AC Longitudinal Study_OMB 12-06-06_Section A

OMB Clearance
Package – Section A

Phase III of the
Longitudinal
Evaluation of
AmeriCorps

Cambridge, MA
Lexington, MA
Hadley, MA
Bethesda, MD
Chicago, IL

August 28, 2006

Prepared for
Lillian Dote
Corporation for National and
Community Service
1201 New York Avenue, NW
Washington, DC 20525
Abt Associates Inc.
55 Wheeler Street
Cambridge, MA 02138

Prepared by
Abt Associates Inc.

Contents
A.

Justification
2
A.1.
Circumstances Requiring the Collection of Data ......................................................... 2
A.2.
Purposes and Uses of the Data ..................................................................................... 4
A.3.
Use of Information Technology To Reduce Burden .................................................... 6
A.4.
Efforts To Identify Duplication.................................................................................... 7
A.5.
Small Business ............................................................................................................. 7
A.6.
Consequences of Not Collecting the Information ........................................................ 7
A.7.
Special Circumstances Justifying Inconsistencies With Guidelines in 5 CFR 1320.6. 7
A.8.
Consultation Outside the Agency................................................................................. 7
A.9.
Payments or Gifts to Respondents................................................................................ 9
A.10. Assurance of Confidentiality........................................................................................ 9
A.11. Questions of a Sensitive Nature ................................................................................. 10
A.12. Estimate of Response Burden..................................................................................... 10
A.13. Estimate of Total Capital and Startup Costs/Operation and Maintenance Costs to
Respondents or Record-Keepers ................................................................................ 11
A.14. Estimates of Costs to the Federal Government .......................................................... 12
A.15. Changes in Burden ..................................................................................................... 12
A.16. Plans for Publication, Analysis, and Schedule ........................................................... 12
A.17. Approval To Not Display Expiration Date................................................................. 13
A.18. Exceptions to Item 19 of OMB Form 83-1 ................................................................ 13

Phase III of the Longitudinal Evaluation of AmeriCorps
Corporation for National and Community Service

Contents

i

A. Justification
A.1. Circumstances Requiring the Collection of Data
The Longitudinal Evaluation of AmeriCorps is a quasi-experimental study designed to measure the
outcomes and impacts of national service on individuals who served in AmeriCorps programs funded by
the Corporation for National and Community Service. The Corporation for National and Community
Service (the Corporation) is a government corporation established in 1993 by the National and
Community Service Act. The Corporation’s mission is to engage Americans of all ages and backgrounds
in community-based national service that is intended to address the Nation’s education, public safety,
human, and environmental needs in order to achieve direct and demonstrable results. The Corporation’s
programs and public-private partnerships involve national and community-based service organizations,
corporations, foundations, colleges, universities, K-12 schools, and State and municipal governments.
One of the major goals of the Corporation is that the lives of those who serve will be improved through
their service experience. A central purpose of the agency and its programs is to foster citizenship and
development for those who serve. The Corporation has an interest in the study of AmeriCorps because the
results of this study will enable the Corporation to determine the extent to which service in AmeriCorps:
1) fosters a lifetime of citizenship and civic engagement among members who serve (Executive Order
13331 of February 27, 2004); and 2) enhances skills development and educational level and opportunities
for participants (National and Community Service Act of 1990, Section 199H 42 USC 12655). Also, this
study complements the Corporation’s strategic plan, specifically addressing our guiding principles: to
measure and continually improve our programs’ benefits to service recipients, participants, community
organizations, and our national culture of service; and to support continued civic engagement, leadership,
and public service careers for our programs’ participants and community volunteers.
This longitudinal study is based on the hypothesis that participation in national service may lead to
measurable outcomes among AmeriCorps participants, as compared to a comparison group who did not
participate in AmeriCorps. The proposed phase of the study (Phase III) is intended to assess the impacts
of participation in AmeriCorps on its members’ civic engagement, education, employment, and life skill
outcomes seven years after program enrollment. The longitudinal study is tracking individuals who
enrolled in AmeriCorps during the 1999-2000 program year.
There were two earlier phases of the study, also conducted by Abt Associates for the Corporation, which
examined shorter-term impacts of participation in service and found statistically significant, positive
outcomes for members. In particular, the effects of participation were especially strong for measures of
civic engagement, a key priority for the Corporation. This new phase of the study, Phase III, will assess
whether the outcomes identified in previous phases are sustained over time, and whether new areas of
AmeriCorps impact will appear.
AmeriCorps is the national service initiative funding a network of community-based programs. In
exchange for a year of full-time service (1,700 hours per year, or intensive part-time), AmeriCorps
members receive a stipend and earn an education award that may be used to pay for higher education,
vocational training, or existing student loans. This study is looking at two AmeriCorps programs:

