2019 Request for Information (consultation and planning)

NSFG

OMB Attachment F2

OMB No. 0920-0314

Notice Type: Special Notice - RFI
Posted Date: Friday, March 22, 2019
Response Date: Tuesday, April 30, 2019
Archiving Policy: Automatic, 15 days after response date
Archive Date: 05/08/2019
Original Set Aside: N/A
Set Aside: N/A
Classification Code: A -- Research & Development
NAICS Code: 541720 -- Research and Development in the Social Sciences and Humanities
Solicitation Number: 2019-RFI-NSFG
Notice Type: Special Notice - RFI
Synopsis:
DEPARTMENT OF HEALTH AND HUMAN SERVICES (DHHS)
Centers for Disease Control and Prevention (CDC)
National Center for Health Statistics (NCHS)
Request for Information (RFI) No. 2019-RFI-NSFG
National Survey of Family Growth - NSFG
I. Introduction:
This is a Request for Information (RFI) per FAR 15.201(e) for market research purposes. This RFI
seeks to obtain ideas for the next implementation of the National Survey of Family Growth (NSFG). It
is not a request for proposal and does not commit the Government to issue a solicitation, make an
award, or pay any costs associated with responding to this notice. All responses shall remain with
the Government and will not be returned. As part of the RFI process, any contractor who provides a
favorable response may be invited to NCHS for an oral presentation.
Response Due Date:
By 3:00 PM EST on Tuesday, April 30, 2019 via email to [email protected].
Communication:
1. All correspondence is to be with the Contracting Officer, Nathan Amador, via email to
[email protected].
2. The Government will accept questions regarding this RFI. All questions will be reviewed and
answered as decided by the Government, and must be submitted via email to [email protected] by
3:00 PM EST on Wednesday, April 17, 2019.
3. Please ensure that any e-mail responses have "2019-RFI-NSFG" in the subject line. Any responses
received after the specified response due date and time will not be accepted or acknowledged.
4. Responses are limited to 25 pages, double-spaced, in Times New Roman 12 font, and 1" margins.
Responses using multimedia or video are limited to ten (10) minutes in duration.

1

NSFG

OMB Attachment F2

OMB No. 0920-0314

II. Background:
The National Survey of Family Growth (NSFG) is conducted by the National Center for Health
Statistics (NCHS), part of the Centers for Disease Control and Prevention (CDC). The NSFG is
currently a survey of men and women ages 15-49 in the household population of the United States. It
serves the data needs of NCHS as well as several other co-sponsors within CDC and other agencies
within DHHS. NCHS conducted the NSFG six times on a periodic basis between 1973 and 2002, with
Cycles 1-5 (1973, 1976, 1982, 1988, 1995) based on women 15-44 and Cycle 6 (2002) based on
independent samples of men and women 15-44. In 2006, NSFG shifted to a continuous fieldwork
design, which apart from a hiatus between June 2010 and September 2011, has continued to the present
time. The NSFG age range was expanded to 15-49 beginning in September 2015.
The NSFG is a principal source of national estimates of factors affecting pregnancy and birth rates,
including
sexual activity, cohabitation, marriage, and divorce, contraceptive use for both birth control and disease
prevention, miscarriage and stillbirth (pregnancy loss), infertility, wanted and unwanted births, adoption,
use of family planning clinics, use of medical services for family planning and infertility. This is the
original focus of the survey and remains its main mission, but the survey has evolved to include
measurement of STD and HIV risk behaviors and to gather analogous information from men, including
fatherhood involvement.
Data and documentation for each of the NSFG Public-Use File (PUF) releases from 1973 through 20152017 are accessible on the main NSFG website, within “Questionnaires, Datasets and Related
Documentation”:
https://www.cdc.gov/nchs/nsfg/
The following is a direct link to the NSFG data/documentation page for the 2015-2017 NSFG PUF
released in December 2018:
https://www.cdc.gov/nchs/nsfg/nsfg_2015_2017_puf.htm
As with all NSFG PUF releases, this webpage includes links to the full survey questionnaires, along
with extensive documentation for each of 3 data files provided for public use:
a. female respondent file (one observation for each female respondent)
b. pregnancy (interval) file (one observation for each pregnancy reported by female respondents)
c. male respondent file (one observation per male respondent).
In particular on the 2015-2017 NSFG page, see:
1. Codebooks linked under “Webdoc interactive codebook,” with separate entries for each PUF
variable showing unweighted frequencies and other basic information.
2. File indexes linked under Appendix 1 of the User’s Guide, listing each PUF variable in order of
file layout, with a brief description.
3. Restricted-use analytic variables (those available only through the NCHS Research Data
Center to minimize the risk of disclosure for individual-level data), with summaries linked under
Appendix 7 of the User’s Guide.
While every NSFG PUF reflects protections to minimize the risk of disclosure for individual-level data,
the 2015-2017 NSFG PUF release required the suppression or modification of a significantly higher
number of variables due to concerns about their disclosure risk, relative to prior NSFG PUF releases.
Of particular note, many century-month date variables for key life events were suppressed, leaving only
2

