2023 Nonresponse Analysis

December 22, 2022
MEMORANDUM FOR

Mary Hyde
Director, AmeriCorps Office of Research and Evaluation

Through:

Kyra M. Linse
Survey Director, Current Population & American Time-Use Surveys
Associate Directorate for Demographic Programs

From:

Anthony G. Tersine, Jr.
Chief, Demographic Statistical Methods Division

Subject:

Evaluating Nonresponse Bias in the 2021 Civic Engagement and
Volunteering Supplement to the Current Population Survey Revision

The purpose of this memorandum is to report on analysis of various nonresponse estimates
computed for the 2021 Civic Engagement and Volunteering Supplement to the Current
Population Survey and to provide nonresponse bias analysis tables for those estimates. The
revision reflects a sponsor requested supplement name change from “Volunteering and Civic
Life (VCL)” to “Civic Engagement and Volunteering (CEV)” and corrects educational attainment
estimates provided in Tables 4 and 7.
If you should have any questions about this document, please contact Jana Hatch at 301-7632230 or [email protected] or Emily Hood at 301-763-0284 or [email protected]
or David Hornick at 301-763-4183 or [email protected].
The U.S. Census Bureau reviewed this data product for unauthorized disclosure of confidential
information and approved the disclosure avoidance practices applied to this release. CBDRBFY22-POP001-0150.
Attachment: Evaluating Nonresponse Bias in the 2021 Civic Engagement and Volunteering
Supplement to the Current Population Survey
cc:
Laura Schlachter
Andrea Robles
Tim Marshall
Michael Brennan

(AMCR)
(ADDP)

Greg Weyland
David Watt
Lorelei De Vos
Sunhak Kim

(DSD)

DRB Clearance Number CBDRB-FY22-POP001-0150
census.gov

David Hornick
Emily Hood
Rebecca Hoop
Jana Hatch
Weimin Zhang
Kiauna Womack

(DSMD)

DRB Clearance Number CBDRB-FY22-POP001-0150

Demographic Statistical Methods Division
Sample Design and Estimation

Evaluating Nonresponse Bias in the 2021 Civic
Engagement and Volunteering Supplement to
the Current Population Survey
December 22, 2022

Jana Hatch

James Farber, ADC
David Hornick, Lead Scientist
Emily Hood, Team Lead

DRB Clearance Number CBDRB-FY22-POP001-0150
census.gov

Table of Contents
Executive Summary......................................................................................................................... 1
1.
Introduction....................................................................................................................... 2
1.1 Overview of the CPS .......................................................................................................... 2
1.2

Overview of the 2021 Civic Engagement and Volunteering Supplement to the Current
Population Survey ............................................................................................................. 3

1.3

Discussion of Nonresponse in the 2021 Civic Engagement and Volunteering Supplement
to the Current Population Survey ..................................................................................... 3

2.

3.
4.
5.
6.
7.

2.1

Methodology ..................................................................................................................... 5
Data ................................................................................................................................... 5

2.2

Weights ............................................................................................................................. 5

2.3

Universe for the Estimates ................................................................................................ 6
Limitations ......................................................................................................................... 6
Response Rates ................................................................................................................. 6
Respondent Distributions................................................................................................ 21
Conclusions...................................................................................................................... 30
References ....................................................................................................................... 31

List of Tables
Table 1: 2021 Civic Engagement and Volunteering Unit Response Rates ...................................... 7
Table 2: Response Rates for September 2021 Current Population Survey Households ................ 9
Table 3: Response Rates for 2021 Civic Engagement and Volunteering Supplement
Households...................................................................................................................... 13
Table 4: Response Rates for 2021 Civic Engagement and Volunteering Supplement Households
for Characteristics Only Available for Responding Current Population Survey
Households...................................................................................................................... 18
Table 5: Respondent and Nonrespondent Distributions for September 2021 ............................ 23
Table 6: Respondent and Nonrespondent Distributions for 2021 Civic Engagement and
Volunteering Supplement Households ........................................................................... 26
Table 7: Respondent and Nonrespondent Distributions for 2021 Civic Engagement and
Volunteering Supplement Households for Characteristics Only Available for Responding
CPS Households............................................................................................................... 29

DRB Clearance Number CBDRB-FY22-POP001-0150

Executive Summary
This report presents results of an analysis of various nonresponse estimates computed for the
2021 Civic Engagement and Volunteering Supplement (CEV) to the Current Population Survey
(CPS). The sample included all households who completed a CPS interview. A nonresponse bias
analysis was conducted to determine whether nonresponse among different demographic
groups may have biased estimates. Overall response rates, demographic subgroup response
rates, and demographic respondent and nonrespondent distributions were investigated.
Key findings for the 2021 CEV:


The CPS household weighted response rate was 75.22 percent. The CEV person
weighted response rate was 65.63 percent. Since the basic CPS nonresponse rate is a
household-level rate and the CEV nonresponse rate is a person-level rate, they cannot
be combined to derive an overall nonresponse rate.



For the CPS household estimates, excluding blanks (no responses), there are significant
differences in the response rates and respondent distributions for each of the variables
investigated. Excluding the blanks and missing values, the largest difference in response
rates is seen for region and age of reference person (there is no significant difference
between their ranges of response rates). The largest difference between respondent
and nonrespondent distributions is also within region and age of reference person.



For the CEV person estimates, excluding blanks, there are significant differences in the
response rates and respondent distributions for each of the variables that were
investigated except type of living quarters and urban/rural status. The age of
respondent has the largest difference in response rates. The largest difference between
respondent and nonrespondent distributions is also within age of respondent.



For CEV person estimates for characteristics only available for CPS respondents, there
are significant differences in the response rates and respondent distributions for each of
the variables investigated. Excluding blanks, the largest difference in response rates as
well as respondent and nonrespondent distributions is seen within household type.

1
DRB Clearance Number CBDRB-FY22-POP001-0150

1.

