NSFG 2009 ATTACH K-Contract

NSFG 2009-2012 Attachment K OMB No. 0920-0314

ATTACHMENT K:

SCOPE OF WORK FOR CYCLE 7 OF THE

NATIONAL SURVEY OF FAMILY GROWTH

(Selected excerpts relevant to the current work of the contractor)

I. Introduction

II. Background

Contract Phases

References

III. Purpose of Contract

IV. Scope of Work: Specifications for Phase I, Cycle 6

1. Communication with NCHS

2. Confidentiality and Data Security

3. Sample Design

4. The Pretest

5. Preparation for Main Study Data Collection

6. Main Study Data Collection

7. Data Preparation

8. Data Files and Final Reports

V. Scope of Work: Specifications for Phase II, Cycle 7

Acronyms

ACASI Audio Computer-Assisted Self Interviewing

CAPI Computer-Assisted Personal Interviewing

CDC Centers for Disease Control and Prevention

CRQ CAPI Reference Questionnaire

DHAP Division of HIV-AIDS Prevention of CDC

EDOC Effective Date of Contract

HIV Human Immunodeficiency Virus, the Virus that causes AIDS

NCHS National Center for Health Statistics

NSFG National Survey of Family Growth

NICHD National Institute for Child Health and Human Development

OPA Office of Population Affairs

PSU Primary Sampling Unit

STDs Sexually Transmitted Diseases

I. Introduction

The National Center for Health Statistics (NCHS) Division of Vital Statistics requires the planning and conduct of Cycle 6 of the National Survey of Family Growth (NSFG), with options to conduct Cycle 7 and Cycle 8. A Cycle consists of all the activities-- including planning, pretest, CAPI programming, fieldwork, data processing, file preparation, and documentation-- for a single national survey.

The National Survey of Family Growth (NSFG) is part of a series of surveys based on national probability samples of the civilian noninstitutional population of women that began in 1955. In 1955 and 1960, the surveys were conducted by private organizations; in 1965 and 1970, they were conducted by university researchers with federal funding. Since then, these surveys have been conducted 5 times by the NCHS--in 1973 and 1976 with samples of ever-married women 15-44 years of age; and in 1982, 1988, and 1995 with samples of women of all marital statuses 15-44 years of age.

In each Cycle of the NSFG, women have been interviewed in person in their own homes by highly trained professional female interviewers. In Cycles 1-4, interview length averaged 60-70 minutes. In Cycle 5 (1995), interview length averaged 105 minutes. The 1995 NSFG was also the first to be conducted using Computer-Assisted Personal Interviewing (CAPI) and Audio Computer-Assisted Self Interviewing (Audio CASI). CAPI and Audio CASI will be used again in Cycle 6.

II. Background

The original purpose of this series of surveys was to collect data from national samples of women that would help to explain trends and differentials in birth and pregnancy rates. The NSFG still serves this function, but also provides data needed by other federal programs, such as

• those of the Office of Population Affairs concerned with teenage sexual activity and pregnancy and the use of Title X Family Planning Services;

• the Healthy People 2000 and 2010 objectives on Family Planning, HIV Prevention, Sexually Transmitted Diseases, Smoking, and Maternal and Child Health;

• NICHD to inform and shape its extramural (grant) research programs related to fertility and reproductive health in the US, and to provide a data resource for private and university-based researchers;

• the HIV Prevention program of the CDC, which needs reliable information on the sexual and drug-related behaviors that increase the risk of HIV and STD transmission.

The NSFG is a principal source of national estimates of factors affecting pregnancy and birth rates, including sexual activity, cohabitation, marriage, and divorce, frequency of intercourse, contraceptive use for both birth control and disease prevention, miscarriage and stillbirth, infertility, wanted and unwanted births, and use of medical services for family planning and

infertility. Thus, Cycle 6 will require the collection of a wide range of sensitive data that must be as comparable as possible with data collected in earlier cycles.

