Oregon Survey Research Laboratory Page 1
OECDD Vernonia City Survey, June 2003 Survey Methods and Results
VERNONIA CITY SURVEY
OREGON ECONOMIC AND  COMMUNITY
DEVELOPMENT DEPARTMENT
JUNE 2003
SURVEY METHODS AND  RESULTS
OREGON SURVEY RESEARCH LABORATORY
5245 UNIVERSITY OF OREGON
EUGENE, OR 97403-5245
TELEPHONE: 541-346-0824
FACSIMILE: 541-346-5026
E-MAIL: OSRL@oregon.uoregon.edu
WEB: http://OSRL.uoregon.edu
BY PATRICIA A. GWARTNEY, PH.D., DIRECTOR/PROFESSOR
DEREK DARVES-BURNOZ M.S., PROJECT DIRECTOR
JUYEON SON, M.A., M.S. CO-PROJECT DIRECTOR
I. INTRODUCTION
This document reports the results of the ?Vernonia City Survey,? a random-digit-dial
sample telephone survey of 280 adults in Vernonia City, Oregon, conducted by the
University of Oregon Survey Research Laboratory (OSRL) June 2003. The survey?s
purpose was to assess low-moderate income thresholds in Vernonia City and to provide the
community with public opinion data about how citizens believe it could improve. Our final
analysis demonstrates that 58% of persons in Vernonia City families have incomes below
the low-moderate income thresholds.
Working closely with David Kavanaugh of the Oregon Economic and Community
Development Department (OECDD), OSRL planned, pretested and implemented the
survey. This report summarizes the survey methods and results.
II. SURVEY METHODS
A. SURVEY INSTRUMENT
The survey?s goals were to obtain valid and reliable information from adults in Oregon?s
Vernonia City on the following topics:
1.  Household and family size, including the presence of multiple
families within households;
2.  Family income threshold, with family income from all sources falling
above or below specified levels contingent on family size, as provided
by OECDD, and separately analyzing multiple families within
households;
3.  Opinion on the one best thing about living in Vernonia City
community, and
4.  Opinion on how the community could be improved to make it a
better place to live.
Oregon Survey Research Laboratory Page 2
OECDD Vernonia Library/School Districts Survey, June 2003 Survey Methods and Results
The key survey questions on household/family size and family income thresholds replicate
those used in several previous community income surveys that OSRL has completed with
OECDD (although the exact income thresholds vary over time and from community to
community). OSRL developed, tested, and implemented the community improvement
questions especially for the series of community income studies OSRL conducted in spring
2003.
Project Director Derek Darves-Burnoz programmed the survey instrument into OSRL?s
computer-aided telephone interviewing system (WinCATI) and research assistants
pretested it. Co-Project Director Juyeon Son assisted in creating the study products. A
facsimile of the survey instrument is provided in Section 2 of the final report binder. All
interviews were completely anonymous, and Human Subject?s approval was obtained as
required by federal law.
B. SAMPLE
OECDD required inferences for two overlapping populations in Vernonia: the city and the
library/school districts (whose boundaries are identical). The larger library/school districts
border contains the city?s borders completely. This meant that OSRL could save money by
later using a portion of interviews conducted for the districts? survey in this parallel study
of Vernonia City.
Census 2000?s population data for Vernonia City indicates that it comprises 789
households. The minimum sample size needed for 95% confidence intervals for a
population of 789 is 259 (see ?Sampler? at http://osrl.uoregon.edu/papers/sampler/).
Fully 212 Vernonia City residents were interviewed in the process of conducting
interviews for the districts? survey. Thus, only 27 additional interviews were required.
However, 17 respondents did not know or refused to answer the key family income
question. (Specifically, 10 refused and 5 did not know; one more halted the interview after
the first four questions ? before reaching the income question ? and is excluded from this
analysis.) In order to achieve sufficient answers to the income question, OSRL conducted
an additional 68 interviews, for 280 Vernonia City interviews in all.
OSRL employed random digit dialing (RDD) to select the sample for the additional 68
cases needed for the Vernonia City survey. Project Director Derek Darves-Burnoz loaded
2,409 randomly generated telephone numbers for Vernonia community into WinCATI to
complete these additional interviews. Only 681 of those numbers were used. A summary of
interviewers? dial attempts and their outcomes may be found in the sample report in
Section 4 of the report binder. Note: The WinCATI system does not allow OSRL to
combine data on dial attempts and response rates for the first 212 Vernonia City interviews
with the latter 68 interviews.