NSFG

OMB Attachment F2

OMB No. 0920-0314

the year of the events for public use, while in prior PUFs, month, year, and century-month variables for
these events were available for public use. The increased number of variable suppressions and
modifications can be seen by comparison of the Restricted-Use Analytic Variables described in
Appendix 7 of the 2015-2017 NSFG PUF, relative to Appendix 7 for the 2013-2015 NSFG linked here:
https://www.cdc.gov/nchs/nsfg/nsfg_2013_2015_puf.htm
In order to illustrate the range of analytic uses for NSFG data, the NSFG webpage also provides listings
of NCHS publications and other reports and articles based on NSFG data. These bibliographies,
arranged by NSFG survey years, are linked below:
https://www.cdc.gov/nchs/nsfg/nsfg_products.htm
https://www.cdc.gov/nchs/nsfg/nsfg_bibliography.htm
The current continuous survey design for the NSFG calls for a nationally representative sample of men
and women 15-49 years of age with oversamples of Hispanic and Black respondents, and those aged 1519. The NSFG sample excludes current residents of military bases and institutions (e.g., long-term
hospitals, jails, prisons). College students temporarily away from their homes living in campus housing
are included by sampling them at their home address; they can be interviewed either at home or at
college. In each sample year, about 15,000 households are screened for eligible members, to produce a
minimum of 5,000 main interviews. All interviews currently take place in respondents’ homes,
conducted in-person by trained female interviewers. Interviews are administered using computer
assisted personal interviewing (CAPI) with laptop computers, with a portion of the more sensitive
questions administered via audio computerized self-administered interview (ACASI). The average
interview length using current survey questionnaires is approximately 75 minutes for females and 50
minutes for males -- under the OMB-approved 80 minutes for female and 60 minutes for male
interviews. Spanish is the only language besides English that is currently offered for conducting NSFG
interviews.
The NSFG is currently based on a stratified multi-stage area probability sample, with about 35 PSUs
selected for each annual sample. In total, a sample of 213 PSUs were selected for the current contract’s
8-year fieldwork period. While each single year of NSFG fieldwork provides a national sample, NCHS
has only released NSFG public-use files every 2 years because a minimum of 2 years of data collection
are needed for sufficient statistical power.
The NSFG is currently conducted using survey design features that were first put into place in 2006.
The NSFG is conducted using a responsive design approach, with extensive use of paradata, to manage
the fieldwork, control costs, and reduce bias in the resulting sample. Data collection for each annual
sample is conducted in 4 quarters (12 weeks each). Weeks 1-10 of each quarter are referred to as Phase
1, and weeks 11-12 are referred to as Phase 2. The incentive protocol and sampling differ for Phase 2,
for which a subsample of non-final cases is selected, and an increased incentive is offered. For more
details on this design please see the following reports:
https://www.cdc.gov/nchs/data/nsfg/NSFG_2013-2015_Summary_Design_Data_Collection.pdf
https://www.cdc.gov/nchs/data/nsfg/NSFG_2013-2015_Sample_Design_Documentation.pdf
https://www.cdc.gov/nchs/data/nsfg/NSFG_2013-2015_Sampling_Error_Estimation_Codes.pdf
https://www.cdc.gov/nchs/data/nsfg/NSFG_2013-2015_Weighting_Design.pdf