Introduction
The Office of Management and Budget (OMB) provides guidelines for conducting a
nonresponse bias study when the expected unit response rate of a survey is below 80
percent (OMB, 2006). The Current Population Survey (CPS) household response rates
have historically been above 80 percent,1 but the response rates for the Civic
Engagement and Volunteering Supplement (CEV)2 are below this threshold.
This document provides results from our evaluation of nonresponse in the 2021 CEV to
the CPS. Its purpose is to determine the existence of potential nonresponse bias in the
2021 CEV.
1.1

Overview of the CPS
The monthly CPS collects primarily labor force data about the civilian
noninstitutional population living in the United States. The institutional
population, which is excluded from the population universe, is composed
primarily of the population in correctional institutions and nursing homes (98
percent of the 4.0 million institutionalized people in Census 2010). Interviewers
ask questions concerning labor force participation about each member 15 years
old and over in sample households. Typically, the week containing the
nineteenth of the month is the interview week. The week containing the twelfth
is the reference week (i.e., the week about which the labor force questions are
asked).
The CPS uses a multistage probability sample based on the results of the
decennial census, with coverage in all 50 states and the District of Columbia. The
sample is continually updated to account for new residential construction.
When files from the most recent decennial census become available, the Census
Bureau gradually introduces a new sample design for the CPS.
Every ten years, the CPS first-stage sample is redesigned 3 reflecting changes
based on the most recent decennial census. In the first stage of the sampling
process, primary sampling units (PSUs)4 were selected for sample. In the 2010
sample design, the United States was divided into 1,987 PSUs. These PSUs were
then grouped into 852 strata. Within each stratum, a single PSU was chosen for

1

2

3
4

Starting in 2020, data collection faced extraordinary circumstances due to the onset of the COVID-19
pandemic as the Census Bureau suspended in-person interviews and closed both telephone contact centers.
This supplement was previously known as the Volunteering and Civic Life Supplement. It was changed to
better reflect the supplement’s focus on civic engagement and the sponsoring agency’s (Americorps) website
content.
For detailed information on the 2010 sample redesign, please reference Bureau of Labor Statistics (2014).
The PSUs correspond to substate areas (i.e., counties or groups of counties) that are geographically
contiguous.

2
DRB Clearance Number CBDRB-FY22-POP001-0150

the sample, with its probability of selection proportional to its population as of
the most recent decennial census. In the case of strata consisting of only one
PSU, the PSU was chosen with certainty.
1.2

Overview of the 2021 Civic Engagement and Volunteering Supplement to the
Current Population Survey
In September 2021, in addition to the basic CPS questions, interviewers asked
supplementary questions of randomly selected household member(s) 16 years
and older on being civically active. The universe for this supplement is persons
in households eligible for the basic CPS. Depending on the number of persons
found within an eligible household, one, two, or three respondents are randomly
selected for the CEV (U.S. Census Bureau, 2019b). This supplement allows for
proxy response.
The key estimates include:
 Number and percentage of persons 16 and over who volunteered in the
past year.
 Percentage of volunteers by range of hours spent volunteering in the past
year.
 Percentage of volunteers by number of organizations involved with.
 Number and percentage of persons aged 18 and over who voted in the
last local election (such as mayor or school board).
Key domains, or characteristics for which the key estimates are created, include:
 Sex
 Race / ethnicity
 Age
 Education levels

1.3

Discussion of Nonresponse in the 2021 Civic Engagement and Volunteering
Supplement to the Current Population Survey
Some degree of nonresponse bias and variance is a normal feature of almost all
statistical surveys. The CEV produces civic engagement and volunteering
estimates using the answers from responding persons. These civic engagement
and volunteering estimates will be biased if answers from respondents differ
from the potential answers of nonrespondents. The magnitude of the bias is a
function of the response rate and differences between respondents and
nonrespondents.
There were two ways that a person could be a nonrespondent to the CEV:

3
DRB Clearance Number CBDRB-FY22-POP001-0150





The entire household did not respond to CPS (the occupants were not
found at home after repeated calls or are unavailable for some other
reason).
The household responded to CPS but did not respond to the CEV
interview.

Because the CEV response rate is directly linked to the CPS response rate, the
CPS and CEV attempt to minimize nonresponse bias by increasing response rates
and adjusting weights for potential differences between respondents and
nonrespondents. We try to increase response rates within CPS by conducting
personal visit interviews for new and returning sample units, mailing advance
letters for all sample units, providing a Spanish language questionnaire for
potential respondents who do not speak English, allowing interpreters for
potential respondents who do not speak English or Spanish, training field
representatives to gain respondent cooperation, allowing proxy respondents in
special circumstances, and mailing follow-up letters to nonresponding
households. We also help minimize nonresponse bias by reducing respondent
burden for the CEV by limiting the length of the survey.
We reduce the effects of respondent/nonrespondent differences through
noninterview weighting adjustments. These adjustments group respondents and
nonrespondents into adjustment cells, and the weights of the nonrespondents
are reallocated to the respondents within the adjustment cells.
CPS noninterview adjustment cells are formed by noninterview cluster (NICL)
and central city status. The NICLs are created based on sample PSUs that are
similar in metropolitan status and population size within the same state (U.S.
Census Bureau, 2006). Metropolitan status is defined as metropolitan or
nonmetropolitan. Within metropolitan PSUs, a further breakdown into “central
city” and “not central city” is defined. This results in 127 NICLs and 214
adjustment cells. These variables were chosen for the noninterview adjustment
cells because they are thought to be correlated with the CPS variables of
interest.
CEV noninterview adjustment cells are defined using a special logistic regression
process. We use a logistic regression to group respondents and nonrespondents
into noninterview cells by like characteristics to calculate the nonresponse
adjustment.
Despite the measures taken to reduce nonresponse bias, there is likely still some
amount of nonresponse bias that we cannot correct without knowing the civic
engagement and volunteering activity of the nonrespondents.

4
DRB Clearance Number CBDRB-FY22-POP001-0150

2.

Methodology
2.1

Data
The data for this nonresponse bias analysis are from the September 2021 CEV to
the CPS and the September 2021 CPS. The U.S. Census Bureau conducts the CPS
every month, although this file has only September data. The September survey
uses two sets of questions, the basic CPS and a set of supplemental questions.
The CPS, sponsored jointly by the Census Bureau and the U.S. Bureau of Labor
Statistics, is the country’s primary source of labor force statistics for the entire
population. The Corporation for National and Community Service sponsors the
supplemental questions for September.
For a small number of variables, we had complete household information for all
sample households, including respondents and nonrespondents. These variables
were primarily limited to geographic and sampling data. There are also some
variables with partial information for the nonrespondents. Normal CPS
processing uses previous responses to demographic questions (when available)
and does not re-ask those that are unlikely to change from interview to
interview. Any variables that have never been answered are imputed using the
hot deck imputation method. Hot deck imputation assigns a value collected for a
person with similar characteristics to the missing value. Where possible,
nonrespondent actual values were used as opposed to edited or imputed values
in the comparison to respondents. The two exceptions are for tenure and
presence of children, because these characteristics were only available for CPS
respondents.

