Supporting Statement

OMB No. 0920-0314 Feb 8, 2008 NSFG 0208 Change package Moien-D3.doc

NATIONAL SURVEY OF FAMILY GROWTH, CYCLE 7

Non-substantive change

OMB No. 0920-0314

Expires: April 30, 2009

February 8, 2008

Contact Information:

William D. Mosher, Ph.D., Statistician

Project Officer, National Survey of Family Growth

National Center for Health Statistics/CDC

3311 Toledo Road, Room 7318

Hyattsville, MD. 20782

301-458-4385

301-458-4034 (fax)

[email protected]

Supporting Statement for Request for Clearance:

NATIONAL SURVEY OF FAMILY GROWTH, CYCLE 7

Abstract

This is a request for a non-substantive change to the National Survey of Family Growth (NSFG) ---OMB No. 0920-0314, for Years 3 and 4 of Cycle 7 (June 2008-June 2010). Clearance was granted for 3 years on April 14, 2006. The changes are two-fold. First we seek to make minor changes to the questionnaires. Second, this request documents the completion of an experiment in incentives and sets an incentive schedule for the survey. The details of the incentive experiment are provided in section A.9. and Attachment C. There is also a small decrease in the burden due to a more accurate estimate of the time required to administer the questionnaires.

The NSFG is conducted by the National Center for Health Statistics (NCHS), Centers for Disease Control and Prevention (CDC). The survey provides nationally representative data on factors related to birth and pregnancy rates, sexually transmitted diseases, and family formation including marriage, divorce, and adoption. The data are used by NCHS and seven other DHHS programs that support the survey. The survey is administered in person, in English and Spanish. In each year of interviewing, about 4,400 men and women 15-44 years of age will be interviewed. In the first 6 quarters (18 months), fieldwork has gone well, and we are achieving these targets.

We are not opening up new content areas with the requested changes; they are incremental, but they do fill gaps that have been revealed in the analysis of the Cycle 6 data and in our own preliminary assessments of the questionnaire. The changes requested here constitute less than one percent of the data file. Interview length is still under our original burden estimate, so these changes do not increase the previously approved interview length, or burden. The new questions involve educational attainment; reasons for non-use of contraception; and attitudes toward cohabitation.

The new continuous interviewing design we introduced in 2006 is working well so far. Cost efficiency and labor productivity are higher than in the 2002 (Cycle 6) survey. No important changes in the design are anticipated. In 4 ½ calendar years of data collection, a national sample of over 19,000 interviews will be collected in 110 Primary Sampling Units, providing the largest national sample ever collected in the NSFG. This will make possible national estimates with as little as 2 or 2.5 years of data, while estimates for small groups can be obtained by cumulating data for a group of consecutive years. Once clearance for these changes is received, we will hold interviewer training in June of 2008 and implement the new questions in July, 2008. July, 2008 is the beginning of Year 3 of the survey. The survey web site is at http://www.cdc.gov/nchs/nsfg.htm

A. Justification

1. Circumstances Making the Information Collection Necessary

The National Center for Health Statistics (NCHS), under its duties specified in 42 U.S.C. 242k, Section 306(b) of the Public Health Service Act, Paragraph 1H (Attachment A1), conducts the National Survey of Family Growth to

(1) supplement the data from birth certificates with survey data on factors related to “family formation, growth, and dissolution,” and

(2) to serve a variety of data needs in public health programs (listed below).

The survey was fielded periodically—in 1973, 1976, 1982, 1988, 1995, and 2002. In surveys through 1995, the NSFG was a multi-purpose statistical survey based upon a national sample of women, and focused on factors affecting pregnancy and birth rates, including marriage, divorce, and cohabitation, adoption, sexual behavior, childbearing and pregnancy outcomes, infertility and infertility services, contraceptive use, and family planning services.

In Cycle 6, for the first time, the NSFG also interviewed an independent national sample of men, to obtain data on men’s behavior and attitudes related to having and raising children, marriage and family formation, and reproductive health. In 2002, NCHS and its contractor, the Institute for Social Research (ISR) at the University of Michigan, conducted Cycle 6, the largest cycle of the National Survey of Family Growth (NSFG) ever conducted, including 7,643 women and 4,928 men 15-44 years of age. The weighted response rate was 79 percent, comparable with previous cycles of the survey.

