henderson, & Fabricius,
1998); however, high school and college students might still have incomplete knowledge
about some of the complex cognitive words, such as conclude, assume, interpret, and
infer (Booth & Hall, 1994). The reason that children do not appropriately use and fully
understand mental state terms until adolescence might be that many of the mental terms
are polysemous, and the sentences that carry these terms are usually complicated and
involved in false belief.
39
Booth and Hall (1994) pointed out that mental state terms develop in a gradual
manner, but that the acquisition of meaning might be different depending on its
conceptual difficulty, levels of uncertainty, and abstractness (Schwanenflugel, Fabricius,
& Noyes, 1996). The examples proposed in Booth, Hall, Robinson and Kim's article
(1997) explained how abstractness and conceptual difficulty could range among different
levels. For example, "know" in "I know the plate is dirty" is the lowest perception level,
but "know" in "I know he will win the award" involves making presuppositions about the
truth and belongs to the highest evaluation level. Therefore, the same mental state words
may represent different internal mental state levels, and it makes mental state terms even
more complicated to understand and takes longer for children to fully grasp the meaning.
However, knowledge of mental state terms increases with age until adulthood as people
organize and categorize these metacognitive words.
Mental State Terms and Literacy
Literacy leads people to represent language in a new way and to provide a way to
think about language in a written format (Olson, Antonietti, Liverta-Sempio, & Marchetti,
2006). In order to fully understand the text, it is critical to understand what is given in the
text and what is the meaning under the surface. Mental state terms serve as a bridge for
readers to correctly comprehend the text. Olson and Torrance (1986) conducted a
two-year longitudinal study beginning when participants were 5 1/2 years-old. Four
metacognitive and metalinguistic tasks were administered and reading and vocabulary
test scores were collected. Results showed that performance on metalinguistic and
metacognitive tasks was strongly correlated to the oral and literate competence,
40
especially reading ability. In addition, the measures on metacognitive and metalinguistic
verbs were intercorrelated. Thus, mental state terms are related not only to pragmatic and
social conversation but also to comprehension, literacy, and metacognitive strategy
development.
The association between metacognition and reading comprehension is strong, and
it is evident that metacognition is critical in reading comprehension training and in
enhancing overall learning efficiency. Booth and Hall (1994) investigated the relationship
between cognitive words, think and know, and the development of metacognition.
Twenty-four short stories, each with one cognitive word, were presented to 5th graders,
7th graders, 10th graders, and college students. Subjects were asked to choose the best
answer from four sentences; a replacement cognitive word was contained in each
sentence. The findings revealed that cognitive words were significantly related to
metacognitive strategies and might enhance metacognitive strategy acquisition. They
further pointed out that children with more cognitive words might be more likely to
process the metacognitive information efficiently and effectively and to maintain and
generalize reading strategies.
Mental state terms also play an important role in literacy. Since most of the
mental state terms exist in literature and textbooks, not dialogue, teachers need to assist
children to comprehend the sentences correctly without building misleading concepts of
knowledge. Olson and Astington (1993) suggested that teachers should ask questions to
facilitate children's thinking such as "Is that an assumption or a hypothesis?" in all
discourse genres from younger ages. In addition, teachers need to talk about thought by
41
using mental state terms and to encourage students to use such language often in class
instead of directly teaching metacognitive and metalinguistic words. Moreover, teaching
students to interpret text through thinking how a statement is to be taken is essential
because texts do not always clearly state the inferential and metaphorical meanings.
Development ofMental State Terms in Preschoolers
Children begin to produce mental state terms around two years of age. In Shatz,
Wellman and Silber's study (1983), they first transcribed a speech sample of one child
collected from age 2:4 to 4:0 years, and then gathered language samples from another
thirty children for six months beginning at age 2:0 to provide stronger evidence. They
focused on the frequency and functional use of know and think. The results showed that
most of the children produced at least one mental verb at 2 years old; however, these
words were not used to refer to internal mental states. They pointed out that younger
children did not use these terms to make reference to mental states, resulting from a lack
of awareness of mental states instead of deficits in linguistic abilities. Therefore, younger
children tended to use think and know for conversational purposes, not for mental
reference.
Johnson and Maratsos (1977) recruited 32 preschoolers, between 3 and 4 years
old, and used a story of a hidden-object trick with follow-up questions to examine the
understanding of know and think. The results indicated that 4-year-olds could distinguish
that think might be false and know must be true, but 3-year-olds showed confusion with
mental state verbs with external events.
42
Another study done by Moore and Davidge (1989) used a similar hiding game
performed by two toy bears with 60 preschoolers, ages 3 to 6 years-old. They examined
three mental terms including, think, know, and sure. In this study, children had to find the
candy from two boxes according to what the two bears said. The bears said the same
sentence but used different mental terms, for example, "I think it is in the red box" or "I
am sure it is in the blue box." The findings showed that children began to differentiate
know and think around four-years-old, but failed to distinguish the difference between
know and sure even by 6-years-old. Therefore, they claimed that younger children could
differentiate the mental terms between high and low certainty, such as know and think,
but failed to make a distinction between two high certainty ones, sure and know.
Forty-five children (ages 3, 4, and 7) and a group of 15 adults participated in
Babu and Mishra's study (2000). Three tasks, including comprehension, verbal-choice,
and pragmatic tasks, were given to examine the understanding of the semantic and
pragmatic meaning of think and know. In the comprehension task, the introductory
context was provided, so children could utilize the context as reference to answer
presupposition and explanation questions. The verb-choice task required children to
choose from one of the two cognitive words that best described the character's mental
state based on the story content. Finally, the hiding game, which was similar to the
previous research, was used in the pragmatic task, and children were asked to provide
justification for their answers. The results revealed that there was a developmental trend
in the understanding of know and think, and the pragmatic understanding was well
developed while the semantic understanding was not completed by age 7.
43
Booth and his colleagues (1997) conducted a study to analyze the utterances with
know of four children (ages between 2 to 5) and their parents. The utterances were
analyzed by semantic or pragmatic properties and referring to self or others. This study
revealed that the percentage of use of high-level cognitive words increased with age for
both children and parents, and children reflected the distinction between metal states on
having greater use of different cognitive words. In addition, children tended to use
cognitive words to refer to themselves rather than to others. Apparently, children
produced the cognitive words that they heard from their parents; thus, parental cognitive
word input was important and related to children's production of these words.
Development ofMental State Terms in School-Age Children andAdolescents
Several studies were conducted to compare the understanding of mental state
terms between preschoolers and young school-age children. In an early study done by
Johnson and Wellman (1980), they used a series of hidden-object tasks with 4 and
5-year-olds and first and third graders to examine the development of the following
mental verbs: remember, know, and guess. After each task, the experimenter asked three
questions about each mental verb, such as "Do you know/remember/guess what is there?"
The results revealed that the four-year-olds could distinguish internal mental state from
external events, but were confused with remembering, knowing and guessing. In addition,
the five-year-olds did show some distinctions of mental state terms, but the understanding
was more advanced in school-age children.
Another study conducted by Moore, Brynt, and Furrow (1989) recruited 69
children (ages 3, 4, 5, 6, and 8), and used a similar hiding game with think, know, and
-------------
44
guess statements. The results indicated that children differentiated think and know by age
4, but that the distinction between think and guess was not complete even at 8-years-old.
However, 4-year-olds showed some understanding of certainty levels in mental state
terms. For example, 4-year-olds understood know as an indicator of certainty, and that
think and guess were indicators of uncertainty.
In the Booth and Hall study (1995), they explored the six levels of cognitive
meaning of know including perception, recognition, recall, understanding, metacognition,
and evaluation. Four groups of children (ages 3, 6, 9, and 12) participated in this study.
After each story was carried out, five basic questions related to the story content and six
questions associated with the six levels of cognitive meaning of know were asked. The
findings indicated that children developed the hierarchical meaning of know with age,
and there was a significant increase in lower level meaning than higher-level meaning
throughout the age range. In addition, they stated that there was a correlation between
knowledge of mental state terms and standardized vocabulary measures for all ages
except 3-year-olds. Furthermore, the distinction between mental state terms was
correlated with the theory-of-mind task; thus, mental state terms are a good way to
understand the development of theory of mind.
There are relatively few studies done with school-age children; however, this
period is likely to show important changes in metacognition since children extend and
elaborate their understanding of mental states from low to high cognitive levels
(Schwanenflugel, Fabricius & Noyes, 1996). Thus, children during these years may have
more advanced, clear, and distinct understanding of mental state terms and this might
45
result from the fact that most of these words exist in textbooks, literature, and poems
(Peskin & Olson, 2004). Schwanenflugel, Henderson and Fabricius (1998) examined the
changes in semantic organization of knowing on 9-, and ll-year-olds and adults.
Participants were asked to read 34 scenarios and a list of 17 verbs related to knowing
followed by each scenario. Subjects had to circle at least three verbs to describe how they
would use their mind in each situation. Seventeen mental verbs included: input verbs,
examine and recognize; memory verbs, remember and understand; and constructive
processing verbs, estimate and guess. They found out that children were in the process of
developing a constructive theory of mind, but were still less constructive than adults. In
addition, adults put similar uncertain verbs together such as guess, estimate, and question,
because these verbs were co-occurring in the same scenarios, but children did not view
these mental processes in the same way as adults did.
Olson and his colleagues (2006) used five metacognitive verbs and four
metalinguistic verbs to create three versions of stories containing knowledge of folk
psychology, history, and mathematics for each verb and asked participants to choose the
best mental verb for each story. The study included 3rd, 5th, 7th, and 9th graders and a
group of undergraduates. A developmental trend was shown in the acquisition of
metarepresentational verbs; however, even undergraduates did not reach 100% correct. In
addition, metalinguistic verbs were mastered before metacognitive verbs in all groups of
participants. Some of the verbs such as assume, conclude, and doubt were more difficult
than others.
46
In another school-age study by Astington and Olson (1990), 6th, 8th, 10th, 12th
graders and undergraduates participated. Twelve stories were used to test the
understanding of higher-level metalinguistic verbs: assert, predict, imply, and
metacognitive verbs: hypothesize, assume, and conclude. Subjects were asked to choose
the most appropriate verbs to replace say and think in each story. The finding showed that
the knowledge of these mental state terms increased over the high school years and the
understanding of these terms was much improved by the end of high school. In addition,
the scores of this task were correlated with reading vocabulary and critical thinking tests
in 8th and 12th graders. Booth and Hall (1994) modified Astington and Olson's study
(1990) to examine six levels of meaning in think and know with 5th, 7th, 10th graders and
college students. Twenty-four short stories were used to examine the understanding of
low and high frequency cognitive verbs. The finding indicated that cognitive word
knowledge increased with age. Moreover, developmental differences were observed in all
groups and cognitive levels except metacognition. Additionally, the largest increase was
between 5th and 10th graders, and the possible reason might be that older children have
more exposure to mental state terms in textbooks and literature.
Mental State Terms in Writing
Very few studies examined the use of mental state terms in writing; however,
writing is an important piece in the language and learning system. Westby (1990) stated
that children could not function independently in reading and writing if there is no
awareness of internal mental states such as remembering, knowing, thinking, and
guessing. It is important to analyze the presence of mental state terms because these
47
words reflect the awareness of mental states and metacognition development (Greenhalgh
& Strong, 2001). Nippold, Ward-Lonergan, and Fanning (2005) investigated semantic,
syntactic, and pragmatic development in persuasive writing. In semantic analysis,
adverbial conjuncts, abstract nouns, and metacognitive/metalinguistic verbs were
analyzed; however, early developing metaverbs, such as know, think, feel, and ask, were
excluded in this study. The results indicated that use of mental state verbs in writing
increased with age but the use of these verbs remained low even in adults.
There is no study to date comparing use of mental state terms between narrative
and expository writing. Most of the research in mental state verbs focused on
comprehension tasks instead of production. However, language is both expressive and
comprehensive; therefore, examining only one aspect of language use cannot offer a
complete picture of language development. Through examining the use of mental state
verbs in writing, researchers could understand how children actually use these words
when children compose two different types of writing. In addition, it is critical to know if
the use of mental state terms increases with age and what kind of metal state terms they
use while writing different content.
Summary ofMental State Verbs
Mental state verbs contain metalinguistic verbs, language for thinking, and
metacogtitive verbs, language about thinking. Olson and Torrance (1986) stated that
metalinguistic and metacognitive verbs are the heart of language and thought, and the
thoughts of a child and adult are closely related to the sentences that they produce and
comprehend. However, when children first use mental state verbs, such as know and
48
think, it does not necessarily mean that they fully grasp the meaning of mental state
verbs (Johnson & Maratsos, 1977; Moore & Davidge,1989; Shatz, Wellman & Silber,
1983;). Certain mental state verbs are learned as early as five-years-old; however, more
complex and advanced mental state verbs such as conclude and assume cannot be fully
understood by adolescents (Olson, Antonietti, Liverta-Sempio, & Marchetti, 2006).
There is a developmental trend in using mental state verbs, and one explanation
is that the information-processing load increases with age to allow people to manage
several meanings at once (Wing & Scholnick, 1986). Mental state verb use is related to
the development of theory of mind and language, and these words develop rather late
compared to other verbs such as get and have. Thus, examining how school-age children
and adolescents use mental state verbs provides insight into the relationship between
language and thought. In addition to comprehension tasks, writing provides a natural
method to understand the actual use of mental state verbs through development.
Derivational Morphology
In addition to abstract nouns and mental state verbs, acquiring derived words is
another important concept in school-age and adolescent language development.
Morphologically complex words used in the textbooks increase beginning at 3rd grade,
and learning derivational morphology is a gradual and complex process (Nippold, 2007).
In addition, children's knowledge of derivational morphology is related to word
knowledge and reading comprehension (Larsen & Nippold, 2007; White, Power & White,
1989; Wysocki & Jenkins, 1987).
49
Derived words are complicated but sometimes learners can follow certain rules to
figure out the meaning of the words during reading. For example, happiness and
loneliness are both nouns and derive from the adjectives, happy and lonely. However, not
all of the derived words are salient to learners or can be decomposed, and it makes
learning of derivational morphology more difficult. What is derivational morphology?
How do children acquire it? A brief literature review is provided in the following section.
What Is Derivational Morphology?
Morphological knowledge includes the acquisition of inflectional morphology,
such as plural and past tense markers, and derivational morphology, such as prefixes and
suffixes. Inflectional morphology is acquired earlier in a fixed manner and follows more
systematic rules compared to derivational morphology (Mann, 2000). Children pick up
derivational morphology later than inflectional morphology because of the greater
complexity of derivatives. Deviational morphology does not always follow consistent
rules and might not generalize the same meaning from the same base or root. In addition,
derived forms might change phonology, orthography, and semantic and syntactic
characteristics from the base form. These issues make derived words more difficult to
learn.
Champion (1997) pointed out that children increase their vocabulary remarkably
from 3rd grade through middle school, and knowing how to utilize rules of derivational
morphology to learn novel words is an important factor. According to Biemiller and
Slonim (2001), children acquire about 5200 root words by the end of 2nd grade and gain
an average 3200 additional root words a year between 3rd and 5th grade. Since the
50
understanding of root words leads children to expand their vocabulary more efficiently, it
is expected that children will learn about twice as many words as the root words, based
on Anglin's estimates (1993). Children show gradual improvements in derivational
morphology from 3rd or 4th grade through high school and college due to increased
exposure to complex words in literature, textbooks, and lectures. Therefore,
understanding how children acquire derivational morphology is critical because children
encounter more and more complex words throughout the grades, and the ability to use
and analyze complex words provides an important foundation for reading and writing.
The relationship between base form and derived form could be transparent or
opaque. Opaque words are more difficult to learn than transparent ones because
decomposing opaque words into meaningful constituents to generate correct meanings is
not easy. Moreover, some derivatives change phonologic and orthographic characteristics
from base forms, and this makes the derivatives even more difficult to learn. In addition
to the relationship between base forms and derivatives, there are two classes of
derivational suffixes including neutral and nonneutral suffixes. Tyler and Nagy (1989)
indicated that neutral suffixes, such as -ness, -er, -ment, and -ize do not involve changing
the vowel quality, stress or phonological characteristics; thus, most of the derived words
formed from these neutral suffixes are transparent and acquired earlier. On the other hand,
non-neutral suffixes, such as -ive, -ous, and -ity change in phonology, and these
derivatives are often not transparent from the base forms.
In addition to the difference in acquisition, productivity is very distinct between
inflectional morphology and derivational morphology. An inflectional morpheme, such as
51
-s (plural), -s (possessive), -s (3rd-person singular), and -ed ( past tense), creates a change
in the function of the word. For example, the ed in walked indicates past tense.
Inflectional morphology is constrained by grammar; therefore, it shows limited variation
in productivity. However, derivational morphology displays greater productivity but most
of the derivatives are highly predictable if people use a decomposition strategy. Although
it is assumed that people tend to use a decomposition method to understand complex
words, there is still no consistent explanation as to how people read complex words by
using a morphological decomposition strategy.
The Process ofDerivational Morphology
Carlisle (1987) pointed out that derived words are difficult to learn because
derived forms are not as common as inflected forms. In addition, these words do not
follow the rules consistently, and the relationship between the base and derived form is
sometimes opaque. For example, application is derived from apply. In addition, learners
have to acquire some word-specific knowledge to correctly use a particular suffix to
convey a derivational meaning. Since derived words are often complicated and difficult
to learn, how do people process these words? Will people process derivational
morphology based on certain rules or use different strategies to categorize, organize, and
store derivationally complex words?
In the Alegre and Gordon (1999) study, they tried to use the "gang effect" which
was found in inflectional morphology to explain how derivational morphology is
processed in an associative memory structure. In inflectional morphology, the "gang
effect" means that people would produce generalized words by using similar patterns that
52
are already stored in memory. For example, to produce the past tense, it would be easier
to generate stink-stank if people know the pattern of drink-drank or sink-sank.
Alegre and Gordon (1999) investigated 6 non-neutral affixes including
-ion,-al(n),-al(v), -ity, -ous, and -ic and 6 neutral affixes including -ize, -en, -ness, -able,
-ment, and -er .Both neutral and non-neutral affixes displayed gang effects, but did not
extend in the same way. In their second study, forty-four undergraduate students filled out
a questionnaire containing 69 novel derived words including four gang affixes (-ion,
-al(v), -al(n}, and -en) and three non-gang affixes (-er, -ness, and -able) from the first
study and their base forms. Participants first rated naturalness of the base form and then
rated the likelihood of the base form carrying a particular affix on a 7-point scale. The
results revealed that when the novel words did not contain any attested forms or violated
the gang patterns, the acceptability of the novel words would be much lower. In addition,
the existence of gang effects had a strong influence on judgments of the acceptability of
the novel word. If an affixed form was similar to members of the gang cluster, it would
be more acceptable.
In addition to using gang effects to explain the process of derivational
morphology, Verhoeven and Perfetti (2003) indicated three models to explain
morphological process in reading. The first model is the decomposition hypothesis.
According to this hypothesis, people first have to analyze the constituent components and
then find out the meaning of the base word. However, the full-listing hypothesis claims
that complex words would represent in memory and the mental lexicon individually no
matter how complex the word is. The third hypothesis combines the first two assumptions
53
and develops into an interactive model. It proposes that people could define the word
based on the frequency of the word and its regularity in orthography. According to the
hypothesized process of derivational morphology reviewed above, learners categorize
and organize affixes based on their meanings and functions and further use that
knowledge to decompose a derived word into meaningful components to gain meaning.