Phase III of the Longitudinal Evaluation of AmeriCorps
Corporation for National and Community Service

Collection of Information Employing Statistical Methods

2

•

AmeriCorps*State and National Programs are operated by national multi-state or local nonprofit organizations and community-based nonprofit organizations. Approximately 67,500
State and National members serve annually in all 50 states, three territories, and seven tribes.
The minimum age for participating is 17; half of the members are in the age range from 22 to
30. There is considerable flexibility in program structure and services provided. FY 2005
funding from AmeriCorps*State and National programs totaled $275 million. State and
National programs address community needs in one of four issue areas: education, public
safety, human services, and the environment. Types of service AmeriCorps members perform
include:
Tutoring teens and elementary school students;
Assisting crime victims or starting neighborhood crime watches;
Turning vacant lots into neighborhood parks;
Leading community health awareness campaigns;
Providing assistance and companionship to homebound elderly or individuals with
disabilities; and
Providing homeland security and public safety services.

•

The National Civilian Community Corps (NCCC) is a ten-month, team-based full-time
residential program for men and women between the ages of 18 and 24. It combines the best
practices of civilian service with the best aspects of military service, including leadership
training and team building. Members serve in diverse teams of 10 to 14 individuals. Priority
is given to projects in homeland security and disaster relief. Teams are based at regional
campuses and are assigned to projects in their respective regions. Approximately 1,000
members serve annually in AmeriCorps*NCCC.

The purpose of the Longitudinal Evaluation of AmeriCorps is to measure longitudinally the outcomes and
impacts of national service on members who serve in AmeriCorps*State and National and the National
Civilian Community Corps (NCCC). In order to be able to generalize to member experiences, the
Longitudinal Evaluation of AmeriCorps is based on a representative sample of first-term, full-time
members from the State and National and the NCCC programs. The study addresses questions relating to
how the experience of serving others may affect members in terms of the kind of person and citizen she or
he will become. In addition to these questions regarding character and ethics, it investigates hypothesized
effects for members in their education and employment experience and in selected life skills. This project
longitudinally measures the outcomes and impacts of national service on members. In order to do this, the
study uses comparison groups to assess the effects of service on members. The comparable individuals
are those who inquired about AmeriCorps, but did not actually enroll in the program.
To summarize, the Longitudinal Evaluation of AmeriCorps study has two components:
•

A national impact study of over 1,700 AmeriCorps*State and National members, comparing
changes in outcomes for a sample of State and National members from a nationally
representative sample of programs with changes in outcomes for a national comparison
group. The program group was recruited from a sample of 108 programs that were randomly
selected from the universe of State and National programs in operation during the 1999/2000
program year. The national comparison group was selected from the pool of individuals
making inquiries about AmeriCorps through the Corporation’s national inquiry line. The

Phase III of the Longitudinal Evaluation of AmeriCorps
Corporation for National and Community Service

Collection of Information Employing Statistical Methods

3

comparison group members were screened to ensure that the individuals both had a
propensity to serve and had not enrolled in an AmeriCorps program.
•

A national impact study of over 476 AmeriCorps*NCCC members entering the program
between September 8, 1999 and January 31, 2000, comparing changes in outcomes for the
population of NCCC members with changes in outcomes for a national comparison sample.
The program sample consists of all members from three of the five NCCC region campuses.
The comparison was recruited from a national wait-list pool of applicants for the NCCC
program who applied for the program at approximately the same time as the treatment group
members. These are individuals who either were alternates to the program or were invited to
enroll but declined.