In Cycle 7, now underway, NCHS is collecting data to carry out its own responsibilities, and for other agencies and programs in DHHS that contribute funding for the NSFG:

• the National Institute for Child Health and Human Development, Public Health Service, under 42 U.S.C. 241 (Section 301 of the Public Health Service Act);

• the Office of Family Planning, Office of Population Affairs (OPA), DHHS, under 42 U.S.C. 300a (Section 1009 of Title X of the Public Health Service Act);

• the Adolescent Family Life Program of the Office of Adolescent Pregnancy Programs, Office of Population Affairs, DHHS, under 42 U.S.C. 300z (Section 2001 of the Public Health Service Act);

• the Division of HIV/AIDS Prevention (DHAP) of the National Center for HIV, Sexually Transmitted Disease, and Tuberculosis Prevention (NCHSTP, Section 301 of the Public Health Service Act);

• the Division of Reproductive Health of the CDC, under Section 301 of the Public Health Service Act;

• the Division of Sexually Transmitted Disease Prevention of the Centers for Disease Control and Prevention (CDC), under Section 301 of the Public Health Service Act.

• the Children’s Bureau of the Administration for Children, Youth, and Families, Office of Human Development Services, under PL 96-272, the Adoption Assistance and Child Welfare Act of 1980; and

• the Office of the Assistant Secretary for Planning and Evaluation (OASPE), under Section 301 of the Public Health Service Act.

The data needs of the agencies that will be served by Cycle 7 were described in detail in the full clearance package (approved in April 2006). Attachment B1 of the main package discussed in detail the need for information collected in the female questionnaire, and Attachment B2 justified the topics covered in the male questionnaire, focusing particularly on the sensitive information that they collect. Last year, we submitted Attachment B3, which summarized the new questions introduced in Year 2. With this request, we have included Attachment B4, which lists the new questions being collected, gives the justifications for their inclusion, and estimates their impact on interview length. The new questions involve educational attainment; reasons for non-use of contraception; and attitudes toward cohabitation. Eight questions are being added to both the male and female questionnaires; two additional questions are being added to the female questionnaire. The updated Female questionnaire is shown in Attachment I and the updated Male questionnaire is shown in Attachment K.

The purpose of the current continuous interviewing design is to collect the data more frequently, with larger sample sizes, at a lower cost per completed interview. Results in the first 18 months suggest that we are attaining this goal. Response rates are comparable to Cycle 6, and the principal measure of productivity—the number of interviewer hours per completed interview—has improved from 11.3 hours per completed interview to 9.0 hours in the first 18 months, a 20% improvement. This design makes the survey more manageable and more cost-efficient, and will soon provide data for every year, which is essential for monitoring trends adequately.

In addition, this design gives the NSFG’s sponsoring agencies a chance to respond to needs for data on new topics, by adding or modifying a limited number of questions in any year, instead of having to wait up to 7 years for a new cycle of the survey. This year’s changes include several examples of our being responsive to new data needs, increasing the usefulness of the survey to its federal agency sponsors and their constituents.

2. Purpose and Use of Information Collection

The National Survey of Family Growth responds to the congressional mandate for NCHS to collect and publish reliable national statistics on “family formation, growth, and dissolution” (Sec. 306(b), paragraph 1(H) of the Public Health Service Act), and a number of aspects of health status and health care.

From Cycle 6 (2002), NCHS alone has produced 11 statistical reports, containing more than 330 statistical tables and over 550 pages of printed results. PDF files of all 11 of these reports may be downloaded from: http://www.cdc.gov/nchs/nsfg.htm. A bibliography of the 73 known publications (as of January 2008) from the 2002 survey is shown here as Attachment D1 and is also on the web site under “NSFG Bibliography,” at:

http://www.cdc.gov/nchs/data/nsfg.htm

In addition, the Cycle 6 (2002) public use data were released on fully documented CD-ROMS and on downloadable files on the NCHS/NSFG web page. Additional data files (including contextual data) are available for use through the NCHS Research Data Center.

9. Explanation of any Payment or Gift to Respondents

The full memorandum describing the incentive test and results was sent to OMB on November 19, 2007 and is shown as Attachment C. This section gives a brief account of our findings and experience with incentives in the NSFG.

History of testing incentives in the NSFG

Cycle 5 Pretest, 1993--A field experiment was conducted in the Cycle 5 pretest in 1993. A $20 incentive was found to produce a significantly higher response rate (67.4%) than no incentive (58.9%--a difference of 8.5 percentage points), as well as lower field costs. This $20 incentive was used in the Cycle 5 Main Study in 1995. A response rate of 78.6% was obtained in the Main Study. (Vital and Health Statistics, Series 1, No. 36, page 9; and Mosher et al, 1994, American Statistical Association, 1994 Proceedings of the Survey Research Methods Section.)