Contract Phases

This contract will be divided into 3 phases:

Phase I consists of Cycle 6, with main study interviewing in 2001.

Phase II consists of Cycle 7 of the NSFG, with main study interviewing in 2004.

Phase III consists of Cycle 8 of the NSFG, with main study interviewing in 2007.

(NOTE: Phase II, Cycle 7 was activated; Phase III, Cycle 8, was cancelled.)

The decision to proceed with Phase II, and then with Phase III, will be made by the Government at its sole discretion--based on the Government’s evaluation of the contractor’s performance in Phase I, the availability of funding, whether the options specified for Cycle 7 make it possible to measure what we need to measure in Cycle 7, and other factors. If all three survey phases are done by the same contractor, NCHS expects considerable cost savings and time savings compared with awarding 3 separate contracts to conduct Cycles 6, 7, and 8 of the NSFG.

IV. Scope of Work: Specifications for Phase I, Cycle 6

The Contractor shall conduct Cycle 6 of the National Survey of Family Growth (NSFG), based upon national samples of men 15-49 and women 15-44. Contractor shall furnish all the necessary personnel, materials, services, facilities, and do all other things necessary for or incidental to the performance of the work as described in the following tasks, in accordance with the Project Schedule and List of Deliverables:

3. Sample Design

Design and select as appropriate a sample of eligible men and women for the survey.

The sample coverage of the NSFG encompasses men and women in the noninstitutionalized civilian population of the United States--women 15-44 and men 15-49--living in households or group quarters (including college dormitories). Age eligibility for the sample will be defined as of the date of interview.

Main Study Sample Design: The Contractor shall design an independent national sample of women 15-44 years of age and of men 15-49 years of age to yield the numbers of completed interviews shown in Table 1. The target response rate in the main study is 80 percent for women and 75 percent for men. The Project Officer shall decide, based on the number of completed cases, the costs of data collection, and the recommendations of the Contractor, when to stop fieldwork.

4. The Pretest

1. Program, using CAPI software approved by the Project Officer, and implement, using appropriate computer hardware, all data collection instruments for pretesting. These include at a minimum a (1) screener interview (screener); (2) the extended interview schedule for males; and (3) the extended interview schedule for females. Unformatted drafts of these instruments will be furnished by NCHS. The designed average length of the female questionnaire shall be 80 minutes--60 minutes CAPI and 20 minutes Audio CASI. The designed average length of the male questionnaire shall be 60 minutes--40 minutes CAPI and 20 minutes Audio CASI.

5. Pretest (Not shown here.)

6. Main Study Data Collection

a) Implement main study interviewer and supervisor training programs for the main study.

b) Mail an advance letter signed by the Director of NCHS to each NSFG sampled person. (See Series 1, No. 36, pages 57-59) A copy of the letter and pamphlet shall be given directly to each eligible respondent at the beginning of the extended interview if he or she has not received them.

c) Conduct all screener and all extended interviews in a 7-month period.

Specifically:

(1) The final unweighted completion rate shall be at least 80 percent., where the completion rate is defined as (total unweighted number of completed extended interviews) divided by (the total unweighted number of NSFG-eligible persons). The denominator of this rate in Cycle 6 shall exclude individuals who have died or left the U.S, or are in prisons or the military. It shall include all others.

(2) The Contractor shall propose, and describe in detail, a plan to achieve an 80 percent completion rate, based on an understanding that 80% completion rates are not commonly achieved in national surveys of males that are more than 1 hour long and are on sensitive topics.

(3) The contractor shall design and implement a system for monitoring field work that is capable of producing status reports at the following intervals:

First, response rates, like those in Series 1, No. 36, page 82, twice a week (e.g., Monday and Thursday), based on cases received by the home office no more than 24 hours before.

Second, the number of hours worked, the number of cases completed, and the cost of field work, by region, based on information that is no more than 5 working days old.

Failure to deliver such reports to the Project Officer in a timely way shall be considered a serious deficiency.

