Predicting aggression in police officers using the MMPI-2

This study attempted to determine if any of four MMPI-2 scales formulated to measure various forms of hostility or aggression is useful in predicting aggressive behavior among police officers. In our analysis of 80 officers from two police departments, we found that some scales predicted supervisor ratings of officer performance in one of the departments but not in the second department. In one department, officers with higher ANG scores or HOS scores, or lower O-H scores, were judged by supervisors to be more assertive. Officers from the same department with higher HO scores or lower O-H scores received moire justified citizen complaints. We discuss the importance of predicting aggression in police officers, the difficulty of making this prediction using MMPI-2 scores, and differences between the two departments that might have impacted the outcome of this research. Authors' Note: Communications should be addressed to the authors at James Madison University, School of Psychology, MSC 7401, Harrisonburg VA 22807.     

Predicting Police Aggression: Comparing Traditional and Non-Traditional Prediction Models

Psychologists can help to reduce inappropriate aggressive behavior through careful screening of police officer candidates. The purpose of this study was to identify whether the IAT Reasoning Test (IAT), a measure of trait aggression, and a Monetary Delay Discounting Task (MDDT), a measure of behavioral control, could predict on-the-job police aggression better than the MMPI-2. We administered the MMPI-2, the IAT, and the MDDT to 85 police officers. Three prediction models were created using scores on the IAT and the MDDT, and scales from the MMPI-2. Model 1 included the IAT and the MDDT. Model 2 included MMPI-2 scales Hostility (HO), Overcontrolled Hostility (OH), and Anger (ANG). Model 3 included MMPI-2 scales Frequency (F), Hysteria (HY), and Psychopathic Deviate (PD). We found that Model 1 was the only predictor of supervisors?? ratings of on-the-job aggression, however, the predictive utility of this model was attributable primarily to the MDDT predictor. Based on these findings, we suggest that using measures of behavioral control during preemployment screening may help to identify potentially aggressive police officer candidates.     

Does the stereotypical personality reported for the male police officer fit that of the female police officer

This research, using the content, validity and clinical scales found in the MMPI-2, investigates the differences in the psychological makeup of male versus female police recruits. The research design involves testing of police recruits on the first day of their academy experience. The literature on the police personality suggest that there exists a stereotype for male officers, a finding replicated in this study. The findings of this research also suggest that female officers do not generally fit the stereotype found in the study of male officers. An earlier version of this paper was present at the 1996 Annual Meeting of the Academy of Criminal Justice Sciences, March 12–16, 1996, Las Vegas, Nevada.     

The effects of tenure on police officer personality functioning

The study compared pre-employment and current MMPI/MMPI-2 Basic Scale scores of small- and mid-size city Texas municipal police officers to assess longitudinal personality change. Participants had at least five years continuous, “street level” police experience. Twenty-three veteran police officers from a mid-size city and 19 small-city police officers volunteered to participate. Pre-employment MMPI/MMPI-2 information was obtained from police department records. Current MMPI-2 data was collected in small group or individual administrations. Significant increases in K-corrected T-Scores occurred on Scale F and eight of the Basic Scales; L and K decreased significantly. Mean scores remained within the normal range. Mid-size city officers produced higher K and 5 scores than small-city officers. Authors' Note: We gratefully acknowledge the assistance of Rick Bradstreet, Ph.D., and E. S. Collins, Chief of Police, in the data collection process. The results of this study were presented at the Annual Convention of the American Psychological Society in Denver, Colorado, in June 1999.     

An investigation into the usefulness of the MMPI and MMPI-2 in municipal and state police candidate selection

The Minnesota Multiphasic Personality Inventory (MMPI) and the later revision (MMPI-2) have been two of the most frequently employed instruments in the selection of law enforcement officers (Bartol, 1996). In this study, the following three sets of data were collected and analyzed: municipal law enforcement officer candidates tested with the MMPI, municipal law enforcement officer candidates tested with the MMPI-2, and state police trooper candidates tested with the MMPI-2. These three data sets were evaluated for their effectiveness at predicting “pass/fail” status on police candidate interview using multiple linear regression. A combination of subscales were found to be associated with classification as either pass or fail on interview.     