Cycle 6 Pretest, 2001---In a field experiment in the Cycle 6 Pretest in 2001, a $20 incentive was compared with a $40 incentive. The response rate for those offered $20 was 62%; for those offered $40, it was 72%. The effect of incentives at this level was larger for women than for men. Women offered $20 had a response rate of 62%, while women offered $40 had a response rate of 81%. (NCHS, Series 1, No. 42, page 13, tables J and K.) Those receiving the higher amount were also less likely to express objections or reluctance to the interview than those receiving $20. (Ibid, tables L and M.).

Cycle 6 Main Study---In the Cycle 6 Main Study in 2002, a $40 incentive was used, but response rates were still lagging in certain groups after 7 months of interviewing. We requested, and received from OMB, permission to use an $80 incentive in a half-sample of the remaining cases, for 4 weeks at the end of interviewing in Cycle 6 (February, 2003). That $80 incentive raised our (weighted) response rate from 64% to 79%. (Only 7 percent of Cycle 6 respondents received $80; 93% received $40.) (Series 1, No. 42, pages 40 and 50; presentation to OMB, January 9, 2006, slides 18-20). Furthermore, the sample in the last 4 weeks had a higher proportion of married women, Hispanic men and women, and full-time workers of both sexes (presentation to OMB, Jan 9, 2006, slides 19-20).

Cycle 7, Year 1 (June 2006-June 2007)---The Cycle 6 experience, showing cost-effective increases in response rates and a more representative sample, led us to propose using the $40 incentive with an $80 non-response follow-up in Cycle 7. However, the NCHS Associate Director for Science suggested that we conduct an experiment to see if it was necessary to increase the payment from $40 to $80, or whether we could increase it from $40 to $50 and get the same results.

So, during the first 10 weeks of each 12-week quarter, we offered all respondents a $40 incentive. During the last 2 weeks of each quarter, under this experiment, we took a half-sample of the remaining respondents. Half of these (about 3 percent of the sample) received an extra $10 (for a total of $50), and half (another 3 percent of the sample) were offered an extra $40 (for a total of $80).

In our memorandum of November 19, 2007, we offered evidence (presented in detail in Attachment C) that the higher amounts are more effective. Given the costs of continuing to run the experiment, we sought permission to end it, and simply offer $80 to the 6-7 percent of respondents who are interviewed in the last 2 weeks of each quarter (aka “the double-sample phase”). Ending the experiment was approved by the National Center for Health Statistics Research Ethics Review Board (ERB) on August 29, 2007, and by OMB on November 21, 2007.

11. Questions of a Sensitive Nature

The survey was approved by the NCHS “Research Ethics Review Board” (the name NCHS uses for its IRB) in January of 2006. A continuation for 1 year (the maximum continuation) was approved on December 21, 2006, and again on November 28, 2007. The questions submitted here will be submitted to the NCHS ERB in early February. We expect that the Board will review the questions at its March, 2008 meeting.

The justifications for sensitive (and non-sensitive) questions in the NSFG questionnaires for males and females are discussed in detail in Attachments B1 and B2 of the main OMB supporting statement, approved by OMB in April of 2006. Justifications for the new questions in year 2 were approved in April 2007 (Attachment B3). Justifications of the new questions are shown in Attachment B4, included with this request. Note that the most sensitive questions in the survey are asked at the end of the interview---in the self-administered Audio Computer-Assisted Self-Interviewing (ACASI) section. In ACASI, the respondent listens to the questions over headphones, or reads them on the screen of the notebook computer, and enters his or her own answers into the computer. The computer is then locked and the interviewer cannot get into it—thus ensuring the respondent’s complete privacy.

12. Estimates of Annualized Burden Hours and Costs

The mean interview length was estimated at about 80 minutes for females and 60 minutes for males. Experience in the first 6 quarters of interviewing shows that the average interview length thus far is actually less than expected. The burden table is being adjusted accordingly.

Female ­­­­­_ Male______

Actual Estimate Actual Estimate

Year 1 68.2 80 minutes 49.7 60 minutes

Year 2 * 74.6 80 minutes 54.2 60 minutes

Year 3 (est) ** 76.1 80 minutes 55.7 60 minutes

*Based on interviews through Feb 4, 2008 (about 7 months of interviewing).