Development ofDerivational Morphology in School-age Children
Children begin to analyze derivational words when they are in elementary school.
Carlisle (1988) pointed out that there is a developmental trend in learning derivational
morphology and that children increase their knowledge of derivational morphology
dramatically between 4th and 8th grade. In order to understand how school-age children
acquire derivational morphology, Windsor (1994) recruited 3rd to 8th graders and 40
adults as participants. Twenty-six nonsense picture pairs were used as stimuli in
production and comprehension tasks. Six different derivational meanings were tested
including diminutive X, to make X, without X, character of X, approximate X, and can be
X-ed. In the production task, participants had to provide derived words to complete the
sentences presented by the experimenters. In the comprehension task, children had to
choose one correct answer from four words derived from the same root after the
experimenters showed the picture and read the story to them.
The results showed that both school-age children and adults performed better on
the comprehension task than on the production task, but that adults tended to utilize
particular suffixes to express specific derivational meanings. For example, adults tended
to use -like and -ish for "approximately X", -ful for "character of X", -ize to express "to
54
make X" and -ette to convey "diminutive X". In contrast to adults, children made more
errors in the use of suffixes by producing compound words and creating nonsense words
without suffixes. In addition, accuracy on the comprehension task for older children was
close to the accuracy level of adults; however, children's accuracy of suffixation
production was significantly lower than the adults'. Therefore, derivational morphology
takes time to develop and suffixation production improves slower than comprehension.
In Lewis and Windsor's (1996) follow-up study, they investigated how children
used morphological knowledge when they encounter unfamiliar real derivational words.
The results demonstrated that in the production task, children tended to use productive
suffixes more frequently than less productive ones. For example, in this study, children
preferred to use -let to express "diminutive X", -ize for "to make X", -Jul for "character of
X", and -ish for "approximately X". Overall, the accuracy in comprehension was
consistently better than that in production, which replicated the findings of Windsor's
(1994) study. In addition, productivity had a great influence on choosing suffixes in both
comprehension and production, and school-age children were aware of derivational
components.
Freyd and Baron (1982) examined the acquisition of derivational morphology on
8th graders and superior 5th graders. A vocabulary test that included words such as ethics
and limber and derived words such as movement and imagination were given. In their
study, superior 5th graders did better on derived words than simple words because
children analyzed words based on the base forms and suffixes. In the meaning learning
test, superior 5th graders did better when the pseudowords were derivationally related, for
55
example, skaf = steal, skafist = thief, but 8th graders did not show a difference. However,
in order to succeed in learning derivational morphology, the ability to recognize the
relationship between base and derived forms is not enough.
Tyler and Nagy (1989) proposed that a full understanding of derivational
morphology requires three steps. First, learners must have the ability to recognize the
relationship between base and derived forms. Second, learners need to understand that
suffixes also represent syntactic categories. For example, -ize is a verb marker and -ion is
a noun marker. The last step is to develop distributional knowledge; thus, learners know
some constraints in the use of suffixes. For example, -ness could be attached to adjectives
but not to verbs. In Tyler and Nagy's study (1989), the findings revealed that 4th graders
had a basic ability to recognize the base form from an unknown derived word, but that
knowledge of syntactic categories developed later than the ability to recognize the
relationship between base and derived forms. Moreover, 6th graders were in the process
of acquiring distributional knowledge of suffixes, but the major progress in this
knowledge occurred even after 8th grade.
Leong (2000) used a different method, a computerized priming procedure, to
determine the accuracy and reaction time of processing morphological relationships in
4th, 5th, and 6th graders. The experimenter used four conditions including no change,
Olihographical change, phonological change, and both Olihographical and phonological
change for the test materials. Ten words embedded in short sentences in each condition
would appear on the computer screen. The results showed that the reaction time was
getting shorter and the accuracy was getting higher from 4th to 6th graders, and the
56
results from this computerized morphology task were consistent with the scores on
standardized spelling tests.
The second experiment utilized the same stimulus materials and procedures on the
same participants to determine if students could accurately and rapidly generate the
derived words from the base forms provided on the computer screen. The results revealed
that there was a developing trend from 4th graders to 6th graders in both base and derived
form tasks. Students showed higher accuracy and shorter reaction times when producing
the base form from a less complex derived words such as deriving happy from happiness
and deriving a less complex derived words from the base form such as generating
knowledgeable from knowledge.
Nippold and Sun (in press) administered the word knowledge task including 15
derived adjectives and derived nominals on typically developing 5th and 8th graders.
Each question provided one sentence as contextual support for the missing target word
and a choice of four derived words. The results revealed that 8th graders outperformed
5th graders on the task and the derived nominals were more difficult than the derived
adjectives for both groups. In addition, word frequency, root word opacity, familiarity of
the root word, and the multi-derivative nature of certain words (such as dictatorship)
seemed to contribute to the complexity of learning the meaning of derivatives.
According to the literature reviewed above, children develop derivational
morphology mainly during the school age years, and knowledge of derivational
morphology increases with age (Nippold, 2007). Why is it important to study derivational
morphology in school age children? Green and the colleagues (2003) pointed out that
57
morphologically complex words account for approximately 60 % of new word learning in
school-age children and students encounter affixed words while age increases (White,
Power & White, 1989). Additionally, a child's vocabulary size increases dramatically
from 3rd to 8th grade and morphological generalization contributes to the large
vocabulary growth (Larsen & Nippold, 2007; White, Power & White, 1989; Wysocki &
Jenkins, 1987).
Derivational Morphology in Vocabulary Learning and Spelling
Bertram, Laine, and Virkkala (2000) investigated Finnish children's knowledge of
derivational morphology and if frequency and affix productivity played an important role
in learning vocabulary. Thirty-two 3rd graders participated in the first experiment and 70
target words including 35 high-frequency and 35 low-frequency words were used as
materials. The experimenters pronounced each word and showed a stimulus card with the
target word to children simultaneously. The children had to provide a definition for each
target word orally. The results showed that high frequency words were defined better than
low frequency words and that words with high productivity suffixes were better identified
than those with low productivity suffixes. In addition, morphological structure had a
positive impact on defining words, since children showed better identification in words
with low productivity suffixes than low-frequency monomorphemic words. Therefore,
when sufficient morphological information is provided by the word, readers can derive
meaning even though it is a complex word.
Larsen and Nippold (2007) asked 50 typically developing 6th graders to explain
15 low-frequency morphological complex words using a dynamic assessment procedure
58
in which different levels of scaffolding were provided by the researchers. Additionally,
scores on standardized vocabulary test (PPVT-III) and reading comprehension (Oregon
Statewide Assessment) were obtained. They pointed out that children's performance on
the task was positively correlated to their literacy skills. Moreover, although children
used the strategy to analyze the morphologically complex words, some degrees of the
scaffolding were needed. Since scaffolding is needed in learning morphologically
complex words, White, Sowell and Yanagihara (1989) advocated the importance of
teaching some most common prefixes and suffixes to promote students' ability to analyze
these words.
In addition to vocabulary acquisition, derivational morphology plays an important
role in spelling. In the Carlisle (1988) study, four tests including the Wide Range
Achievement Test-Spelling subtest, the morphology test, the spelling test, and the suffix
test were administered to 4th, 6th, and 8th graders. The morphology test contained two
subtests that were Derived forms and Base forms subtests; the stimulus words were
divided into four types (no change, orthographic change, phonological change, and both
change) to indicate the transformation between the base form and the derived form. In the
spelling test, participants had to spell the base and derived forms from the derived forms
subtest of the Morphology test. In the suffix test, the participants had to add a suffix to 30
nonsense words using the doubling final consonant rule, the final -y rule, and the final -e
rule. The results showed that there was a developmental trend in learning derivational
morphology and that children were able to use the relationship between the base and
derived forms to spell derivatives even though children still had not mastered suffix rules.
59
In the Carlisle (1987) study, typically developing children, 4th, 6th, and 8th
graders, and 9th graders with learning disabilities were recruited. Four tests including the
Wide Range Achievement Test (WRAT), the Test of Morphological Structure (TMS), the
Spelling Test (ST), and the Test of Suffix Addition (TSA) were delivered to compare the
performance of typically developing students and students with learning disabilities.
The finding showed that spelling base and derived words was difficult because
spelling required knowledge of phonological awareness, syntactic, orthographical rules,
and morphology. In addition, students made more errors in producing words with
phonological change and both phonological and orthographic change. More difficulties
were found with 9th graders with LD since they tended to learn derived words as whole
words and did not know how to use morphological rules in spelling derived words.
Derivational Morphology in Writing
Most of the research was done to examine the acquisition of derivational
morphology through comprehension tasks in which children had to generate the meaning,
recognize the base and derived forms, and categorize the suffixes. However, how children
practically use the derived words has not received enough attention. Rubin, Patterson,
and Kantor (1991) stated that implicit morphological knowledge is one of the underlying
factors for writing since both oral and written language production require the same
linguistic structure and implicit knowledge. Therefore, children who do well on oral
morphological tasks would also do well on writing and morphological analysis since
implicit knowledge would reflect the explicit levels. One of the findings in their study
showed that morpheme use, both inflectional and derivational morphology, in writing
60
showed very similar patterns to the use of morphological knowledge in spoken language.
In addition, children who demonstrated strong implicit and explicit morphological
knowledge made fewer errors in spelling and morphemic errors in writing.
Limited studies were done to examine the use of derived words in children's
writing. Green and colleagues (2003) investigated morphological development in
narrative writing in 3rd and 4th graders. Participants were showed two pictures in
different time periods and then asked to write stories about the pictures. The findings
showed that children used fewer derived words than inflections, and that the accurate use
of inflections was higher than the use of derivatives. In addition, 4th graders used
derivational forms more accurately than 3rd graders, and they have mastered most of the
inflectional forms used in their writing but not for derivational morphology. Green and
colleagues indicated that morphological development in writing was similar to
development in oral language, and a majority of the 4th graders did not use derivational
morphology in writing. Additionally, most of the derived words used in writing fell into
six derivational categories: -11, -ed, -1, -ing, -luI, -er, and only phonologically transparent
derived words were used in 3rd and 4th graders' writing. Therefore, opaque derived
forms are more difficult to learn and use in both oral and written language.
Reading and writing are important to morphological learning. Children expand their
morphological knowledge from exposure to print starting in 2nd or 3rd grade. They
master inflectional morphology first and begin to learn derivational morphology during
these early school years. Carlisle (1996) investigated how commonly and accurately 2nd
and 3rd graders with and without learning disabilities used int1ections, derivations, and
61
compound words in spontaneous writing. In addition to the writing task, children were
asked to complete sentences with the correct forms while the base forms were provided;
they then spelled out the word. The results showed that 2nd graders used fewer words and
were less accurate in using morphologically complex words such as derived forms and
compound words. In addition, the use of morphological markers increased from 2nd to
3rd grade, but 2nd graders with LD were less accurate than their peers. Also, there was a
strong correlation between the sentence completion task and the accuracy of using
morphological forms in spontaneous writing. The results suggested that there might be a
transitional period between 2nd to 3rd grade, and that children consolidated their
knowledge of inflectional morphology and began to use derived forms in their
spontaneous writing during this period.
A recent study by Ravid (2006) examined two genres, narrative and expository,
and two modalities, speech and writing, to detect the use of different nouns. Nouns were
divided into ten levels based on abstractness and semantic~pragmatic content. The ten
levels were concrete nouns (ball, bike), proper names (Tiffany, Mike), collection/location
(library, class), role nouns (cousin, neighbor), generic nouns (people. somebody),
temporal nouns (week, months), event nouns (game, party), imaginable abstract nouns
(yells, kicks), abstract nouns (control, purpose), and derived abstract nominals (discussion,
pressure). Twenty native Hebrew speakers from each age group (4th, 7th, 11 th graders
and adults age between 25 to 35) participated in this study (N= 80). Participants were
asked to write and talk about their personal experience of problems between people
62
(narrative) and to discuss the same topic in both spoken and written modalities
(expository discussion).
The results indicated that nominal density (the mean number of lexical nouns per
clause) increased with age especially in adolescents and adults. In addition, use of nouns
was denser, more linguistically complex and rich in expository than narrative gemes
across all age groups. Overall, participants used more abstract and high-level nouns in
expository and in written texts than narrative and spoken texts. In addition, role,
collective/location, imaginable abstract, and generic nouns were the four mostly used
nouns in narrative texts; on the other hand, generic, imaginable abstract, derived abstract,
and abstract nouns were mostly shown in expository texts.
Plag, Dalton-Puffer, and Bayyen (1999) studied three types of corpus including
written, spoken-formal and spoken-daily conversation from British National Corpus.
They focused on the productivity of fifteen suffixes including abstract nouns, -ity, -ness,
-ion, participant nouns, -er, -ist, measure partitive nouns -Jul, derived verbs, -ize, and
derived adjectives, -able, -free, -Jul, -ish, -less, -like, -type, -wise in the corpus. The
findings showed that the suffixes yielded more types in the written than in the spoken
corpus and derived nouns made a larger contribution to the vocabulary size. However,
because the sample size of each corpus was uneven, the productivity of suffixes varied
depending on the corpus.
Summary ojDerivational Morphology
Derivatives are acquired later because derived words are more abstract, and these
words are commonly seen in written and formal language, not daily conversation (Nagy,
63
Diakidoy, Anderson, 1993). In addition, the abstractness of the derivatives, available
knowledge of word structure, and meta1inguistic competence explain the late acquisition
of morphologically complex words (Nippo1d, 2007; Nippold & Sun, in press). According
to research, children develop knowledge of derivational morphology between 4th grade
and high school, and they make major gains between 4th and 7th grades (Nippold, 2007;
Nagy, Diakidoy, Anderson, 1993; Wysocki & Jenkins, 1987). Because of exposure to
print and learning in content areas, children have an enormous opportunity to encounter
derivatives and to use either context abstraction or morphological generalization to learn
new words. However, even high school students and adults were not completely accurate
on derivational morphology tasks, especially for production tasks (Nippo1d, 2007).
Production of derivatives is more difficult than comprehension. However,
according to the literature, development of derivational morphology shows a similar
pattern in oral and written language. When children have sufficient implicit knowledge,
that knowledge is reflected in explicit production. Thus, instead of testing comprehension
of derivatives, understanding how people use derived words gives researchers, teachers,
and clinicians a full picture of development of derivational morphology, which is an
important element in later language development. Writing is planned and formal and is
essential for school-age children to succeed in school. Thus, writing provides valuable
and different insights into the use of language. How do school-age children use
derivatives in their writing? The literature reviewed above showed that inflections are
easier and acquired earlier than derivatives, and that children use infections more often
and accurately in writing. However, how children use derivatives in writing was rarely
64
studied. Based on Ravid's study (2006), when genres were taken into account, children
used more concrete nouns in narratives, but more abstract and morphologically complex
words in expository. Therefore, genre needs to be considered when investigating derived
words in writing.
Research Rationale and Hypotheses of the Current Study
Three types of literate words - abstract nouns, mental state verbs, and
derivatives - were examined in students' essays in the present study. These three types of
words were indicators of later language development and are more common in written
language than in oral language. The literate lexicon differs from the core lexicon. Use of
the literate lexicon is a sign of advanced cognitive and language development (Ravid,
2004). Many research studies focused on how words are learned, such as incidental
learning (Oetting, Rice & Swank, 1995), context abstraction (Cain, Oakhill & Lemmon,
2004; Chaffin, Morris & Seely, 2001), and morphological generalization (Larsen &
Nippold, 2007; White, Power & White, 1989; Wysocki & Jenkins, 1987), but how
children actually use the words in production was rarely discussed. Word use in writing
provides a natural way of examining how children use words other than in speaking.
Since content, syntactic structure, and word use varies depending on modalities, writing
provided a distinct view to examine language development. Although modality may have
an impact on language production, only written discourse was included since everyday
conversation may not represent what children have mastered (Ravid & Zilberbuch, 2003),
65
and the literate lexicon is more often shown in written language (Nippold, 2007; Ravid,
2004).
Based on the research reviewed above, it was important to take genre into account
while comparing children's language performance. Writers alter the organization of
content, syntactic structure, and vocabulary based on different genres. Prater and Padia
(1983) stated that writing skill changed among different genres, and the practice in school
had an impact on student's writing performance. Moreover, writing in different genres
enhanced the relationship between language and cognition and the development of
complex syntactic structures (Blair & Crump, 1984; Scott, 2005). Since a major
difference in genres exists between narrative and expository, use of the literate lexicon
compared across these two types of discourse was selected.
Language development in school-age children and adolescents often receives
much less attention than development in young children. However, language continues to
develop and adds new meaning to old words throughout life. Later language is different
from early language development in speed, salience, and substance (Nippold, 2007) and
is worthy of study. Thus, comparing young children, school-age children, and adolescents
supplied evidence that was needed in understanding the subtle and gradual language
development beyond early childhood. In addition to later language development, writing
was another key variable examined in the present study. Writing plays a critical role in
later language development and academic success. Writing is a type oflanguage
production, but it is less studied compared to speaking. Additionally, research on
children's writing has often focused on cohesion, syntactic structures, and content
66
organization. However, the lexicon, an important and basic component in language use,
was often missing from the research. Therefore, the current study provided research
evidence needed in both later language development and lexical use in writing.
Later language development is a gradual process; thus, widely separated age
group comparisons are needed to detect the subtle differences (Nippold, 1995; Nippold,
1993). In addition, a higher grades comparison is required because children begin to
display their distinctive knowledge between narrative and expository around 4th grade
and by that time, their writing shows more written-like structures and mature word use
that is different from spoken language. Therefore, a comparison among 5th, 8th and 11 th
graders' writing allows researchers to determine if there are quantitative and qualitative
differences in the use of literate words between the three grade levels. Additionally, the
present study investigated if literate lexicon use showed a developmental trend and was
different based on genre selected.
The purpose of the present study was to provide evidence of literate language
development in school-age children and adolescents for clinicians and educators.
Different modalities provide distinct and valuable information in language processes, and
examining writing offered a more complete understanding of language use. In order to
acquire writing and the literate lexicon, students need direct instruction and practice to
master these advanced and complex language skills (Blair & Crump, 1984). Thus, the
present study provided a foundation for future instruction planning and raised attention to
literate lexicon development in writing.
67
Although the present study examined the literate lexicon in writing in typically
developing children, it provided a norm for children with language difficulties. Children
with spoken language problems often have difficulties in reading and writing, given the
reciprocal relationship between spoken and written language (ASHA, 2001). In addition
to the difficulties in writing, the literate lexicon was rarely used by children with
language impairments compared to the matched peers (Greenhalgh & Strong, 2001).
Thus, the present study can be used as guidance in assessment and intervention in literate
language use in writing for school-age children and adolescents who might be at risk of
language difficulties and those who have language impairments.
In sum, the major research question of this study was to examine how literate
words -- abstract nouns, mental state verbs, and derivatives - were used in narrative and
expository writing in three age groups including 5th, 8th, and 11th grade. The hypotheses
of this developmental study are as follows:
1. Older children will use more abstract nouns, mental state verbs, and derivatives than
younger children in both narrative and expository writing.