4) Respondents in the Main Study shall be interviewed under conditions of maximum privacy, normally in the respondent’s home, but elsewhere if the respondent prefers. Pending OMB approval, respondents shall be paid an incentive of $20 to complete the interview.

7. Data Preparation

Recodes: In addition to variables computed by the CAPI program during the interview, prepare computer programs for the creation of about 500 recoded variables (hereafter, recodes) to be constructed from and added to the raw data tapes. The programs shall be written in a language approved by NCHS (e.g., SAS, PL/1). Detailed specifications for these recodes will be provided by NCHS project staff; illustrative specifications may be found on the Public Use CD-ROM and on the NCHS Web Site. When notified by the Project officer, execute these programs and add the resulting variables to the respondent (men, women, combined) files. Prepare a brief written report on this task, and include it, or a summary of it, in the Series 2 report.

Imputation: Prepare specifications for imputing values for non-response categories on the approximately 500 recodes , using a hot-deck or other procedure approved by the Project Officer. Prepare and submit for approval specifications for machine editing and case-by-case adjudication of imputed values for recodes. Include and code an imputation “flag” variable for each recode. Specifications for Cycle 5 Recodes are given in the Series 2 report for Cycle 5. The Contractor shall prepare a report on imputation and imputation edits, and include it in the Cycle 6 Series 2 report.

Variance Estimation: The Contractor shall design and execute a procedure for computing design-based variance estimates for NSFG statistics. The estimates produced should be similar to those shown in the Series 2 report for Cycle 5. The variance estimation procedure can be a Taylor Series, or replication, or some other technique, but the plan for variance estimation must be submitted for review and approval by the Project Officer. The contractor shall also estimate the imputation variance for the respondent, interval, male, and combined files.

8. Data files and Final Reports

Offerors shall review the 1995 NSFG Data file and documentation, to become familiar with NSFG conventions and expectations regarding the content, layout, code outline, and documentation of data files. NCHS welcomes contractor suggestions on ways to improve the layout and usability of these data files and documentation.

Documentation: Data file documentation shall be submitted with each file, conforming to the standards shown in the NCHS public use documentation (tape or CD-ROM) for Cycle 5. Documentation shall be reviewed by NCHS project staff.

Series 1 Report: In collaboration with the Project staff, prepare a final report on the survey that is modeled closely on Series 1, No. 36. Submit for the Project Officer’s review an outline of the report and a list of Appendixes, a first draft, and a final draft. Submit electronic files of the entire report for further editing by NCHS before publication. Include a 10-15 page Summary Report containing recommendations (not for publication) for the design of the next Cycle of the NSFG.

Series 2 Report: In collaboration with the Project Officer or representative, prepare a report on the sample design, weighting, imputation, and variance estimation for the survey that is modeled closely on Series 2, No. 124.

V. SCOPE OF WORK: Specifications for Phase II, Cycle 7

Before any work begins, or any government funds are expended on Phase II of this contract, the Project Officer must give written approval. The Project Officer shall recommend to the Contracting Officer as to whether to:

(a) proceed with Phase II as outlined below, (hereinafter, the “baseline design”), or

(b) cancel Phases II and III completely, or

(c) modify, enhance, or improve Phase II, as outlined in Options 1-5 below.

The Contractor is not to engage in any labor on Phase II or charge any costs to Phase II until the the Contracting Officer has authorized such labor in writing in accordance with this paragraph.

If a decision is made to modify, enhance, or improve Cycle 7, those modifications may include any, or some combination of, the following.

(There were 5 options. Only the first, an increase in sample size, was implemented through the use of continuous interviewing. Four other options were not implemented because the NSFG’s sponsors decided that a successful transition to continuous interviewing was the most urgent priority. The other options (not implemented) were: a military sample, an incarcerated sample, a biomarker, and a 120-day follow-up.)

The Contractor shall furnish the necessary personnel, materials, services, facilities, and otherwise do all things necessary for or incident to the performance of the work as follows:

Objectives (Baseline Design): The Contractor shall conduct Cycle 7 of the National Survey of Family Growth in 2004, with a sample of about 11,800 women 15-44 years of age and 7,200 men 15-49 years of age, distributed by age and race/ethnicity in the same way as in Cycle 6.