2.2

Weights
In the detailed weighting process for the CPS, base weights were adjusted with
the weighting control factor (WCF), which accounts for subsampling in the field
but does not include any nonresponse/noninterview or population coverage
adjustments. This subsampling-adjusted base weight is the weight used
throughout this report for household calculations for CPS.
When computing rates and distributions in this report for CEV persons, the CEV
adjusted base weights, which are the noninterview-adjusted weights from CPS,
were used. These weights do not take into account any adjustment for CEV subsampling within the household, and they do not adjust for CEV nonresponse
using logistic regression. Note that CEV weights are person level, not household
level, and include the CPS noninterview adjustment, which inflates weights back
up to the eligible weighted CPS sample.

5
DRB Clearance Number CBDRB-FY22-POP001-0150

The CEV also created weights based on self- or proxy-response, but these were
not considered to be in the scope of this analysis.
All numbers presented in the report are weighted unless otherwise noted. All
estimates in this report are rounded according to Census Bureau Disclosure
Review Board policies.
2.3

Universe for the Estimates
We analyzed nonresponse for persons aged 16 years and older in households. Our
analysis focused on person nonresponse within respondent demographics,
housing unit (HU) characteristics, and geography.

3.

Limitations
There are some limitations to this analysis which may affect the results. In particular:
1. Using past data to assign subgroup variables to nonrespondents is not necessarily
accurate for persons. Due to inmovers and outmovers, it is possible for
demographic variables that we get from past data to be out of date. However, we
do not believe our results need to be 100 percent accurate to show major
differences between respondents and nonrespondents. This assumes that the
demographics of neighborhoods do not change much in one and a half years.
2. Nonrespondents for CPS are never given the opportunity to respond to the CEV.

4.

Response Rates
The response rates tell us the percentage of eligible sample cases that responded to the
CPS and the CEV. It is useful to compare response rates for different subgroups to
understand the magnitude of potential biases.
We produced weighted and unweighted response rates for the 2021 CEV by key
domains and variables. Since the basic CPS nonresponse rate is a household-level rate
and the CEV supplement nonresponse rate is a person-level rate, we cannot combine
these rates to derive an overall nonresponse rate. Nonresponding households may have
more or fewer persons than interviewed ones, so combining these rates may lead to an
under- or overestimate of the true overall nonresponse rate for persons for the CEV
supplement.
Response rates are defined as:
𝑅𝑅 =

∑∈ 𝑤𝑅𝐷
∑∈ 𝑤 𝐷
6

DRB Clearance Number CBDRB-FY22-POP001-0150

where:
𝑤 = the appropriate weight (1 if unweighted) for the response rate calculation
𝑅 = the response indicator (1 for respondents, 0 for nonrespondents)
𝐷 = the domain indicator (1 if within domain of interest, 0 otherwise)
𝑠 = the set of all eligible households
Eligible households are all sample housing units (HUs) that did not receive Type B or
Type C (out-of-scope) outcome codes. Persons within group quarters (GQs) are treated
as individual HUs. The CPS interview data contains all eligible and non-eligible HUs, and
the CEV interview data contains only persons in eligible HUs to the CPS.
For the September 2021 CPS, there were approximately 58,500 occupied HUs eligible
for the household analysis. Of the 58,500 occupied households, 43,500 were
interviewed. Within the 43,500 households that were interviewed for CPS, 83,500
persons were eligible for the CEV. Due to subsampling within households, 73,000
persons were selected for interview and 48,000 responded to the CEV.
Table 1 shows that the weighted percentage of households where at least one person
responded to CPS is 75.22 percent. Out of all persons eligible and selected in those
responding households, 65.63 percent (weighted) responded to the CEV.
Table 1: 2021 Civic Engagement and Volunteering Unit Response Rates
Response Category

Count

Weighted
Sum*

Unweighted
Response
Rates

Weighted
Response
Rates

Sampled CPSA Households
67,500 139,000,000
Eligible CPSA Households
58,500 121,000,000
CPSA Household Response
43,500
91,010,000
74.97%
75.22%
CEVB Eligible Households
43,500
CEVB Eligible Persons
83,500 234,200,000
CEVB Selected PersonsC
73,000 202,900,000
B
CEV Person Response
48,000 133,200,000
65.79%
65.63%
Source: U.S. Census Bureau internal data from September 2021 Current Population Survey interviews.
A
CPS: Current Population Survey
B
CEV: Civic Engagement and Volunteering Supplement
C
The CEV Supplement subsamples within eligible households when three or more respondents are eligible.
Response rates are calculated out of selected eligible persons.
* May not sum to totals due to rounding. For Current Population Survey (CPS) households, CPS household
weights prior to noninterview adjustments were used. For Civic Engagement and Volunteering
Supplement (CEV) households, the CEV base weights, which are the noninterview-adjusted weights from
CPS, were used.

7
DRB Clearance Number CBDRB-FY22-POP001-0150

Table 2 shows weighted response rates for all CPS households by domain. The standard
error column shows the standard error of the response rate. Standard errors are
conditional on the sample and represent expected variability in the response process,
rather than traditional sampling error. Replicate weights were used to calculate the
standard errors to account for the sample design. The CPS uses the successive
difference replication method to calculate replicate weights. For detailed information
on variance estimation, please reference U.S. Census Bureau (2019a).

8
DRB Clearance Number CBDRB-FY22-POP001-0150

Table 2: Response Rates for September 2021 Current Population Survey Households
Characteristic
Type of Living Quarters
Housing Unit
Non-Housing UnitA
BlankB
Principal City Status
Principal City within CBSA/MSAC
Residual within CBSA/MSAC
Outside of a CBSA/MSAC
Region
Northeast
Midwest
South
West
Urban/Rural Status
Urban
Rural
Missing
Race of Reference Person
White Only
Black Only
Asian Only
Other Race/Two or More Races
Blank
Gender of Reference Person
Male
Female
Blank
Hispanic Origin of Reference Person
Hispanic
Non-Hispanic
Blank
Age of Reference Person
15-29
30-39
40-49
50-59
60-69
70+
Blank or Less than 15
Overall

