Measuring Sexual Harassment in the Military: The Sexual Experiences Questionnaire (SEQ—DoD)

Abstract

The harassment of women in the military is receiving increased attention from both policymakers and the popular media. Stimulated originally by the Navy's Tailhook episode in 1991, and more recently by revelations of widespread sexual harassment of female Army recruits, there appears to be a growing consensus that the harassment of female military personnel is a problem with profound consequences for both individuals and the armed services more generally. At the same time, there are few reliable estimates of the actual nature, prevalence, and severity of this problem; those that do exist are marked by a variety of shortcomings.

This article attempts to address the issue. We begin by describing the development and measurement characteristics of a military version of the Sexual Experiences Questionnaire (SEQ—DoD; Fitzgerald, Gelfand, & Drasgow, 1995; Fitzgerald et al., 1988), which was administered to more than 28,000 military personnel as part of the Department of Defense (DoD) 1995 study of gender issues in the services. Following analysis of the structure of the instrument, we examine incidence rates for the effects of gender, race/ethnicity, armed service, and rank. We then discuss options for scoring the SEQ—DoD and conclude with a discussion of the difficult question, “Who should be counted as sexually harassed?”

Notes

1 It is critical to differentiate the behavioral and organizational phenomenon of sexual harassment from the legal concept of the same name; we use the term here in its former sense, emphasizing that legal decisions rest on a variety of factors that surveys can inform but not assess.

2 See Berdahl, Magley, and Waldo (1996) and Waldo, Berdahl, and Fitzgerald (1998) for a discussion of the concept of gender harassment as applied to men.

3 The great majority of this research has examined the most common situation, that is, male-to-female harassment; recently, Berdahl, Magley, and Waldo (1996) reported the same general structure for the sexual harassment of men, albeit with some gender-specific variations.

4 Given the extremely widespread nature of low-level garden variety, offensive behavior, the question might be more appropriately framed as “Who doesn’t count?” One possible approach to this problem might then be to examine groups of individuals who have experienced one, two, or more instances of a single behavior (e.g., sexual hostility) and determine when the consequences begin to “kick in.” This does not, of course, resolve the policy question of whether such behavior should be tolerated unless and until it causes measurable damage.