NLSY27_Pretest_OMB_Justification_Part B

National Longitudinal Survey of Youth 2027 Pretest

OMB Control Number 1220-xxxx

OMB Expiration Date: N/A

SUPPORTING STATEMENT FOR

National Longitudinal Survey of Youth 2027 Pretest

OMB CONTROL NO. 1220-XXXX

2. COLLECTIONS OF INFORMATION EMPLOYING STATISTICAL METHODS

1. Describe (including a numerical estimate) the potential respondent universe and any sampling or other respondent selection methods to be used. Data on the number of entities (e.g., establishments, State and local government units, households, or persons) in the universe covered by the collection and in the corresponding sample are to be provided in tabular form for the universe as a whole and for each of the strata in the proposed sample. Indicate expected response rates for the collection as a whole. If the collection had been conducted previously, include the actual response rate achieved during the last collection.

This statement covers the National Longitudinal Survey of Youth 2027 (NLSY27) Pretest. This Pretest will ensure all systems, protocols, and procedures are operating successfully for Round 1 data collection and experiment with certain aspects of data collection to improve efficiencies and reduce respondent burden.

The target population for the NLSY27 is youth aged 11–16 years as of December 31, 2026, in households in the United States (population size 25,933,328 youth per American Community Survey Public Use Microdata Sample 2023,1 year).

Because the NLSY27 Pretest is occurring 1 year prior to the cohort formation in the Round 1 survey, the age eligibility criteria will be adjusted appropriately. The Pretest is expected to yield approximately 800 completed youth interviews with youth aged 11–16 years as of December 31, 2025, along with 518 surveys from their parents or caregivers. We present planned Pretest sample sizes starting with the 13,381 housing units (HUs) selected from occupied and screened HUs (i.e., households) to arrive at the desired 800 youth completes below.

Exhibit 1. NLSY27 Pretest Sample Sizes and Assumed Rates

	Count	Rate, %
Housing Unit (HU) sample	13,381
Occupied HUs	12,177	91.0
Screened HUs	6,088	50.0
Households with age-eligible youth	1,218	20.0
Households selected for interview	895	73.5
Average eligible youth per household	1.39
Eligible youth	1,244
Household screening and consent	931	74.9
Youth interviews	800	85.9
Eligible parents or caregivers	576	71.9
Parent/caregiver interviews	518	90.0

Note: Numbers and percentages are rounded.

The Pretest will be conducted in nine areas that will be selected purposively. Deliberately selecting Pretest areas allows a focus on ensuring population representation, accommodating data collection preferences for where the Pretest locations are, and avoiding 2026 Census test sites.¹ The Pretest study will include areas within the 48 contiguous states that meet the criteria of having significant Black, Hispanic/Latino, and rural youth populations as well as geographic spread across U.S. regions. This focus is informed by the decision to oversample the Black and Hispanic populations in line with the NLSY79 and NLSY97 studies. Additionally, steps are being taken to ensure significant representation of rural households. For the Pretest, we also aim to observe logistical or response-oriented factors that may be associated with these subpopulations, such as issues related to connectivity, to ensure that these subgroups are effectively included in the study design.

Although Pretest areas will not necessarily be equivalent in administrative or geographic structure to the primary sampling units in the NLSY27 Round 1 sample, the Pretest will use the same administrative structure for the secondary sampling units (SSUs) as will be used in Round 1: Census Block Groups (CBGs). When the Pretest areas are defined, the CBGs within them will be assessed first for their Address-Based Sampling (ABS) frame coverage. ABS frames are a high-quality, cost-effective household sampling frame for in-person, mail, and multi-mode surveys of populations in residential households. The foundation of ABS frames is the U.S. Postal Service’s Computerized Delivery Sequence file, which is made available through non-exclusive license agreements with qualified private companies (Iannacchione, 2011).²

An in-person interviewing ABS coverage rate (i.e., the estimated proportion of occupied housing units in a Secondary Sampling Unit (SSU) with a locatable mailing address) will be calculated to subset³ the pretest segments down to those with sufficient ABS coverage. The SSUs with sufficient ABS coverage will be stratified by race, ethnicity, and urbanicity, and a random selection of approximately 50 SSUs (each SSU will contain around 280 or fewer sampled households) will be drawn across the Pretest areas. The stratified SSU sample allocation will be designed to achieve the desired diversity of Pretest sample members.

The NLSY27 Pretest ABS frame will be limited to residential addresses, geocoded into selected SSUs, which will include city-style addresses but exclude non-locatable addresses like OWGM⁴ PO boxes (McMichael & Brown, 2018)⁵, highway contract boxes, and rural route boxes. The NLSY27 Pretest ABS frame will also exclude “No-Stat” addresses⁶.

No estimation of population characteristics will be done with the Pretest data, and subsequently no defined accuracy is needed.

2. Describe the procedures for the collection of information including:

• Statistical methodology for stratification and sample selection,

  • Estimation procedure,

  • Degree of accuracy needed for the purpose described in the justification,

  • Unusual problems requiring specialized sampling procedures, and

  • Any use of periodic (less frequent than annual) data collection cycles to reduce burden.

The Pretest will be conducted in nine areas that will be selected purposively. Within each area, the SSUs (which will be CBGs) will be stratified by race, ethnicity, and urbanicity, and a random selection of approximately 50 SSUs will be drawn across the Pretest areas. Given the target sample of selecting 13,381 households, around 280 households or less will be selected per SSU and contacted in the Pretest.