Impacts to be assessed include, but are not limited to: civic engagement, propensity to engage in volunteer
service, and the degree to which the program affected members’ personal and professional development.
The research design is to measure these outcomes over time for a nationally representative cohort of
AmeriCorps members and their comparison group in the longitudinal study. The baseline data collection
(1999-2000; OMB 3045-0060, expired September 30, 2002) enabled us to describe the status of
AmeriCorps members prior to their participation in service. Both the post-program survey (2000-2001;
OMB 3045-0070, expired August 31, 2003) and the post-program supplemental survey (2003-2004;
OMB 3045-0077, expired August 31, 2003) assessed the short-term impacts of participation on member
outcomes. The current request for OMB approval covers the 2006-2007 follow-up survey of panel of
members in the longitudinal study approximately six years after the cohort completed their term of
service.

A.2. Purposes and Uses of the Data
This submission request for approval pertains to the Phase III follow-up survey. There are two
overarching research questions for the study:
1.

How do outcomes change over time?

2.

What factors explain variation in these outcomes at different stages of time?

To answer these questions, one objective of this study is to trace changes in AmeriCorps members’ lives
after their participation in the program. We will then be able to produce an accurate understanding of the
nature of change in the outcomes. We also will be able to model the effects of key demographic and
policy-relevant factors on these outcomes.
The second objective of this study is to attribute changes in AmeriCorps members to participation. As in
all impact studies, the study derives impact estimates by comparing what happened to AmeriCorps
members to what would have happened in the absence of the program. This quasi-experimental study
design uses the experiences of a comparison group of similar individuals who did not join AmeriCorps as
a proxy for what would have happened to AmeriCorps members in the absence of the program, after
controlling for demographic characteristics and prior service experiences. For the State and National
programs, a comparison group was constructed by drawing a sample of 1,524 individuals who made

Phase III of the Longitudinal Evaluation of AmeriCorps
Corporation for National and Community Service

Collection of Information Employing Statistical Methods

4

inquiries to the Corporation for National and Community Service about enrolling in national service, but
who did not enroll.
For the NCCC program, a comparison group (N=401) was constructed from a national wait list of eligible
individuals who were interested in entering the program, had completed an interview with program staff,
and were determined eligible. However, the individuals in the NCCC comparison group declined to enter
or were not accepted into the program due to the limited number of slots. As a way to control for
differences in the two groups at baseline, the study compares changes in outcomes for AmeriCorps
members with changes in outcomes over the same time period for comparison group members after
controlling for demographic characteristics and prior service experiences.
The outcomes in this study focus on the following domains:
•
•
•
•
•

Civic values and involvement;
Political participation;
Educational skills, aspirations, and achievements;
Employment skills, aspirations, and achievements; and
Life skills, values, and behaviors.

These outcomes were derived from previous work with national service programs, a review of the
literature of national service outcomes, site visits to individual AmeriCorps programs, and consultation
with our Technical Working Group.
The research design is to measure these outcomes over time for a nationally representative cohort of
AmeriCorps members and their comparison group in the longitudinal study. The baseline data collection
(1999-2000; OMB 3045-0060, expired September 30, 2002) enabled us to describe the status of
AmeriCorps members prior to their participation in service. Both the post-program survey (2000-2001;
OMB 3045-0070, expired August 31, 2003) and the post-program supplemental survey (2003-2004;
OMB 3045-0077, expired August 31, 2003) assessed the short-term impacts of participation on member
outcomes. The current request for OMB approval covers the 2006-2007 follow-up survey of panel of
members in the longitudinal study approximately six years after the cohort completed their term of
service.
The Phase III follow-up survey will provide additional information about respondents’ backgrounds,
AmeriCorps program experiences (for the treatment group only), experiences following AmeriCorps
program completion (or analogous time period for the comparison group), and civic engagement, political
engagement, education and employment attainment and success, and life skills. Finally, we expect that the
attitudinal and behavioral questions will yield valuable information about AmeriCorps alumna and will
provide important information for future phases of the longitudinal study.

