Non-substantive Change Request Justification

Request for Approval of a Non-Substantive Change to the

National Survey of Family Growth

OMB No. 0920-0314

(Expiration: 12/31/2024)

Contact Information:

Anjani Chandra, Ph.D.

Principal Investigator and Team Lead

National Survey of Family Growth Team

Division of Vital Statistics/Reproductive Statistics Branch

CDC/National Center for Health Statistics

3311 Toledo Road, Room 5414

Hyattsville, MD. 20782

301-458-4138

301-458-4034 (fax)

[email protected]

March 14, 2022

1. Circumstances making the collection of information necessary

This request is for a non-substantive change to the National Survey of Family Growth (NSFG)

(OMB No. 0920-0314, Exp. Date 12/31/2024), conducted by the National Center for Health Statistics (NCHS), Centers for Disease Control and Prevention (CDC).

NCHS, under its duties specified in 42 U.S.C. 242k, Section 306(a and b)(1)(h) of the Public Health Service Act, conducts the NSFG to collect and disseminate “statistics on family formation, growth, and dissolution” among a nationally representative sample of reproductive-age women and men in the U.S. household population. The NSFG supplements and complements data from birth and fetal death certificates by monitoring factors (such as sexual activity, contraception, marriage and cohabitation, and infertility) that affect birth and pregnancy rates. In addition, the NSFG serves a variety of data needs in public health programs that sponsor and depend on it, including several divisions within CDC/NCHHSTP and CDC/NCCDPHP.

Our recent OMB reinstatement package, approved on December 2, 2021, specified a multi-mode, multi-phase survey design that builds on the most successful features of the continuous fieldwork design used for NSFG from 2006-2019. During this period, NSFG had been conducted solely through in-person interviews, with a portion of the survey conducted using audio computer-assisted self-interview (ACASI). After data collection ended in September 2019 and a new contract was awarded in September 2020, the reinstatement supporting statements requested approval to resume data collection in January 2022, and to conduct several methodologic studies in order to offer an alternative to the in-person interview mode. These mode alternatives were considered partly in response to the COVID-19 pandemic, and partly to address ongoing challenges with maintaining acceptable response rates, minimizing nonresponse bias, and reducing overall respondent burden.

NSFG resumed data collection in January 2022 and is scheduled to proceed through December 2029 with annual, continuous data collection with a multi-mode, multi-phase design including online and in-person data collection. The COVID-19 pandemic has impacted in-person data collection directly, with periods when data collection is not permissible in many—and at times most—counties in the NSFG sample. The pandemic has had a similar indirect impact on in-person data collection, with nationwide labor shortages making it difficult to attain and to retain field interviewers. People may be less likely to open the door and to agree to sit down with an interviewer due to COVID-related concerns. The Delta and Omicron variants have demonstrated the extreme extent of these impacts, and that they can be cyclical and long-lasting.

As a result, we have had to modify the NSFG data collection procedures for Quarter 1 that began on January 11. While the original multi-mode, multi-phase design laid out in the reinstatement request specified that field interviewers would begin in-person data collection during Phase 2 (weeks 5-12 of the quarter), Quarter 1 is proceeding with only web data collection for Phases 1 and 2. Field interviewers will conduct Phase 3 interviews with higher incentives for a subsample of non-respondents from Phases 1 & 2, as originally planned, and will follow up with 100% of web survey breakoffs from the prior phases. In addition, the mode comparison study originally planned for Quarters 1 and 2 of 2022, which depends on full staffing of interviewers, will be shifted to Quarters 2 and 3 of 2023. (Further rationale for this shift is provided later in this document.)

Due to the ever-evolving changes of the COVID-19 pandemic, in preparation for later data collection quarters, the NSFG multi-mode design needs to further reduce reliance on in-person data collection in periods when in-person data collection is limited or not possible. Increasing participation to modes other than in-person (face-to-face) can be achieved through two main strategies: offering additional non-in-person modes and motivating sample members to complete the survey in these modes. We describe these strategies further below and given the timing of this request, we seek OMB approval to conduct two experiments to test these approaches in Quarter 3, which begins on June 22, 2022. Quarter 2 that begins on March 30, 2022 would proceed with the previously approved multi-mode design.