1. The Project Officer will provide a brief written description of the Cycle 7 Main Study sample design for review and comment by the contractor. All interviewers shall be female. The expected average length of interview will be 80 minutes for females and 60 minutes for males. The interviews shall be conducted with Computer-Assisted Personal Interviewing (CAPI) and Audio CASI technology. Interviewers shall use Global Positioning Systems technology to specify the current address of the respondent.

2. The contractor shall design and conduct a Pretest, consisting of at least 100 completed interviews for males and 100 interviews for females, in a manner approved by the Project Officer, before conducting the Main Study interviews. A shorter version of the Pretest Evaluation report shall be presented orally and recorded in the minutes of the monthly management meetings, and reported formally in the Series 1 report.

3. Within 2 weeks after implementation of Cycle 7, the Project Officer will submit a list of proposed deletions, technical corrections, additional edits, and other changes in Computer-Assisted Personal Interview (CAPI) and Audio CASI programs, for review and comment by the contractor. These may include changes in up to 25% of the questions, including 15% that are minor (do not affect skip patterns in existing questions) and 10% that are major (do affect skip patterns in existing questions). The contractor shall submit to the Project Officer in writing an estimate of the time and cost of making the proposed revisions and testing them before the pretest, and the time and cost of fixing any problems discovered after the pretest.

4. If Cycle 7 of the NSFG is conducted using a design similar to the Cycle 6 survey, with 19,000 or fewer respondents, that shall be referred to here as the baseline design.

5. The following materials from the Cycle 6 survey can and should be used with only minor revisions, both in the baseline design and in option 1:

the CAPI program, the Audio CASI program, the Introductory materials (advance letter for respondents, parental consent forms, etc), interviewer and supervisor manuals, training materials, questionnaire specifications manuals, data receipt manuals and procedures, field status report forms, codebooks, coding and editing specifications, recode programs, imputation specifications, and contextual data tape plans.

6. The CAPI questionnaire and program, and the Introductory Materials, are deliverable items. Written documentation of the other items shall be in the minutes of the monthly meetings and in the Series 1 Report, the Series 2 report, and the tape (or file) documentation.

7. Once per month, the contractor and the NCHS Project staff shall have a formal meeting to review the progress of the project, via conference call, Envision (video conference), or in-

person visit. At a minimum, 4 of these meetings per year shall be in person at NCHS. The contractor shall submit written minutes of the decisions made at those meetings within 4 business days, via e-mail or overnight mail.

8. Conduct main study fieldwork and data preparation as described for Phase I of the contract, but limit the deliverables to those listed in the Cycle 7 contract below.

G6. PROJECT OFFICER AND ALTERNATE PROJECT OFFICER

William D. Mosher, Ph.D. is hereby designated as the Project Officer for this contract.

The Alternate Project Officer is Joyce C. Abma, Ph.D.

H 9. SYSTEMS SECURITY

Offerors are required to present a detailed outline of their proposed AIS security program in their proposals. The offerors security proposal must be sufficiently detailed to show an understanding of what is required. A general statement that the offeror will comply with applicable requirements is not acceptable. Consult the following chapters of the Automated Information Systems Security Handbook (available at www.hhs.gov/progorg/oirm/secprog.html.) in determining the AIS Security requirements: (Chapters 3-10).