Unweighted
Households*

Weighted
Households*

55,500
2,500
80

116,200,000
4,591,000
174,400

75.10%
80.80%
8.935%

0.2615%
0.9003%
3.472%

B
A
C

19,000
28,000
11,000

39,780,000
64,320,000
16,890,000

72.51%
76.08%
78.32%

0.4202%
0.3315%
0.8082%

C
B
A

10,000
11,500
22,000
15,000

21,230,000
26,740,000
46,050,000
26,970,000

67.85%
77.27%
75.71%
78.16%

0.6528%
0.5006%
0.4176%
0.4491%

C
A, B
B
A

44,500
13,000
400

96,660,000
23,520,000
811,900

74.49%
78.30%
72.90%

0.2830%
0.5586%
2.694%

B, C
A, C
C

38,000
5,500
2,300
1,400
11,000

77,470,000
12,220,000
5,370,000
2,357,000
23,570,000

89.08%
84.84%
88.27%
86.64%
20.58%

0.2088%
0.5197%
0.7121%
1.085%
0.5452%

A
C
A, B
A, B, C
D

24,500
26,000
8,100

50,880,000
53,370,000
16,750,000

87.71%
86.90%
0.06620%

0.2338%
0.2570%
0.03190%

A
B
C

5,800
43,500
9,000

14,060,000
88,210,000
18,730,000

86.06%
88.58%
4.139%

0.5547%
0.1935%
0.2925%

B
A
C

5,300
8,000
7,700
8,300
9,000
9,500

10,630,000
16,690,000
16,250,000
17,500,000
18,340,000
19,420,000
22,170,000
121,000,000

84.74%
85.60%
86.68%
87.90%
89.86%
93.67%
16.17%
75.22%

0.5962%
0.4498%
0.4530%
0.4027%
0.3603%
0.3185%
0.4301%
0.2586%

E
D, E
C, D, E
C
B
A
F

10,500
58,500

Weighted
Response Rate (%)

Standard Error
(%)

Significance
Grouping×

Source: U.S. Census Bureau internal data from September 2021 Current Population Survey interviews.
A
Non-Housing Units include quarters within rooming or boarding homes; non-permanent units in transient
hotels, motels, etc.; unoccupied sites for mobile homes, trailers, or tents; group quarters in school
dormitories; and other units that are not defined to be housing units.
B
Blank indicates that the living quarters type was either not identified or was identified with an invalid
code.
C
CBSA/MSA: Core-Based Statistical Area/Metropolitan Statistical Area
* May not sum to totals due to rounding. For weighted percent of total sample, reference

9
DRB Clearance Number CBDRB-FY22-POP001-0150

Table 5.
×
Within each characteristic, response rates identified with the same letter are not significantly different at
the α=0.10 level. A indicates the highest response rates, B indicates the next highest rates, etc. P-values
were adjusted for multiple comparisons within each demographic characteristic using the Tukey-Kramer
method (NIST/SEMATECH, 2013).

Excluding the blanks and missing values, one of the largest differences in response rates
for the CPS subgroups is 10.31 percent, seen for region, where the West has a response
rate of 78.16 percent versus 67.85 percent for the Northeast.5 Also having a large
spread in response rates (8.93 percent)6 are the age groups, having a range of 93.67
percent for reference persons aged 70+ to 84.74 percent for reference persons aged 1529.7
Additionally, households in living quarters that are non-HUs have a higher response rate
than households in HU living quarters. Households outside of a core-based statistical
area/metropolitan statistical area (CBSA/MSA) have the highest response rate within
principal city status, rural has a higher response rate than urban, White only has a
higher response rate than Black only,8 and non-Hispanic has a higher response rate than
Hispanic.
The response rate for blanks within the demographic subgroups is low because these
demographic items are collected during the interview, resulting in a large portion of the
household nonrespondents falling within these blank categories instead of the
categories where they belong. Any household with a blank value within the
demographic subgroups above indicates that the household has not previously
responded to the CPS or never provided responses to those demographic questions in
previous interviews. The nonresponse in the non-blank demographic categories is from
households which had previously responded to the CPS and provided a valid response
(non-blank) within the demographic category.

5

6
7

8

The response rates for the Midwest and the West are not significantly different and the response rates for the
Midwest and the South are not significantly different.
The response rate ranges for region and age of reference person are not significantly different.
The response rate for reference persons aged 15-29 is not significantly different than the response rates for
reference persons aged 30-39 and reference persons aged 40-49. The response rates for reference persons
aged 30-39 and reference persons aged 40-49 are not significantly different. The response rates for reference
persons aged 40-49 and reference persons aged 50-59 are not significantly different.
The response rate for other race/two or more races is not significantly different than the response rates for
White only, Black only, and Asian only.

10
DRB Clearance Number CBDRB-FY22-POP001-0150

Table 2 shows standard errors which facilitate hypothesis testing of differential response
rates. However, the practical significance of response rate differences is usually driven
more by the magnitude of the difference. Therefore, excluding blanks, if the
nonrespondents are different from respondents, region and age of reference person
have the most potential for bias.

11
DRB Clearance Number CBDRB-FY22-POP001-0150

Table 3 shows weighted response rates for all CEV selected persons by domain.