Experiment to test an additional mode for responding to household screener. Most of the nonresponse in household surveys relying on mailed invitations to a web survey is at the household screener stage. Adding a paper screener is a promising method to reduce screener nonresponse and/or reduce reliance on in-person follow-up. We propose an experiment to be conducted in Quarter 3 of 2022 (starting on June 22, 2022) in which we would mail a paper screener (Attachment 1) to 50% of the sample on the third mailing (approximately one week after the first mailing). This paper screener will be sent along with a letter that invites them to complete the paper screener and also reminds them about the web screener option (Attachment 2). For the 50% of households not receiving the paper screener, the letter to remind them to complete the web screener is shown in Attachment 3. All households will also receive the NSFG Family Facts sheet (as included in our reinstatement package approved in December 2021), a tool to help sample members understand how the survey data will be used. Due to complexities imposed by the NSFG target population aged 15-49 and the consent protocol for minors, the paper screener would be used to select only an eligible adult household member to participate in the main survey. If this experiment is approved in time for Quarter 3 administration, results would be used to evaluate primarily response rates, demographic composition, and cost indicators.

The NSFG mailing protocol has five potential mailings for the household screener. These are listed in Table 1, showing the third mailing (in blue) that will include the paper screener for 50% of the nonresponding households.

Table 1. NSFG screener mailings with and without the paper screener.

	Without a paper screener	With a paper screener
1st mailing	Invitation and Q&A brochure, $2 visible	Same
2nd mailing	Pressure sealed reminder	Same
3rd mailing	Large envelope with letter and Family Facts sheet	Include paper screener, letter, Family Facts sheet, and postage-paid envelope
4th mailing	Postcard reminder	Same
5th mailing	UPS Mail Innovations mailing with letter	Same

Experiment to test an earlier delivery of the higher incentives used in Phase 3. When in-person data collection is not possible or is severely limited, the NSFG Phase 3 protocol may be an effective way to increase participation, essentially to incentivize participation from the beginning. The Phase 3 protocol includes an additional $5 prepaid screener incentive and an additional $40 prepaid main survey incentive. We propose to conduct an experiment in which half the sample would be offered the Phase 3 incentives at the normal time during Quarter 3 (weeks 37-40 of Year 1 ), and half would be offered these incentives at the start of Quarter 3 in Phase 1. The experimental incentive condition will be assigned to 50% of the sample at the address level. As this Phase 3 protocol is part of the multi-mode, multi-phase protocol already approved and in place for Quarters 1 and 2, most materials and procedures have been developed for implementation. In areas where in-person data collection is not possible, starting with the Phase 3 incentives may attract greater attention to the survey request and be more effective when used at survey launch rather than after numerous prior mailings offering a lower incentive. By increasing the response rate, the cost of higher incentives can be offset by reducing the starting sample size and requiring fewer mailings per completed survey. In areas where in-person data collection is possible, the cost efficiency can be even greater, by reducing the number of sample addresses that require in-person data collection (interviewer hours, miles travelled, and other expenses).

The Phase 3 $5 prepaid screener incentive will be provided in the third screener reminder mailing to increase its effectiveness (in addition to the initial mailing with $2). This incentive treatment is not being designed explicitly to interact with the paper screener experiment, but when paper screeners are assigned, it will coincide with the mailing that includes the mailed screener.

Experimental design for the paper screener and incentives experiments. A 2x2 experimental design will be used in Quarter 3 of 2022, crossing the paper screener experimental condition and offering the Phase 3 incentive from the survey start experimental condition. Table 2 shows the expected sample allocation, using equal allocation for each treatment.

Table 2. Experimental design and approximate sample sizes for each condition, Quarter 3 of 2022.

Condition		Control	Experimental
		No Paper Screener	Paper Screener	Total
Control	Start Quarter with Phase 1&2 Incentives, Followed by Phase 3 Incentives	1,216	1,216	2,432
Experimental	Start Quarter with Phase 3 Incentives	1,216	1,216	2,432
	Total	2,432	2,432	4,864

Planned analysis and implementation. Analysis of the two experiments will be done in Quarter 4. Production outcomes will be monitored via the project dashboard. If this analysis yields a recommendation to implement one or both treatments, our likely start will be Quarter 2 of 2023 to allow time for analysis and additional protocol approval. As noted earlier, the mode comparison study has been shifted to Quarters 2 and 3 of 2023. This timing is intended to coincide with any potential protocol changes based on the experimental results so that the findings of the mode comparison can better generalize to the design going forward.

2. Purpose and use of the information collection

The NSFG responds to the congressional mandate for NCHS to collect and publish reliable national statistics on “family formation, growth, and dissolution” (Sec. 306(a and b), paragraph 1(H) of the Public Health Service Act) as well as vital statistics on births and deaths, and a number of aspects of health status and health care. The NSFG collects and publishes the most reliable, and in most cases the only, national data to monitor such major topics as: contraceptive use and effectiveness, infertility and use of infertility services, unintended births, self-reported pelvic infection and sexually transmitted disease, sterilization, expected future births, marriage and cohabitation, the sexually active population, and the use of and need for family planning services. Under the continuous data collection design planned for the survey in this reinstatement request, the NSFG will be able to maintain adequate sample sizes for reliable time series for nationally representative statistics on these major topics at an affordable cost.