Using the good cop/bad cop profile with the MMPI-2

The Minnesota Multiphasic Personality Inventory (MMPI) has been widely used in a variety of ways to screen candidates for law enforcement positions. This study extends the use of the MMPI Good Cop/Bad Cop (GC/BC) profile (Blau, Super, & Brady, 1993) to the MMPI-2. The MMPI-2 profiles of 39 veteran police officers were used to predict their performance (No Apparent Problems, Borderline, or Serious Problems Possible), and these predictions were compared with supervisors’ ratings of the officers’ actual performance. The MMPI-2 predictions were accurate for 46% of the officers, a rate that was significantly better than chance (p=.024). Based on the current data, the best selection outcome would be obtained by accepting officers whose MMPI-2 profiles place them in the No Apparent Problems or Borderline groups, and rejecting officers whose profiles suggest Serious Problems Possible. This could be accomplished simply by rejecting any officer who obtained a score above 65T on any of the clinical scales. This selection strategy would have resulted in the acceptance of 22 officers, 19 of whom were highly rated by their supervisors, and the rejection of 17 officers, 11 of whom were rated as borderline or poorer by their supervisors. It would also result in the erroneous rejection of 6 officers who were highly rated by supervisors. AUTHOR NOTE: An earlier version of this paper was presented at the 1994 annual meeting of the Society for Police and Criminal Psychology, in Roswell, New Mexico. The authors wish to thank Jim Herndon, Ph.D., for this comments on this study.     

The MMPI-2 and statisfactory police academy performance: Differences and correlations

The Minnesota Multiphasic Personality Inventory (MMPI) and its later revision (MMPI-2) have been the most frequently employed psychological instrument in the selection of law enforcement officers (Bartol, 1996). In this study, state police trooper cadets were given a broad number of measures including the MMPI-2 prior to academy admission. Their performance at the police academy was then analyzed and related to their performance on the MMPI-2. Using step-wise multiple linear regression. Hy3 and Sc4 scores produced significant negative correlation’s with academy final grade point average. Student’s unpaired 1-tests between successful and unsuccessful cadets revealed differences between the groups for several MMPI-2 subscales including, but not limited to, K, MA, and SI, Lastly, logistic regression revealed no single subscale, or combination of subscales, which significantly predicted classification of the cadets as either successful or unsuccessful based on their MMPI-2 scores alone. Author Note: Elements of this article have been presented at the 27^th Annual Conference of the Society of Police and Criminal Psychology in Portland, OR, October 1998.     

Associations between MMPI-2-RF (Restructured Form) and Inwald Personality Inventory (IPI) Scale Scores in a Law Enforcement Preemployment Screening Sample

The MMPI-2 is widely used in assessments of police officer candidates. The MMPI-2-RF, composed of 338 of the 567 MMPI-2 items, is designed to measure the core clinical constructs assessed by the MMPI-2, but in a more psychometrically sound and efficient manner (Ben-Porath and Tellegen 2008/2011). Law enforcement comparison group data are embedded in the scoring software for the inventory, and reported in the User’s Guide for the MMPI-2-RF Police Candidate Interpretive Report (PCIR; Corey and Ben-Porath 2014). In the present concurrent validity study, conceptually based hypotheses regarding associations between MMPI-2-RF scales and scores on the Inwald Personality Inventory (IPI, Inwald et al. 1982; Hilson Research 2006), a test specifically designed to screen law enforcement officers, were tested with a sample of 277 male, full-time police officer candidates from four separate police departments of varying sizes examined after a conditional offer of employment was tendered. The majority of participants were Caucasian (69%), followed by African-American (27%), or Other/Mixed ethnicity (4%). Our hypotheses were largely borne out, lending support to the construct validity of the MMPI-2-RF and the ability of scores on the inventory to assess variables relevant to police candidate evaluations.     

Use of the MMPI-2 to predict the employment continuation and performance ratings of recently hired police officers

The MMPI-2 is one of the most frequently employed instruments for the selection of police officers. Serafino and Serafino (1997) collected data which involved information about employment continuation and ratings by supervisors of 32 police officers who had recently been hired and who had been given the MMPI-2 during the hiring process. In this study, the Paranoia Obvious (Pa) and Paranoia Subtle (Ps) scales proved to be the significant. Pa Subtle correlated with removal whereas Pa Obvious correlated with rating. Higher scores on Pa Subtle correlated significantly with being removed from the job, whereas low scores on the Pa Obvious correlated with higher ratings of performance by supervisors. Discussion of the results involved the fact that Subtle Pa scores would suggest paranoid tendencies not easily detected during the interview. Since most high Pa Obvious individuals would have been eliminated in the hiring process, expression of this tendency was at a low level after being hired but if present resulted in low ratings. Significant predictors were noted to be very much a function of the type of criterion variable employed in the study.     