12
DRB Clearance Number CBDRB-FY22-POP001-0150

Table 3: Response Rates for 2021 Civic Engagement and Volunteering Supplement Selected Persons
Unweighted
Weighted
Weighted
Standard
Significance
Characteristic
Persons*
Persons*
Response
Error (%)
Grouping×
Rate (%)
Type of Living Quarters
70,000
195,300,000
A
65.60%
Housing Unit
0.3657%
3,200
7,603,000
A
66.60%
Non-Housing UnitA
1.425%
<15
3,1830
A+
37.65%
BlankB
19.56%
Principal City Status
22,000
64,580,000
66.59%
A
Principal City within CBSA/MSAC
0.5749%
36,500
110,400,000
64.66%
B
Residual within CBSA/MSAC
0.4748%
14,500
27,940,000
67.26%
A, B
Outside of a CBSA/MSAC
1.141%
Region
11,500
35,520,000
C
63.39%
Northeast
0.8040%
15,000
44,440,000
A
70.17%
Midwest
0.7036%
27,000
76,620,000
C
63.56%
South
0.5713%
19,500
46,340,000
B
66.43%
West
0.6178%
Urban/Rural Status
55,000
160,800,000
A
65.44%
Urban
0.3761%
17,500
40,830,000
A
66.54%
Rural
0.7840%
500
1,321,000
A
61.18%
Missing
3.516%
Race of Respondent
56,000
152,700,000
A
67.34%
White Only
0.4273%
7,200
21,930,000
B
61.86%
Black Only
0.7494%
3,800
11,780,000
B
61.42%
Asian Only
1.048%
2,100
4,766,000
A, B
65.92%
Other Race/Two or More Races
1.470%
3,700
11,790,000
C
54.67%
Blank
1.180%
Gender of Respondent
35,000
96,710,000
B
65.02%
Male
0.4052%
38,000
106,100,000
A
66.25%
Female
0.3864%
50
92,060
C+
5.401%
Blank
3.709%
Hispanic Origin of Respondent
9,000
29,440,000
B
62.21%
Hispanic
0.7065%
63,500
171,800,000
A
66.71%
Non-Hispanic
0.3822%
600
1,682,000
C
15.30%
Blank
2.312%
Age of Respondent
3,000
8,379,000
E
50.52%
16-19
1.107%
4,100
11,390,000
D
58.70%
20-24
0.9970%
11,000
30,240,000
C
67.08%
25-34
0.6025%
11,500
32,390,000
C
67.30%
35-44
0.6075%
9,900
28,160,000
C
66.93%
45-54
0.6155%
12,000
32,830,000
B
70.42%
55-64
0.5864%
11,000
29,990,000
A
74.53%
65-74
0.5648%
7,700
21,200,000
A
74.81%
75+
0.7041%
2,900
8,347,000
F
0.09620%
Blank or Less than 16
0.06830%
73,000
202,900,000
65.63%
Overall
0.3627%
Source: U.S. Census Bureau internal data from September 2021 Current Population Survey interviews.
A
Non-Housing Units include quarters within rooming or boarding homes; non-permanent units in transient
hotels, motels, etc.; unoccupied sites for mobile homes, trailers, or tents; group quarters in school
dormitories; and other units that are not defined to be housing units.
B
Blank indicates that the living quarters type was either not identified or was identified with an invalid code.
C
CBSA/MSA: Core-Based Statistical Area/Metropolitan Statistical Area

13
DRB Clearance Number CBDRB-FY22-POP001-0150

* May not sum to totals due to rounding. For weighted percent of total sample, reference

14
DRB Clearance Number CBDRB-FY22-POP001-0150

+
×

Table 6.
Exercise caution: The sample size is extremely small, leading to unreliable estimates.
Within each characteristic, response rates identified with the same letter are not significantly different at
the α=0.10 level. A indicates the highest response rates, B indicates the next highest rates, etc. P-values
were adjusted for multiple comparisons within each demographic characteristic using the Tukey-Kramer
method (NIST/SEMATECH, 2013).

For the CEV person estimates, there is no significant differences among the urban/rural
status response rates or the type of living quarters response rates. Excluding the blanks
and missing values, the largest difference in response rates for the CEV subgroups is
25.29 percent, seen for the age of respondent, where ages 75+ has a response rate of
74.819 percent versus 50.52 percent for ages 16-19. This large difference may be
attributed to the greater number of subgroups within the age category compared to the
other characteristics. Comparatively, the next largest difference in response rates is
6.7810 percent for region, between the Midwest at 70.17 percent and the Northeast at
63.3911 percent.
Additionally, persons living in a principal city have a higher response rate than persons
living in the residual of the city (although both groups were not significantly different
from those living outside of principal city area), White only respondents have a higher
response rate than Black only and Asian only respondents12, female respondents have a
higher response rate than male respondents, and non-Hispanic has a higher response
rate than Hispanic.
Again, although

9
10
11
12

The response rate for ages 75+ is not significantly different than the response rate for ages 65 to 74.
The response rate ranges for region and race of respondent are not significantly different.
The response rate for the Northeast is not significantly different than the response rate for the South.
The response rate for other race/two or more races is not significantly different than the response rate for
White only, Black only, and Asian only.

15
DRB Clearance Number CBDRB-FY22-POP001-0150

Table 3 shows standard errors which facilitate hypothesis testing, the practical
significance of response rate differences is driven more by the magnitude of the
difference than the sample size. Therefore, excluding blanks, if the nonrespondents are
different from respondents, age of respondent has the most potential for bias.

16
DRB Clearance Number CBDRB-FY22-POP001-0150

Table 4 shows weighted response rates for all CEV households by domain for
characteristics that were only available for CPS respondents.

17
DRB Clearance Number CBDRB-FY22-POP001-0150

Table 4: Response Rates for 2021 Civic Engagement and Volunteering Supplement Persons for
Characteristics Only Available for Responding Current Population Survey Households
Unweighted
Weighted
Weighted
Standard
Significance
Characteristic
Persons*
Persons*
Response
Error (%)
Grouping×
Rate (%)
Tenure (Edited)
52,000
142,100,000
Owned or Mortgage
66.19%
0.4313%
A
20,500
58,800,000
Rented for Cash
64.31%
0.5284%
B
800
1,996,000
No Cash Rent
65.14%
2.517%
A, B
Family Income
10,500
28,340,000
Less than $30,000
75.40%
0.6055%
A
20,000
53,810,000
$30,000-$74,999.99
72.77%
0.5185%
B
15,500
43,240,000
$75,000-$149,999.99
71.89%
0.5879%
B, C
9,000
26,450,000
$150,000+
69.97%
0.7584%
C
3,400
9,751,000
Blank or Don’t Know
42.76%
1.565%
D
15,000
41,330,000
Refused
45.72%
0.6542%
D
Household Type
41,500
114,500,000
Husband/Wife Primary Family
64.47%
0.4424%
B
Unmarried Householder
12,500
36,440,000
60.60%
0.6223%
C
Primary Family
19,000
51,890,000
Primary Individual
71.77%
0.4816%
A
Group Quarters
50
125,700
53.49%
11.68%
A, B, C +
Child(ren) Present (Edited)
54,000
149,800,000
No
66.81%
0.4039%
A
19,000
53,130,000
Yes
62.32%
0.5259%
B
Educational Attainment of Respondent (ages 25+, excludes blank ages)
Less than High School
4,900
14,370,000
65.72%
0.8231%
B
High School Graduate
17,500
48,310,000
66.06%
0.5359%
B
9,800
26,140,000
Some College
71.63%
0.6537%
A
College Degree
30,500
85,980,000
72.31%
0.4339%
A
Nativity and Citizenship of Respondent
Native Citizen
64,000
172,500,000
66.23%
0.3715%
A
Foreign-Born Citizen
4,900
16,440,000
63.12%
0.9359%
B
Foreign-Born Non-Citizen
4,300
14,030,000
61.23%
1.031%
B
Measure of Labor Force Participation of Respondent
Employed
44,000
123,000,000
64.17%
0.4404%
B
Unemployed
1,500
5,014
67.02%
1.518%
A, B
Not in Labor Force
27,000
74,380,000
68.39%
0.4330%
A
Blank
150
500,800
2.567%
1.313%
C
73,000
202,900,000
65.63%
Overall
0.3627%
Source: U.S. Census Bureau internal data from September 2021 Current Population Survey interviews.
* May not sum to totals due to rounding. For weighted percent of total sample, reference