Pretest data collection will comprise three main components: (1) household screening, (2) youth interviews, and (3) parent/caregiver surveys. Each household will be contacted by mail or in-person to complete the Eligibility Screener. Respondents will be able to complete the Eligibility Screener by web, mail, telephone, or in-person. Households identified with eligible youth will be asked to complete the Household Screener by web or in-person.

After completing the Household Screener, all eligible youth in the household who are either Hispanic or non-Hispanic Black will be selected for the study, while approximately 75% of eligible youth from other racial or ethnic groups will be selected. One parent/caregiver for each household with selected youths will be asked to complete the parent/caregiver survey. The screening respondent will be immediately notified whether anyone in the household is selected to complete the youth and parent/caregiver interviews.

Parent/caregiver and youth respondents will complete consent or assent procedures for their participation at the start of their interviews. If the interviewer is unable to complete the parent/caregiver and/or youth interviews during the same visit, they will schedule an appointment to complete the interviews at another time. Cases will be kept open for the entire Pretest data collection period to allow for repeated follow-up attempts.

As outlined in Section 4, the Pretest will include a youth video interview experiment. For this experiment, households will be randomly assigned to an experimental or control group upon completion of the Household Screener. In the experimental group, if a selected youth is unable to complete the interview during the in-home visit and does not schedule a follow-up in-person interview, the case will be referred to the Video Interviewing Team for contact and completion. In the control group, face-to-face interviewers will continue visiting the household, with the video interview option becoming available only in Week 8 of data collection.

We anticipate that 670 youths and their households will be included in the experiment, based on expected 1,244 eligible youths from 895 screened households, assuming 54% of them are contacted during face-to-face data collection and complete the Household Screener. These cases are eligible for the experiment and will be randomized into the two treatments with approximately 335 youths and for each treatment. Assuming an average youth response rate of approximately 64%, we can detect an 8.9 percentage point decrease with 80% power using a Pearson’s chi-square test with a one-sided significance level of 0.05. This assumes a design effect of 1.2 due to an intracluster correlation (ICC) of 0.5 among youths within households. This detectable difference, which means identifying a drop in the response rate from 64% to 55.1% or lower, is within a reasonable range for this type of experiment.

3. Describe methods to maximize response rates and to deal with issues of non-response. The accuracy and reliability of information collected must be shown to be adequate for intended uses. For collections based on sampling, a special justification must be provided for any collection that will not yield "reliable" data that can be generalized to the universe studied.

No statistical method has been established to maximize response rates or address issues of non-response in the Pretest. No statistical inference about the target population will be made with this data because the purpose of the Pretest is only for testing methods, procedures, and materials in preparation for Round 1.

4. Describe any tests of procedures or methods to be undertaken. Testing is encouraged as an effective means of refining collections of information to minimize burden and improve utility. Tests must be approved if they call for answers to identical questions from 10 or more respondents. A proposed test or set of tests may be submitted for approval separately or in combination with the main collection of information.

The U.S. Bureau of Labor Statistics will embed the following experiment in the Pretest. The purpose of evaluating the video interviewing mode experiment is to plan for future cost efficiencies and to maximize response rates and the representativeness of participants in the national cohort.

Video Interviewing for Youth, Experiment:

Youth who do not complete an interview during the in-person interviewer visit will have the option to complete their interview virtually, but the timing of when video interviews are offered depends on whether the household is assigned to the experimental or control group of the NLSY27 Pretest video interview experiment. For cases in the experimental group, if they are unable to complete the interview when the interviewer is in the house, the interviewer will offer the video interview option. The remaining half of the sample will be offered the video interview option in Week 8 of data collection and no earlier. Households will be randomly assigned to the control or experimental group upon completion of the Eligibility Screener.

The primary questions underlying this experiment are the following:

• Does offering video interviewing earlier in data collection increase youth response rates?

• Does video interviewing encourage quicker response by youth?

• Are there significant differences in data quality based on mode of completion (video interview vs. in-person)? We will assess data quality by comparing key data quality metrics such as item missingness and data edits across the two modes of data collection completion.

• Does offering video interviewing earlier in data collection decrease non-response follow-up thereby decreasing data collection costs?

5. Provide the name and telephone number of individuals consulted on statistical aspects of the design and the name of the agency unit, contractor(s), grantee(s), or other person(s) who will actually collect and/or analyze person(s) who will actually collect and/or analyze the information for the agency.

RTI International developed the Pretest plan. The following individuals is the statistical contact:

• Dan Liao, RTI, NLSY27 Director of Statistics, 301-816-4605

1 https://www.census.gov/newsroom/press-releases/2024/2026-census-test-site-selection.html

2 Iannacchione, V. G. (2011). The changing role of address-based sampling in survey research, Public Opinion Quarterly, 75(3), 556–575. https://doi.org/10.1093/poq/nfr017

3 Round 1 will not exclude low coverage SSUs; rather, those SSUs will be field enumerated.

4 OWGM stands for “Only Way to Get Mail.”

5 McMichael, J., & Brown, D. (2018). PO boxes on address-based sampling (ABS) frames: Under- or over-coverage or both? In JSM Proceedings, Survey Research Methods Section. Alexandria: American Statistical Association.

6 A "No-Stat" address refers to an address that the United States Postal Service (USPS) designates as not receiving regular mail delivery.

NLSY27 Pretest OMB Part B Draft Page 5 of 5