Supporting Statement Part B

SUPPORTING STATEMENT – PART B

B.  COLLECTIONS OF INFORMATION EMPLOYING STATISTICAL METHODS

1.  Description of the Activity

This study’s objectives/research questions directly support Recommendation 2.3 from the Independent Revision Commission on Sexual Assault in the Military (IRC): “Implement community-level prevention strategies unique to Service members’ environments.” This study will be the first empirical research to provide critical findings on prevention gaps and needs at the community level (as well as the installation and unit levels) that are unique to Army Soldiers’ specific environments.

Correspondingly, this study will also directly support the related IRC Recommendation 2.3a, “The Services and the National Guard Burau should resource and implement prevention strategies at organizational and community levels,” under POAM milestones 7-15.

The study’s objectives/research questions are also likely to also directly influence a number of additional IRC recommendations, including 2.1c / 2.3a / 2.4 / 3.2 / 3.5 / 3.5a / 3.5b / 3.6 / 4.4c / 4.4d.

The purpose of this project is to better understand Soldiers’ experiences with sexual harassment (SH) and gender discrimination (GD) to provide the Army with a more nuanced picture of the environment in which SH and GD events occur than can be identified through survey data. The results will be used to update existing sexual harassment and gender discrimination prevention programming. As such, we will be systematically collecting qualitative data via focus group discussions from active-duty Soldiers (under the rank of flag officer) across six CONUS installations. These focus group discussions cover a number of topics including unit members’ understanding of what behaviors constitute SH and GD; factors that contribute to the climate for SH and GD (i.e., risk and protective factors); reporting and intervention by unit members and leaders; and recommendations for what information prevention training should include.

Below is the proposed sample size by pay grade and gender. We plan to oversample females in each pay grade category. As these are focus groups, response rates do not apply.

Pay Grade	Gender	Number
E1-E4	Male	85
E1-E4	Female	165
O1-O2	Male	85
O1-O2	Female	165
E5-E6	Male	85
E5-E6	Female	165
O3-O4	Male	85
O3-O4	Female	165
E7-E9	Male	50
E7-E9	Female	110
O5-O6	Male	50
O5-O6	Female	110
Installation Commander (O6)	Male	6
Total		1326

2.  Procedures for the Collection of Information

Focus groups will be conducted separately for men and women, officers and enlisted, and by rank (junior, mid-level, and senior). There are a total of six focus group protocols: junior enlisted/officer (E1-E4/O1-O2), mid-level enlisted/officer (E5-E6/O3-O4), and senior enlisted/officer (E7-E9/O5-O6). These focus groups are intended to be 60 minutes. A 30-minute discussion protocol tailored to installation commanders (generally at the O6 level) is also included. In total, we anticipate including:

• Jr. Enlisted (E1-E4)/Jr. Officers (O1-O2): up to 500 individuals

• Mid-Level Leaders (E5-E6/O3-O4): up to 500 individuals

• Senior Level Leaders (E7-E9; O5-O6): up to 320 individuals

• Installation Commanders: up to 6 individuals

These installations vary in their risk of experiencing sexual harassment (identified from a prior Army sponsored RAND study, Organizational Characteristics Associated with Risk of Sexual Assault and Sexual Harassment in the U.S. Army; research report available here: https://www.rand.org/pubs/research_reports/RRA1013-1.html): two installations are high-risk for both men and women, one is high-risk for women only, one is high-risk for men only, and two are non-high-risk for both men and women. Within installations, climate data from the Defense Organizational Climate Survey (DEOCS) (for additional information, see: https://www.opa.mil/research-analysis/opa-surveys/defense-organizational-climate-survey) will be used to identify units that are in the “top” and “bottom” quartiles in terms of a composite measure of culture and climate characteristics associated with sexual harassment. This composite measure was based on a nearing-completion Army-sponsored RAND study (Determining the Climate Predictive of Sexual Harassment and Sexual Assault in the U.S. Army: Identifying and Developing Measures of Unit-Level Climate; note this research report is undergoing final stages of peer review and has not been cleared for public release). These climate/culture characteristics include cohesion, fairness, inclusion, leadership support (immediate supervisor), transformational leadership (organizational leader/senior NCO), and workplace hostility.

At each installation, Soldiers will be randomly selected with the help of an installation POC based on their 1) gender, 2) officer/enlisted status, 3) pay grade, and 4) unit membership. The analysis strategy will rely on variation across these different groups, as well as across the different installations (selected based on risk) to better understand soldiers’ experiences and the environments in which SH and GD events do or do not occur.

Site visits will occur over two days. A total of 32 focus group discussions are possible; if there are no Soldiers who meet the eligibility criteria for a particular focus group discussion, then that discussion will be eliminated (e.g., if there are no female senior officers in the selected units). Each focus group will have 6-8 participants. One discussion will be held with the installation commander; they may choose to include members of their support staff if they wish. A mock schedule is included with this information collection request.

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

Qualitative data from the focus groups will be analyzed using a COTS software designed for qualitative data (e.g., NVivo). We will upload transcribed notes from the focus groups and interviews into the qualitative software program for coding. A codebook will be developed using an iterative process to further refine the codebook for coding of the interview data using a thematic analysis. A preliminary set of transcripts will be coded to identify any emerging themes and explore the need for creating new codes. Following refinement of the codebook, we will code the remainder of the transcripts, using a joint coding approach in which each coder independently codes each transcript once inter-rater reliability is established. We will use a thematic analysis to identify emerging themes and patterns across the data. We do not anticipate reporting numbers (e.g., percentages) that could be interpreted as prevalence rates, but we will use frequencies to identify the most-commonly mentioned themes to present in the final report.

4.  Tests of Procedures

Focus group protocols and study procedures were reviewed by roughly six Army and Air Force fellows at RAND. These fellows are active-duty service members who spend a year at RAND and are generally between the ranks of O3 and O6. In addition to written feedback, we also held a group discussion with the full research team and the fellows.

5.  Statistical Consultation and Information Analysis

No additional consultation apart from soliciting public comments through the Federal Register was conducted for this submission.

RAND Arroyo Center researchers will collect and analyze the information.