Phase III of the Longitudinal Evaluation of AmeriCorps
Corporation for National and Community Service

Collection of Information Employing Statistical Methods

5

Exhibit 1
Data Collection Rounds
Tool

Timinga

Baseline Survey
(1999-2000)

P

Focus

P

Members: Within days of enrolling
U

U

Comparison Group: 3–4 months after
inquiring about AmeriCorps (roughly when
they might have enrolled)
U

Post-Program
Survey (2000-2001)

U

State and National Members: 1–2 months
after completing service (approximately 1
year after baseline survey)
U

U

•
•
•

Prior service experience
Other background characteristics
Attitudinal information related to
outcomes

•

Attitudinal information related to
outcomes
Information on AmeriCorps program
experience (members only)

•

NCCC Members: During final weeks of
service (approximately 10 months after
baseline survey)
U

U

Comparison Group: 12–15 months after
baseline survey
U

Post-Program
Supplemental
Survey (PPSS)
(2003-2004)

U

Members: 3 years after baseline survey
(approximately 2 years after most members
completed their service)
U

U

Comparison Group: 3 years after baseline
survey
U

Phase III (20062007)

U

Members: 7 years after baseline survey
(approximately 6 years after most members
completed their service)
U

U

•
•
•
•
•
•

Additional background information to
address selection bias
Social networking behavior
Additional information on program
experience (members only)
Limited data on post-program activities
Attitudinal information related to
outcomes
Limited data on post-program activities

Comparison Group: 7 years after baseline
survey
U

a

U

A note on survey timing: The duration of AmeriCorps programs was generally between 10 and 12 months. Cases were
released for the post-program and post-program supplemental interviews at 12 and 36 months after baseline interview.
Most respondents were interviewed within a few weeks of survey release. In some instance, it took longer (up to five
months) to locate and interview respondents.

A.3. Use of Information Technology To Reduce Burden
The Phase III follow-up study will use Computer-Assisted Telephone Interviewing (CATI) and a web
survey to collect information for all study respondents. The CATI and web methods of data collection
reduces error and minimizes the number of times survey respondents have to be re-contacted for
additional or clarifying information.
Many of the respondents have access to computers and the Internet. Completing a web-based version of
the survey will be an additional option for all respondents. Based on experience with similar populations,
we estimate approximately 25 percent of respondents will elect to respond using this option. The total
time it takes to complete the web-based survey is expected to be consistent with that for the CATI
version. However, the use of this technology may reduce burden in three ways: (1) respondents can
complete the survey at any time reducing the need to schedule with a CATI interviewer, (2) respondents
can complete the survey anywhere they have access to the Internet, reducing the burden of being

Phase III of the Longitudinal Evaluation of AmeriCorps
Corporation for National and Community Service

Collection of Information Employing Statistical Methods

6

accessible by phone, and (3) respondents will be able to visually refer to questions and response
categories as needed rather than receiving the information orally.

A.4. Efforts To Identify Duplication
Efforts to avoid duplication include a review of the Corporation’s administrative agency reporting
requirements and of existing studies of Corporation programs. Existing data can, and will, be used where
possible. However, no data exist to answer research questions about program outcomes or impacts on
members. Previously approved surveys do not provide the data about members and program effects that
will be assessed using this Phase III follow-up. More specifically, the Phase III follow-up will provide
new information about members’ experiences, attitudes, and behaviors, as well as determine long-term
impacts of AmeriCorps.

A.5. Small Business
This information collection will not have a significant economic impact on a substantial number of small
entities.

A.6. Consequences of Not Collecting the Information
This submission is for the Phase III follow-up survey, which is necessary to characterize the longitudinal
outcomes of AmeriCorps members and their counterparts. Less frequent data collection would not allow
this study to answer the research questions on the full range of possible program impacts. Less frequent
data collection also would limit the Corporation’s ability to assess how the AmeriCorps experience affects
members at various points in time.

A.7. Special Circumstances Justifying Inconsistencies With
Guidelines in 5 CFR 1320.6
There are no special circumstances required for the collection of information on the Phase III follow-up
study.

A.8. Consultation Outside the Agency
In accordance with the Paperwork Reduction Act of 1995 (Pub.L. No. 104-13) and Office of Management
and Budget (OMB) regulations at 5 CFR Part 1320 (60 FR 44978, August 29, 1995), the Corporation
published a notice in the Federal Register announcing the agency’s intention to request an OMB review of
data collection activities. The notice was published on February 3, 2006 on page 5817 and provided a 60day period for public comment. A copy of the Federal Register notice for this information collection is
included in Attachment B. There were no public comments submitted to the Corporation in response to
the Federal Register notice.
•

In the planning process of this study, and in previous rounds of data collection, face-to-face
meetings with two technical working groups (TWGs) were conducted:

Phase III of the Longitudinal Evaluation of AmeriCorps
Corporation for National and Community Service

Collection of Information Employing Statistical Methods

7

•

The methodology TWG of researchers and practitioners in the national service field. Its
members are listed in Exhibit 1.

