Use of Force in Policing

One of the hallmarks of policing is the ability to use force (Fyfe, Reference Fyfe1988; Skolnick & Fyfe, Reference Skolnick and Fyfe1993). According to Payne (Reference Payne2012),

The use of force is a hallmark of policing but becomes problematic when excessive force is used (Fyfe, Reference Fyfe1988; Skolnick & Fyfe, Reference Skolnick and Fyfe1993). Excessive use of force is defined in fuzzy terms, usually as the use of more force than is necessary or reasonable (Skolnick & Fyfe, Reference Skolnick and Fyfe1993). This is why there are so many disagreements about whether particular police actions were appropriate as well as one of the reasons why there are many disappointments following investigations into police use of force. But it is also clear that some uses of force are excessive, such as in the cases that Ruggs et al. describe.

Throughout this commentary, we assume that an important criterion for police departments and the communities in which they serve is fewer fatal shootings of unarmed citizens. We do not suggest that this is the only important criterion in police work, nor do we suggest that police organizations should sacrifice quality on other performance criteria. However, as we will demonstrate next, women officers are less likely to use excessive force than are male officers, and thus police organizations should attempt to recruit and hire more women.

Sex Differences in Use of Force

Lonsway et al. (Reference Lonsway, Wood, Fickling, De Leon, Moore, Harrington and Spillar2002) summarized the literature on the use of excessive force and both civil liability cases (e.g., “police brutality” civil lawsuits) and citizen complaints about force in the United States. Their review clearly demonstrates that, compared with female officers, male officers (a) are significantly more likely to have the excessive force complaints against them sustained, (b) have significantly higher civil liability payouts, and (c) have a significantly higher rate of citizen complaints made against them. For example, at the Los Angeles Police Department in the 1990s, men made up approximately 80% of the patrol force but 96% of payouts in excessive force cases; conversely, women composed approximately 20% of the patrol force and only 4% of payouts. As misconduct worsened, the disparity between male and female officers widened, such that there was a 23:1 general ratio in payouts, which increased to 32:1 when only assaults and battery were considered, which furthered increased to 43:1 when only killings by officers were considered (Spillar et al., Reference Spillar, Harrington, Wood, Aguirre, Aguilar, Yick and Meza2000). These sex-based differences are seen across multiple indicators of excessive force for several large and midsized metropolitan police forces in the United States (Lonsway et al., Reference Lonsway, Wood, Fickling, De Leon, Moore, Harrington and Spillar2002; Rabe-Hemp, Reference Rabe-Hemp2008; see Porter & Prenzler, in press, for similar results in Australia).

In addition to evidence that male officers are more likely to use extreme force than are women officers, the use of routine force appears to exhibit similar although much weaker sex differences. Bolger (Reference Bolger2015) conducted a meta-analysis examining the situational, suspect, and officer characteristics that influence use-of-force decisions; Bolger found that there was a weak effect for male officers using force more than women officers did. This reflects the police use-of-force literature, which generally shows in individual studies that either (a) women officers are significantly less likely to use force than are male officers (Bazley, Lersch, & Mieczkowski, Reference Bazley, Lersch and Mieczkowski2007; Hoffman & Hickey, Reference Hoffman and Hickey2005; Schuck & Rabe-Hemp, Reference Schuck and Rabe-Hemp2007) or (b) female officers have lower rates of use of force than male officers, but the difference is not statistically significant (e.g., Paoline & Terrill, Reference Paoline and Terrill2005). Interestingly, there is evidence that situational and suspect characteristics associated with use of force differ across male and female officers (Paoline & Terrill, Reference Paoline and Terrill2005).

The effects of sex on use of force are enhanced when considering officer pairs (Schuck & Rabe-Hemp, Reference Schuck and Rabe-Hemp2007). Schuck and Rabe-Hemp found that officer pairs consisting of two women were the least likely to use force and the most likely to use less extreme force, compared with other officer pairings. Further, within mixed-sex pairings, pairs in which the female officer had the bigger role in the encounter (e.g., was the arresting officer) used less force than did mixed-sex pairs in which the male officer had the larger role. Male–male pairs were the most likely to use force and the most likely to use more force, followed by male-led mixed-sex pairs.

In sum, these studies demonstrate that women officers are (a) less likely than are male officers to be implicated in cases of extreme force, (b) somewhat less likely to use routine force, and (c) less likely to use force when paired together or as leaders of a mixed-sex pair, compared with other pairings. Together, these studies suggest that more women officers would reduce the number of events that are at the extreme end of the use-of-force continuum.

