0720-hrbs_ssb_3.23.2023

NOTE: Complete Part B for Survey ICR Requests

SUPPORTING STATEMENT – PART B

B.  COLLECTIONS OF INFORMATION EMPLOYING STATISTICAL METHODS

If the collection of information employs statistical methods, it should be indicated in Item 17 of OMB Form 83-I, and the following information should be provided in this Supporting Statement:

1.  Description of the Activity

The potential respondent universe is all active-duty service members in the Army, Air Force, Marine Corps, Navy, Space Force, and Coast Guard below the rank of flag officer (O7 and above). Trainees and students at the miliary academies are also excluded. We will use a stratified random sampling procedure based on service branch, pay grade, and gender. Please see the attached sampling plan for details.

2.  Procedures for the Collection of Information

As noted above, we will use a stratified random sampling procedure based on service branch, pay grade, and gender. Sample selection is based on a maximum margin of error of 3% on one of strata dimension and 5% on two dimensions (e.g. gender X service branch). Expected response rates are based on those from the 2018 HRBS. Given differences in expected response rates and overall population sizes, we will oversample as follows:

• Women are sampled at twice the rate of men.

• Service members from the Coast Guard are sampled at a rate that is 1.5 times that of the other branches.

• Service members from the Marine Corps are sampled at a rate that is 1.75 times that of the other branches.

• Junior enlisted (E1-E4) service members are sampled at a rate that is 1.15 times that of the other paygrades.

Please see the attached sampling plan for details.

3.  Maximization of Response Rates, Non-response, and Reliability

The HRBS is now a confidential survey, so we are able to target follow-up reminders to only those participants who have not yet completed the survey. In addition, we will use an imputation procedure to address missing data among respondents. Based on our experience with the 2018 HRBS and with the DoD Women’s Reproductive Health Survey (WHRS) the vast majority of this missing is due to survey breakoff (i.e., not completing all the items in the survey). As in the 2018 HRBS and WRHS, we will also utilize analytic weights in all study analyses ensuring that the results generalize to all active-duty service members (excluding trainees and flag officers). Analytic weights are the product of non-response weights (designed to account for unequal probabilities of response across respondents) and sample weights (designed to account for the stratified random sampling approach).

4.  Tests of Procedures

We will pretest the instrument using a sample of military fellows (Army and Air Force) who are part of a year-long program at RAND. The size of this cohort will be 9 or fewer.

5.  Statistical Consultation and Information Analysis

Statistical aspects of the design and data analysis (RAND Corporation):

• Sarah Meadows, Ph.D., RAND Corporation (Study lead)

• Stephanie Holiday, Ph.D., RAND Corporation (Study co-lead)

• Terry Schell, Ph.D., RAND Corporation

• Matthew Cefalu, Ph.D., RAND Corporation (until May 20, 2022)

• Rebecca Collins, Ph.D., RAND Corporation

• Robin Beckman, MPH, RAND Corporation

Data collection:

• Westat is the survey vendor.

• Staff include: Michael Hornbostel (Study lead), April Fales, Jackson Sauls, Bryan Davis, Karin Wilson, Nicholas Petho, Mary Kliesch, and Reina Sprankle

2