2. Children and adolescents will use more abstract nouns, mental state verbs, and
derivatives in expository than narrative writing across all age groups.
3. Older children will use more advanced abstract nouns such as circumstance and
consequence, more later-developing mental state verbs, such as assert, intend, and claim,
and more complex derivatives, such as confrontation, intervention, and negativity, and,
than younger ones.
68
4. Children will use more advanced abstract nouns, more later-developing mental state
verbs, and more complex derivatives, and in expository writing than in narrative writing.
69
CHAPTER III
METHOD
Participants
Recruiting and Sampling
One elementary school, one middle school, and one high school in the same
school district from a suburban area in Western Oregon agreed to participate in the
present study. Only those students who signed a student assent form participated in the
study. The present study was a developmental investigation of typically developing
school-age children and adolescents; therefore, participants who received special
education services, had an individualized education plan (IEP), or whose first language
was not English were excluded. Based on the teachers' reports, all participants in the
present study attended regular classes and demonstrated normal language, cognitive,
behavioral, and social-emotional development. Thus, they were defined as typically
developing children and adolescents.
There were 45 students in 5th grade, 44 students in 8th grade, and 62 students in
11th grade during the recruitment period. However, participants whose chronological age
was not in the typical age range (10-11 years for 5th grade, 13-14 years for 8th grade, and
70
16-17 years for 11th grade) were excluded from the study. Thus, one 5th grade, two 8th
grade, and four 11 th grade students were excluded. The researcher randomly selected 40
children for each age group from the rest of the pool, examining each participant's
written language samples in the two gemes. Therefore, the present study included eighty
language samples (40 narrative and 40 expository) for each of 5th, 8th and 11 th grades,
for a total of 240 samples.
Demographics
The participants (N= 120) included forty 5th grade (mean age = 11;1 [years;
months], range = 10;5-11;8), forty 8th grade (mean age = 14;2, range = 13;4-14;10), and
forty 11th grade (mean age = 17;5, range = 16;8-17;10) students. There were 23 males
and 17 females in 5th grade, 20 males and 20 females in 8th grade, and 15 males and 25
females in 11 th grade. The ethnicity proportion in the school district was 91 % Caucasian,
5% Hispanic, 2% Native American, 1% Asian, and 1% African American. The gender
ratio in the school district was 52% male and 48% female, and 54% of the students in the
school district were economically disadvantaged. All participants reported they were
native English speakers.
Materials
Experimental Materials in Writing
Students completed two essays, a narrative piece and an expository piece. They
were allowed 20 minutes per essay. Prior to beginning each essay, they were given an
instruction sheet along with a booklet that contained four pages of lined paper to
71
complete their writing. The order of the writing tasks was counterbalanced to avoid
possible order effects. If the students in the first class wrote the narrative first and the
expository second, the students in the second class wrote the expository first and the
narrative second. The order of the tasks rotated in this way throughout the study.
Instructions for the Narrative Essay
The title of the narrative essay was "What happened one day." The title was set to
be general and participants were asked to write from their personal experiences or to
imagine a story. At the beginning of the instruction sheet, the participants were asked to
write a story that was either true or imaginary about something funny, sad, or scary that
happened to themselves or to people around them. An outline was provided on the
instruction sheet to help the participants generate a narrative. The outline, which was
similar to a story grammar, included the settings, characters, plot, problems, attempts,
outcome, and thoughts. On the instruction sheet, participants were informed of the
twenty-minute time limit and were asked to use complete sentences and correct grammar,
spelling, and punctuation to complete the essay. The instruction sheet for the narrative
essay is provided in Appendix A.
Instructions for the Expository Essay
The title of the expository essay was "The Nature of Friendship." The title was set
to be familiar to participants because domain knowledge has an impact on writing
performance and ability. McCutchen (1986) pointed out that children generated more
coherent and longer texts if they knew the topics well. "Friendship" is important to
people and is a familiar and common topic for students to write about. Moreover, this
---------------------------
72
topic allowed participants to inform, explain, describe, or define their subject to potential
readers. The format was similar to the instructions for the narrative. An outline was
presented that contained seven questions designed to prompt participants to think about
friendship. Each point consisted of a wh- question such as "why is friendship important
to people." The outline provided prompts to help participants write an expository essay.
The instruction sheet for the expository narrative essay is provided in Appendix B.
Teacher Questionnaire
In addition to examining students' essays, the type of teacher instruction they had
received was important to examine (Duin & Graves, 1987; Laflamme, 1997; Zarry, 1999).
Therefore, four teachers, two from 5th grade, one from 8th grade, and one from 11th
grade, agreed to fill out a 6-page questionnaire during or after the writing task. The
teachers of 8th and 11 th grades were instructors in language arts and the 5th grade
teachers were general teachers. Each teacher received $5 for filling out the questionnaire.
There were no time limits in completing the questionnaire.
The questionnaire was non~standardizedand was designed by the primary
investigator. The questionnaire contained two questions about writing frequency in each
genre, five questions related to writing instruction, three questions related to the feedback
provided to the students, and ten rating questions to address their students' word
knowledge and genre familiarity in writing.
The writing instruction questions were based on Hayes and Flower's writing
model (1984) and previously reviewed studies about the literate lexicon in speaking and
writing (Curenton & Justice, 2004; Greenhalgh & Strong, 2001; Nippold, Ward-Lonergan
73
& Fanning, 2005). The feedback questions were based on Hayes and colleagues' text
evaluation model and possible problem detection in writing (Hayes, Flower, Schriver,
Stratman & Carey, 1987). The teacher questionnaire was designed to provide exploratory
infoffi1ation about the kinds of feedback teachers provided on students' writing and
teachers' writing instruction in general. The teacher questionnaire is provided in
Appendix C.
Procedures
The two investigators visited the three schools during the academic year. The task
was administrated in several regular writing classes while the teachers were in the
classrooms. One investigator first explained the study to the students and addressed the
importance of their participation. Each student was asked to complete a written narrative
and a written expository essay within one regular class time. Participants were provided
with a one-page instruction sheet and a booklet of four pages of lined paper for each
essay. After participants received the instruction sheet, one of the investigators read the
instructions aloud to the students and answered any questions regarding the study. The
order of the writing tasks was counterbalanced. If the students in the first class wrote the
narrative first followed by the expository essay, the students in the second class wrote the
expository first followed by the narrative essay, to avoid any possible order effects.
At the end of twenty minutes, participants were asked to complete the sentence
they were currently writing. Then, the two investigators collected the instruction sheets
and booklets. The instruction sheets and booklets for the other essay were distributed to
74
the class after all of the first essays were collected. Then the investigators repeated the
same procedures as the first essay. Participants were asked to remain in their seats until
all students had finished the writing tasks. After all participants had completed the
writing tasks, the investigators offered a gift certificate to an ice cream shop for
participating in the study.
Coding and SALT Program
Transcription and SALT Program
A total of 240 essays (120 narrative and 120 expository) were keyboarded
verbatim into the Systematic Analysis of Language Transcripts program (SALT) (Miller
& Chapman, 2002), following SALT conventions. This work was completed by the
primary investigator and two graduate research assistants who were familiar with SALT
procedures. The SALT program is designed to analyze morphemic, syntactic, pragmatic,
and semantic aspects of language in spoken and written language samples (Weston,
Shriberg & Miller, 1989). It is the most widely used tool to process language samples
(Hill, 2001). In order to analyze the data, the transcription was further segmented into
T-units, a common procedure used in the analysis of written and spoken discourse. A
T-unit is defined as an independent (main) clause and any dependent (subordinate)
clauses that are attached to it (Hunt 1965). When an utterance did not qualify as aT-unit,
fragment codes or parentheses were placed. All T-units and fragments were included in
the analysis set in which the total body words were used to analyze literate word use in
the present study.
75
Each T-unit or fragment was further coded for three types of literate words and
different types of clauses such as nominal clause, relative clauses, and adverbial clauses
for the present and future analyses. The primary investigator and two graduate research
assistants followed the coding system of clauses (Mansfield, 2007) and literate words,
and solved any inconsistencies among the coders and investigators. The written language
samples were coded by one coder and double-checked by another coder to detect any
coding errors. Each code was tagged right after each target word and clause and followed
the SALT conventions.
The spelling and spacing errors were corrected in transcription for the SALT
analysis; however, the original spelling and spacing errors were unchanged on the
students' test booklets for possible future analysis. No participant or school names were
used in coding. Each participant was assigned a number, and only chronological age,
gender, and grade were coded for the analysis purpose of the present study. In addition,
any identifying information such as names and school names in the essays were replaced
by codes to protect participants' confidentiality.
Inclusion Criteria for the Literate Lexicon
Three types of literate words were the main focus of the present study; thus, the
inclusion criteria were needed to code the target words accurately and consistently.
Abstract Nouns
Abstract nouns indicate intangible objects unlike concrete nouns, which have
distinct boundaries and stable relations among the components (McGhee-Bidlack, 1991).
An abstract noun often refers to states, events, concepts, feelings, qualities, and things
76
that have no physical existence. Additionally, anything that cannot be perceived through
the five physical senses (as with concrete nouns) is defined as an abstract noun. Examples
of abstract nouns are listed in Table 3.1.
Table 3.1 Coding System for Three Types of Literate Words
Codes of literate lexicon Examples
ABN- abstract nouns freedom, authority, challenge, dignity
DRN-derived nominals happiness (ness), magician (ian), payment (ment),
relationship (ship), creation (tion), maintenance
(ance/ence), personality (ity), director (or/er)
DRA-derived adjectives enjoyable (able), critical (al), beautiful (ful), facial (ial),
childish (ish), creative (ive), careless (less), curly (ly),
bigger (er), biggest (est)
EMCV know, think, guess
Early-developing MCV
EMLV say, tell, ask
Early-developing MLV
LMCV understand, discover, realize, reflect, intend, deserve,
experience, figure out (figure), wonder, pretend, assume,
Later-developing MCV remember, doUbt, infer, hypothesize, perceive, recall,
comprehend, analyze, observe, predict
LMLV explain, assert, concede, conclude, imply, predict, interpret,
Later-developing MLV confirm, argue, persuade, agree, deny, claim, admit,
confess, define, describe, criticize, suggest
Derivatives
According to Nippold and Sun (in press), there was a significant difference in
understanding the meaning of derived nominals and derived adjectives while school-age
children were asked to do a forced-choice derivational morphology task. Therefore, in
order to further understand if there is a difference in understanding and producing
different types of derived words, the current study divided derivatives into derived
nominals and derived adjectives.
77
The inclusion criteria were based on the types of suffixes listed in Nippold's book
(2007, pp. 51-52). In addition, the decomposition strategy was used to analyze possible
derivatives, and the coders followed the rule that the root words must be able to stand
alone without the suffixes. Derived nominals included suffixes that formed from a verb or
an adjective to a noun and gerunds, which is a verbal that ends in -ing and functions as a
noun. For example, there were two derived nominals in the T-unit from a 11 th grader's
narrative essay: " along withjinishing [DRN] this accomplishment [DRN] we would be
looked upon as a hero[ABN]". Derived adjectives included not only the words with
suffixes to form adjectives but also adjectives of emotion/feeling that are formed from the
-ed / -ing forms of verbs. For example, there were two derived adjectives formed from
the verbs, love and want, in the T-unit from a 11th grader's expository essay:
"Friendship[DRN] is important to people because it makes them feel wanted/O[DRA]
and loved/O[DRA] in this world[ABN]". Another example of a T-unit from an 11 th grade
expository essay contained two derived adjectives including one formed from the -ing
form of the verb and one with a suffix to create an adjective: "There are always people to
talk to and have an entertaining [DRA] and enjoyable [DRA] conversation [DRN] with in
the presence[DRN] offriendship[DRN]".
Prefixes, derived verbs, and derived adverbs were excluded from the present
study because prefixes were comparatively easier than suffixes, and derived verbs and
derived adverbs were less frequent than derived nominals and derived adjectives. Some
suffixes used as inclusion criteria in the present study are listed in Table 3.2.
78
Mental State Verbs
Mental state verbs included metacognitive and metalinguistic verbs. According to
the previous literature review, early developing mental state verbs such as think and say
emerge around 4 years old. However, when children advance in linguistic complexity and
cognitive development and expand their vocabulary through schooling and exposure to
print, they have more mental state verbs to describe their thinking and mental state.
Table 3.2 Samples of Suffixes Used to Create Nouns and Adjectives
Suffixes Noun Examples Suffixes Adjective Examples
-age marriage, leakage -able enjoyable, eatable
-ance/ence difference, importance -al identical, critical
-er/or editor, speaker -ent excellent, dependent
-ess actress, waitress -ful beautiful, thankful
-hood childhood, neighborhood -ible reversible, edible
-Ian magician, musician -ial facial, racial
-Ism criticism, racism -ish selfish, reddish
-ist artist, activist -lve active, creative
-ness happiness, sadness -less homeless, hopeless
-ment payment, investment -like childlike,
-ship friendship, relationship -ous nervous, obvious
-tion attraction, creation -some troublesome, burdensome
-th truth, death -y fluffy, scary
(Source: Nippold, M. (2007). Later language development: School-age chIldren,
adolescents, and young adults 3rd ed. Austin: Pro-Ed.)
The metacognitive and metalinguistic verbs listed in Table 3.2 were drawn from
the previous studies (Astington & Olson, 1990; Nippold, Ward-Lonergan & Fanning,
79
2005; Olson, Antonietti, Liverta-Sempio & Marchetti, 2006; Wilson, 1999). In order to
detect qualitative differences in mental state verb use, the present study separated
metacognitive verbs from metalinguistic verbs and further distinguished the early and
later developing metalinguistic and metacognitive verbs. According to the studies
previously reviewed, know, think, guess, say, ask, and tell developed during preschool
(Babu & Mishra, 2000; Booth, Hall, Robinson & Kim, 1997; Johnson & Maratsos, 1977;
Shatz, Wellman & Silber, 1983). Therefore, the present study categorized these mental
state verbs as early developing metacognitive and metalinguistic verbs. Mental state
verbs other than the early developing ones were coded as later developing metacognitive
and metalinguistic verbs. Examples of metacognitive and metalinguistic verbs are listed
in Table 3.2.
Coding/or the Literate Lexicon
The investigators set up a system for coding the three types of literate words in
SALT. The coding system is shown in Table 3.2. Each code was attached after the target
word and followed the SALT program coding system. One word was counted in only one
category. For example, relationship is a derived nominal and an abstract noun, but it was
labeled only as a derived nominal in the present study. Counting one word in one
category avoided the possibility of inflating the numbers of abstract nouns and derived
nominals since most of the derived nominals are abstract nouns such as happiness,
relationship and decision.
80
Interrater Reliability in Coding
A separate investigator, who was familiar with the SALT program but did not
code the transcripts the first time, reviewed the coding of literate words. Ten percent of
the total transcripts were randomly selected for the reliability check. A total of24
transcripts, 4 transcripts in each grade/genre cell, were reviewed. The level of
disagreement between two investigators was 2% for abstract noun, 3% for derivatives,
and 9% for mental state verbs. The disagreements were discussed and a second round of
interrater reliability was conducted. Another 24 transcripts which differed from the 1st
round interrater reliability check, including 4 transcripts in each grade/genre cell, were
randomly selected for the second round of interrater reliability check. The level of
disagreement between the investigators was 1% for abstract noun, less than 1% for
derivatives, and 2% for mental state verbs. All disagreements in coding were resolved
through discussion, yielding 100% agreement.
Data Analysis
The present study aimed to provide both quantitative and qualitative data
regarding how school-age children and adolescents use literate words in two genres of
writing. Additionally, teacher questiolmaires were analyzed to provide supplemental
information about teachers' views and the instruction they provide in genre writing and in
the use of literate words in writing.
81
Quantitative Analysis
The SALT program (Miller & Chapman, 2003) was used to generate the total
number of words (TW), total number of utterances (TTU), mean length of utterance in
words (MLTU), and the frequency counts for each type of literate word. TW was used as
the foundation to compare the use of literate words since participants produced essays of
different lengths. TTU and MLTU were used to detect the quantitative differences that
participants produced in different genres of writing. All T-units and fragments
(incomplete sentences disconnecting from the main clauses) were included in the data
analysis. Thus, each participant had 10 scores including TW, TTU, MLTU, ABN (abstract
nouns), DRN (derived nominals), DRA (derived adjectives), EMCV (early developing
metacognitive verbs), EMLV (early developing metalinguistic verbs), LMCV (later
developing metacognitive verbs), and LMLV (later developing metalinguistic verbs).
Percentages instead of frequency counts were used for each literate word type to
control for differences in the number of utterances that the writers produced (Nippold et
aI, 2005). Seven raw scores including ABN, DRN, DRA, EMCV, EMLV, LMCV, and
LMLV, were entered into the SPSS system (SPSS, 2005) and were transformed to
percentages such that each raw score was divided by the total words (TW) that each
participant produced for each genre. For example, one participant used ABN five times in
narrative writing and the total words he/she produced in narrative writing was 100 words.
Therefore, his/her ABN score in narrative writing was 0.05.
82
Content Analysis
In order to have a full understanding of the subtle changes in the use of literate
words in school age children and adolescents, content analysis was essential. The
differences in word use in writing can be subtle and not represented in statistical analyses.
In the present study, the use of different literate words was listed, categorized, and
organized to further compare word use. All literate words that were used in writing in this
study were recorded.
In content analysis, each abstract noun was recorded only once without repeating
in other age groups or the other genre to discover the difference in abstract noun use. For
example, the word accident was also used in 8th and 11 th graders' writings, but it was
only recorded in the column of abstract nouns used in 5th grade narrative since the use of
abstract nouns was substantial and the space for recording was limited. Derivatives were
additionally analyzed by suffixes categorizing in derived nominal and derived adjective
in order to represent the actual word use under each suffix. The recording of derivatives
was based on the types of suffixes listed in Nippold's book (2007, pp. 51-52). Mental
state verbs were categorized based on the coding system of the present study, which
contained four categories EMCV, EMLV, LMCV, and LMLV All derivatives and mental
state verbs used in writings were listed to show the actual differences of word use in
writings. Content analysis provides valuable information about the actual word use
difference in addition to the statistical analysis. Therefore, utilizing both quantitative and
content analysis offer a more complete picture of how school-age children and
adolescents use literate words in different genres of writing.
83
Teacher Questionnaire Report
The purpose of the teacher questionnaire was to provide supplemental
information about school-age children's writing. Since teacher instruction is related to
children's writing performance and ability (Duin & Graves, 1987; Laflamme, 1997; Zarry,
1999), it was necessary to understand how teachers provided instruction in writing,
including the frequency, genre difference, word instruction, and type of feedback they
provided to students. Four questionnaires were collected in the present study. A summary
of the teachers' responses (two 5th grade, one 8th grade, and one 11th grade) is reported
in the results section.
84
CHAPTERIY
RESULTS
Quantitative Results
A two-way (2x3) mixed analysis of variance (ANOYA) was conducted to analyze
the effects of age and genre on the use of literate words in writing. The independent
variables included one between-subjects variable, age group, with three levels (5th grade,
8th grade, and 11 th grade) and one within-subjects variable, genre, with two levels
(narrative and expository). Eta Square (11 2), which is used to report effect sizes, was also
calculated. The significance level was set at p < .05.