Police Demographics and the Subsequent Practical Effect on Use-of-Force Interactions

Even small differences in use of force by male and female officers are amplified and take on practical importance when considered in light of the sex disparity within police forces in the United States and beyond (see Prenzler & Sinclair, Reference Prenzler and Sinclair2013, for a review of sex demographics in English-speaking police forces around the world). In 2015, of the approximately 688,000 police and sheriff's patrol officersFootnote ² who were employed in the United States, 13.6% were womenFootnote ³ (Bureau of Labor Statistics, 2016). This ratio of male to female officers (6.35:1) has a staggering practical effect on policing behaviors within the communities they serve. Bolger (Reference Bolger2015) found that male officers were 1.11 times more likely to use force than were female officers, indicating that for every 100 times female officers used force, male officers used force 111 times. But practically, given the ratio of male to female officers, this means that for every 100 times that one female officer used force, just over six male officers used force 111 times each or about 705 times in total. This means that the somewhat small elevation in use of force by male officers compared with female officers results in an extra 105 instances of force than would happen if male officers used force at the same rate that female officers do.

We can also estimate the reduction in the number of police contacts that use force if women accounted for 50% of patrol forces. Paoline and Terrill (Reference Paoline and Terrill2005) showed that women used force in 2% fewer of their contacts than did male officers (58% of contacts by male officers used coercive authority vs. 56% by female officers, nonsignificant). According to the U.S. Department of Justice, in 2008 there were nearly 67 million contacts between police and community members. Assuming that male and female officersFootnote ⁴ engage with the public at a rate approximately equal to their proportion in the police force, approximately 9.1 million of these encounters were with female officers and 57.9 million were with male officers. Using Paoline and Terrill's (Reference Paoline and Terrill2005) descriptions of the rate of use of force by male and female officers, this means that of the 67 million contacts made by police in 2008, approximately 33.6 million were conducted by male officers and used force while 5.1 million were conducted by female officers and used force. However, if male and female officers each composed half of the police population at all ranks, male officers would use force approximately 19.4 million times per year, whereas female officers would use force approximately 18.8 million times per year—an overall reduction of 500,000 instances of use of force in a year's time from the current number of instances of force.

The above analyses account for use of force whether routine or excessive. However, male officers account for a disproportionately large share of excessive force complaints (Lonsway et al., Reference Lonsway, Wood, Fickling, De Leon, Moore, Harrington and Spillar2002; Spillar et al., Reference Spillar, Harrington, Wood, Aguirre, Aguilar, Yick and Meza2000). Thus, the effects of the above analysis regarding police demography would be even further amplified if only excessive force were considered.

Job Analysis and Competency Modeling Focusing on Sex-Based Differences

We have shown that women and men behave differently as police, at least in terms of use of force. Further, the causes of use of force appear to differ across male and female officers (Paoline & Terrill, Reference Paoline and Terrill2005). Beyond the policing literature, there is evidence that men are moderately more prone to both verbal and physical aggression than are women (Archer, Reference Archer2004). The aggression literature shows that the gap between men's and women's use of aggression is reduced considerably when provocation is taken into account (Bettencourt & Miller, Reference Bettencourt and Miller1996). Additionally, the aggression literature demonstrates that men and women differ in their perceptions of the cost of aggression to themselves and the target (Eagly & Steffen, Reference Eagly and Steffen1986).

Although these factors have been documented as sex differences, we do not believe that these are essentialist differences between men and women. Certainly, there are both biological and socialization factors that influence aggressive tendencies (see Hyde, Reference Hyde2014, for a review of theories of sex similarities and differences), both of which would be influential long before people enter the workforce. Of course, both women and men make good patrol officers who use force appropriately. But the sex differences that have been documented in police use of force and in aggression more broadly suggest that attending to how women police could provide insights into better models for selection and training that ultimately reduce shootings of unarmed citizens.

It is important to note that although women have assuredly been included in job analysis and competency modeling in police organizations to date, the demographics of police organizations suggest that job analyses and competency models for police officers are highly influenced by the perspectives and experiences of male officers. Even with careful attention to demographic representation in job analyses (and even oversampling of underrepresented groups), men's perspectives will overshadow women's. Not only are male officers more common in the patrol ranks than are women, but they are also more common among the leadership ranks; these leaders—as subject matter experts—are likely to offer a gendered interpretation of effective policing, whether due to availability heuristics, confirmation bias, stereotypes, implicit biases, or just personal experience. This will cause job analysts to determine the frequency of and competencies for this (male) interpretation of effective policing. Further, there is ample evidence that women's comments are overlooked, suppressed, and dismissed in mixed-group discussions when women are not a majority (or more) of the group membership (cf. Carli, Reference Carli, Ryan and Branscombe2013), which is problematic for the inclusion of women's perspectives if group discussions are used anywhere in the job analysis and competency modeling process. (These problems are further amplified when women who are also underrepresented minorities are considered.) Finally, police organizations on a whole are gendered as masculine, so that even notions of teamwork and communication outside of the contact with suspects are masculine, creating a culture of masculine expectations and a masculine organization beyond each individual's perspective (Acker, Reference Acker1992).

Thus, we propose that it is important to examine how men and women police and where differences in policing behavior, perceptions, and attitudes lie. It is also important to create competency models of police work that are explicitly female and feminized (i.e., with women as the observed members, with women as the subject matter experts). Then applying these models to the selection of all recruits, regardless of sex, could result in lower use of force across all members of police organizations, which ultimately should result in fewer shootings of unarmed citizens and other forms of police brutality.