3

NSFG

OMB Attachment F2

OMB No. 0920-0314

For background on the initial development and implementation of the features of the NSFG survey
design, see:
https://www.cdc.gov/nchs/data/series/sr_01/sr01_048.pdf
Planning and Development of the Continuous National Survey of Family Growth:
Describes planning for and implementation of the transition from a periodic to a continuous
survey, prior to the release of the first data from continuous interviewing.
https://www.cdc.gov/nchs/data/series/sr_02/sr02_150.pdf
The 2006-2010 National Survey of Family Growth: Sample Design and Analysis of a
Continuous Survey:
Describes the sample design and weighting and variance estimation procedures under the
continuous design, prior to the release of the first data from continuous interviewing.
https://www.cdc.gov/nchs/data/series/sr_02/sr02_158.pdf
Responsive Design, Weighting, and Variance Estimation in the 2006-2010 National Survey
of Family Growth:
Presents fieldwork results and weighting, imputation, and variance estimation procedures
corresponding to the first release of data (2006-2010) under the continuous design.
III. Information Requested:
The Government expects the responder to address the “General” and “Specific” for Topic (A-E) listed
below:
General:
Responses should reflect the goal of maintaining or improving the survey’s:
1. interview production (sample size);
2. response rates;
3. coverage;
4. cost control; and
5. ability to meet analytic requirements;
in 2022 and beyond in the face of likely continuation of decreasing receptivity of the population to
participation in household surveys. (For 1-5 above, reference links in part II. Background)
Specific:
1. Estimates must be nationally representative by race and Hispanic origin (Hispanic, non-Hispanic
black, non-Hispanic other), age groups (15-19; 20-29; 30-39; 40-49) and sex
2. A minimum of 5,000 interviews must be completed each year
3. A response rate of 70% should be considered as a target
4. Some means of nonresponse-bias assessment is necessary
5. Should be able to detect differences over time (adjacent 2-year time periods) and between
important subgroups with two years of data, for key estimates
6. Include discussion of costs for each area (only rough estimates / ranges)
7. Some areas overlap, and responses can portray this overlap
8. If experiments / field tests during data collection are thought to be useful, without sacrificing
basic output, please describe
9. If an option in the list of areas to address would not be advisable, discussion of that nature is
welcome.
4

NSFG

OMB Attachment F2

OMB No. 0920-0314

All Topics (A-E) below must be addressed, but equal length of response for each topic is not required.
A. MODES OF INTERVIEWING
How different interview modes (in addition to or instead of in-person) could be used to maintain or
enhance the survey’s response rates, coverage, and data quality. Including: how could current NSFG
content and visual aids be adapted to smaller screens and from hard copy to digital? For example,
female respondents write on a 9” x 11” paper “Life History Calendar” as an aid to recalling detailed
contraceptive use histories and histories of dates of relationships and other life events.
B. DATA LINKAGE
Linkage of NSFG data with administrative data (such as birth certificate data, Medicaid, National
Death Index):
The goals for data linkage may include reduction of respondent burden as well as improvement in
data quality (for example, by getting some information directly from birth records or obtaining
service payment information from CMMS). Validation of survey responses is of lesser potential
value to NSFG, but can be considered. Looking at current NSFG survey content, discuss possible
data linkages, with particular attention to logistical challenges, informed consent issues, risk to
response rates, and increased disclosure risk for survey participants.
C. DISCLOSURE RISK
Balancing disclosure risk and data accessibility:
The most recently released 2015-2017 NSFG public-use files (PUF) include significantly more
variable suppressions and modifications than prior NSFG PUF releases. In particular, the majority of
“century-month date” variables for key life course events of pregnancy, marriage, divorce, and
cohabitation have been suppressed from the PUF in order to address risks of potential matching with
publicly available, administrative data. These original CM date variables, along with a number of
other variables that were collapsed or categorized in some way for public use, are now restricted-use
variables only accessible with an approved research proposal to use the Research Data Center. For
more detail on these protections undertaken to minimize disclosure risk associated with the NSFG
2015-2017 PUF, including the use of statistical perturbation, see the User’s Guide section on this
subject, as well as Appendix 7.
Discuss potential additional or alternative approaches for making such restricted-use data more
accessible to researchers while still protecting the confidentiality of survey participants. If relevant,
include discussion of any major future developments in approaches to data protection.
D. BIOMARKERS
Discuss collection of biomarkers for NSFG survey participants either in the household or another
site. Pay particular attention to biomarkers that would enhance NSFG utility or complement existing
survey content. Discuss challenges including: impact on clearance procedures stemming from
biomarker collection; changes to recruitment, training, and retention of interviewers (for example,
based on skills needed for specific samples collected); risk to response rates; and any other critical
aspects for cost and implementation.
E. OTHER (optional, but must stay within 25 page limit)

5

NSFG

OMB Attachment F2

OMB No. 0920-0314

Include other opportunities/ideas/innovations, that are not included in this list, if any, that would
benefit the NSFG and maintain at least the basic requirements as specified in “General” and
“Specific” above.
Place of Performance:
Contractor's Facility; and
National Centers for Health Statistics 3311 Toledo Road Hyattsville, MD 20782; and
Multiple Locations in the United States
Contracting Office Address:
Mailstop TCU 4
2900 Woodcock Blvd
Atlanta, Georgia 30341
Primary Point of Contact:
Nathan R. Amador, JD
Contracting Officer, Branch 4, Team 3, in support of NCHS and NIOSH
Office of Acquisition Services (OAS)
Office of Financial Resources (OFR)
Office of the Chief Operating Officer (OCOO)
Centers for Disease Control and Prevention (CDC)
e-mail: [email protected]

6