General Comparison between Age Groups and Genres
Total Number ofWords (TW)
In the 5th graders, the mean of the total number of words was 170.68 (SD = 80.26)
in narrative (n = 40) and 139.58 (SD = 60.31) in expository (n = 40). In the 8th graders,
the mean of the total number of words was 245.4 (SD = 83.94) in narrative (n = 40) and
208.4 (SD = 66.08) in expository (n = 40). In the 11th graders, the mean of the total
number of words was 336.73 (SD = 71.71) in narrative (n = 40) and 285.5 (SD = 75.88)
in expository (n = 40).
85
There was a significant main effects for genre, F(l, 117) = 32.15, p < .00, Y]2
= .04 , but no interaction between age group and genre was found, F(2, 117) = .073,p =
0.49. Participants produced more words in narrative than in expository writing. There
was a significant main effects for age group, F(2, 117) = 62.17, P < .00, Y]2 = .42.
Follow-up tests were conducted to evaluate pairwise differences among the means using
a Tukey HSD test (atp < .05). There were significant differences obtained among the
three groups' comparisons. Eleventh graders produced significantly more words overall
in both narrative and expository writing than did 8th graders, and 8th graders produced
significantly more words in both narrative and expository writing than did 5th graders.
Measures of the total number of words in each age group and in the two genres are
reported in Table 4.1.
Table 4.1 Two-Way, Mixed-Effects Analysis ofVariance on TW
Source df F Y]2 P
Between Subjects
Age 2 62.17** 0.42 <.00
Error Between 117
Within Subjects
Genre 1 32.15** 0.04 <.00
Genre * Age 2 0.73 0.00 .49
Error Within 117
Total 239
*p < .05. ** P < .01.
86
Total Number ofUtterances (TTU)
For the 5th graders, the mean of the total number of utterances was 19.10 (SlJ =
8.86) in narrative (n = 40) and 11.53 (SD = 5.02) in expository (n = 40). For the 8th
graders, the mean of the total number of utterances was 23.80 (SD = 9.93) in narrative (n
= 40) and 15.18 (SD = 6.50) in expository (n = 40). For the 11 th graders, the mean of the
total number of utterances was 30.70 (SD = 7.99) in narrative (n = 40) and 22.43 (SD =
7.30) in expository (n = 40).
There was a significant main effects for genre, F(l, 117) = 100.62,p < .00, Y]2
= .17, but no interaction between age group and genre was found, F(2, 117) = 0.14, p =
0.87. Participants produced more utterances in narrative than in expository writing. A
significant main effect was found for age group, F(2, 117) = 31.97, p < .00, Y]2= .22. The
follow-up Tukey HSD test (at p < .05) indicated that significant differences were obtained
among the three groups' comparisons. Eleventh graders produced significantly more
utterances overall in both narrative and expository writing than did 8th graders, and 8th
graders produced significantly more utterances in both narrative and expository writing
than did 5th graders. Measures of the total number of utterances in age groups and genres
are reported in Table 4.2.
Mean Length ofUtterance in Words (MLTU)
For the 5th graders, the mean ofMLTU was 9.14 (SD = 2.22) in narrative (n = 40)
and 12.33 (SD = 3.00) in expository (n = 40). For the 8th graders, the mean ofMLTU was
11.19 (SD = 3.93) in narrative (n = 40) and 14.53 (SD = 3.13) in expository (n = 40). For
87
the 11th graders, the mean of MLTU was 11.27 (SD = 2.07) in narrative (n = 40) and
13.26 (SD = 2.54) in expository (n = 40).
Table 4.2 Two-Way, Mixed-Effects Analysis of Variance on TTU
Source df F 11 2 p
Between Subjects
Age 2 31.97** 0.22 <.00
Error Between 117
Within Subjects
Geme 100.62** 0.17 <.00
Geme * Age 2 0.14 0.00 .87
Error Within 117
Total 239
*p < .05. ** p < .01.
A significant main effect was obtained for geme, F(l, 117) = 75.46,p < .00, 11 2
= .18; however, there was no interaction between age group and geme, F(2, 117) = 1.69,
p = 0.19. Participants produced longer utterances in expository than in narrative writing.
A significant main effect was found for age group, F(2, 11 7) = 9.41, p < .00, rl = .07.
The follow-up Tukey HSD test (at p < .05) showed that there were significant differences
among the three age groups' comparisons. The results revealed that 11 th graders
produced longer utterance in narrative writing than did 8th graders but produced slightly
88
shorter utterances in expository writing than did 8th graders. 8th graders produced longer
utterances in two genres of writing than did 5th graders. Measures of the mean length of
utterance in words in age groups and in the two genres are reported in Table 4.3.
Table 4.3 Two-Way, Mixed-Effects Analysis of Variance on MLTU
Source Df F 11 2 P
Between Subjects
Age 2 9.41 ** 0.07 <.00
Error Between 117
Within Subjects
Genre 1 75.46** 0.18 <.00
Genre * Age 2 1.69 0.01 .19
Error Within 117
Total 239
*p < .05. ** P < .01.
Literate Words Use across Age Groups and Genres
Abstract Nouns (ABN)
For the 5th graders, the mean of the use ofABN was 3.89 (SD = 2.10) in narrative
(n = 40) and 9.15 (SD = 2.87) in expository (n = 40). For the 8th graders, the mean of the
use ofABN was 4.17 (SD = 1.74) in narrative (n = 40) and 8.26 (SD = 2.54) in expository
89
(n = 40). For the 11 th graders, the mean of the use ofABN was 4.95 (SD = 1.59) in
narrative (n = 40) and 9.85 (SD = 1.99) in expository (n = 40).
A significant main effect was found for geme, F(l, 117) = 319.50,p < .00, 11 2
= .53; however, there was no interaction between age group and geme, F(2, 117) c= 1.72,
p = 0.18. Participants used more abstract nouns in expository than in narrative writing.
There was a significant main effect for age group, F(2, 117) = 5.72,p = .004,112 = .02.
The follow-up Tukey HSD test (at p < .05) revealed that there were significant
differences in the use of abstract nouns between the 8th graders and 11 th graders and
between 5th graders and 11th graders, which means that 11 th graders used more abstract
nouns than 8th graders and 5th graders in both narrative and expository writing. However,
the comparison of the use of the abstract nouns between 5th graders and 8th graders was
not significant. Measures of the use ofABN in age groups and gemes are reported in
Table 4.4.
Derived Nominals (DRN)
For the 5th graders, the mean of the use ofDRN was 0.68 (SD = 1.15) in narrative
(n = 40) and 5.40 (SD = 3.86) in expository (n = 40). For the 8th graders, the mean of the
use ofDRN was 1.13 (SD = 0.83) in narrative (n = 40) and 6.50 (SD = 3.14) in expository
(n = 40). For the 11th graders, the mean of the use of DRN was 1.62 (SD = 0.95) in
narrative (n = 40) and 6.14 (SD = 2.24) in expository (n = 40).
A significant main effect was obtained for geme, F(I, 117) = 246.53, p < .00, 112
= .52, but no interaction between age group and geme was found, F(2, 117) = .70,p =
0.501. Participants used more derived nominals in expository than in narrative writing.
90
Table 4.4 Two-Way, Mixed-Effects Analysis of Variance on ABN
Source Df F 11 2 P
Between Subjects
Age 2 5.72** 0.02 <.01
Error Between 117
Within Subjects
Genre 1 319.50** 0.53 <.01
Genre * Age 2 1.72 0.01 .18
Error Within 117
Total 239
*p < .05. ** p < .01.
There was a significant main effect for age group, F(2, 117) = 3.41, p = .04, 11 2
= .01. The follow-up Tukey HSD test (atp < .05) revealed that there was no significant
difference obtained from the comparisons between the three age groups in the use of
derived nominals. Measures of the use ofDRN in age groups and genres are reported on
Table 4.5.
Derived Adjectives (DRA)
For the 5th graders, the mean of the use ofDRA was 1.81 (SD = 1.35) in narrative
(n = 40) and 1.18 (SD = 0.86) in expository (n = 40). For the 8th graders, the mean of the
use ofDRA was 1.97 (SD = 1.32) in narrative (n = 40) and 1.42 (SD = 0.61) in expository
91
(n = 40). For the 11 th graders, the mean of the use ofDRA was 1.88 (SD = 0.96) in
narrative (n = 40) and 1.71 (SD = 0.89) in expository (n = 40).
Table 4.5 Two-Way, Mixed-Effects Analysis of Variance on DRN
Source· df F 112 P
Between Subjects
Age 2 3.41 0.01 .04
Error Between 117
Within Subjects
Genre 1 246.53** 0.52 <.01
Genre * Age 2 0.70 0.00 .50
Error Within 117
Total 239
*p < .05. ** p < .01.
A significant main effect was found for genre, F(I, 117) = 11.77, P < .00, 112 = .05,
but there was no interaction between age group and genre, F(2, 117) = l.11,p = 0.33.
Participants used more derived adjectives in narrative than in expository writing. There
was no significant main effect for age group, F(2, 117) = 1.69, p = .19, which means
there was no difference obtained from the comparisons between the three age groups in
the use of derived adjectives. Measures of the use of DRA in age groups and genres are
reported in Table 4.6.
92
Table 4.6 Two-Way, Mixed-Effects Analysis of Variance on DRA
Source df F 11 2 p
Between Subjects
Age 2 1.69 0.01 .19
Error Between 117
Within Subjects
Genre 11.77** 0.05 <.01
Genre * Age 2 1.11 0.01 .33
Error Within 117
Total 239
*p < .05. ** p < .01.
Early Developing Metacognitive Verbs (EMCV)
For the 5th graders, the mean of the use ofEMCV was 0.69 (SD = 0.78) in
narrative (n = 40) and 1.96 (SD = 1.44) in expository (n = 40). For the 8th graders, the
mean of the use ofEMCV was 0.81 (SD = 0.78) in narrative (n = 40) and 1.90 (SD = 1.02)
in expository (n = 40). For the 11th graders, the mean ofthe use ofEMCV was 0.83 (SD
=0.55) in narrative (n = 40) and 2.18 (SD = 1.06) in expository (n = 40).
A significant main effect was found for genre, F(l, 117) = 94.27, p < .00, 112 = .29,
but there was no interaction between age group and genre, F(2, 117) = 0.37, p = 0.69.
Patiicipants used more early developing metacognitive verbs in expository than in
narrative writing. There was no significant main effect for age group, F(2, 117) = 0.76, p
93
= .47, which means there was no difference obtained from the comparisons between the
three age groups in the use ofEMCV. Measures of the use ofEMCV in age groups and
genres are reported on Table 4.7.
Table 4.7 Two-Way, Mixed-Effects Analysis of Variance on EMCV
Source df F T]2 P
Between Subjects
Age 2 0.76 0.00 .47
Error Between 117
Within Subjects
Genre 1 94.27** 0.29 <.01
Genre * Age 2 0.37 0.00 .69
Error Within 117
Total 239
*p < .05. ** P < .01.
Early Developing Metalinguistic Verbs (EMLV)
For the 5th graders, the mean of the use of EMLV was 1.67 (SD = 2.06) in
narrative (n = 40) and 1.30 (SD = 1.75) in expository (n = 40). For the 8th graders, the
mean of the use of EMLV was 1.41 (SD = 1.67) in narrative (n = 40) and 1.08 (SD = 0.87)
in expository (n = 40). For the 11th graders, the mean of the use of EMLV was 0.85 (SD
=0.82) in narrative (n = 40) and 0.88 (SD = 0.80) in expository (n = 40).
94
There was no significant main effect found for genre, F(1, 117) = 1.49,p = .23,
and the interaction between age group and genre was not significant, F(2, 117) = 0.50, p
= 0.61. There was no significant difference in the use of EMLV between narrative and
expository writing. There was a significant main effect for age group, F(2, 117) = 3.71, p
= .027, 11 2 = .03. The follow-up Tukey HSD test (atp < .05) revealed that there was a
significant difference in the use of EMLV between 5th graders and 11 th graders, which
means that 5th graders used more EMLV than 11 th graders in both narrative and
expository writing. However, the comparisons of the use of EMLV between 5th and 8th
graders and between 8th and 11 th graders were not significant. Measures of the use of
EMLV in age groups and genres are reported on Table 4.8.
Table 4.8 Two-Way, Mjxed-Effects Analysis of Variance on EMLV
Source df F 11 2 P
Between Subjects
Age 2 3.71 * 0.03 .03
Error Between 117
Within Subjects
Genre 1 1.49 0.01 .23
Genre *Age 2 0.50 0.00 .61
Error Within 117
Total 239
*p < .05. ** p < .01.
95
Later Developing Metacognitive Verbs (LMCV)
For the 5th graders, the mean of the use ofLMCV was 0.25 (SD = 0.46) in
narrative (n = 40) and 0.94 (SD = 1.01) in expository (n = 40). For the 8th graders, the
mean of the use ofLMCV was 0.56 (SD = 0.71) in narrative (n = 40) and .81 (SD = 0.88)
in expository (n = 40). For the 11th graders, the mean of the use of LMCV was 0.85 (SD
=0.59) in narrative (n = 40) and 1.06 (SD = 0.76) in expository (n = 40).
A significant main effect was found for genre, F(l, 117) = 17.30, p < .00,112 = .06,
but there was no interaction between age group and genre, F(2, 117) = 2.84,p = 0.06.
Participants used more later developing metacognitive verbs in expository than in
narrative writing. There was a significant main effect for age group, F(2, 117) = 4.47, P
= .013, 112 = .04. The follow-up Tukey HSD test (at p < .05) revealed that there was a
significant difference in the use of LMCV between 5th graders and 11 th graders, which
means that 11th graders used more LMCV than 5th graders in both narrative and
expository writing. However, the comparisons of the use ofLMCV between 5th and 8th
graders and between 8th and 11 th graders were not significant. Measures of the use of
LMCV in age groups and genres are reported in Table 4.9.
Later Developing Metalinguistic Verbs (LMLV)
For the 5th graders, the mean of the use ofLMLV was 0.10 (SD = 0.31) in
narrative (n = 40) and 0.09 (SD = 0.28) in expository (n = 40). For the 8th graders, the
mean of the use ofLMLV was 0.13 (SD = 0.26) in narrative (n = 40) and 0.12 (SD = 0.30)
in expository (n = 40). For the 11th graders, the mean of the use ofLMLV was 0.19 (SD
=0.27) in narrative (n = 40) and 0.15 (SD = 0.25) in expository (n = 40).
96
Table 4.9 Two-Way, Mixed-Effects Analysis of Variance on LMCV
Source df F 112 P
Between Subjects
Age 2 4.47* 0.04 .01
Error Between 117
Within Subjects
Genre 1 17.30** 0.06 <.01
Genre * Age 2 2.84 0.02 .06
Error Within 117
Total 239
*p < .05. ** P < .01.
There was no significant main effect for genre, F(1, 117) = 0.39,p = .53; the
interaction between age group and genre was not significant, F(2, 117) = 0.07,p = 0.93.
There was no significant difference in the use of LMLV between narrative and expository
writing. There was no significant main effect for age group, F(2, 117) = 1.53, P = .22,
which means there was no difference obtained from the comparisons between three age
groups in the use of LMLV. Measures of the use of LMLV in age groups and genres are
reported in Table 4.1 O.
Summary ofQuantitative Results ofLiterate Word Use
A table of descriptive statistics of TW, TTU, MLTU, ABN, DRN, DRA, EMCV,
EMLV, LMCV, and LMLV is reported in Appendix D.
97
Table 4.10 Two-Way, Mixed-Effects Analysis of Variance on LMLV
Source df F 11 2 P
Between Subjects
Age 2 1.53 0.01 .22
Error Between 117
Within Subjects
Genre 1 0.39 0.00 .53
Genre * Age 2 0.07 0.00 .93
Error Within 117
Total 239
*p < .05. ** p < .01.
In addition, in order to provide a clear comparison between the use of literate
words in narrative and expository writing across 5th, 8th, and 11th graders, a summary of
the use of each literate word in the two genres across the three age groups is reported in
Figure 4.1. Three lines represent each grade, the X axis represents the percentages 0 fuse
and the Y axis represents each literate word in both genres.
Content Analysis Results
Word Lists from Two Genres across Three Age Groups
In addition to the quantitative data, the content analysis data provided valuable
information and different insights into the results.
98
Figure 4.1 Percentages of Each Literate Word Use in Two Genres across Three Groups
Each Iiterate word used in two genres
---------
ORA EMCV EMLV LMCV LMLV ABN
expexp
·~5th
----------i . 8th ;
i -,- Ilthl
L __~ __:::J
expexpexpexp
ORN ORA EMCV EMLV LMCV LMLV
expnaffnaffnannaffnaffnaIl'
ORN
naff
ABN
!
i
I
I
~-~--
Abstract Nouns
Some examples of abstract nouns used in narrative and expository writing at the
three grade levels are listed in Table 4.1. Older students tended to use more complex and
advanced abstract nouns in narrative and expository writing compared to younger ones.
In addition, the use of abstract nouns was topic related and genre relevant. The abstract
nouns used in expository writing were less varied than in narrative writing, and the
younger students used less diverse abstract nouns in both narrative and expository writing
compared to the older ones. The examples of abstract nouns used in the two genres of
writing across the three grade levels are listed in Table 4.11.
99
Table 4.11 Examples ofAbstract Nouns Used in Writings across Three Age Groups
Grade 5 narrative Grade 5 expository =1
accident, carnival, chill, contact, dream, advice, answer, apology, category, cause,
fight, friend, fun, mistake, prank, problem choice, culture, drama, enemy, gift goal,
section, sense, shock, story, subject, gossip, heritage, history, interest, lie
sudden, time, trick, trouble, try, way, word lifestyle, moment, nature, paragraph,
problem, reason, recess, religion, rule,
science, secret, suicide, trick, trouble, turn
vibe
Grade 8 narrative Grade 8 expository
abuse, anniversary, attack, balance, basics, adventure, aspect, attitude, basis, belief,
blast, bond, community, control, distance, breakdown, company, downfall, effect,
effort, errand, lesson, nature, offense, encounter, gossip, insult, interlude,
outcome, peace, point, practice, reply, memory, mind, moment, mood, pride,
slope, support, surprise, suspense, torture, quality, thought, trait, trust, weight
view, wit
Grade 11 narrative Grade 11 expository
anatomy, antic, attempt, chaos, character, acquaintance, adventure, apology, aspect,
curfew, disaster, disuse, divorce, force, attempt, aura, burden, challenge, character,
issue, journey, mercy, moral, motive, choice, circumstance, comment,
multitude, mystique, neglect, objective, compassion, crisis, culture, desire, impact,
occasion, ordinary, practice, prank, insult, knowledge, leisure, moral, obstacle,
process, promise, proof, relief, result, outcome, remark, remedy, research,
revenge, spectrum, tissue, traction, respect, struggle, unity, value
triumph, twist, unison, venture, wreck
Derived Nominals
Older students used more varied suffixes in both narrative and expository writing'
than younger ones. In addition, older students used more advanced and complex derived
nominals in their writings. For example, 11th graders used accomplishment,
acknowledgement, encouragement, and resentment as derived words with the
suffix -ment, but 8th graders only used excitement as the -ment suffix derived nominal.