18
DRB Clearance Number CBDRB-FY22-POP001-0150

Table 7.
+
Exercise caution: The sample size is small, leading to unreliable estimates.
×
Within each characteristic, response rates identified with the same letter are not significantly different at
the α=0.10 level. A indicates the highest response rates, B indicates the next highest rates, etc. P-values
were adjusted for multiple comparisons within each demographic characteristic using the Tukey-Kramer
method (NIST/SEMATECH, 2013).

For the CEV person estimates of characteristics only available for CPS respondents,
excluding the blanks, refusals, and GQs, there were significant differences in response
rates within all of the characteristics. The largest difference in response rates for the
subgroups is 11.2 percent, seen for the household type variable, where the rates range
from 60.60 percent for unmarried household, primary individuals to 71.77 percent for
primary individuals.
For the tenure category, owned or mortgage has a response rate higher than rented for
cash, but both rates are not significantly different than the response rate for no cash
rent. Additionally, persons with a family income less than $30,000 have a higher
response rate than persons with other family incomes, persons with no children present
in their household have a higher response rate than when children are present. The
response rates for persons with a college degree and persons with some college are not
significantly different, but both are higher than the other education groups 13. Native
citizens have a higher response rate than those who are foreign-born, whether or not
they are a citizen14. Persons not in the labor force have a higher response rate than
employed persons, but both of these groups are not significantly different from
unemployed persons.
Again, although

13

14

The response rate for persons with less than high school education is not significantly different from the
response rate for persons who have graduated from high school.
The response rate for foreign-born citizen is not significantly different from the response rate for foreign-born
non-citizens.

19
DRB Clearance Number CBDRB-FY22-POP001-0150

Table 4 shows standard errors which facilitate hypothesis testing, the practical
significance of response rate differences is driven more by the magnitude of the
difference than the sample size. Therefore, excluding blanks, if the nonrespondents are
different from respondents, among the categories in

20
DRB Clearance Number CBDRB-FY22-POP001-0150

Table 4, household type has the most potential for bias.
5.

Respondent Distributions
Respondent and nonrespondent distributions show the relative percent of members of
a domain subset within respondents and nonrespondents separately. This is different
than the response rates, which are the relative percent of respondents within the
different domain subsets. We used chi-square tests to determine if the respondent and
nonrespondent distributions differed.
Respondent distributions are defined as:
𝑅𝐷 =

∑∈ 𝑤𝑅𝐷
∑∈ 𝑤𝑅

This definition assumes the same eligibility criteria, weights, and indicators as the
response rate calculations in the previous section. Nonrespondent distributions use the
same formula, but with the 𝑅 variable indicating nonrespondents instead of
respondents. The chi-square test statistics were calculated using replicate weights to
account for the sample design.

21
DRB Clearance Number CBDRB-FY22-POP001-0150

Table 5 shows the percent of total sample distribution as well as comparisons of
respondent and nonrespondent distributions for CPS households within the different
domain subgroups.

22
DRB Clearance Number CBDRB-FY22-POP001-0150

Table 5: Respondent and Nonrespondent Distributions for September 2021
Current Population Survey Households
Household
Respondent
Percentage*

Household
Nonrespondent
Percentage*

96.06%
3.795%
0.1441%

95.91%
4.760%
0.01712%

96.53%
2.940%
0.5296%

201.7 (2)
[32.21.(1)

< 0.0001
< 0.0001]

32.88%
53.16%
13.96%

31.70%
53.77%
14.54%

36.47%
51.31%
12.22%

56.41 (2)

< 0.0001

17.55%
22.10%
38.06%
22.29%

15.83%
22.70%
38.31%
23.16%

22.77%
20.27%
37.31%
19.65%

217.2 (3)

< 0.0001

79.89%
19.44%
0.6710%

79.12%
20.23%
0.6503%

82.24%
17.02%
0.7338%

42.81 (2)
[35.86 (1)

< 0.0001
< 0.0001]

64.03%
10.10%
4.438%
1.948%
19.48%

75.83%
11.39%
5.209%
2.244%
5.329%

28.23%
6.179%
2.101%
1.051%
62.44%

17,110 (4)
[76.41 (3)

< 0.0001
< 0.0001]

42.05%
44.11%
13.84%

49.03%
50.96%
0.01218%

20.86%
23.33%
55.82%

30,180 (2)
[6.188 (1)

< 0.0001
0.0129]

11.62%
72.90%
15.48%

13.29%
85.85%
0.8517%

6.535%
33.59%
59.88%

8.785%
13.80%
13.43%
14.47%
15.16%
16.05%
18.32%
100%

9.900%
15.70%
15.47%
16.90%
18.12%
19.98%
3.938%
100%

5.411%
8.017%
7.221%
7.067%
6.203%
4.100%
61.98%
100%

Percentage
of Total
Sample*

Characteristic
Type of Living Quarters
Housing Unit
Non-Housing UnitA
BlankB
Principal City Status
Principal City within CBSA/MSAC
Residual within CBSA/MSAC
Outside of a CBSA/MSAC
Region
Northeast
Midwest
South
West
Urban/Rural Status
Urban
Rural
Missing
Race of Reference Person
White Only
Black Only
Asian Only
Other Race/Two or More Races
Blank
Gender of Reference Person
Male
Female
Blank
Hispanic Origin of Reference Person
Hispanic
Non-Hispanic
Blank
Age of Reference Person
15-29
30-39
40-49
50-59
60-69
70+
Blank or Less than 15
Overall

Chi-Square
Statistic (df)

P-value

18,330 (2)
[21.20 (1)

< 0.0001
< 0.0001]

19,410 (6)
[321.9 (5)

< 0.0001
< 0.0001]

Source: U.S. Census Bureau internal data from September 2021 Current Population Survey interviews.
A
Non-Housing Units include quarters within rooming or boarding homes; non-permanent units in transient
hotels, motels, etc.; unoccupied sites for mobile homes, trailers, or tents; group quarters in school
dormitories; and other units that are not defined to be housing units.
B
Blank indicates that the living quarters type was either not identified or was identified with an invalid code.
C
CBSA/MSA: Core-Based Statistical Area/Metropolitan Statistical Area
* May not sum to totals due to rounding.
[] The values within brackets are the chi-square statistic, df, and p-value when the blanks/missings are
excluded from the chi-square test.