MODIFICATION TO CONDUCT CYCLE 7, IMPLEMENTED IN 2004 & 2005:

(THERE WERE OTHER MODIFICATIONS IN THE CONTRACT, BUT THIS IS THE PRINCIPAL MODIFICATION THAT IMPLEMENTS CYCLE 7):

Cycle 7 (Phase II) Contract Modification

The purpose of this modification is to revise and specify the scope of work for Cycle 7 (Phase II of the contract). This modification implements the intent we expressed in Modifications 7 and 17. There is no increase in cost; we are using the originally negotiated amount of $17,775,231. Many of the features of Phase II (Cycle 7) specified in this modification are explicitly designed to reduce and control costs. This modification adjusts the schedule of deliverables for Cycle 7, updates the invoicing details and task descriptions for Cycle 7, and updates key personnel. It is expected that the contractor’s formal proposal for this modification will include the following elements:

Insert the following in “Section C—Statement of Work:

V. SCOPE OF WORK: Specifications for Phase II, Cycle 7”—

This modification updates the “Objectives (Baseline Design)” as follows:

The Contractor shall conduct Cycle 7 of the National Survey of Family Growth starting in

May 2006 with continuous data collection through May 2010. The total sample size yield

after four years of continuous interviewing is expected to be about 17,400 interviews

(about 10,600 female interviews and 6,800 male interviews) with oversamples of

Black, Hispanic, and 15-24 year old respondents.

1. Questionnaire: NCHS Project Staff will provide written specifications (a CAPI

Reference Questionnaire or “CRQ”) for each section of the questionnaire.

The specifications will be delivered to the contractor in batches in Calendar year 2005.

After a review period involving discussions by telephone, e-mail, exchange of documents, and other means between NCHS and the contractor to ensure proper understanding, the Project Officer shall instruct the contractor to update the Cycle 6 CAI program to reflect the specifications.

Blaise testing and program fixes are expected to end in February 2006.

The expected average length of interview will be 80 minutes for females and 60 minutes

for males. The interviews shall be conducted with Computer-Assisted Personal

Interviewing (CAPI), with a self-administered portion included. The contractor shall

evaluate and report to the Project Officer on the use of:

(a) Audio-CASI versus text CASI, as well as

(b) touch-screen capability for this portion of the interview.

The contractor shall provide to the project officer information as to the relative advantages and costs of these innovations, and recommend to the project officer whether these innovations should be implemented. An updated CAPI program revision is expected to be released on a yearly basis, with the first revised version reflecting up to 25% change from the Cycle 6 version. Subsequent revisions will be limited to those identified as critical data requirements of funders and the target research/data user communities.

2. Sample Design: Each year, the sample will consist of 8 self-representing areas and 25

non-self-representing areas (a total of 33 areas). Each year, a new set of 25 non-self- representing areas will be used, total (25x 4) + 8 108 areas in 4 years of interviewing.

3. Use of Commercial address listings.--Commercial address lists will be used where possible. It is assumed that about 85% of the sample segments will utilize commercial listings and 15% (mostly rural areas) will require traditional Address listings. As in

Cycle 6, a double-sample of non-respondents will be used.

4. CAPI and Audio CASI Hardware: All hardware shall be reviewed and approved

by the Project Officer before use and will meet or exceed the hardware specifications

used for the Cycle 6 Main Study.

5. Study Materials: As stated in the original contract, the contractor shall adapt the following Cycle 6 materials for use in Cycle 7. It is expected that these materials can and should be used with only minor revisions:

• the introductory materials, including advance letters for respondents,

consent forms, and brochures,

• interviewer and supervisor manuals,

• training materials, and

• data processing programs.

6. Interviewers: In order to reduce costs related to recruiting, hiring, training, and supervision of interviewers, the contractor shall hire a smaller interviewer corps than in Cycle 6. In Cycle 7, these interviews shall be permanent full-time staff. As in Cycles 1-6, all interviewers shall be female.

7. Training: Interviewer training will be similar in content and length to Cycle 6 training, but with extra emphasis on mock interviews. In an effort to contain training costs,

the contractor shall investigate the use of home-based training using CD-ROM’s for portions of the General Interviewer Training and some portions of the study-specific training, with the review and approval of the Project Officer.

8. Data Collection: The contractor shall conduct field work and data preparation as

generally described for Phase I of the contract (and its modifications), except for differences described in this modification.