In other words, although the three age groups used the same suffix in their writing, older
students used more complex and diverse derived words than younger ones.
100
Based on the types of suffixes used in the two essays, participants used different
types of suffixes in the two genres; however, there was no specific trend in using certain
suffixes in specific genres. Similar to the abstract nouns, the use of derived nominals was
topic dependent and genre related. Examples of derived nominals used in the two genres
of writing across the three grade levels are listed in Table 4.12.
Table 4.12 Examples of Derived Nominals Used in Writing across Three Age Groups
Grade 5 narrative Grade 5 expository
af: festival er: teacher, partner
ence: silence ing: feeling, liking, cheating, socializing
er: runner, stroller, skater ship: relationship, friendship, hardship
ing: camping, shopping, setting tion/ion: action
ness: darkness th: truth
th: death
tion/ion: vacation, admission
Grade 8 narrative Grade 8 expository
--
af: arrival af: betrayal
ance: entrance ance/ence: difference, confidence,
can: meXIcan experience, influence, acceptance
cy: emergency, pregnancy er/or: scooter, killer, counselor
ee: employee ing: saying, feeling, understanding
er/or: manager, instructor, investigator, ist: racist
ing: beginning, disputing, screaming ite: opposite
ist: guitarist, terrorist ity: personality, profanity, popularity
istic: characteristic ness: loneliness
ite: favorite ship: friendship, relationship
ity: opportunity ty: honesty, property, (dis)loyalty
ment: excitement th: truth, death
ship: friendship tion/ion: tension, opinion" confrontation,
th: youth imagination, connection, communication
tion/ion: inspection, companion
ure: creature
Grade 11 narrative Grade 11 expository
age: leverage age: marriage
af: arrival, festival af: material, betrayal
ance/ence: experience, importance ance/ence: acceptance, (re)assurance,
ant: occupant perseverance, acquaintance, variance
dom: boredom
erlor: murderer, thrower, scavenger,
laughter, director, creator
hood: childhood, neighborhood
ing: feeling, surrounding
ity: reality, opportunity, curiosity,
responsibility
ment: acknowledgement, accomplishment,
equipment, tournament
ness: loneliness, listlessness, staleness
ry: rivalry
th: youth, strength, death
tion/ion: abrasion, generation,
competition, decoration, anticipation,
destination, protection, connection
ure: creature
y: recovery
Derived Adjectives
101
ency: sufficiency
erlor: laughter, applauder, factor, crusher
hood: neighborhood
ian: guardian
ible: variable
ing: (mis)understanding, belonging
itylty: unity, difficulty, opportunity,
flexibility, popularity, variety
ment: accomplishment, commitment,
encouragement, excitement, resentment
ness: willingness, bitterness, faithfulness
ship: hardship, friendship, relationship,
companionship
I
th: strength, death
tion/ion: dedication, definition,
interaction, introduction, obligation,
y: jealousy
Similar to the use of derived nominals, older students used more complex and
varied derived adjectives compared to younger ones. In addition, 8th graders and 11 th
graders tended to use more different types of suffixes in narrative than in expository
essays. It was noteworthy that 11th graders used multi-derivative words that had
undergone two or more derivations from their roots in their expository essays, such as
unbreakable, impenetrable, and inseparable.
In general, participants utilized certain suffixes more frequently across the three
age groups, including comparative er, superlative est, -ful, -y, -ly, and -al. The examples
of derived adjectives used in the two genres across the three age groups are listed in Table
4.13.
102
Table 4.13 Examples of Derived Adjectives Used in Writings across Three Age Groups
Grade 5 narrative
al: fictional, nocturnal, tropical
Jul: beautiful
ible: terrible
ing: wrapping, fishing, freezing
ly: friendly
ous: nervous
y: dirty, freaky, scary, bouncy
comparative er: later
superlative est: biggest, funniest, loudest
Grade 8 narrative
able: unbelievable, unthinkable
al: emotional, burial
ant: hesitant
ary: elementary
ed: stunned, dazed, designated, annoyed
Jul: wonderful, thankful
ible: responsible
ing: upsetting, rewarding, tempting
ish: blackish, girlish
ive: sensitive
less: speechless, harmless
ous: cautious
some: troublesome
y: freaky, muddy, squeaky, handy, trusty
comparative er: closer, happier, slower
superlative est: biggest, largest, bravest
Grade 11 narrative
Grade 5 expository
able: enjoyable, valuable
al: personal, delusional
Jul: respectful, thoughtful
ish: selfish
ly: lonely
some: lonesome
y: bloody, rainy, dirty
comparative er: easier, funnier, closer
superlative est: greatest
Grade 8 expository
able: miserable, valuable, (un)comfortable
al: serial, personal
ed: bored, depressed, strained
ent: different
Jul: respectful, trustful, hurtful
ing: boring, lasting, standing
ite: favorite
ive: active, positive
ly: lonely, daily, (un)friendly
ous: prevIOUS
y: lucky, funny
comparative er: quieter
superlative est: wildest, greatest, smartest
Grade 11 expository
able: unbreakable, enjoyable, bearable,
sociable, impenetrable, inseparable
al: social, emotional, ethical, essential
Jul: respectful, stressful, truthful, doubtful
ive: supportive
less: harmless, endless, selfless, careless
ly: friendly, lonely
y: goofy, healthy
comparative er: happier
superlative est: dullest
able: valuable, comfortable
al: hysterical, skeptical
ar: particular
ed: scared, occupied
Jul: successful, fateful, powerful
ible: responsible, terrible
ic: athletic
ing: amusing, clipping, freezing
ish: skittish
ive: intensive, consecutive, creative
less: fruitless, breathless, harmless
ly: sickly
ous: glorious, humorous, monstrous I I
some: irksome~----
103
y: sneaky, spooky, cheesy, yellowy,
creepy, blurry
I
comparative er: dirtier
superlative est: scariest
Mental State Verbs
Across all three age groups and both gemes, participants used a wide variety of
early developing metalinguistic verbs (EMLV) such as say, talk, tell, yell, ask, call,
scream, whisper, lie and answer and similar early developing metacognitive verbs
(EMCV) such as want, hope, love,jeel, think, know, guess, need, hate, care, wish, and
give up. In the use oflater developing metalinguistic verbs (LMLV), older students used a
slightly greater variety ofLMLV in both narrative and expository writing such as confide,
recite, encourage, describe, discuss, and define. Overall, the later developing
metalinguistic verbs used by the three groups appeared homogeneous and were similar in
the two gemes.
The three age groups used more varied later developing metacognitive verbs
(LMCV) in the two gemes compared to the use of later developing metalinguistic verbs.
Additionally, the three age groups used similar LMCV in the two gemes; however, older
students used a slightly greater variety ofLMCV in both genres such as disown, shape,
digest, attempt, determine, and cherish. Although older students used a slightly greater
variety of later developing mental state verbs, the mental state verbs used in the three age
groups were similar and less varied. In general, understand, realize, remember,figure
(out), and believe were commonly used later developing metacognitive verbs and agree
and explain were frequently used later developing metalinguistic verbs in the two gemes
104
across the three age groups. The later developing metacognitive and metalinguistic verbs
used in participants' writings are listed in Table 4.14.
Table 4.14 Later Developing Metacognitive and Metalinguistic Verbs Used in Writings
across Three Age Groups
Grade 5 narrative Grade 5 expository
LMLV: mention, blurt, swear, agree, admit LMLV: agree, disagree, explain, express,
LMCV: realize, understand, doubt, dare, greet. mention
believe, ignore, find out, forget, learn, LMCV: understand, trust, respect, count
expect, wonder, decide, picture on, learn, enjoy, share, depend on,
appreciate, believe, cheer up, rely on,
comfort, remind, decide, hope support,
find out, ignore, betray
Grade 8 narrative Grade 8 expository
LMLV: explain, agree. confess, deny, LMLV: agree, admit, confide, express,
reply, question, blame, introduce, refer argue, blame, gossip, discuss
LMCV: figure, realize, figure out, LMCV: understand, remember, figure out
remember, decide, dare, believe, trust, doubt, cheer up, trust, share, enjoy, ignore,
enjoy, determine, discover, notice, expect, respect, figure out, believe, support,
ignore, plan, forget, share, hope, count on, forget, judge, mistake, consider, cherish,
learn, find out wonder, count on, rely on
Grade 11 narrative Grade 11 expository
LMLV: explain, suggest, question, reply, LMLV: explain, argue, describe, express,
mention, rumor, praise, inform, confide, define, apologize, confide, state, mention,
recite, discuss, apologize, discourage encourage, reply
LMCV: realize, figure out, notice, assume LMCV: understand, remember, figure out,
remember, understand, attempt, learn, realize, share, betray, enjoy, trust, imagine,
imagine, pretend, decide, suppress, forget, judge, believe, learn, commit, respect, rely
plan, hope, believe, discover, ignore, find on, defy, consider, ignore, appreciate,
out, expect, shape, wonder, digest, refuse, depend on, count on, cheer up, support,
reflect forget, remind, shape, find out, overlook,
disown, treasure
Questionnaire Results
Teachers of each grade whose students participated in the present study, including
two from fifth grade, one from eighth grade, and one from eleventh grade, agreed to fill
out a 6-page questionnaire during or after the writing task. The teachers of 8th and 11 th
105
grade students were instructors in language arts. The 5th grade teachers were general
education teachers. All teachers were instructors of each grade's writing class. The
questionnaire results are reported in the following subcategories.
Frequency
Fifth grade participants were asked to write a narrative essay once every two
weeks, and an expository essay once every two weeks to once a month. The 8th grade
teacher reported a similar frequency with once every two weeks for students to write a
narrative and an expository essay. However, the 11 th grade participants were asked to
write a narrative and an expository essay once per term (every three months).
Instruction
There were five questions in the instruction section. All teachers reported that the
most common type of instruction they provided for students' writing was planning,
including idea generation, organization, and goal setting. In addition, they gave students
vocabulary instruction, especially the vocabulary related to the topic and conjunctions
such as while, because, and however, before writing according to four teachers' reports.
One 5th grade teacher and the 8th grade teacher indicated that vocabulary related
to the topic and conjunctions was important for students to know before they wrote their
essays. The other 5th grade teacher believed that conjunctions and adverbs such as
extremely and suddenly were critical in students' writing, and the 11 th grade teacher
pointed out that vocabulary related to the topic and metacognitive and metalinguistic
verbs were essential for students to know before writing.
106
The last question concerned the method that was used to teach students to
distinguish the difference between gemes. One 5th grade teacher and the 11 th grade
teacher reported that they provided lectures on differences between gemes. Class
activities were reported by the other 5th grade teacher, and writing samples were the most
commonly used method to teach students the geme differences for the 8th grade teacher.
Types ofFeedback
There were three questions related to the feedback that the teachers provided to
students about their writing. All four teachers reported that spelling, grammar, and
organization were the comments that were often provided to students. Three out of four
teachers indicated that their students experienced difficulties in spelling, grammar, and
organization when writing a narrative and expository essay, and the 8th grade teacher
pointed out that spelling, faulty text structure, and organization were difficulties in 8th
graders' narrative and expository writing.
Word Knowledge and Genre Ratings
Ten rating questions related to students' geme and word knowledge while
composing narrative and expository essays were answered by teachers based on a
five-point Likert scale including strongly disagree, moderately disagree, neutral,
moderately agree, and strongly agree. Three out of four teachers moderately agreed that
word use is essential for students to compose a narrative or expository essay, but one
teacher strongly disagreed. Three out of four teachers moderately agreed that their
students have adequate word knowledge to write good narrative or expository essays;
however, one teacher strongly disagreed.
107
In the rating questions of vocabulary instruction in writing, one teacher strongly
agreed, two teachers moderately agreed, and one teacher was neutral that their students
need vocabulary instruction to complete a good narrative or expository essay. Two
teachers strongly agreed, one teacher moderately agreed, and one teacher was neutral
about providing sufficient vocabulary instruction to their students to complete an essay.
In the rating questions of students' geme knowledge and writing ability, two
teachers strongly agreed and two teachers moderately agreed that their students
understand the difference between writing a narrative and an expository essay. Two
teachers strongly agreed, one teacher moderately agreed and one teacher was neutral
about providing sufficient practice for students in writing two gemes of essays.
In the rating questions of difficulties in composing two gemes of essays, one teacher
strongly disagreed, two teachers moderately disagreed, but one teacher moderately agreed
that their students had difficulties composing a narrative essay. However, two teachers
moderately agreed, one teacher moderately disagreed, and one teacher was neutral about
their students having difficulties composing an expository essay. In the rating questions
of using words in composing two gemes, three teachers moderately disagreed and one
teacher strongly disagreed that their students had difficulties in using appropriate words
to write a narrative essay. On the other hand, two teachers moderately disagreed, one
teacher moderately agreed, and one teacher was neutral about their students having
difficulties using appropriate words to write an expository essay. The report of teacher
questionnaire was summarized in Table 4.15.
108
Table 4.15 Summaries of Teacher Questionnaire Report
Frequency
Instruction
Feedback
Rating
(Only
focused on
narrative and
expository
essays)
Narrative: Once every two weeks (3) - Once a month (1)
Expository: Once every two weeks (3) - Once a month (1)
Writing: planning (idea generation, organization and goal setting) (4).
Vocabulary instruction commonly provided: vocabulary reacted to the
topic and conjunction (4).
Vocabulary important to know: vocabulary related to topic (3),
conjunctions (3), Adverbs (1), mental state verbs (1).
Teaching method used to distinguish genre differences: writing samples
(1), class activities (1), lectures (2).
Commonly provided to students: spelling, grammar, disorganization
Students 'difficulties in narrative writing: spelling (4), grammar (3),
disorganization (4), faulty text structure (1).
Students 'difficulties in expository writing: spelling (4), grammar (3),
disorganization (4), faulty logic and inconsistencies (l).
1. Word use is essential in writing:
Strongly agree (1), Moderately agree (2), Strongly disagree (1).
2. My Students have adequate word knowledge to write a good essay:
Moderately agree (3), Strongly disagree (1).
3. My students need vocabulary instruction to write a good essay:
Strongly agree (1), Moderately agree (2), Neutral (1).
4. Provide sufficient vocabulary instruction for students to write a good
essay:
Strongly agree (2), Moderately agree (l), Neutral (1).
5. My students understand the difference between two genres:
Strongly agree (2), Moderately agree (2).
6. Provide sufficient practice for students to write these two types of
essay:
Strongly agree (2), Moderately agree (1), Neutral (1).
7. My students have difficulties composing a narrative essay:
Moderately agree (1), Moderately disagree (2), Strongly disagree (1).
8. My students have difficulties composing an expository essay:
Moderately agree (2), Neutral (1), Moderately disagree (1).
9. My students have difficulties using appropriate words while writing a
narrative essay:
Moderately disagree (3), Strongly disagree (1).
10. My students have difficulties using appropriate words while writing an
expository essay:
Moderately agree (1), Neutral (1), Moderately disagree (2).
109
CHAPTER V
DISCUSSION
The purpose of the present study was to examine how children used literate words
including abstract nouns, mental state verbs, and derivatives in two different writing
genres, narrative and expository. The first aim was to evaluate the use of three types of
literate words across the three groups. Second, the use of three types ofliterate words was
compared between narrative and expository writing. Third, content analysis was used to
represent the literate word use; in this way, subtle differences in literate word use in two
genres and three age groups were detected in addition to quantitative analyses. It was
predicted that older children would use more abstract nouns, mental state verbs, and
derivatives than younger children. In addition, it was predicted that all three age groups
would use more abstract nouns, mental state verbs, and derivatives in expository writing
than in narrative writing. Lastly, it was predicted that older children would use more
advanced abstract nouns, later developing mental state verbs, and complex derivatives,
and than younger children, and more so in expository writing than in narrative writing.
110
Analysis of Writing Productivity in Age Groups and Gemes
Writing productivity was analyzed in terms of total number of words (TW), total
number of utterances (TTU), and mean length of utterance in words (MLTU). Past
research had indicated that older children produced more and longer utterances in writing
than younger ones (Nippold, 2007; Nippold, Ward-Lonergan & Fanning, 2005). Given
cognitive and linguistic development, it was expected that older children in the present
study would write more and generate longer sentences in their writings. Not surprisingly,
the results of the present study indicated that older children wrote more and generated
longer utterance than younger children, except that 8th graders (M= 14.53, SD = 3.13)
produced slightly longer utterances in expository writing than did 11 th graders (M = 13.26,
SD= 2.54).
The explanation of this unexpected result in MLTU might be that development in
writing is not a straight line with the increase of age. Since the present study only focused
on the literate word use in writing, syntactic structure was not examined in this study.
Eighth graders and 11 th graders produced similar length of utterance in words, but longer
utterances did not necessarily equal to more complex syntactic structures. Therefore,
more studies are needed to examine the complexity of syntactic structure between 8th and
11th grader's writings. Additionally, it is critical to rule out that the production of similar
length of utterance between 8th and 11 th grades in the present study was a single case or
there was a pattern of drop off in writing productivity in high school.
111
Literate Word Use across Three Age Groups
Abstract Nouns
Abstract nouns were used more by older children in narrative writing but not
necessarily in expository writing. In addition, the significant differences in the use of
abstract nouns were found in the comparison between 8th graders and 11th graders and
between 5th and 11th graders. According to Nippold's statement (2007), abstract thought
is one of the differences between early and later language development. Since the use of
abstract nouns reflects abstract thinking, it is expected that the use of abstract nouns will
increase between childhood and adulthood (Nippold et aI, 2005). Because the word
learning process heavily focuses on abstract concepts in adolescents and adults, older
children have a larger repertoire of abstract nouns. Therefore, the finding that the use of
abstract nouns increased from 5th grade to 11 th grade supported the development of
abstract thinking although not all of the results reached significant levels. It was
surprising to find that 5th graders used slightly more abstract nouns in expository writing
than did 8th graders.
Although the changes in quantity might be subtle, changes in quality of language
use show differences (Nippold, 2007; Nippold, Mansfield & Billow, 2007; Scott, 1988).
Older students tended to use slightly more complex and advanced abstract nouns in
narrative and expository writing compared to younger ones. For example, 5th graders
used accident, mistake, problem, trouble, advice, heritage, history, interest; 8th graders
used offense, outcome, suspense, torture, wit, adventure, breakdown, downfall; and 11th
graders used chaos, multitude, mystique, ordinary, spectrum, triumph, acquaintance,
112
circumstance, obstacle. Although the use of abstract nouns was heavily topic dependent,
it was clear that older children used more complex abstract nouns compared to younger
ones. Moreover, the development in linguistic and cognitive ability and exposure to more
schooling contribute a larger repertoire of abstract nouns in older children.
Derivatives - Derived Nominals and Derived Adjectives
According to research, children develop knowledge of derivational morphology
between 4th grade and high school, and they make major gains between 4th and 8th
grades (Nippold, 2007; Nagy, Diakidoy, Anderson, 1993; Wysocki & Jenkins, 1987).
Therefore, it was expected that older children would use more derivatives in their writing.