No changes to NSFG’s survey content are proposed in this nonsubstantive change request. The in-person version of the main survey will still include a computer-assisted self-interview (CASI) component at the end, and the online version of the main survey will by definition remain self-administered in entirety.

9. Explanation of any payment or gift to respondents

Of the two modifications described in this non-substantive change request, the one that involves offering an additional mode for responding to the household screener will not result in the receipt of any additional incentives. Participants in the paper screener experiment will receive the incentives in the amount and timing described in the approved reinstatement package for NSFG.

The experiment to test earlier offering of the previously approved Phase 3 incentives will lead to more participants receiving the higher incentive amounts. Unlike the 2006-2019 NSFG design, each quarterly data collection is started without interviewer administration. Due to COVID-19, self-administration may be the only method of data collection in some areas, as previously described. As noted in section 1, if the experiment demonstrates that response rates are higher with the earlier offering of Phase 3 incentives, the cost of higher incentives can be offset by reducing the starting sample size and requiring fewer mailings per completed survey. In areas where in-person data collection is possible, the cost efficiency can be even greater, by reducing the number of sample addresses that require in-person data collection.

10.   Protection of the Privacy and Confidentiality of Information Provided by Respondents

The following burden and confidentiality statements will be included in the paper screener questionnaire (Attachment 1), just as it is already included in the Blaise-programmed screener questionnaire previously approved:

CDC estimates the average public reporting burden for this collection of information as 3 minutes per response, including the time for reviewing instructions, searching existing data/information sources, gathering and maintaining the data/information needed, and completing and reviewing the collection of information. An agency may not conduct or sponsor, and a person is not required to respond to a collection of information unless it displays a currently valid OMB control number. Send comments regarding this burden estimate or any other aspect of this collection of information, including suggestions for reducing this burden to CDC/ATSDR Information Collection Review Office, 1600 Clifton Road, MS D-74, Atlanta, GA 30333; ATTN: PRA (0920-0314).

Assurance of Confidentiality – We take your privacy very seriously.  All information that relates to or describes identifiable characteristics of individuals, a practice, or an establishment will be used only for statistical purposes.  NCHS staff, contractors, and agents will not disclose or release responses in identifiable form without the consent of the individual or establishment in accordance with section 308(d) of the Public Health Service Act (42 U.S.C. 242m) and the Confidential Information Protection and Statistical Efficiency Act (Title III of the Foundations for Evidence-Based Policymaking Act of 2018 (Pub. L. No. 115-435, 132 Stat. 5529 § 302)).  In accordance with CIPSEA, every NCHS employee, contractor, and agent has taken an oath and is subject to a jail term of up to five years, a fine of up to $250,000, or both if he or she willfully discloses ANY identifiable information about you.

12. Estimates of annualized burden hours and costs

Table 12A shown below from our approved reinstatement request shows that the estimated annualized burden associated with the household screener interview. The proposed paper screener experiment is not expected to alter this estimate because the mailed paper screener is estimated to take the same amount of time as the web-mode screener and is intended to move us closer to the targeted number of completed screeners. The proposed incentive experiment would also have no impact on the estimated burden for household screeners or main interviews for males or females because it too is intended to move us closer to the targeted numbers of completed screeners and main interviews.

12.A Estimated Annualized Respondent Table

Respondents	Form	Number of Responses	Responses per Respondent	Average Burden/ Response (in hours)	Total Burden Hours
Household member	Screener Interview	13,500	1	3/60	675
Household Female 15-49 years of age	Female Interview	2,750	1	75/60	3,438
Household Male 15-49 years of age	Male Interview	2,250	1	50/60	1,875
Household member	Screener Verification	1,350	1	2/60	45
Household Individual 15-49 years of age	Main Verification	500	1	5/60	42
Household Female 15-49 years of age	Respondent debriefing questions about calendar	325	1	3/60	16
Household member	Phase 4 nonresponse follow-up questions	375	1	5/60	31
TOTAL	6,122

The average response burden cost for the NSFG was estimated to be $183,415, and this would be unchanged by the modifications described in this request. (Wage information is from the Bureau of Labor Statistics: http://www.bls.gov/news.release/empsit.t19.htm.)

12.B Estimated Annualized Respondent Costs

Total Burden Hours	Respondent Wage Rate per Hour	Total Respondent Costs
6,122	$29.96	$183,415

15. Explanation for program changes and adjustments

The proposed modifications described in this submission do not change the estimated average burden hours from the previously approved clearance (see Table 12A).

List of attachments

Attachment 1 – Paper Screener

Attachment 2 – Reminder letter to be mailed with paper screener

Attachment 3 – Reminder letter to be mailed without paper screener