Personality differences between women police recruits, their male counterparts, and the general female population

Presently police manager and their recruiting staff are using an increasing number of screening tools including psychological testing during the hiring processes. Prior research suggests that gender is often ignored in the review of testing results. This could lead to problems in the interpretation of results. This research investigates the existence, if any, in the pre-socialization personality differences between male and female police recruits as well as between female police recruits and the general population of females. Subjects (N=104) were from a multi-agency training facility in a southern state. Subjects completed the MMPI-2 at the beginning of their police academy training to control for any socialization effects the training may have. The diagnostic clinical scales, the content scales, and the gender scales of the MMPI-2 were subjected to independent discriminant function analyses to determine if differences between male and female recruits exist. Scale 5 (Mf), GM, and GF were analyzed via t-tests to determine if differences between the female recruits and women in the normative sample exist. All discriminant functions were significant at the .05 level. Scales that differentiated male and female recruits were: Ma (Scale 9), ASP, FRS, BIZ, Mf (Scale, 5), and GF. Female recruits differed from the female normative sample on only GF (recruits were lower). It is suggested that future research focus on the predictive power of the content scales for continuation in the police field. It is also suggested that future research investigate the flexibility of the gender constructs throughout the policing career. Author Notes: Send all enquires to Larry A. Gould, Ph.D., Dept. of Criminal Justice, Northern Arizona Univerity, Box 15005, Flagstaff, AZ 86011-5005. 1. A special thank you is given to Northern Arizona University for providing the funding of the parent project from which this research was accomplished. I also wish to extend my gratitude to those whose time and effort went into the collection and computer entry of the data—Dr. Larry Gould and Rachel Rowland. In addition, I would like to thank Dr. Clay Moore for the statistical “chats”.     

An Examination of Authoritarian Personality Traits among Police Officers: The Role of Experience

This study compared authoritarian traits as determined by MMPI-2 scores between inexperienced versus experienced police officers. The purpose of this comparison was to investigate whether experienced police officers possessed higher levels of authoritarian traits which may be related to years on the job. Results found that inexperienced and experienced police officers tend to be psychologically healthy and do not possess high levels of authoritarian traits. Some specific differences emerged with inexperienced police officers demonstrating higher scores on ASP (antisocial practices) but lower scores on HY (emotional reactions to stress). Possible reasons for these results are discussed.     

Impression Management in Police Officer Candidacy on the MMPI-2

It is well documented that candidates in employment situations attempt to present overly positive pictures of themselves that may be inaccurate. Police officer candidacy situations are not an exception. They are more critical to the public welfare, however, because of the sensitive nature of police officers as an interface between government and the community. For this reason, the study of impression management using devices that are employed in police selection is of great importance. This study used 36 college students to whom the MMPI-2 was administered, first under standard conditions (control condition) in which the students responded as they ordinarily would and second, under a set of special instructions (experimental condition) which instructed them to respond as if they were police officer candidates. The two profiles were compared. As hypothesized, the Lie (L) and Correction (K) scales were elevated in the experimental condition. An inspection of the clinical scales revealed that all tended to be lower in the experimental condition than in the control condition, with the exception of Masculinity-Femininity (Mf), Psychopathic Deviate (Pd), Paranoia (Pa), and Hypomania (Ma). Implications of these findings for officer candidate selection are discussed.     

A study of the interrelationships of several validity scales used in police selection

Validity scale data have received attention recently as providing valuable information about potential problematic police officer candidates. In this study, validity scale data from a number of selection instruments were obtained using 42 state police officer candidates. The scale used were the MMPI-2 L scale, the MMPI-2 K scale, the PAI Positive Impression Scale (PIM), the PAI Defensiveness Index (DI), the IPI Guardedness scale, the Hilson Life Adjustment Profile Lack of Candor scale, and the Inwald Survey 2 Denial of Shortcomings scale. Intercorrelations among the scales were developed and a factor analysis was performed. Factor analysis revealed two factors to be present. One is associated with the Hilson scales and is appropriately named guardedness or defensiveness. The other is associated with the MMPI-2 K scale and the PAI scales and is appropriately named social desirability. The MMPI-2 L scale loaded significantly on both factors and seemed to be the most general of the validity scales in terms of its characteristics. Implications of these analyses for police selection are discussed.     

Authoritariansim as indicated by measures of acceptance of authority and directiveness among police officers,students and townspeople

Conclusion Recently theory and scales measuring authoritarianism and cynicism of police have come into close scrutiny. In particular, Langworthy (1987:28,33) has concluded from his review of 21 empirical studies that police cynicism appears to have several different dimensions and that the “failure to detect cynicism or a factor closely related to it is a product of reliance on a very flawed instrument the Niederhoffer index.” To address that criticism of a faulty instrument, this research has utilized scales measuring directiveness and proauthority. For reasons of alpha coefficients, only the scores on the pro-authority scale were analyzed. Mean scores on the pro-authority scale between law officer, students, townspeople and police officers were not significantly different. When investigating variables relating to the high scores of proauthority of police officers, rank was statistically significant while age was not. For future research, separate population should be continued as well as random sampling. In addition, variables of age, sex, ethnic, rank and departmental/organizational dimensions should be explored.     