** including the newly proposed questions.

Estimated Annualized Respondent Table

Respondents /Instrument	No. of Respondents	Responses per Respondent	Average Burden/Response (in hours)	Total Burden Hours
PRETEST
Screener	403	1	5/60	34
Males	109	1	1	109
Females	133	1	1.33	177
MAIN STUDY
Screener	7,250	1	5/60	604
Males	1,957	1	56/60	1827
Females	2,393	1	1.27	3,039
VERIFICATION	725	1	5/60	60
TESTING OF ADDITIONAL QUESTIONS	2,000	1	10/60	333
TOTAL				6,183

15. Explanation for Program Changes or Adjustments

The estimated annual burden has decreased by 275 hours due to a better estimate of the time needed to conduct the survey. The new estimate is based on the first 18 months of the survey. This decrease includes the time needed for the 8 extra questions documented in this request.

B. Statistical Methods

3. Methods to maximize Response rates and Deal with Nonresponse

Summary—As discussed in detail in the full supporting statement for Cycle 7, we are using Advance Letters, highly trained interviewers, the NCHS web site, 800 numbers at both the University of Michigan and at NCHS, customized follow-up letters to address particular concerns, special interviewer training to avert non-response, and active monitoring of fieldwork by the University of Michigan fieldwork staff (Nicole Kirgis, Robert Groves and others) to encourage cooperation with the survey.

Given that even these measures do not attain an 80% response rate, we have requested and obtained OMB permission to use a $40 cash incentive (called a “token of appreciation” in the respondent materials). The request was based on data from (1) the experiment in the Cycle 5 pretest which showed that $20 was more cost-effective than no incentive; (2) the Cycle 6 pretest, which showed that $40 was more cost-effective than $20; (3) the Cycle 6 main study which showed that $80 was more effective than $40 in reaching non-respondents at the end of fieldwork; and (4) Year 1 of Cycle 7, which showed that $80 was more effective than $50 in reaching non-respondents. All of these experiments were approved by the NCHS IRB and by OMB.

We also show in Attachment C that $80 was more effective than $50 in recruiting certain groups into the sample: women with college educations, women and men who had never had (or fathered) a birth, men and women living in single-family homes, and Hispanic males.

These results suggest that busy, college-educated, childless people, and others who are not at home very much, are not as well represented in the standard Phase I sample as they

should be. It takes the $80 amount to bring more of these people into the sample.

LIST OF ATTACHMENTS

(Only attachments B4, C, D1, I and K are included with this request )

A. Authorizing legislation

A1. NSFG Authorizing Legislation

A2. Office of Family Planning Authorizing Legislation

A3. Adolescent Family Life Authorizing legislation

A4. NICHD Authorizing legislation

A5. CDC’s Division of HIV/AIDS Prevention (DHAP) Authorizing Legislation

A6. Children's Bureau (ACF) Authorizing Legislation

B. Justifications for Questions in the Survey

B1. Justification of the Female questionnaire by topic

B2. Justification of the Male questionnaire by topic

B3. Justification of New Questions for Year 2 of interviewing

B4. Justification of New Questions for Year 3 of interviewing -- NEW

C. Memo to Margo Schwab of OMB, dated November 19, 2007 -- NEW

D. Partial list of publications from the Survey

D1. List of publications from the 2002 NSFG -- NEW

D2. Chronological list of publications from the 1995 NSFG

D3 List of publications from the 1995 NSFG by topic

E. Memoranda from other offices and agencies on their use of the NSFG

E1. NCHS Public Affairs Officer

E2. Healthy People 2010 Health Objectives on Family Planning, HIV, STDs

E3. Office of Population Affairs.

E4. NICHD, NIH

E5. Children's Bureau, ACF, DHHS.

E6. HIV/AIDS Prevention Program, CDC

E7. OASPE (Office of the Assistant Secretary for Planning and Evaluation)

E8. Division of Reproductive Health, CDC

E9. Office of Planning, Research, and Evaluation, ACF

F. Consultation outside the agency:

F1. Federal Register Notice on the NSFG Main Study, Sept 13, 2005

F2. The NSFG: Current Uses and Future Prospects, April, 2001

F3. A Concept Overview for a Continuous Measurement Design for the National Survey of Family Growth, by Robert Groves, March, 2002

F4. Questions and Answers on Continuous Interviewing, 2003

F5. A Summary of Options for the Design of Cycle 7 of the National Survey of Family Growth, March, 2004.

G. Respondent Materials for Cycle 7 of the NSFG

H. Summary of Changes to the FEMALE Questionnaire for NSFG Cycle 7

I. FEMALE Questionnaire -- NEW

J. Summary of Changes to the MALE Questionnaire for NSFG Cycle 7

K. MALE Questionnaire -- NEW

L. Scope of Work for Cycle 7 of the NSFG

M. IRB Approval Form for the NSFG Cycle 7

The NSFG MAIN STUDY Protocol was submitted on December 1, 2005.

Approval is expected by March, 2008.

6