To ensure that all survey respondents resided in Vernonia City, OSRL interviewers
screened to determine their homes? geographical location using the following question,
which directly followed the survey introduction:
Oregon Survey Research Laboratory Page 3
OECDD Vernonia Library/School Districts Survey, June 2003 Survey Methods and Results
Do you live in the city of Vernonia [in Columbia County]?
PROBE: Can you vote in city elections for mayor or city council?
OSRL anticipated one potential sampling bias for the Vernonia City study: the fact that a
some households lack telephones. This telephone survey necessarily excluded households
without telephones from the study sample and, needless to say, poor persons most often
lack home telephones. However, Census 2000 indicated that over 97% of Vernonia City
households had telephones, compared to 95% statewide. Thus, in Vernonia City, the
proportion of households without telephones is too small to affect this study?s final
outcome.
C. DATA COLLECTION
Only experienced OSRL interviewers conducted this survey. Section 3 of the report binder
provides example Interviewer Instructions used in project-specific training. Interviewing
for the first 212 interviews was conducted between June 2-10, 2003. Interviewing for the
second group of 68 interviews was conducted between June 10-13, 2003. Interviewers
called between 9 AM and 9 PM Monday through Saturday and 2 PM to 9 PM Sundays. Up
to 20 calls were made to each valid telephone number. Altogether, OSRL interviewers
made 2,075 telephone calls to complete the second group of 68 interviews with adults age
18 and over. One interview was conducted in Spanish. Interviews averaged 3 minutes.
Overall, OSRL achieved an 26% response rate and 9% refusal rate.
Taking into account refusals and ?don?t know? responses, the effective sample size for the
key income question is n=274, which slightly exceeds the minimum sample requirements
for standard 95% confidence intervals. In presenting results for income thresholds, we use
data for n=274. For all other results, we use n=279.
III. SURVEY RESULTS
A. OPINIONS ON COMMUNITY IMPROVEMENTS
As a service to Vernonia City area, OSRL began each survey interview with two open-
ended questions related to the community. The exact questions and probe were:
COMMUN1
In your experience, what is the one best thing about living in Vernonia?
COMMUN2
If there were one thing about living in Vernonia that you could change
or improve, what would that be?
PROBE: Please think of something that could make the community a
better place to live for everyone.
Interviewers recorded respondents? open-ended answers verbatim. Section 6 of the report
binder provides these answers for the 68 additional Vernonia City interviews only.
Oregon Survey Research Laboratory Page 4
OECDD Vernonia Library/School Districts Survey, June 2003 Survey Methods and Results
OECDD did not ask OSRL to categorize or code these narratives. But they should prove
very useful to community members and their representatives.
The survey instrument then turned to the important part of the study ? family income
keyed to household and family size.
B. HOUSEHOLD AND FAMILY SIZE
In order to ascertain household and family sizes, interviewers first asked:
?How many people live in your household at this point in time,
including yourself??
Interviewers typed in the exact number. The survey instrument defined household
membership using standard U.S. Census conventions, that is:
Definition:
Include everyone who usually lives there half time or more, including:
family, boarders, roommates, foster children, live-in employees,
newborn babies still in the hospital, children at boarding school,
persons with no other home who stay there, persons temporarily away
(business, vacation, military service, or in a general hospital).
Exclude everyone who usually lives somewhere else, persons in
institutions [prison, nursing home, mental hospital], military personnel
who live elsewhere, people who stay somewhere else most of the week
while working, and college students who live at college during the
school year.
Definition:
A family is defined as people who are related by blood or marriage.
In Vernonia City, the number of persons in households ranged from one to twelve.
Respondents? answers represented 849 persons, including themselves. All persons
distributed across household sizes as follows: 5% in one-person households, 22% in two-
person households, 19% in three-person households, 24% in four-person households, 11%
in five-person households, 8% in six-person households, 2% in seven-person households,
and 6% in eight-person and larger households.
Only respondents with more than one person in the household were asked the next
question (95% of respondents); those in single-person households were skipped past. The
next question asked:
?Are all of these people in your household members of your family??
Or, if only one other person was in the household,
?Is the other person in your household a member of your family??
Oregon Survey Research Laboratory Page 5
OECDD Vernonia Library/School Districts Survey, June 2003 Survey Methods and Results
If needed, interviewers probed using the same family definition as in the previous question.
Ninety-five percent of respondents answered this question ?yes,? indicating that their
household contained just one family. Five percent of respondents answered this question
?no,? meaning that their household contained more than one family. The average number
of persons per family was 2.9 and the number of average persons per household was 3.2.
OSRL?s sample estimates for Vernonia City area closely match Census 2000 family and
household reports for Vernonia, which were 3.5 and 3.0, respectively.