23
DRB Clearance Number CBDRB-FY22-POP001-0150

The chi-square tests for CPS households showed significant differences (at the α=0.10
level) between respondent and nonrespondent distributions for all characteristics.
Simply looking at the distributions for the race, gender, Hispanic origin, and age of
reference person, it is apparent that there are large differences between the
respondent and nonrespondent distributions, which correspond to the magnitude of the
chi-square test statistics (17,110, 30,180, 18,330, and 19,410, respectively). When
blanks are excluded from the chi-square test, the chi-square statistic decreases but the
distributions are still significantly different for all characteristics. Note: The chi-square
tests only indicate that the distributions of respondents and nonrespondents differ but
do not necessarily indicate a nonresponse bias problem. These differences will only
cause bias if the respondents and nonrespondents report differing rates of civic
engagement and volunteering activity.
Even though there are significant differences between the respondents and
nonrespondents, the differences might not be large enough to cause meaningful
differences in estimates. Furthermore, weighting adjustments might also minimize the
impact of some differences. Because the CPS noninterview adjustments take NICL and
central city status into account, the principal city status and region differences may be
reduced within those adjustments.

24
DRB Clearance Number CBDRB-FY22-POP001-0150

Table 6 shows the percent of total sample distribution as well as comparisons of

respondent and nonrespondent distributions for CEV persons within the different
domain subgroups.

25
DRB Clearance Number CBDRB-FY22-POP001-0150

Table 6: Respondent and Nonrespondent Distributions for 2021 Civic Engagement and
Volunteering Supplement Persons
Characteristic
Type of Living Quarters
Housing Unit
Non-Housing UnitA
BlankB
Principal City Status
Principal City within CBSA/MSAC
Residual within CBSA/MSAC
Outside of a CBSA/MSAC
Region
Northeast
Midwest
South
West
Urban/Rural Status
Urban
Rural
Missing
Race of Respondent
White Only
Black Only
Asian Only
Other Race/Two or More Races
Blank
Gender of Respondent
Male
Female
Blank
Hispanic Origin of Respondent
Hispanic
Non-Hispanic
Blank
Age of Respondent
16-19
20-24
25-34
35-44
45-54
55-64
65-74
75+
Blank or Less than 15
Overall

Percentage
of Total
Sample*

Person
Respondent
Percentage*

Person
Nonrespondent
Percentage*

96.24%
3.747%
0.01568%

96.19%
3.802%
0.008997%

96.33%
3.641%
0.02845%

31.82%
54.41%
13.77%

32.29%
53.60%
14.11%

17.51%
21.90%
37.76%
22.84%

Chi-Square
Statistic (df)

P-value

2.502 (2)
[0.4895 (1)

0.2862
0.4841]

30.93%
55.95%
13.16%

8.753 (2)

0.0126

16.91%
23.41%
36.57%
23.11%

18.65%
19.01%
40.04%
22.31%

70.83 (3)

< 0.0001

79.23%
20.12%
0.6511%

78.99%
20.40%
0.6070%

79.67%
19.59%
0.7355%

3.559 (2)
[1.796 (1)

0.1687
0.1802]

75.23%
10.80%
5.805%
2.349%
5.811%

77.19%
10.18%
5.432%
2.359%
4.840%

71.50%
11.99%
6.516%
2.329%
7.665%

185.9 (4)
[77.84 (3)

< 0.0001
< 0.0001]

47.66%
52.30%
0.04537%

47.21%
52.79%
0.003733%

48.51%
51.63%
0.1249%

80.53 (2)
[15.28 (1)

< 0.0001
< 0.0001]

14.51%
84.66%
0.8287%

13.75%
86.05%
0.1932%

15.95%
82.00%
2.042%

4.129%
5.615%
14.90%
15.96%
13.88%
16.18%
14.78%
10.45%
4.113%
100%

3.178%
5.022%
15.23%
16.37%
14.15%
17.36%
16.78%
11.91%
0.006030%
100%

5.945%
6.748%
14.27%
15.19%
13.36%
13.93%
10.95%
7.658%
11.96%
100%

236.6 (2)
[36.91 (1)

< 0.0001
<0.0001]

4,779 (8)
[722.3 (7)

< 0.0001
<0.0001]

Source: U.S. Census Bureau internal data from September 2021 Current Population Survey interviews.
A
Non-Housing Units include quarters within rooming or boarding homes; non-permanent units in transient
hotels, motels, etc.; unoccupied sites for mobile homes, trailers, or tents; group quarters in school
dormitories; and other units that are not defined to be housing units.
B
Blank indicates that the living quarters type was either not identified or was identified with an invalid
code.
C
CBSA/MSA: Core-Based Statistical Area/Metropolitan Statistical Area
* May not sum to totals due to rounding.
[] The values within brackets are the chi-square statistic, df, and p-value when the blanks/missings are
excluded from the chi-square test.

26
DRB Clearance Number CBDRB-FY22-POP001-0150

The chi-square tests for CEV persons showed significant differences (at the α=0.10 level)
for the distributions of all variables except urban/rural status and type of living quarters.
Simply looking at the distributions for age of the respondent, it is apparent that there
are large differences between the respondent and nonrespondent distributions, which
correspond to the magnitude of the chi-square test statistic (4,779).
As mentioned for CPS household respondent distributions, the chi-square tests only
indicate that the distributions of respondents and nonrespondents differ but do not
necessarily indicate a nonresponse bias problem. Furthermore, weighting adjustments
might minimize the impact of some differences. Because the CEV noninterview
adjustments take NICL and central city status into account, the principal city status and
region differences may be reduced within those adjustments.

27
DRB Clearance Number CBDRB-FY22-POP001-0150

Table 7 shows the percent of total sample distribution as well as comparisons of
respondent and nonrespondent distributions for CEV persons within the different
domain subgroups for characteristics that were only available for CPS respondents.

