آزار و اذیت جنسی زنان پرسنل فعال وظیفه: اثرات بر روی رضایت شغلی و تمایل برای باقی ماندن در ارتش

Abstract This paper examines the relationship between sexual harassment and the job satisfaction and intended turnover of active-duty women in the US military. Using single-equation probit models, we find that experiencing a sexually harassing behavior is associated with reduced job satisfaction and heightened intentions to leave the military. However, bivariate probit results indicate that failing to control for individuals’ unobserved, time-invariant characteristics leads single-equation estimates to be overstated. Similarly, controlling for women's views about whether they have been sexually harassed reduces the single-equation estimates of the effect of the harassing behavior itself on job satisfaction and intentions to leave the military.

1. Introduction In 1995, approximately 195,000 women (13 percent of the total force) were on active-duty in the US military. This represents a six-fold increase since 1973 when the all-volunteer force was established (DoD, 1996). Intrinsic differences between military and civilian employment make sexual harassment a particularly salient issue for the US military. Military personnel often live on bases and are on duty 24 h/day. This high degree of proximity can blur professional and personal relationships and may increase both the incidence and subsequent costs of sexual harassment (Department of Defense, DoD).2 In particular, sexual harassment has been linked to a reduction in unit cohesion and combat readiness (Rosen and Martin, 1997), and some have predicted that in the future the military may find “the equal opportunity climate of its units is one of its primary criteria of mission effectiveness” (Knouse, 1991, p. 386). Our objective is to examine the relationship between sexual harassment, job satisfaction and intended turnover of active-duty women in the Armed Forces. We begin by incorporating measures of unwanted gender-related behaviors into single-equation models of job satisfaction and intentions towards future military employment. This allows us to compare our results directly to those in the literature. This strategy, however, implicitly assumes that reports of sexually harassing behaviors are exogenous, which is unlikely. In particular, unobserved heterogeneity may influence reports of sexually harassing behaviors on the one hand and reported satisfaction with and intentions to remain in military employment on the other. We therefore adopt two alternative strategies for accounting for the role of unobserved characteristics. First, we specify a bivariate probit model that accounts for any correlation between the error terms in sexual harassment and job satisfaction equations. Second, we explicitly control for women's views about whether they have in fact been sexually harassed. Overall, 70.9 percent of active-duty women reported experiencing some type of sexually harassing behavior in the previous 12 months. Single-equation estimates indicate that sexually harassing behavior is associated with reduced job satisfaction and heightened intentions to leave the military. However, failing to control for individuals’ unobserved, time-invariant characteristics (such as personality) causes single-equation estimates to be overstated. Bivariate probit results indicate that sexually harassing behavior does not significantly increase dissatisfaction with military employment once the correlation in unobserved factors associated with reporting a sexually harassing behavior and job satisfaction are taken into account. Similarly, directly controlling for women's views about whether they have been sexually harassed substantially reduces the estimated negative effect of the behavior itself on job satisfaction and indicates that there is, in general, no significant effect of sexually harassing behaviors on intentions to remain in military employment. Perceptions of harassment are driven by the unwanted, gender-related behaviors that women experience as well as by institutional arrangements (including the availability of training and formal complaint channels) in the workplace. Women who view their experiences as sexual harassment suffer additional consequences over and above those associated with the behavior itself. This is at odds with previous evidence that women exposed to sexually harassing behaviors report similar negative consequences whether or not they label their experiences as sexual harassment (Magley et al., 1999). In the next section, we summarize the previous literature on the job satisfaction, intentions to quit, and the role of sexual harassment. Section 3 provides the details of the data used in the analysis, while Section 4 examines the determinants of reported unwanted gender-related behaviors in military employment. Subsequently, the estimation results from the single-equation models are discussed. The potential endogeneity of reported sexual harassment is examined in Section 6, while our conclusions are presented in Section 7.

Conclusions This paper examines the relationship between sexual harassment, job satisfaction and intended job change using data from a 1995 US DoD survey of active-duty women serving in the military. Overall, 70.9 percent of women on active-duty in the US military report experiencing some type of sexually harassing behavior in the 12 months prior to the survey. Using single-equation probit models we find a strong positive relationship between experiencing a sexually harassing behavior and dissatisfaction with military employment and intention to leave the military. These results are consistent with previous results for women employed as lawyers in the US (Laband and Lentz). However, unobserved heterogeneity in individual characteristics causes single-equation estimates of the effect of sexually harassing behavior to be overstated. This heterogeneity may stem from aspects of an individual's personality (including perceptions of, tolerance for, or willingness to report unpleasant events in the workplace) that are likely to affect both reports of sexually harassing behaviors and women's satisfaction with military employment. Once the correlation in the unobserved factors associated with reporting sexual harassment and job satisfaction are taken into account, experiencing a sexually harassing behavior does not in and of itself significantly increase dissatisfaction with military employment. Furthermore, women's views about the unwanted gender-related behaviors they experience are closely related to subsequent outcomes. Women who view their experiences as sexual harassment have significantly higher levels of overall job dissatisfaction and heightened intentions to leave the military than women who experience unwanted, gender-related behavior, but who do not believe themselves to have been sexually harassed. The estimated negative effect of the sexually harassing behavior itself on overall job satisfaction is substantially reduced (and the effect on intentions to remain in military employment eliminated) once these views are taken into account.