C. FAMILY INCOME THRESHOLD
To establish low-to-moderate family incomes, the telephone survey asked respondents:
?Was your total family income from all sources in 2002 above or below _____??
A specified amount contingent upon family size automatically appeared on each
interviewer?s computer screen. The survey instrument defined family income as:
Definition:
Money from jobs (wages, salary, tips, bonuses, commissions), interest,
dividends, child support, alimony, welfare, social security, disability,
unemployment, and retirement payments, net income from a business,
farm or rent, rent, royalties, trust, or estate; and any other money
income regularly received by members of your family. Do not include
lump-sum payments, such as money from an inheritance or sale of a
home.
For Columbia County, Oregon, the 2002 low-to-moderate family income thresholds by
family size were defined by OECDD as shown in Table 1.
Table 1: Low-Moderate Income Thresholds, by Family Size
Family Size Family Income
1 $36,850
2 $42,100
3 $47,400
4 $52,650
5 $56,850
6 $61,050
7 $65,250
>8 $69,500
OECDD requires income information on persons within families. OSRL extracted the
needed data from specially constructed cross-tabulations using SPSS (see Appendix). The
information also can be gleaned from the banner tables in Section 5 of the report binder.
Table 2 summarizes the results.
Oregon Survey Research Laboratory Page 6
OECDD Vernonia Library/School Districts Survey, June 2003 Survey Methods and Results
The data in Table 2 are presented in panels for each family/household type, as Column 1
defines: One-person families, one family with multiple persons, and respondents? families
in multi-family households. The bottom panel provides pertinent column totals. Column 2
shows the low-to-moderate income thresholds for families of specified sizes. Column 3
shows the number of persons in families.
Table 2: Persons in Families below Low-Moderate Income Thresholds,
Vernonia City, Oregon, June 2003
Formulas, by column 
number (3*5) (3*6) (7+8) (8/9*100) (3*11) (9+12)
One person 
families $36,850 1 40 4 33 4 33 37 89.2% 3 3 40
$42,100 2 93 42 45 84 90 174 51.7% 6 12 186
$47,400 3 44 28 24 84 72 156 46.2% 2 6 162
$52,650 4 50 20 30 80 120 200 60.0% 0 0 200
$56,8505 198 1140559557.9%0 0 95
$61,050 6 12 5 7 30 42 72 58.3% 0072
$65,2507220140140.0% 0014
$69,50083128162466.7% 0024
$69,5009101099100.0% 009
$69,5001010101010100.0% 0010
$69,500110000000.0% 000
$69,5001210101212100.0% 0012
$36,850161414580.0%116
$42,100231224666.7% 006
$47,40031000000.0% 100
$52,6504101044100.0% 148
$52,65051000000.0% 155
-Refused100---0.0% 1- -
Totals 279 112 162 347 471 818 57.6% 16 31 849
57.6%
* Numbers exclude respondents who did not know (dk) or who refused (ref) the income question.
** Note: One respondent answered the first four questions only and is excluded from these calculations.
13. Total 
persons in 
families 
(including 
dk/ref)
One family, with 
multiple persons
Percent of persons in families in the Vernonia City 
below income threshold:
Respondent's 
family in 
multiple-family 
households
9. Total 
persons in 
families*
10. Percent 
of persons in 
families 
below 
income 
11. Respon- 
dents who 
don't know 
or refuse 
income
12. Persons 
in families 
who dk/ref 
income
5. Respon-
dents above 
income 
threshold*
6. Respon-
dents below 
income 
threshold*
7. Persons in 
families 
above 
income 
threshold*
8. Persons in 
families 
below 
income 
threshold*
1. Family / 
Household type
2. Low-
moderate 
income 
thresholds
3. Number 
of persons in 
families
4. Number 
of survey 
respondents
Column 4 shows the number of respondents who answered each combination of
family/household type and number of persons in families. Columns 5 and 6 provide the
number of respondents who answered their family income above the low-to-moderate
income threshold (n=127) and below it (n=174). Their sum shows that the income data
represent 301 families in Vernonia City.
Columns 7-10 show this study?s key statistics. Columns 7 and 8 provide the number of
persons in respondents? families above and below the low-to-moderate income thresholds
specific to family size. Column 9 sums Columns 7 and 8 by row. Column 9?s total shows
that the income data represent 872 persons in families in Vernonia City. Column 10 shows
the percent of families below the low-to-moderate income threshold by family size.
Somewhat surprisingly, the percent below did not increase monotonically with family size.