28
DRB Clearance Number CBDRB-FY22-POP001-0150

Table 7: Respondent and Nonrespondent Distributions for 2021 Civic Engagement and Volunteering
Supplement Households for Characteristics Only Available for Responding CPS Households
Percentage of
Person
Person
Chi-Square
Characteristic
Total
Respondent
Nonrespondent
P-value
Statistic (df)
Sample*
Percentage*
Percentage*
Tenure (Edited)
Owned or Mortgage
70.04%
70.63%
68.91%
10.39 (2)
0.0056
Rented for Cash
28.98%
28.39%
30.10%
No Cash Rent
0.9838%
0.9764%
0.9980%
Family Income
Less than $30,000
13.97%
16.05%
9.997%
$30,000-$74,999.99
26.52%
29.40%
21.01%
$75,000-$149,999.99
21.31%
23.34%
17.43%
1,980 (5)
< 0.0001
$150,000+
13.03%
13.89%
11.39%
[36.60 (3)
< 0.0001]
Blank or Don’t Know
4.806%
3.131%
8.004%
Refused
20.37%
14.19%
32.17%
Household Type
Husband/Wife Primary
Family
56.41%
55.40%
58.32%
Unmarried Householder
244.1 (3)
< 0.0001
Primary Family
17.96%
16.58%
20.59%
Primary Individual
25.57%
27.97%
21.01%
Group Quarters
0.06185%
0.05048%
0.08384%
Child(ren) Present (Edited)
No
73.82%
75.14%
71.29%
64.51 (1)
< 0.0001
Yes
26.18%
24.86%
28.71%
Educational Attainment of Respondent (ages 25+, excluding blank ages)
Less than High School
8.221%
7.725%
9.375%
High School Graduate
27.63%
26.10%
31.20%
175.5 (3)
< 0.0001
Some College
14.96%
15.32%
14.11%
College Degree
49.19%
50.86%
45.31%
Nativity and Citizenship of Respondent
Native Citizen
84.98%
85.76%
83.50%
Foreign-Born Citizen
8.103%
7.793%
8.694%
34.45 (2)
< 0.0001
Foreign-Born Non-Citizen
6.914%
6.450%
7.801%
Measure of Labor Force Participation Status of Respondent
Employed
60.63%
59.28%
63.21%
Unemployed
2.471%
2.523%
2.372%
275.8 (3)
< 0.0001
Not in Labor Force
36.65%
38.19%
33.717%
[73.25 (2)
< 0.0001]
Blank
0.2468%
0.009655%
0.6997%
Overall
100%
100%
100%
Source: U.S. Census Bureau internal data from September 2021 Current Population Survey interviews.
* May not sum to totals due to rounding.
[] The values within brackets are the chi-square statistic, df, and p-value when the blanks and refusals are
excluded from the chi-square test.

The chi-square tests for CEV person estimates of characteristics only available for CPS
respondents showed significant differences (at the α=0.10 level) for the distributions of
all variables. Simply looking at the distributions for family income, it is apparent that
there are large differences between the respondent and nonrespondent distributions,
which correspond to the magnitude of the chi-square test statistic (1,980). Even after
29
DRB Clearance Number CBDRB-FY22-POP001-0150

excluding the blanks and refusals, there is still a significant difference between the
respondent and nonrespondent distributions for family income and measure of labor
force participation status of respondent.
As mentioned previously, the chi-square tests only indicate that the distributions of
respondents and nonrespondents differ but do not necessarily indicate a nonresponse
bias problem. Furthermore, weighting adjustments might minimize the impact of some
differences.
6.

Conclusions
This analysis found evidence of potential nonresponse bias for both CPS households and
CEV persons. For CPS, there is potential nonresponse bias for all investigated
characteristics. For CEV, there is potential nonresponse bias for all investigated
characteristics except type of living quarters and urban/rural status.
Excluding the blanks and missing values, one of the largest differences in response rates
for the CPS subgroups is 10.31 percent, seen for region, where the West has a response
rate of 78.16 percent versus 67.85 percent for the Northeast.15 For respondent and
nonrespondent distributions within CPS households, the largest differences are seen
within region and age of reference person.
Excluding the blanks and missing values, the largest difference in response rates for the
CEV subgroups is 25.29 percent, seen for the age of respondent, where ages 75+ has a
response rate of 74.8116 percent versus 50.52 percent for ages 16-19. For respondent
and nonrespondent distributions within CEV persons, the largest difference is also seen
within age of respondent.
Among the estimates for the CEV persons for characteristics only available for CPS
respondents, one of the largest differences in response rates and between the
respondent and nonrespondent distributions are seen within household type.
Using the information learned from this analysis, discussions should be had with the
sponsor regarding enhancements to the CEV weighting process. However, nonresponse
weighting adjustments are already performed using all of the key domains and are
expected to correct for differences between respondents and nonrespondents.
Research could be done into the possible inclusion of other characteristics into the
logistic regression for the CEV, such as veteran status.
Note: The weighting process for basic CPS does not allow us to expand upon the
characteristics used.

15
16

The response rate ranges for region and age of reference person are not significantly different.
The response rate for ages 75+ is not significantly different than the response rate for ages 65 to 74.

30
DRB Clearance Number CBDRB-FY22-POP001-0150

Another recommendation would be to reevaluate or shorten the CEV questionnaire to
potentially allow for more supplement respondents. Specifically, reducing the number
of respondents selected per household and reducing the number of callbacks might
help.
7.

References
Bureau of Labor Statistics. (2014). “Redesign of the Sample for the Current Population
Survey.” http://www.bls.gov/cps/sample_redesign_2014.pdf
NIST/SEMATECH. (2013). “NIST/SEMATECH e-Handbook of Statistical Methods.”
http://www.itl.nist.gov/div898/handbook/prc/section4/prc471.htm
Office of Management and Budget. (2006). “Standards and Guidelines for Statistical Surveys.”
https://georgewbushwhitehouse.archives.gov/omb/inforeg/statpolicy/standards_stat_surveys.pdf
U.S. Census Bureau. (2019a). Current Population Survey: Design and Methodology. Technical
Paper 77. Washington, DC: Government Printing Office. https://www2.census.gov/programssurveys/cps/methodology/CPS-Tech-Paper-77.pdf
U.S. Census Bureau. (2019b). “September 2019 Volunteering and Civic Life Supplement
Technical Documentation.” https://www2.census.gov/programssurveys/cps/techdocs/cpssept19.pdf

All online references last accessed on August 23, 2022.

31
DRB Clearance Number CBDRB-FY22-POP001-0150