•

The instrument design TWG of individuals with substantive knowledge in fields related to the
intended outcomes; and staff from another Corporation contractor who developed a related
instrument for an earlier study. Technical Working Group members are listed in Exhibit 2.

Consultations with TWG members included questionnaire development, establishing outcomes of
interest, and evaluation design and plan.
Exhibit 2
Planning Process: Members of Technical Working Groups
Methodology Technical Working Group Members
Alan Melchior, Brandeis Center for Human Resources
D. Wayne Osgood, Penn State University
Antonio Perez, Milwaukee Community Service Corps
James Perry, University of Indiana
Thomas Smith, former Vice-President at Public/Private Ventures
Carol Weiss, Harvard University Graduate School of Education
Abe Wandersman, University of South Carolina
Instrumentation Design Technical Working Group Members
Robert Higgins, Walt Whitman Center for Culture & Politics of Democracy, Rutgers University
JoAnn Intili, Aguirre International
Ivy Jones, Peace Games
Alan Melchior, Brandeis Center for Human Resources
D. Wayne Osgood, Penn State University
Carol Weiss, Harvard University Graduate School of Education

Exhibit 3 identifies the TWG members for the current study, Phase III. An in-person meeting was
facilitated that focused on instrument design for Phase III and analysis plan of longitudinal data. In
addition to feedback and guidance from the Phase III TWG, pilot tests were conducted with six
individuals: three who served in AmeriCorps and three who inquired about AmeriCorps or other types of
national service. First, cognitive interviews with four individuals were conducted exploring new survey
items, and second, pilot testing the entire instrument was conducted with two individuals.

Phase III of the Longitudinal Evaluation of AmeriCorps
Corporation for National and Community Service

Collection of Information Employing Statistical Methods

8

Exhibit 3
Phase III: Members of Technical Working Groups
Instrumentation Design and Evaluation Plan Technical Working Group Members
Greg Duncan, Northwestern University
Rebecca Maynard, University of Pennsylvania Graduate School of Education
Doug McAdam, Stanford University
David Reingold, Indiana University
Carole Uhlaner, University of California, Irvine
Edward Vytlacil, Columbia University
Steve West, Arizona State University

A.9. Payments or Gifts to Respondents
Respondents will receive an incentive designed to encourage respondents to complete the survey, thereby
achieving a high response rate and reducing the likelihood of non-response bias. The $35.00 incentive is
based on precedence in this study, recognition of the average respondent age and employment, and an
interest in obtaining a suitable response rate. In addition, due to the long period of the longitudinal study
and the use of a panel of members over time, incentives will help to ensure members remain in each wave
of data collection. Based upon prior experience and research, the use of a small incentive payment may
also reduce the need for expensive field tracking and locating services. The previous waves of the
longitudinal study suggest AmeriCorps members are a highly mobile group in the years following the
completion of their term of service. Tracking and locating services may be needed to locate respondents,
and small incentive payments to encourage participation have been shown to reduce some of the costs of
locating highly-mobile and hard-to-contact individuals.1 In the previous wave of the survey, OMB
approved an incentive payment of $35.00 to every respondent who completed the survey. The team
proposes to maintain the amount from previous waves of this study, consistent with the prior OMB
approval. Reducing the incentive would likely jeopardize the response rate, and would be inconsistent
with the study’s past history.
TP

PT

A.10. Assurance of Confidentiality
Every effort will be made to maintain the privacy and/or confidentiality of respondents. The
confidentiality procedures adopted for this study during all rounds of data collection, data processing, and
analysis consist of the following:
•

TP

1
PT

All study respondents will be assured that the information they provide is confidential and
will be used only for the purpose of this research. To ensure data security, all individuals
hired by our contractor, Abt Associates Inc. are required to adhere to strict standards and sign
an oath of confidentiality as a condition of employment.