Comparison of the MMPI-2 and Niederhoffer Cynicism Scales for a Sample of police academy trainees

This study compared the Niederhoffer Cynicism Scale, as a less costly alternative, to the MMPI-2 measure of cynicism. Data were collected on both scales from a sample of male and female police academy trainees to compare reliability and construct validity. Subgroup analyses by race and sex yielded comparable Cronbach alpha estimates for minority (.71) and majority (.83) males to MMPI-2 norms. The Niederhoffer reliability values were marginally smaller (.73 to .77) but within the acceptable range. Construct validity correlations between both scales were significant for minority and majority males. However, sample size was too small for both minority and majority females groups to use appropriate statistical techniques to determine reliability or validity. Continued use of the Niederhoffer is recommended for minority and majority males. Author's Note: This paper was presented at the 44^th Annual Meeting of the Western Social Science Association in Albuquerque, NM, in April of 2002.     

An Examination of Officer Stress: Should Police Departments Implement Mandatory Counseling?

This study suggests that police departments who promote counseling benefit from officer stress reduction. Officers from sixteen municipal police departments (n = 1,114) across the state of Alabama possessed moderate stress, but were influenced significantly by organizational demographics (including counseling opportunities). A majority of officers believed that stress signs were not predictive of police suitability but remained reluctant to share fears and anxieties with fellow colleagues, suggesting that officers feared the stigma associated with the need for stress counseling. Officers working in supportive counseling climates had significantly less stress, a reduced need for counseling, and a greater willingness to use counseling. Officers who engaged in counseling (at least occasionally) also reported more stress, indicating an awareness of their need for counseling. The authors concluded that police departments should consider requiring mandatory and periodic counseling for all officers, a procedural tactic that camouflages counseling need while concurrently treating the source of officer stress.     

Women in Law Enforcement

ABSTRACT

Despite equal employment opportunity laws, affirmative action policies, and studies which demonstrate the capability of females, women account for only 8% of all municipal police officers in this country. The underrepresentation of social groups in policing is not a new phenomenon and can be explained through the “denial hypothesis.” The present study uses tobit regression techniques to examine the variation in female employment as law enforcement officers in municipal police departments within the State of Florida. The findings indicate that the distribution of sworn female officers does not resemble local labor market conditions. Neither the degree of parity between men and women in local economic conditions nor the availability of a qualified female applicant pool affect agency gender composition.     

Use of the MMPI-2 and the Inwald personality inventory to identify the personality characteristics of dropouts from a state police academy

The MMPI-2 and the Inwald Personality Inventory were employed to investigate the personality characteristics of dropouts from a state police academy. A traditional model of training borrowed from military models was used at the academy rather than a police generated model. Sensitive and independent individuals, more compatible with modern community policing methods may have rejected police work as a result of the experience. 15 academy completers and 9 dropouts were used in the sample. Analyses of the scales of the MMPI-2 and the Inwald Personality Inventory identified variables upon which the two groups differed. The hypothesis that more sensitive, empathic and independent individuals were leaving the academy appeared to be supported.     

Demographic and workplace characteristics which add to the prediction of stress and job satisfaction within the police workplace

Much research has been conducted into aspects of the police workplace that contribute to stress for individual officers. The current paper examines the influence of worksplace and job characteristics on both officers' stress and their job satisfaction. Police officers recruited from two divisions of an Australian state police service (n=749) were surveyed. Results show that there was a positive relationship between organizational support and job satisfaction and a negative relationship between organizational support and job stress. Difficulty in dealing with organizational change led to lower job satisfaction and higher levels of job stress. Working long hours led to increased job stress but it did not lead to lower job satisfaction. In contrast, shiftwork led to lowered job satisfaction but it did not lead to job stress. Of particular interest in this study was the finding that the job content factor of dealing with dangerous and unpredictable duties was not predictive of job stress but in fact led to higher job satisfaction. Authors' Note: Jeremy Davey, DipT, Bed, Med, is Deputy Director, Centre for Accident Research and Road Safety, School of Psychology and Counselling, Queensland University of Technology, Beams road, Carseldine, Qld 4034, Australia.