9. Deliverables: The deliverables will be limited to those listed in the updated version of the Schedule of Deliverable Items for Phase II (NSFG Cycle 7) included in this modification.

Section F—Deliveries or Performance

F.2. DELIVERABLES

Schedule of Deliverable Items for Phase II (NSFG Cycle 7) --This modification updates the Schedule of Deliverable Items as follows:

Deliverable Description Number of copies

1. Monthly Activity Report Summary of Past Month’s 1 by email

fieldwork activities, costs

for each month of interviewing

2. Final Production CAPI with ACASI, on CD 1 on CD

Questionnaire (English and Spanish)

3. Interviewer & Supervisor Agenda and Manuals, 1 on CD &

Training Materials Submitted annually 1 on paper

4. Field Status Report Costs, production, quality control. 1 by email

Submitted as requested by Project Officer.

5. Final Field Materials Advance letter, brochure, consent 1 on CD

forms, etc. Annually.

6. Draft File Lay-outs Female, male, interval 1 by email/CD

7. Preliminary Respondent Female and male data and 1 on CD

& Interval file data documentation with selection

& documentation and non-response weights and

preliminary recodes.

8. FINAL Respondent & Female & male data files & 1 on CD

Interval file data documentation with post-stratified

& documentation weights & final imputed recodes.

9. Interviewer Observations from completed 1 on CD

Observations interviews

10. Section-level timing data Section lengths for each section 1 on CD

of the male and female questionnaires.

11. Trend Analysis File 300 pre-selected variables 1 on CD

(primarily recodes) from

Cycles 1-7 for females & Cycles 6-7

for males. (Updated once/year).

12. DRAFT Series 1 Report Cycle 7, year 1 base report will be 1 on email

updated with the “annual report” in

deliverable 18.

13. REVISED Series 1 Report Revisions based on NCHS 1 by e-mail

Review

14. FINAL Series 1 Report Final Series 1 Report 1 on CD

Will be prepared for

publication after year 2.

15. DRAFT Series 2 Report Cycle 7, year 1 base report. 1 electronic

16. REVISED Series 2 report Revisions based on 1 electronic NCHS review.

17. FINAL Series 2 report Final Series 2 report. 1 electronic

Will be prepared for

publication after year 2.

18. Annual Report Update to Series 1 and 2 1 electronic

Reports for each year of

data collection.

19. Contextual Data and Contractor shall load 1 on CD

Documentation contextual data onto the

data with Deliverable 8.

20. Quality Profile Report Each year by key indicators 1 electronic

(about 10 variables)

10. (Omitted.)

11. Management: Once per week, the contractor and the NCHS Project Staff shall have a management meeting via conference call to review the progress of the project. Two in-person meetings will be held per year at NCHS.

12. Section H.1, HHSAR 352.270-5---Key Personnel:

The following Key Personnel will continue to serve in their positions:

Robert M. Groves, Ph.D (Project Director);

William Axinn, Ph.D. (Deputy Project Director),

Mick Couper, Ph.D. (Director of Questionnaire Development), and

James Lepkowski, Ph.D. (Chief Mathematical Statistician).

Krishna Winfrey, Field Director, has left the University of Michigan, and the contractor proposes to replace her with Nicole Kirgis (vita attached).

The contractor also proposes to replace Erik Austin (Director of Data Processing) with Dr. Peter Granda, who worked extensively on Cycle 6.

Both replacements appear to be very well qualified.

13. Task List: To keep up with changes in computer and survey monitoring technology and accounting practices, the contractor proposes to modify slightly the list of 9 “tasks” that were used to categorize work on the project. The project officer’s judgment is that this modification is useful and will give us essentially the same information as the previous set of tasks.

All other terms of the contract remain unchanged.

William D. Mosher, Ph.D.

Project Officer

National Survey of Family Growth

National Center for Health Statistics

3311 Toledo Road, Room 7421

Hyattsville, MD. 20782

301-458-4385

301-458-4033

Alternate Project Officer:

Joyce C. Abma, Ph.D.

301-458-4058

7