For more information see Duffer, Allen P. et al. (1994). "Effects of Incentive Payments on Response Rates and
Field Costs in a Pretest of a National CAPI Survey" Chapel Hill, NC: Research Triangle Institute; and
Providing Incentives to Survey Respondents: Final Report (Council of Professional Associations on Federal
Statistics, September 1993).

Phase III of the Longitudinal Evaluation of AmeriCorps
Corporation for National and Community Service

Collection of Information Employing Statistical Methods

9

•

Hard-copy data collection forms are delivered to a locked area for receipt and processing. Abt
Associates Inc. maintains restricted access to all data preparation areas (i.e., receipt, coding,
and data entry). All data files on multi-user systems will be under the control of a database
manager, with access limited to project staff on a “need-to-know” basis only.

•

Survey data collected via the web will also be protected. The Corporation’s contractor, Abt
Associates Inc., will maintain web-based survey data on a secure server with appropriate
levels of password and other types of protection.

•

Individual identifying information will be maintained separately from completed data
collection forms and from computerized data files used for analysis. No respondent identifiers
will be contained in public use files made available from the study, and no data will be
released in a form that identifies individual grantee staff, service providers, program
participants, or comparison group members.

A.11. Questions of a Sensitive Nature
The questions asked in the Phase III follow-up instrument do not involve questions of a sensitive nature,
and are consistent with prior rounds of data collection. Participants in the survey will be informed that
participation in the interview is voluntary and that they have the option not to participate if they feel
questions are burdensome. Finally, security measures and commitment to ensuring the confidentiality of
the data (see previous section) minimize the encumbrance to respondents of providing this information

A.12. Estimate of Response Burden
Time and cost estimates of the reporting burden for the Phase III follow-up survey appear in Exhibit 4.
Time estimates are based on prior round of data collection and experience with similar instruments in
other studies.
A.12.1. Number of Respondents, Frequency of Response, and Annual Hour Burden

In the 1999/2000 program year, the sample comprised of 1,752 members from a nationally representative
sample of 108 AmeriCorps programs and 476 corpsmembers from three (of five) NCCC campuses. The
Corporation’s evaluation contractor, Abt Associates, has tracked all study participants since the 1999
baseline survey, through the post-program survey in 2000 and the follow-up supplemental survey (PPSS)
in 2002. Abt will be following up with the same study participants in the fall 2006. In all years, the
response rates have been high for both the treatment and comparison group (approximately 80 percent).

Phase III of the Longitudinal Evaluation of AmeriCorps
Corporation for National and Community Service

Collection of Information Employing Statistical Methods

10

Exhibit 4
Estimates of Annualized Burden Hours and Costs
Number of
Respondentsa

Minutes per
Respondent

Response
Burden in
Hours

Cost/Hourb,c

Cost to
Respondents

Members:
Follow-up telephone survey of former
State and National program members.

1,401

55

1284.25

$9.70

$12,457.23

Follow-up telephone survey of former
NCCC members.

380

55

348.33

$9.70

$3,378.83

Follow-up telephone survey of State
and National comparison group
members.

1,219

55

1117.42

$9.70

$10,838.94

Follow-up telephone survey of NCCC
comparison group members.

320

55

293.33

$9.70

$2,845.33

TOTAL

3,320

Comparison Group:

a
b
c

3,043

$29,520

For the Phase III follow-up surveys, numbers are based on an 80% estimated response rate from baseline.
Cost/hour based on Corporation’s estimate of average hourly living allowance for members.
Cost/hour based on estimated average annual earnings of $20,176.00 for the Phase III follow-up, reflecting average 1999 earnings
by the comparison group plus 4% annual increases.