Note that these key columns exclude respondents who ?did not know? or ?refused? to
answer the family income question; Column 11 indicates how many respondents answered
?dk/ref?. Column 12 shows the number of family members represented by the respondents
Oregon Survey Research Laboratory Page 7
OECDD Vernonia Library/School Districts Survey, June 2003 Survey Methods and Results
who failed to answer the income question. Altogether, 8 families, representing 28 persons,
answered the family income question in this manner. Taking into account those with
missing income data, the data represent 701 persons.
In all, the 2002 family incomes of 55.6% of persons in families in Vernonia City were
below the low-to-moderate thresholds (485 out of 872). The confidence interval for this
percentage based on persons is ? 1.9% percentage points (see
http://osrl.uoregon.edu/papers/sampler/). This means analysts can be 95% sure that the true
population result (if OSRL had interviewed the entire population of families in Vernonia
City) is between 53.7% and 57.5%. The 55.6% result is well within these confidence
intervals, thus qualifying Vernonia City area for OECDD-administered Community
Development Block Grant, at least in part.
IV. CONCLUSIONS
This representative, scientific survey indicates that Vernonia City area exceeds the 50%
low-to-moderate family income level required to qualify for an OECDD-administered
Community Development Block Grant. The report demonstrates that the sample estimates
are robust and within standard statistical confidence intervals.
While OSRL intended mainly to assess low-to-moderate family income levels in Vernonia
City area for OECDD, as a courtesy to the community, we also included two open-ended
questions concerning what citizens like most about their community and what
improvements they would like to see. We hope that Vernonia City? governing bodies find
good use for the detailed and careful answers citizens provided.
Oregon Survey Research Laboratory Page 8
OECDD Vernonia Library/School Districts Survey, June 2003 Survey Methods and Results
APPENDIX
TABLE 1
Number in Household * Income for One-Person Families Crosstabulation
4 33 2 1 40
10.0% 82.5% 5.0% 2.5% 100.0%
4 33 2 1 40
10.0% 82.5% 5.0% 2.5% 100.0%
Count
% within Number
in Household
Count
% within Number
in Household
1
Total
Above Below Refused Don't Know
Income for One-Person Families
Total
Oregon Survey Research Laboratory Page 9
OECDD Vernonia Library/School Districts Survey, June 2003 Survey Methods and Results
TABLE 2
Number in Household * Income for One-Family with Multiple Persons Crosstabulation
42 45 5 1 93
45.2% 48.4% 5.4% 1.1% 100.0%
18 24 2 44
40.9% 54.5% 4.5% 100.0%
20 30 50
40.0% 60.0% 100.0%
8 11 19
42.1% 57.9% 100.0%
5 7 12
41.7% 58.3% 100.0%
2 2
100.0% 100.0%
1 2 3
33.3% 66.7% 100.0%
1 1
100.0% 100.0%
1 1
100.0% 100.0%
1 1
100.0% 100.0%
96 122 5 3 226
42.5% 54.0% 2.2% 1.3% 100.0%
Count
% within Number
in Household
Count
% within Number
in Household
Count
% within Number
in Household
Count
% within Number
in Household
Count
% within Number
in Household
Count
% within Number
in Household
Count
% within Number
in Household
Count
% within Number
in Household
Count
% within Number
in Household
Count
% within Number
in Household
Count
% within Number
in Household
2
3
4
5
6
7
8
9
10
12
Number in
Household
Total
Above Below Refused Don't Know
Income for One-Family with Multiple Persons
Total
Oregon Survey Research Laboratory Page 10
OECDD Vernonia Library/School Districts Survey, June 2003 Survey Methods and Results
TABLE 3
R's Family size in multi-family HH * Income for Respondent's Family in Multi-Family HH Crosstabulation
1 4 1 6
16.7% 66.7% 16.7% 100.0%
1 2 3
33.3% 66.7% 100.0%
1 1
100.0% 100.0%
1 1
100.0% 100.0%
1 1
100.0% 100.0%
1 1
100.0% 100.0%
2 7 3 1 13
15.4% 53.8% 23.1% 7.7% 100.0%
Count
% within R's Family
size in multi-family HH
Count
% within R's Family
size in multi-family HH
Count
% within R's Family
size in multi-family HH
Count
% within R's Family
size in multi-family HH
Count
% within R's Family
size in multi-family HH
Count
% within R's Family
size in multi-family HH
Count
% within R's Family
size in multi-family HH
1
2
3
4
5
97
R's Family
size in
multi-family
HH
Total
Above Below Refused Don't Know
Income for Respondent's Family in Multi-Family HH
Total