Older children used slightly more derived nominals in both genres of writing except 8th
graders used slightly more derived nominals than 11 th graders in expository writing. In
the use of derived adjectives, older children used more derived adjectives than younger
ones, except 8th graders used slightly more derived adjectives than 11th graders in
narrative writing. Children increase in the use of derived nominals and derived adjectives
with age and schooling (Nippold, 2007); however, none of the group comparisons
reached significant levels in the present study. The present study partially supported the
finding in the Ravid study (2006) that the use of derived nominals increased with age and
schooling without demonstrating a significant difference between age groups in the
present study.
Because of the abstractness of the derivatives, available knowledge of word
structure and metalinguistic competence, children learn derivatives relatively late
(Nippold, 2007; Nippold & Sun, in press). Although children gain most of the derived
113
word knowledge during the school-age years, it is not necessarily shown in their writing.
In addition, the low use of derivatives in writing might not be able to detect the
age-related differences. Avoiding of spelling errors, general literacy skills, and overall
vocabulary knowledge might contribute to the low use of derivatives in the present study.
When children have sufficient implicit knowledge, that knowledge is reflected in
explicit production (Rubin, Patterson & Kantor, 1991). The implicit knowledge of
derivational morphology of participants was not obtained in the present study; therefore,
the relationship between comprehension and production of derivatives could not be
confirmed. Therefore, present study could not answer questions concerning whether
participants had adequate derivational knowledge to produce a large number of derived
words in writing, if they used alternative words to replace derivatives to avoid spelling
errors, or if their teachers did not emphasize the use of derivatives. However, these
concerns all possibly contribute to the low use of derivatives.
Although the significant difference in age groups was only found in derived
nominals, 5th graders did use less varied and limited suffixes for derived nominals and
adjectives in both written genres. However, 8th graders and 11 th graders used similar and
widely varied suffixes for derived nominals and adjectives in both written genres, which
supported the view that children increase their knowledge of derivational morphology
dramatically between 4th and 8th grade (Carlisle, 1988). According to White and
colleagues (1989), the suffixes that 5th graders used as derived nominals and adjectives
were mostly high frequency suffixes in printed school English for grades 3-9 including
-er, -ion/tion, ness, ing, -al, -ible/able, -ing, -ly, and -yo In contrast, 8th and 11th graders
114
used a wider range of suffixes including high and low frequency suffixes in both derived
nominals and derived adjectives.
It is noteworthy that 5th graders tended to use neutral suffixes such as -ness, -er,
-able, -ly, -ing, which do not involve changing the vowel quality, stress, or phonological
characteristics and are transparent and acquired earlier (Tyler & Nagy, 1989). On the
other hand, 8th and 11th graders used both neutral and non-neutral suffixes and made use
of more complex and advanced derivatives in both genres. It is not surprising to observe a
wide range of suffix use in 8th and 11th graders, since children show gradual
improvements in derivational morphology from 3rd or 4th grade through high school and
college due to increased exposure to complex words in literature, textbooks, and lectures.
Furthermore, the use of morphologically complex words in textbooks increases greatly
beginning in 3rd grade (Nippold, 2007). This also contributes to the learning of
derivatives. Some 8th and 11th graders used multi-derivatives where the derivative has
undergone two or more derivations from its root such as reassurance, misunderstanding,
companionship, unbelievable, uncomfortable, unfriendly, impenetrable and inseparable.
The content analysis provided strong evidence that older children have more knowledge
of derivational morphology and are able to use morphologically complex words explicitly
and appropriately in writing.
Mental State Verbs - Metacognitive and Metalinguistic Verbs
Statistically significant age-group differences were found in the use of early
metalinguistic (EMLV) and later metacognitive verbs (LMCV) in the present study.
Although children showed an increase in their use of early developing metacognitive
------ .._- ----- ._--
115
verbs (EMCV), later developing metacognitive verbs (LMCV), and later developing
metalinguistic verbs (LMLV) with age, only the comparison between 5th and 11 th
graders in the use of LMCV reached a significant level. It was interesting to find that
older children used fewer early developing metalinguistic verbs (EMLV) in both written
genres and that 5th graders used significantly more EMLV in both genres than did 11th
graders. Given the low production of all mental state verbs, children in the present study
did not use this type of literate word often in their writing. Moreover, LMCV and LMLV
were rarely used in all three age groups. Therefore, the findings indicated that the use of
mental state verbs in writing increased with age but that the use of these verbs remained
low even in older students, consistent with what Nippold et ai, 2005) reported.
The increased use ofEMCV, LMCV, and LMLV indicated that knowledge of
metalinguistic and metacognitive verbs increased with age, improved greatly between 5th
and 10th grade and continued to develop into adolescence and adulthood (Astington &
Olson, 1990; Booth & Hall, 1994). Wing and Scholnick (1986) claimed that gains in
information processing load allow people to manage several meanings at once. Therefore,
with development in linguistics and cognition, children are able to process multiple
meanings, to reason, and to think critically to fulfill their communication purposes.
Across all three age groups and both genres, participants used similar early
developing metalinguistic verbs (EMLV) such as say, talk, tell, yell, ask, call, scream,
whisper, lie and answer and early developing metacognitive verbs (EMCV) such as want,
hope, love,jeel, think, know, guess, need, hate, care, Wish, and give up. The declining use
of EMLV indicated that younger children used more early metalinguistic verbs in writing
116
than did older children. It was not surprising to find that the use of EMLV declined from
5th graders to 11 th graders, because older children were expected to use more later
developing mental state verbs in writing. Substantial use of conversational dialogue and
quotations were commonly observed in younger children's writings and possibly
contributed to the use of EMLV in their writings. In contrast, older children used more
EMCV than younger children and the possible explanation may be that there was a
different mechanism underlying the use ofEMCV and EMLV, since these two types of
mental state verbs served different language functions (Olson & Astington, 1993).
Regarding later developing metalinguistic verbs (LMLV), older students used a
slightly greater variety of LMLV such as confide, recite, encourage, describe, discuss and
define, discourage, and confess and a greater variety ofLMCV such as assume, attempt,
suppress, reflect, overlook, disown, and treasure. Fifth graders used less varied and fewer
later developing mental state verbs compared to 8th and 11th graders; however, 8th and
11 th graders used similar later developing mental state verbs. Schwanenflugel and
colleagues (1996) claimed that the acquisition of mental state verbs might be different
depending on their conceptual difficulty, level of uncertainty, and abstractness. The
similar use of later developing mental state verbs in 8th and 11 th graders suggested that
the participants in the present study might have similar knowledge of later developing
mental state verbs although 11th graders used a slightly greater variety of later
developing mental state verbs in writing. According to Booth and Hall (1994), some of
the later developing mental state verbs are more difficult and acquired later than others,
and even college students might have incomplete knowledge about some complex mental
117
state verbs. Therefore, the present study provided evidence that later developing mental
state verbs continue to develop into adulthood and that growth is a gradual process.
Literate Word Use between Narrative and Expository Writing
Abstract Nouns
Not surprisingly, children used more abstract nouns in expository than narrative
writing in all three age groups. The results supported the findings ofRavid (2006, 2004)
and Berman and Nir-Sagiv (2007) in showing that writers used more concrete/imageable
nouns in narrative writing, but used more abstract and low frequency nouns in expository
writing. Expository texts usually contain factual or teclmical information such as
cause-effect explanations and procedural directions (Hadley, 1998), having a
non-temporal, logically-based, and argumentative structure (Ragnarsdottir, Aparici,
Cahana-Amitay, van Hell, & Viguie, 2002). Therefore, the use of abstract nouns reflects
the writer's abstract thinking while composing an expository essay.
According to Paivio's dual coding theory (1991), concrete information activates
both verbal and mental image systems in the brain. Since narrative is personal and
empirical, a mental image can be activated during writing. In this study, the topic of
narrative writing was "What happened one day"; therefore, participants wrote a personal
or imaginary story with a possible vivid image, such as a trip, an accident, or a party. In
order to describe the images, writers used concrete and imageable words to convey the
story. In contrast, it is difficult to generate an image during expository writing since it
involves abstract thought, reasoning, and logical thinking. From the neuro-imaging
studies to the word definition tasks (Jessen et aI, 2000; McGhee-Bidlack, 1991; Nippold,
118
1999; Nippold, Hegel, Sohlberg, & Schwarz, 1999; Noppeney & Price, 2004; Sabsevitz
et aI, 2005; Swaab et aI, 2002), it is evident that abstract words are difficult and take a
longer time to process and acquire, which results from the non-activating mental-imaging
system. Therefore, dual coding theory provided a fundamental explanation for the
different use of abstract nouns between narrative and expository writing.
Children in the present study used slightly less varied abstract nouns in expository
than in narrative writing and the use of abstract nouns were topic and genre related.
Because the title of the expository essay was "Friendship" in the present study, children
tended to use homogeneous abstract nouns such as friend, moment, company, belie}; and
memory. Since the title of the narrative writing varied depending on each writer's story,
abstract noun use varied greatly from topic to topic such as abuse, anatomy, anniversary,
tissue, and accident.
Derivatives - Derived Nominals and Derived Adjectives
In the present study, children used significantly more derived nominals in
expository than in narrative writing, but used significantly more derived adjectives in
narrative than in expository writing. It was noteworthy that children used fewer derived
adjectives in both written genres and fewer derived nominals in narrative writing
compared to the use of abstract nouns; however, the use of derived nominals in
expository writing was numerous. This supported the finding in the Ravid study (2006)
that derived nominals were used greatly in written expository text. One of the
explanations is that the nature of expository discourse leads to the frequent use of derived
nominals. Derivatives are more abstract and are commonly seen in written and formal
119
language (Nagy, Diakidoy, Anderson, 1993; Plag, Dalton-Puffer, & Bayyen, 1999).
Similar to the use of abstract nouns, expository discourse stimulates abstract thinking and
advanced cognitive and linguistic skills that are essential to express the writer's views.
In contrast, narratives are about people engaged in events, including personal
experiences, fictional narratives/stories, and film/book/TV program summaries (Hadley,
1998; Scott, 1988). Derived adjectives act like adjectives to describe or modify a noun or
pronoun, giving more information about what the noun or pronoun refers to. Therefore,
writers tended to use derived adjectives to describe the settings, characters, and plots in
order to provide the audience a complete and vivid picture while reading the story.
Although the comparison in the use of derived nominals and adjectives between two
genres reached significance, there was no specific trend in using certain suffixes in
specific genres based on the content analysis. Moreover, children used less varied
suffixes in expository text than in narrative text. One possible explanation of slightly less
variety of suffixes in derived nominals and adjectives is that the title of the expository
writing, "The Nature of Friendship," narrowed the variety of suffix use and led to similar
and homogeneous use of derivatives in expository writing. In general, similar to the
abstract nouns, the use of derived nominals was topic dependent and genre related, and
no specific trend of using certain suffixes in certain genres was observed.
Mental State Verbs - Metacognitive and Metalinguistic Verbs
It was unexpected that significant differences between two genres were only
found in the use of early and later developing metacognitive verbs, but no difference was
found in the use of early and later developing metalinguistic verbs. According to the
120
results, children used more EMCV and LMCV in expository than in narrative writing.
While writing an expository essay, a writer needs to use appropriate reasoning and logical
thinking to justify his/her standpoint and argument. Since metacognitive verbs are talking
about cognition (Olson & Astington, 1993) and are related to the development of theory
of mind (Astington & Jenkins, 1999; Booth et aI, 1997; Miller, 2004), increasing use of
early and later developing metacognitive verbs strongly reflects a writer's thought
processes, opinions, and ways of thinking about the world.
Given the extremely low use ofLMLV, this type of mental state verb was difficult
and not commonly used in writing in the present study. However, the possibility of
limited knowledge of LMLV or other interfering factors resulting in the extreme low use
ofLMLV could not be answered in the present study. The use of EMLV in two genres
was similar, which means that children did not use this type of literate word specifically
in certain written genres.
There was no pattern observed in the use of certain mental state verbs in specific
written genres except 5th graders used less varied LMCV and LMLV in narrative text.
Similar to the abstract nouns and derivatives, the use of mental state verbs was topic
dependent and genre related and no specific trend of using certain suffixes in certain
genres was observed.
Vocabulary Instruction in Writing
Studies examining the relationship between vocabulary instruction and students'
writing performance mainly focused on teaching content and topic related words instead
---------_. -_._------_. _.
121
of the later acquired and complex literate words (Duin & Graves, 1987; Laflamme, 1997;
Zarry, 1999). It is critical to expand the breadth and depth of students' vocabulary
because it has a great influence on the descriptiveness, accuracy, and quality of their
writing. However, in addition to the content words, literate words are critical in
composing a text, especially for school-age children and adolescents to be able to express
their thoughts, ideas, feelings, and opinions. Vocabulary items that school-age children
encounter are more abstract, complex, longer, contain multiple affixes, and are used to
reflect internal states (Nippold, 2007; Ravid, 2004; Westby, 1990). Therefore, use of an
advanced lexicon reflects later language and cognitive development.
Although a child's lexicon grows with age in different ways without direct
instruction (Cain, Oakhill, & Elbro, 2003; Cain, Oakhill, & Lemmon, 2004; Chaffin,
Morris, & Seely, 2001; Dockrell & Messer, 2004; Larsen & Nippold, 2007; Nippold,
2007; Oetting, 1999;Oetting, Rice, & Swank, 1995; O'Hara & Johnston, 1997; Rice,
Cleave, & Oetting, 2000), effective instruction in vocabulary is necessary. In addition to
reading, writing plays an important role in learning and academic success and is viewed
as schooled language competence (Perfetti & McCutchen, 1987). Thus, how children use
advanced vocabulary and literate words to boost their writing quality and express their
thoughts and feelings appropriately is an important issue in teachers' instruction design.
According to the teacher questionnaires, the most common type of instruction
they provided for students' writing was planning, including idea generation, organization,
and goal setting. In addition, they gave students vocabulary instruction, especially the
vocabulary related to the topic and co~unctions such as while, because, and however,
122
before writing. Therefore, vocabulary instruction in advanced vocabulary and literate
words was weak compared to content and topic related vocabulary in the present study.
Most of the teachers agreed that word use is essential for students to compose a
narrative or expository essay and indicated that vocabulary related to the topic,
conjunctions, and adverbs are important for students to know before writing. It was
interesting to find that only the 11 th grade teacher recognized the importance of
metacognitive and metalinguistic verbs.
Conclusions
The present study provided groundwork in examining the use of three types of
literate words in narrative and expository writing and strong evidence that genre has a
substantial influence on the use of three types ofliterate words in school-age children and
adolescents' writing, except metalinguistic verbs. Although children increased the use of
these literate words in both genres, only abstract nouns, derived nominals, and early
metacognitive verbs were found to reach the significant level in the age group
comparisons. It was interesting to find that there was no interaction effect of age and
genre for any of the literate word use in writings.
This study reported quantitative and content analysis results to offer a full picture
of how school-age children and adolescents use three types ofliterate words in two
genres of writing. The findings detected subtle changes of literate word use in writing
across three age groups, and supported previous studies in order to provide integrated
evidence of later language development and literate lexicon use in writing. In addition,
123
teacher questionnaires in the present study provided exploratory and supplemental
information about teachers' instructions in writing and their views about students'
vocabulary and genre knowledge.
Although this study examined the use of only three types of literate words in
typically developing school-age children and adolescents, it provided valuable
information for educators and clinicians to determine when children are at risk. It is
critical to first understand how typically developing children build up and utilize their
language in order to use the information as a foundation to identify children with
language difficulties. Since writing is viewed as school competence and is important to
academic success, it should be viewed as just as important as reading and other school
competencies. Most of the studies examining writing in children with language
impairments focused on syntactic errors, length of utterance, and the use of morphology
(Gillam & Johnston, 1992; Mackie & Dockrell, 2004; Windsor, Scott, & Street, 2000).
Although Morris and Crump (1982) examined the vocabulary used in writing in children
with learning disabilities, they did not look at the use of literate words. No studies to date
have focused on how children with language impairments use literate words in writing.
Researchers have been advocating for the importance of evaluating children's written
language skills for many years (Singer, 1995; Windsor, Scott & Street, 2000). Thus, the
present study offered a basis for clinicians and educators to detect possible language
difficulties while examining their students' writings.
The present study not only provided valuable information about how typically
developing school-age children and adolescents use literate words in narrative and
124
expository writing but also raised the question about providing vocabulary instruction in
advanced literate words for children's writing. In addition, this study offered a foundation
for clinicians and educators to emphasize the importance of literate word use in writing
and possible ways to detect children at risk through their writings.
Limitations and Future Research
Limitations
Several limitations ofthis study should be noted. First of all, the definitions of
inclusion criteria were partially supported by literature; therefore, it was possible that
some literate words may have been left out from the present study. Although a solid
literature review had been conducted prior to the present study, there was a possibility
that not all literate words were included in the data analysis. The second limitation ofthe
study is that the topic of expository writing, "Friendship." In the present study, this topic
might constrain and narrow the use of certain literate words and may have resulted in a
homogeneous use of literate words. Although participants were familiar with and had
domain knowledge about the topic, children produced very similar and consistent literate
words for this topic. It was possible that children had larger repertoires of abstract nouns,
derivatives, and mental state verbs, but that they only chose to use certain words that fit
best for the topic and genre. Thirdly, more interrater reliability checking was needed for
this study to enhance the internal validity since the inclusion criteria provided only a
general idea of each type of literate word. Increasing the number of raters would enhance
internal validity and strengthen the ability to draw conclusions and results. Moreover,
125
both raters were not blind to this study; therefore, possible bias may have occurred during
the coding.
The present study only focused on the use of literate words in narrative and
expository writing. However, no other tests scores were obtained for further correlation
tests. Although the present study provided a close look and deep analysis in the
production of literate words in writing, knowing the correlation between comprehension
and production of these literate words provides a strong foundation for understanding
how school-age children and adolescents use these later developing words. Given the low
use of derivative and mental state verbs, other tests, such as measures of vocabulary
knowledge, reading ability, and literacy skill, might be able to provide adequate
explanations and control for potentially interfering factors.
Future Research
Future studies should examine different types of literate words and advanced
vocabulary such as words with multiple meanings (e.g., strike, short), words with
multiple grammatical functions (e.g., hard, sweet), adverbs oflikelihood and magnitude
(e.g., possibly, extremely), and Latinate vocabulary of English (e.g. transfer, responsible)
(Bar-Dan & Berman 2007; Nippold, 2007; Paul, 2007) to further understand how children
use later developing literate words in writing. In addition, the participants in the present
study were school-age children and adolescents. Given that literate words continue to
develop into adulthood, it is necessary to include young and older adults in future
research to detect differences in the use of literate words in writing. Nippold and
126
colleagues (2005) investigated adverbial conjuncts, abstract nouns, and metaverbs in the
persuasive writing of school-age children, adolescents, and adults. Therefore, it is
important to examine how school-age children, adolescents, and adults use different
kinds of literate words in different type of discourse such as narrative and expository
writing.
Moreover, the relationship between the use of metacognitive and metalinguistic
verbs in narrative and expository writing needs to be studied further. Although
metacognitive and metalinguistic verbs are correlated (Olson & Torrance, 1986), it is not
known how these two mental state verbs are correlated in writing. This requires more
studies. In addition to mental state verbs, more research needs to be conducted to explore
the relationship between the use ofderived nominals and derived adjectives in different
written genres, such as persuasive and descriptive writing. While conducting future
research, it is suggested that investigators measure participants' implicit knowledge of
literate words such as vocabulary knowledge, literacy skill, and reading comprehension
to provide a full understanding of the use ofliterate words in writing.