A.12.2. Hour Burden Estimates by Each Form and Aggregate Hour Burdens

There are two versions of the Phase III follow-up survey, a version for those who enrolled in an
AmeriCorps program, and another version for the comparison group. The average hour burden per
respondent is 55 minutes. This estimated burden is based on prior rounds of data collection for the
Longitudinal Evaluation of AmeriCorps and the pilot test conducted for the Phase III instrumentation.
A.12.3. Estimates of Annualized Cost to Respondents for the Hour Burdens

Exhibit 3 offers an estimate of reporting burden for 1,401 members of 108 State and National programs
and a national comparison group of 1,219 individuals. As for the NCCC, Exhibit 3 offers estimates of the
reporting burden for 380 NCCC members from 3 campuses and a comparison group of 320 individuals.
The estimate of burden is based on an 80% response rate of the original sample. AmeriCorps and
comparison group respondents will complete a 55-minute survey (Appendix A). The Corporation
estimates a corpsmembers’ average hourly living allowance at $9.70, reflecting the average 1999 earnings
by the comparison group plus 4 percent annual increases. Other than their time to complete the survey,
there are no direct monetary costs to respondents.

A.13. Estimate of Total Capital and Startup Costs/Operation and
Maintenance Costs to Respondents or Record-Keepers
The Phase III follow-up survey does not place any capital equipment or maintenance requirements on
respondents.

Phase III of the Longitudinal Evaluation of AmeriCorps
Corporation for National and Community Service

Collection of Information Employing Statistical Methods

11

A.14. Estimates of Costs to the Federal Government
The estimated cost to the Federal Government for the data collection is $858,588.

A.15. Changes in Burden
This submission to OMB is for an initial request for approval for this survey instrument.

A.16. Plans for Publication, Analysis, and Schedule
Study schedule:
Instrument Development
OMB Clearance
Data Collection
Prepare Data Files for Analysis
Data Analysis
Initial Findings
Draft Report
Final Report

March 2006 – July 2006
August 2006 – December 2006
January 2007 – June 2007
June 2007 – August 2007
August 2007 – October 2007
October 2007 – November 2007
December 2007 – February 2007
March 2008 – May 2008

Plans for Tabulation and Statistical Analysis

To address the two main research questions of this study, we will analyze the data in three ways: 1)
comparison of differences between AmeriCorps members and the comparison group; 2) changes over
time; and 3) comparisons to a national benchmark. In determining current differences between
AmeriCorps member and the comparison group, we will employ Propensity Score Analysis (PSA) to
address any selection bias in our sample. Straight comparisons of treatment and control groups potentially
lead to biased estimates of the program’s effect. Using PSA, treatment cases are compared to comparison
group cases that have a similar probability of selection into treatment to estimate the treatment effect.
To measure changes over time, we plan to take advantage of longitudinal data, and in particular, having
four time points of data (baseline and 12 months after baseline, 36 months after baseline, and seven years
after baseline) on outcomes of interest. We will use standard methods of analyzing growth, as well as
conduct hierarchical multivariate linear modeling (HMLM). The benefit of employing HMLM is that it
takes into account the variance within each person’s growth trajectory as well as the variance between
individuals. HMLM is also advantageous, in comparison to other methods such as structural equation
modeling, because it is robust to varying number and spacing of time points across cases. Hence, at level1, each person’s trajectory, or change over time, is modeled. These individual trajectories become the
outcome at level-2, where we can control for person-level characteristics, such as their propensity to serve
score. Further, we can test whether the trajectory is linear or curved. For example, it can be hypothesized
that seven years after AmeriCorps, civic engagement can decrease, suggesting a quadratic growth model.
To compare the study’s sample to a national benchmark, we will use publicly available datasets, such as
the Current Population Survey (CPS), to determine if our sample differs in such areas as voting and
political participation. As is the case in education research, where researchers compare test scores of a

Phase III of the Longitudinal Evaluation of AmeriCorps
Corporation for National and Community Service

Collection of Information Employing Statistical Methods

12

treatment and control group against national norms, we will investigate how outcomes of interest, such as
voting, differ among the treatment, comparison, and national benchmark. In the example of voting, it is
hypothesized that even after seven years since beginning AmeriCorps participation, the treatment group
should have high levels of voting, the comparison group (because of their initial interest in service and
volunteerism) moderate levels, and the national norm would be the lowest.

A.17. Approval To Not Display Expiration Date
No exemption is requested.

A.18. Exceptions to Item 19 of OMB Form 83-1
The submission describing data collection requires no exemptions to the Certificate for Paperwork
Reduction Act (5 CFR 1320.9).

Phase III of the Longitudinal Evaluation of AmeriCorps
Corporation for National and Community Service

Collection of Information Employing Statistical Methods

13