The teacher questionnaire used in this study is only preliminary and supplemental,
in order to provide a general idea of how teachers view their writing instruction and their
students' vocabulary use in writing. Since only four teacher questionnaires were obtained
in this study, more teachers' input is needed. Moreover, expanding the questionnaire to
deepen the understanding of vocabulary instruction in literate words is essential to know
how school-age children and adolescents acquire and use these later developing literate
words. In addition to the teacher questionnaire, observation of teaching and teacher
127
interviews would provide detailed and valuable information and reflect teachers' actual
opinions and attitudes regarding the instruction in students' writings. Understanding
teachers' vocabulary instruction in writing not only provides useful information about
effective instruction in writing but also verifies the relationship between vocabulary
instruction in writing and students' production of literate words. Future studies should
include not only students' writing samples in different genres and their test scores related
to literacy and word knowledge, but also teachers' instruction in writing in order to have
a full picture of children's writing development and performance.
128
APPENDIX A
INSTRUCTION SHEET -- NARRATIVE ESSAY
At this time, I would like you to write a story. Please write a story about something
funny, sad, or scary that happened to you and a friend. You get to decide what to write
about. It can be anything that was funny, sad, or scary. If you can't think of something
that really happened, you can make it up. It doesn't have to be a true story. You can use
your imagination, if you want. It's up to you.
The outline below will help you organize your thoughts and write a good story. In
your story, be sure to do the following:
Tell where the events took place (the setting).
Tell who the main people are (characters).
Tell everything that happened in the story (plot).
Tell about the problems that came up (problems).
Explain what the characters tried to do (attempts).
Explain how things turned out (outcome).
Tell how everyone felt during the events (thoughts).
Keep this list of points in front of you as you write your story. As you address each
point, try to write a full paragraph of your own ideas. You will have 20 minutes to
complete your work. I have given you a booklet of lined paper to use in writing your
story. Please put your name, age, and grade level on the booklet.
As you do this work, please use your best writing style with complete sentences, and
correct grammar, spelling, and punctuation. If you aren't sure how to spell a word, make
your best guess. Try to write neatly, using a pen or pencil. If you make a mistake, just
cross it out or use an eraser. Keep going until I ask you to stop writing.
Do you have any questions?
The title of your story is: "What Happened One Day"
129
APPENDIXB
INSTRUCTION SHEET -- EXPOSITORY ESSAY
At this time, I would like you to write an essay. Please write an essay on the topic of
friendship. Friendship is very important to people of all ages - children, adolescents, and
adults. Most people say they enjoy spending time with their friends. They like to talk
with their friends in person or on the phone and spend time together.
The outline below will help you organize your thoughts and write a strong essay. In
your essay, be sure to explain the following:
What is friendship?
Why is it important to people?
How can friendship make life more enjoyable?
What kinds of things do friends like to do together?
How can people become good friends?
What kinds of actions can damage friendships?
How can people remain good friends over time?
Keep this list of questions in front of you as you write your essay. As you answer
each question, try to write a full paragraph of your own ideas. You will have 20 minutes
to complete your work. I have given you a booklet of lined paper to use in writing your
essay. Please put your name, age, and grade level on the booklet.
As you do this work, use your best writing style with complete sentences, and
correct grammar, spelling, and punctuation. If you aren't sure how to spell a word, make
your best guess. Try to write neatly, using a pen or pencil. If you make a mistake, just
cross it out or use an eraser. Keep going until I ask you to stop writing.
Do you have any questions?
The title of your essay is: "The Nature of Friendship"
130
APPENDIXC
TEACHER QUESTIONNAIRE
I am a doctoral student in the Communication Disorders and Sciences Program
at the University of Oregon. My adviser, Dr. Marilyn Nippold, and I are
conducting a study to learn how school-age children and adolescents
spontaneously use different types of words in their narrative and expository
writing. In addition to collecting students' essays, it would be valuable if we
could ask teachers some questions about how writing is taught at school. Your
participation is important for understanding the relationship between classroom
instruction and students} writing. Responses will be kept anonymous. We
appreciate your help. Thank you very much!
Best Wishes,
Lei Sun, CDS doctoral student
Dr. Marilyn Nippold, CDS Professor
Your name:
---------------
Schoolname: _
Grade level you are teaching: _
Frequency
1. How often do you ask your students to write a narrative essay?
D More than 3 times a week D 2-3 times a week D Once a week
o Once every two weeks 0 Once a month
o Other _
2. How often do you ask your students to write an expository essay?
131
D More than 3 times a week D 2-3 times a week 0 Once a week
D Once every two weeks D Once a month
D Other _
Instruction
1. What kind of instruction on writing do you most often provide to your
students? (check ONE that applies best)
D Planning (idea generation, organizing, goal setting)
D Translating (sentence generation, word choice)
D Reviewing/Editing (review written texts or plans)
D Others
------------------------------
2. Do you provide vocabulary instruction for students before writing?
D Yes (If YES, please continue to question 3 )
D No (If NO, please skip question 3 and continue to question 4)
3. What kind of vocabulary do you teach before students write their essays?
(check TWO that apply best)
o Vocabulary related to the topic (terminology)
o Metacognitive verbs (assume, hypothesis) and metalinguistic verbs (imply,
predict)
o Conjunctions (while, because, however, on the other hand)
o Adverbs (extremely, suddenly)
o Complex derivatives (discussion, confrontation, frustration)
o Others. _
4. What kind of vocabulary is important to know before students write their
essays based on your experience? (check TWO that apply best)
D Vocabulary related to the topic (terminology)
o Metacognitive verbs (assume, hypothesis) and metalinguistic verbs (imply,
predict)
o Conjunctions (while, because, however, on the other hand)
o Adverbs (extremely, suddenly)
o Complex derivatives (discussion, confrontation, frustration)
o Others
132
5. Which method do you mostly use to help your students understand the
distinction between different genres such as narrative and expository? (check
ONE that applies best)
D Writing samples
DReading
D Provide lectures on differences between genres
D Class activities
D Others. _
Feedback
1. What kind of feedback do you usually provide to your students? (check
THREE that apply best)
D Spelling
D Grammar
D Ambiguities of word use
D Faulty logic and inconsistencies
D Errors of fact and schema violations
D Faulty text structure (genre conventions)
D Incoherence
D Disorganization
D Did not consider audience needs
D Others _
2. What kind of difficulties do you usually see in your students' narrative
writing? (check THREE that apply best)
D Spelling
D Grammar
D Ambiguities of word use
D Faulty logic and inconsistencies
D Errors of fact and schema violations
D Faulty text structure (genre conventions)
D Incoherence
D Disorganization
D Did not consider audience needs
D Others, ~ _
3. What kind of difficulties do you usually see in your students' expository
writing? (check THREE that apply best)
133
D Spelling
D Grammar
D Ambiguities of word use
D Faulty logic and inconsistencies
D Errors of fact and schema violations
D Faulty text structure (genre conventions)
D Incoherence
D Disorganization
D Did not consider audience needs
D Others _
Rating
1. Word use is essential for students to compose a narrative or expository essay.
D Strongly disagree
D Moderately disagree
D Neutral
D Moderately agree
D Strongly agree
2. My students have adequate word knowledge to write a good narrative or
expository essay.
D Strongly disagree
D Moderately disagree
D Neutral
D Moderately agree
D Strongly agree
3. My students need vocabulary instruction to complete a good narrative or
expository essay.
D Strongly disagree
D Moderately disagree
D Neutral
D Moderately agree
D Strongly agree
4. I provide sufficient vocabulary instruction for my students to complete a
good narrative or expository essay.
D Strongly disagree
D Moderately disagree
D Neutral
D Moderately agree
--'--- ._----- -_._------
134
D Strongly agree
5. My students understand the difference between writing a narrative and
writing an expository essay.
D Strongly disagree
D Moderately disagree
D Neutral
D Moderately agree
D Strongly agree
6. I provide sufficient practice for my students to write a narrative or expository
essay.
D Strongly disagree
D Moderately disagree
D Neutral
D Moderately agree
D Strongly agree
7. My students have difficulties composing a narrative essay.
D Strongly disagree
D Moderately disagree
D Neutral
D Moderately agree
D Strongly agree
8. My students have difficulties composing an expository essay.
D Strongly disagree
D Moderately disagree
D Neutral
D Moderately agree
D Strongly agree
9. Generally, my students have difficulties in using appropriate words while
writing a narrative essay.
D Strongly disagree
D Moderately disagree
D Neutral
D Moderately agree
D Strongly agree
10. Generally, my students have difficulties in using appropriate words while
writing an expository essay.
D Strongly disagree
D Moderately disagree
D Neutral
------------------
D Moderately agree
D Strongly agree
Thanks for your participation!
Your answers will be kept confidential and secure!
135
APPENDIXD
DESCRIPTIVE STATISTIC TABLE OF 10 DEPENDENT VARIABLES - MEAN, STANDARD DEVIATION AND RANGE
Age ~ODV TW TTU MLTU ABN DRN DRA EMCV EMLV LMCV LMLV
group Genre
5th Narrative 170.68 19.10 9.14 3.89 0.68 1.81 0.69 1.67 0.25 0.10
Grade (80.26) (8.86) (2.22) (2.10) (1.15) (1.35) (0.78) (2.06) (0.46) (0.31)
[64-392] [8-46] [6-17] [1.09-10.79] [0-5.04] [0-5.71] [0-3.90] [0-8.67] [0-1.54] [0-1.62]
Expository 139.58 11.53 12.33 9.15 5.40 1.18 1.96 1.30 0.94 0.09
(60.31) (5.02) (3.00) (2.87) (3.86) (0.86) (1.44) (1.75) (1.01) (0.28)
[29-346] [5-31] [6-18] [3.19-15.18] [0-20.43] [0-3.45] [0-5.38] [0-9.26] [0-3.61] [0-1.12]
8th Narrative 245.40 23.80 11.19 4.17 1.13 1.97 0.81 1.41 0.56 0.13
Grade (83.94) (9.93) (3.93) (1.74) (0.83) (1.32) (0.78) (1.67) (0.71) (0.26)
[70-433] [4-44] [6-31 ] [1.76-9.48] [0-3.43] [0-7.32] [0-3.11] [0-8.64] [0-3.73] [0-0.94]
Expository 208.40 15.18 14.53 8.26 6.50 1.42 1.90 1.08 0.81 0.12
(66.09) (6.50) (3.13) (2.54) (3.15) (0.61) (1.02) (0.87) (0.88) (0.30)
[48-242] [3-33] [7-25] [3.76-13.17] [1.32-15.34] [0.53-3.45] [0-4.51 ] [0-3.51] [0-3.45] [0-1.19]
11th Narrative 336.73 30.70 11.27 4.95 1.62 1.88 0.83 0.85 0.85 0.19
Grade (71.71) (7.99 (2.07) (1.59) (0.95) (0.96) (0.55) (0.82) (0.59) (0.27)
[182-452] [15-46] [8-17] [1.43-8.63 ] [0-4.18] [0.3-3.87] [0-2.14] [0-3.79] [0-2.13] [0-0.97]
Expository 285.50 22.43 13.26 9.85 6.14 1.71 2.18 0.88 1.06 0.15
(89.96) (7.30) (2.54) (1.99) (2.24) (0.89) (1.06) (0.80) (0.76) (0.25)
[121-433] [7-34] [9-22] [6.19-14.63] [0-10.00] [0-4.05] [0.33-4.84] [0-3.31] [0-2.80] [0-0.73]
(): Standard Deviation
[ ]: Range from Low to High
......
w
0\
137
BIBLIOGRAPHY
Aitchison, 1. (2003). Words in the mind: An introduction to the mental lexicon-
third edition. Oxford: Blackwell.
Alegre, M., & Gordon, P. (1999). Rule-based versus associative processes in derivational
morphology. Brain and Language, 68, 347-354.
Altarriba, 1., Bauer, L., & Benvenuto, C. (1999). Concreteness, context availability, and
imageability ratings and word associations for abstract, concrete and emotion
words. Behavior Research Methods, Instruments, & Computers, 31(4),578-602.
American Speech-Language-Hearing Association. (2001). Roles and responsibilities of
Speech-Language Pathologists with respect to reading and writing in children and
adolescents [Position Statement]. Available from www.asha.org/policy.
Anglin, J. (1993). Vocabulary development: A morphological analysis. Monographs of
the Society for Research in Child Development, 58(10, Serial No. 238).
Astington, J., & Jenkins, 1. (1999). A longitudinal study of the relation between language
and theory-of -mind development. Developmental Psychology, 35(5),1311-1320.
Astington, J., & Olson, D. (1990). Metacognitive and metalinguistic language: Learning
to talk about thought. Applied Psychology: An International Review, 39(1), 77-87.
Babu, N., & Mishra, S. (2000). Development of mental state verbs among Oriya speaking
children: Pragmatics or semantics. Psychological Studies, 45, 24-29.
Bar-Han, L., & Berman, R (2007). Developing register differentiation: The
Latinate-Germanic divide in English. Linguistic, 45(1),1-35.
Berman, R (2004). Genre and modality in developing discourse abilities. In C. Moder &
A. Martinovic-Ziv (Eds.), Discourse across languages and cultures. (pp. 229-356).
Amsterdam: John Benjamins.
Berman, R, & Katzenberger, 1. (2004). Form and function in introducing narrative and
expository texts: A developmental perspective. Discourse Processes, 38(1), 57-94.
138
Berman, R., & Nir-Sagiv, B. (2007). Comparing narrative and expository test
construction across adolescence: A developmental paradox. Discourse Processes,
43(2), 79-120.
Berman, R., & Nir-Sagiv, B. (2004). Linguistic indicators of inter-genre differentiation in
later language development. Journal ofChild Language, 31, 339-380.
Berninger, v., & Swanson, H.L. (1994). Modifying Hayes and Flower's model of skilled
writing to explain beginning and developing writing. Advances in Cognition and
Educational Practice, 2,57-81.
Bertram, R., Laine, M., & Virkka1a, M. M. (2000). The role of derivational morphology
in vocabulary acquisition: Get by with a little help from my morpheme friends.
Scandinavian Journal ofPsychology, 41, 287-296.
Biemiller, A., & Slonim, N. (2001). Estimating root word vocabulary growth in
normative and advanced populations: Evidence for a common sequence of
vocabulary acquisition. Journal ofEducational Psychology, 93(3),498-520.
Blair, T., & Crump, C. (1984). Effects of discourse mode of the syntactic complexity of
learning disabled students' written expression. Learning Disability Quarterly, 7,
19-29.
Bloom, P. (2001). How children learn the meanings of words. Behavioral and Brain
Sciences,24,1095-1103.
Booth, 1., & Hall, W. (1994). Role of the cognitive internal state lexicon in reading
comprehension. Journal ofEducational Psychology, 86(3), 413-422.
Booth, 1., & Hall, W. (1994). Metacognitive development and the cognitive internal state
lexicon (Reading Research Rep. No. 18). College Park, MD: University of Georgia
and Maryland, National Reading Research Center.
Booth, 1., & Hall, W. (1995). Development of the understanding of the polysemous
meanings of the mental-state verb knOw. Cognitive Development, 10, 529-549.
Booth, J., Hall, W., Robinson, G., & Kim, S. (1997). Acquisition of the mental state verb
know by 2- to 5-year-old children. Journal ofPsycholinguistic Research, 26,
581-603.
Brent, M., & Cartwright, T. (2002). Distribution regularity and phonotactic constrains are
useful for segmentation. In G. Altmann (Ed.), Psycholinguistics: Critical concepts
in psychology. New York: Routledge.
139
Bretherton, 1., & Beeghly, M. (1982). Talking about internal states: The acquisition of an
explicit theory of mind. Development Psychology, 18, 906-921.
Cain, K., Oakhill, J., & Elbro, C. (2003). The ability to learn new word meanings from
context by school-age children with and without language comprehension
difficulties. Journal ofChild Language, 30, 681-694.
Cain, K., Oakhill, J., & Lemmon, K. (2004). Individual differences in the inference of
word meanings from context: The influence of reading comprehension, vocabulary
knowledge, and memory capacity. Journal ofEducational Psychology, 96(4),
671-681.
Carlisle, 1. (1987). The use of morphological knowledge in spelling derived forms by
learning-disabled and normal students, Annals ofDyslexia, 37, 90-108.
Carlisle, 1. (1988). Knowledge of derivational morphology and spelling ability in forth,
sixth, and eight graders. Applied Psycholinguistic, 9, 247-266.
Carlisle, J. (1996). An exploratory study of morphological errors in children's written
stories. Reading and Writing: An Interdisciplinary Journal, 8, 61-72.
Chafe, W., & Danielewicz, 1. (1987). Properties of spoken and written language. In R.
Horowitz & S. 1. Samuels (Eds.), Comprehending oral and written language (pp.
83-113). San Diego: Academic Press.
Chaffin, R., Morris, R., & Seely, R. (2001). Learning new word meanings from context:
A study of eye movements. Journal ofExperimental Psychology: Learning,
Memory, and Cognition, 27(1), 225-235.
Champion, A. (1997). Knowledge of suffixed words: A comparison of reading disabled
and nondisabled readers. Annals ofDyslexia, 47, 29-55.
Clark, E. (1993). The lexicon in acquisition. London: Cambridge University Press.
Curenton, S., & Justice, 1. (2004). African American and Caucasian preschoolers' use of
decontextualized language: Literate language features in oral narratives. Language,
Speech, and Hearing Services in School, 35, 240-253.
Curtis, M. (1987). Vocabulary testing and vocabulary instruction. In M. G. McKeown &
M. E. Curtis (Eds.), The nature ofvocabulary acquisition (pp. 37-52). Hillsdale, NJ:
Lawrence Erlbaum Associates.
140
de Villiers, 1., & de Villiers, P. (2000). Linguistic determinism and the understanding of
false beliefs. In P. Mitchell & K. Riggs (Eds.), Children sreasoning and the mind
(pp. 191-228). Hove, UK: Psychology Press.
Dockrell, J., & Messer, D. (2004). Lexical acquisition in the early school years. In R.
Berman (Ed.), Language development across childhood and adolescence (pp.
53-82). Amsterdam: John Benjamins.
Duin, A., & Graves, M. (1987). Intensive Vocabulary Instruction as a Prewriting
Technique. Reading Research Quarterly, 22(3), 311-330.
Dunn, L., & Dunn, D. (2007). The Peabody picture vocabulary test, fourth Edition.
Bloomington, MN: NCS Pearson, Inc.
Flower, L., & Hayes, 1. (1984). Images, plans, and prose: The representation of meaning
in writing. Written Communication, 1(1),120-160.
Freyd, P., & Baron, J. (1982). Individual differences in acquisition of derivational
morphology. Journal ofVerbal Learning and Verbal Behavior, 21,282-295.
Fuste-Herrmann, B., Silliman, E., Bahr, R., Fasnacht, K., & Federico, J. (2006). Mental
State verb production in the oral narratives of English- and Spanish-speaking
preadolescents: An exploratory study of lexical diversity and depth. Learning
Disabilities Research & Practices, 21 (1),44-60.
Gillam, R., & Johnston, 1. (1992). Spoken and written language relationships in
language/learning-impaired and normally achieving school-age children. Journal of
Speech and Hearing Research, 35,1303-1315.
Green, L., McCutchen, D., Schwiebert, C., Quinlan, T., Eva-Wood, A., & Juelis, 1. (2003).
Morphological development in children's writing. Journal ofEducational
Psychology, 95(4), 752-761.
Greenhalgh, K., & Strong, C. (2001). Literate language features in spoken narratives of
children with typical language and children with language impairments. Language,
Speech, and Hearing Services in School, 32, 114-125.
Hadley, P. (1998). Language sampling protocols for eliciting text-level discourse.
Language, Speech, and Hearing Services in School, 29, 132-147.
Hayes, J. (1996). A new framework for understanding cognition and affect in writing. In
C. Levy & S. Ransdell (Eds.), The science ofwriting: Theories, methods,
individual differences, and applications (pp. 1-27). Hillsdale, NJ: Lawrence
Erlbaum Associates, Inc.
141
Hayes, 1., Flower, L., Schriver, K., Stratman, 1., & Carey, L. (1987). Cognitive processes
in revision. In S. Rosenberg (Ed.), Advances in applied psycholinguistics. Vol. 2:
Reading, writing, and language processing (pp. 176-240). Cambridge: Cambridge
University Press.
Hidi, S., & Hildyard, A. (1983). The comparison of oral and written productions in two
discourse types. Discourse Processes, 6, 91-105.
Hill, K. (2001). The development of a model for automated performance measurement
and the establishment of performance indices for augmented communicators under
two sampling conditions. Unpublished doctoral Dissertation, University of
Pittsburgh.
Hunt, K. (1965). Grammatical structures written at three grade levels. NCTE Research
Report No.3. Champaign, IL: National Council ofTeachers of English.
Jessen, E, Heun, R., Erb, M., Granath, D., Klose, 0., Papassotiropoulos, A., & Grodd, W.
(2000). The concreteness effect: Evidence for dual coding and context availability.
Brain and Language, 74, 103-112.
Johnson, C., & Maratsos, M. (1977). Early comprehension of mental verbs: Think and
know. Child Development, 48,1743-1747.
Johnson, C., & Wellman, H. (1980). Children's development understanding of mental
verbs: Remember, know, and guess. Child Development, 51, 1095-1102.
Joshi, R. (2005). Vocabulary: A critical component of comprehension. Reading and
Writing Quarterly, 21, 209-219.
Kellogg, R., Olive, T., & Piolat, A. (2007). Verbal, visual, and spatial working memory in
written language production. Acta Psychologica, 124, 382-397.
Kounios, J., & Holcomb, P. (1994). Concreteness effects in semantic processing: ERP
evidence supporting dual-coding theory. Journal ofExperimental Psychology:
Learning, Memory, and Cognition, 20(4), 804-823.
Kroll, B. (1981). Developmental relationships between speaking and writing. In B. M.
Kroll & R. J. Vann (Eds.), Exploring speaking-writing relationships (pp. 32-54).
Urbana, IL: National Council of Teachers of English.
Laflamme, J. (1997). The Effect of Multiple Exposure Vocabulary Method and the Target
Reading/Writing Strategy on Test Scores. Journal ofAdolescent & Adult Literacy,
40(5),372-384.
142
Langer, 1. (1985). Children's sense of genre: A study of performance on parallel reading
and writing tasks. Written Communication, 2(2), 157-187.
Larsen, 1., & Nippold, M. (2007). Morphological analysis in school-age children:
Dynamic assessment of a word leaning strategy. Language, Speech, and Hearing
Services in Schools, 38, 1-12.
Laufer, B., Elder, C., Hill, K., & Congdon, P. (2004). Size and strength: Do we need both
to measure vocabulary knowledge? Language Testing, 21(2), 202-226.
Leong, C-K. (2000). Rapid processing of base and derived forms of words and grades 4,
5 and 6 children's spelling. Reading and Writing: An Interdisciplinary Journal, 12,
277-302.
Lewis, B., O'Donnell. B., Freebairn, L., & Taylor, H. (1998). Spoken language and
written expression- Interplay of delays. American Journal ojSpeech-Language
Pathology, 7, 77-84.
Lewis, D., & Windsor, J. (1996). Children's analysis of derivational suffix meanings.
Journal ojSpeech and Hearing Research, 39, 209-216.
Mackie, C., & Dockrell, 1. (2004). The nature of written language deficits in children
with SLI. Journal ojSpeech, Language, and Hearing Research, 47,1469-1483.
Mann, V. (2000). Introduction to special issue on morphology and the acquisition of
alphabetic writing systems. Reading and Writing: An Interdisciplinary Journal, 12,
143-147.
Mansfield, T. (2007). Anatomy of a code. Unpublished manuscript.
Marslen-Wilson, W. (2002). Functional parallelism in spoken word-recognition. In G.
Altmann (Ed.) Psycholinguistics: Critical concepts in psychology. New York:
Routledge.
McCutchen, D. (1987).Children's discourse skill: Form and modality requirements of
schooled writing. Discourse Processes, 10,267-286.
McCutchen, D. (1986). Domain knowledge and linguistic knowledge in the development
of writing ability. Journal ojMemory and Language, 25, 431-444.
McCutchen, D., Covill, A., Hoyne, S., & Mildes, K. (1994). Individual differences in
writing: Implications of translating fluency. Journal ojEducational Psychology,
86(2), 256-266.
143
McFadden, T., & Gilllam, R. (1996). An examination of the quality of narratives
produced by children with language disorders. Language, Speech, and Hearing
Services in Schools, 27, 48-56.
McGhee-Bidlack, B. (1991). The development of noun definitions: Ametalinguistic
analysis. Journal ofchild Language, 18,417-434.
Mellet, E., Tzourio, N., Denis, M., & Mazoyer, B. (1998). Cortical anatomy of mental
imagery of concrete nouns based on their dictionary definition. NeuroReport, 9,
803-808.
Miller, C. (2004). False belief and sentence complement performance in children with
specific language impairment. International Journal ofLanguage and
Communication Disorders, 39(2),191-213.
Miller, l, & Chapman, R. (2003). SALT: A computer program for the systematic analysis
of language transcripts [computer software]. Madison, WI: Language Analysis
Laboratory, Waisman Center, University of Wisconsin.
Moore, C., Brynt, D., & Furrow, D. (1989). Mental terms and the development of
certainty. Child Development, 60, 167-171.
Moore, C., & Davidge, l (1989). The development of mental terms: Pragmatics or
semeatics? Journal ofChild Language, 16,633-641.
Morris, N., & Crump, W. (1982). Syntactic and vocabulary development in the written
language oflearning disabled and non-learning disabled students at four age levels.
Learning disability quarterly, 5, 163-172.
Nagy, W., & Anderson, R. (1984). How many words are there in printed school English?
Reading Research Quarterly, 19, 304-330.
Nagy, W., Diakidoy, I, & Anderson, R. (1993). The acquisition of morphology: Learning
the contribution of suffixes to the meanings of derivatives. Journal ofReading
Behavior,25(2),155-170.
Newcomer, P., & Hammill, D. (1997). The test oflanguage development-3. Austin, TX :
Pro-ed.
Nippold, M. (2007). Later language development: School-age children, adolescents, and
young adults (3rd ed.). Austin, TX: Pro-Ed.
Nippold, M. (1999). Word definition in adolescents as a function of reading proficiency:
A research note. Child Language Teaching and Therapy, 15(2), 171-176.
144
Nippold, M. (1995). Language norms in school-age children and adolescents: An
introduction. Language, speech, and Hearing Services in Schools, 26, 307-308.
Nippold, M. (1993). Adolescent language developmental makers in adolescent language:
Syntax, semantics, and pragmatics. Language, speech, and Hearing Services in
Schools, 24, 21-28.
Nippold, M., Hegel, S., Sohlberg, M., & Schwarz, L. (1999). Defining abstract entities:
Development in pre-adolescent, adolescent, and young adults. Journal ofSpeech,
Language and Hearing Research, 42, 473-481.
Nippold, M., & Sun, L. (in press). Knowledge of morphologically complex words: A
developmental study of older children and young adolescents. Language, Speech,
and Hearing Services in School.
Nippold, M., Mansfield, T., & Billow, 1. (2007). Peer conflict explanations in children,
adolescents, and adults: Examining the development of complex syntax. American
Journal ofSpeech-Language Pathology, 16,1-10.
Nippold, M., Ward-Lonergan, 1., & Fanning, J. (2005). Persuasive writing in children,
adolescents, and adults: A study of syntactic, semantic, and pragmatic development.
Language, Speech, and Hearing Services in School, 36, 125-138.
Nittono, H., Suehiro, M., & Hori, T. (2002). Word imageability and N400 in an incidental
memory paradigm. International Journal ofPsychophysiology, 44,219-229.
Noppeney, D., & Price, C. (2004). Retrieval of abstract semantics. NeuroImage, 22,
164-170.
O'Donnell, R. (1974). Syntactic differences between speech and writing. American
Speech, 49, 102-110.
Oetting, J. (1999). Children with SLI use argument structure cues to learn verbs. Journal
ofSpeech, Language, and Hearing Research, 42, 1261-1274.
Oetting, J., Rice, M., & Swank, L. (1995). Quick incidental learning (QDIL) of words by
school-age children with and without SLI. Journal ojSpeech and Hearing
Research, 38, 434-445.
O'Hara, M., & Jolmston, J. (1997). Syntactic bootstrapping in children with specific
language impairment. European Journal ofDisorders pjCommunication, 32,
189-205.
145
Olson, D., & Astington, J. (1993). Thinking about thinking: Learning how to take
statements and hold beliefs. Educational Psychologist, 28(1), 7-23.
Olson, D., Antonietti, A., Liverta-Sempio, 0., & Marchetti, A. (2006). The mental verbs
in different conceptual domains and in different cultures. In A. Antonietti, O.
Liverta-Sempio, & A. Marchetti (Eds.), Theory ofmind and language in
developmental contexts (pp. 31-65). New York, NY: Springer Science Business
Media, Inc.
Olson, D., & Ton-ance, N. (1986). Some relations between children's knowledge of
metalinguistic and metacognitive verbs and their linguistic competencies. In 1.
Gopnik & M. Gopnik (Eds.), Models to modules: Studies in cognitive science from
the McGill workshops (pp. 66-81). Norwood, NJ: Ablex Publishing Corp.
Paul, R. (2007). Language disorders from infancy through adolescence: Assessment and
intervention (3rd ed.). St.Louis: Mosby Elsevier.
Pavio, A. (1991). Dual coding theory: Retrospect and current status. Canadian Journal of
Psychology, 45, 225-287.
Perfetti, c., & McCutchen, D. (1987). Schooled language competence: Linguistic
abilities in reading and writing. In S. Rosenberg (Ed.), Advances in applied
psycholinguistics, reading, writing and language learning, Vol. 2 (pp. 105-141).
New York: Cambridge University Press.
Peskin, 1., & Olson, D. (2004). On reading poetry: Expert and novice knowledge. In R.
A.. Berman (Ed.), Language development across children and adolescence (pp.
211-232). Amsterdam: John Benjamins Publishing Company.
Plag,I., Dalton-Puffer, C., & Bayyen, H. (1999). Morphological productivity across
speech and writing. English Language and Linguistics, 3(2),209-228.
Prater, D., & Padia, W. (1983). Effects of modes of discourse on writing performance in
grades four and six. Research in the Teaching ofEnglish, 17(2), 127-134.
Ragnarsdottir, H., Aparici, M., Cahana-Amitay, D., van Hell, J., & Viguie, A. (2002).
Verbal structure and content in written discourse: Expository and nanative texts.
Written Language & Literacy, 5(1), 95-125.
Ravid, D. (2006). Semantic development in textual contexts during the school years:
Noun scale analysis. Journal ofChild Language, 33, 791-821.
146
Ravid, D. (2004). Derivational morphology revised: Later lexical development in Hebrew.
In R. Berman (Ed.), Language development across childhood and adolescence (pp.
53-82). Amsterdam: John Benjamins Publishing Company.
Ravid, D., & Berman, R. (2006). Information density in the development of spoken and
written narratives in English and Hebrew. Discourse Processes, 41(2), 117-149.
Ravid, D., & To1chinsky, L. (2002). Developing linguistic literacy: A comprehensive
model. Journal ofChild Language, 29, 417-447.
Ravid, D., & Zilberbuch, S. (2003). Morphosyntactic constructs in the development of
spoken and written Hebrew text production. Journal ofChild Language, 30,
395-418.
Rice, M., Cleave, P., & Oetting, 1. (2000). The use of syntactic cues in lexical acquisition
by children with SLI. Journal ofSpeech, Language, and Hearing Research, 43,
582-594.
Rubin, H., Patterson, P., & Kantor, M. (1991). Morphological development and writing
ability in children and adults. Language, Speech, and Hearing Services in School,
22,228-235.
Sabsevitz, D., Meddler, D., Seidenberg, M., & Binder, J. (2005). Modulation of the
semantic system by word imageability. Neurolmage, 27,188-200.
Sadoski, M. (1995). Concreteness Effects in Text Recall: Dual Coding or Context
Availability? Reading Research Quarterly, 30(2), 278-288.
Sadoski, M., Kealy, W., Goetz, E., & Raivio, A. (1997). Concreteness and imagery effects
in the written composition of definitions. Journal ofEducational Psychology, 89(3),
518-526.
Schwanenflugel, P., Akin, C., & Luh, W. (1992). Context availability and the recall of
abstract and concrete words. Memory & Cognition, 20(1), 96-104.
Schwanenflugel, P., Henderson, R., & Fabricius, W. (1998). Developing organization of
mental verbs and theory of mind in middle childhood: Evidence from extensions.
Developmental Psychology, 34(3),512-524.
Schwanenflugel, P., Fabricius, W., & Noyes, C. (1996). Developing organization of
mental verbs: Evidence for the development of a constructivist theory of mind in
middle childhood. Cognitive Development, 11, 265-294.
147
Scott, C. (2005). Learning to write. In H. W. Catts & A G. Kamhi (Eds.), Language and
reading disabilities, 2 (pp. 224-258). Boston, MA: Pearson.
Scott, C. (1994). A discourse continuum for school-age students: Impact of modality and
genre. In G. P. Wallach & K. G. Butler (Eds.), Language learning disabilities in
school-age children and adolescents: Some principles and applications
(pp.l80-218). New York: Macmillan College.
Scott, C. (1988). A perspective on the evaluation of school children's narratives.
Language, Speech, and Hearing Services in Schools, 19,67-82.
Scott, C. (1988). Spoken and written syntax. In M.A Nippold (Ed.), Later language
development: Age 9 through 19 (pp. 49-95). Boston, MA: Little, Brown and
Company.
Scott, C., & Windsor, J. (2000). General language performance measures in spoken and
written narrative and expository discourse of school-age children with language
learning disabilities. Journal ofSpeech, Language, and Hearing Research, 43,
324-339.
Semel, E., Wiig, E., & Secord W. (1995). Clinical evaluation oflanguage fundamentals -
3. San Antonio, TX: The Psychological Corporation.
Shatz, M., Wellman, H., & Silber, S. (1983). The acquisition of mental verbs: A
systematic investigation of the first reference to mental status. Cognition, 14,
301-321.
Slade, 1., & Ruffman, T. (2005). How language does (and does not) relate to theory of
mind: A longitudinal study of syntax, semantics, working memory and false belief.
British Journal ofDevelopmental Psychology, 23, 117-141.
SPSS Inc. (2005). SPSS 13.0 for Windows. [Computer software]. Chicago, IL: SPSS Inc.
Stromqvist, S., Nordqvist, A, & Wengelin, A (2004). Writing the frog story:
Developmental and cross-modal perspectives. In S. Stromqvist & 1. Verhoeven
(Eds.), Relating events in narrative, Volume 2: Typological and contextual
perspectives. (pp.359-394). Hallsdale, NJ: Lawrence Erlbaum Associates.
Swaab, T., Baynes, K., & Knight, R. (2002). Separable effects of priming and
imageability on word processing: An ERP study. Cognitive Brain Research, 15,
99-103.
Tyler, A, & Nagy, W. (1989). The acquisition of English derivational morphology.
Journal ofMemory and Language, 28, 649-667.
148
Verhoeven, L., Aparici, M., Cahana-Amitay, D., van Hell, J., Kriz, S., & Viguie-Simon.,
A (2002). Clause packing in writing and speech. Written Language & Literacy,
5(2), 135-162.
Verhoeven, L., & Perfetti, C. (2003). Introduction to this special issue: The role of
morphology in learning to read. Scientific Studies ofReading, 7(3),209-217.
West, c., & Holcomb, P. (2000). Imaginal, semantic and surface-level processing of
concrete and abstract words: An electrophysiological investigation. Journal of
Cognitive Neuroscience, 12,1024-1037.
Westby, C. (1999). Assessing and facilitating text comprehension problem. In H. W. Catts
& A G. Kamhi (Eds.), Language and reading disabilities (pp. 154-221). Boston:
Allyn & Bacon.
Westby, C. (1994). The effects of culture in genre, structure, and style of oral and written
texts. In G. P. Wallach & K. G. Butler (Eds.), Language learning disabilities in
school-age children and adolescents: Some principles and applications (pp.
180-218). New York: Macmillan College.
Westby, C. (1990). The role of the speech-language pathologist in whole language.
Language, Speech, and Hearing Services in School, 21,228-237.
Westby, C. (1985). Learning to talk-talking to learn: Oral-literate language differences.
In C.S. Simon (Ed.), Communication skills and classroom success (pp. 181-213).
San Diego: College-Hill Press.
Weston, A, Shriberg, L., & Miller, 1. (1989). Analysis of language-speech samples with
SALT and PEPPER. Journal of Speech and Hearing Research, 32, 755-766.
White, T., Power, M., & White, S. (1989). Morphological analysis: Implications for
teaching and understanding vocabulary growth. Reading Research Quarterly, 24(3),
283-304.
White, T., Sowell. J., & Yanagihara, A (1989). Teaching elementary students to use
word-part clues. The Reading Teacher, 42, 302-307.
Wilson, 1. (1999). Using words about thinking: Content analyses of chemistry teacher's
classroom talk. International Journal ofScience Education, 21(10), 1067-1084.
Windsor, 1. (1994). Children's comprehension and production of derivational suffixes.
Journal ofSpeech and Hearing Research, 37,408-417.
149
Windsor, 1., Scott., C., & Street, C. (2000). Verb and noun morphology in the spoken and
written language of children with language learning disabilities. Journal o/Speech
and Hearing Research, 43, 1322-1336.
Wing, C., & Scholnick, E. (1986) Understanding the language of reasoning: Cognitive,
linguistic, and developmental influences. Journal 0/Psycholinguistic Research,
15(5),383-401.
Wysocki, K., & Jenkins, J. (1987). Deriving word meanings through morphological
generalization. Reading Research Quarterly, 22(1), 66-81.
Zarry, L. (1999). Vocabulary enrichment in composition: A grade six study. Education,
120(2),267-272.