Since the 70’s arguments for and against the existence of a personality profile distinctive for police applicants have appeared in the scientific literature. Those who argue for the existence of such a profile favour the dispositional model that states that police candidates show some determined personality traits that distinguish them from the general population. Opponents to such a pre-existing profile for policemen propose the socialization model that states that the profile of a police officer is formed during the training at the academy. Both points of view have special importance for the selection of police candidates. If a personality profile distinctive for policemen exists it is necessary to know its components and use them as criteria to select future applicants. In this way, voluntary and compulsory dropouts can be diminished, whether during the formation or later professional career. If however such a profile does not exist and solely develops as the result of the socialization process during the formation at the academy, the component of the selection process addressing to that profile will no longer prove useful and should be discarded.
Several studies (Burbeck & Furnham, 1985; Johnson, 1990; Wright, Doerner & Speir, 1990) have analysed which are the most widely used psychological tests in the selection of policemen, as well as the reliability and validity of the selection process itself. Concerning the tests, the classical techniques of personality (CPI, 16PF, EPQ, MMPI) and intelligence assessment are used, as well as other questionnaires that assess interests and attitudes. The authors stressed the use of situational tests, but this procedure is expensive and delays the selection process. They also stressed the unhampered decrease in the use of projective techniques because of their lack of reliability and validity. Concerning the review of the selection process itself, the authors noted that the large diversity of applied methodologies makes it difficult to draw any conclusion on which model to prefer above another. In general, several methodological strategies are used to assess the possible existence of the aforementioned personality profile. The most commonly used methodologies are the comparison of the candidate’s profile with that of the general population and the comparison with other groups that share distinctive characteristics, as e.g. with policemen that have continued their career and with policemen that have left the police corps on a voluntarily or non-voluntarily basis.
Comparison with the general population
Several investigations have been performed using the Rokeach Value Survey (e.g. Rokeach, Miller & Snyder, 1971; Griffeth & Cafferty, 1977; Cochrane & Butler, 1980). This questionnaire contains two lists of 18 values, presented in alphabetical order, that the subject has to order as a function of the value it assigns to it. The results with this instrument have not been conclusive at all concerning the appropriateness of the dispositional or the socialization model in the selection of police applicants.
Other authors have argued that the police candidates are more conservative and authoritarian than the general population and therefore aspire to this occupation. Some of these studies (Genz & Lester, 1976) however have not really helped much to clarify this issue as they have used a large set of diverse instruments to assess authoritarianism (Eysenck Social and Political Attitude Inventory, Rokeach Dogmatism Scale, Wilson-Patterson Attitude Inventory, etc.). However, more recent studies (Colman & Gorman, 1982; Austin, Hale & Ramsey, 1987) using the Rokeach Dogmatism Scale reached similar conclusions. That is, the police corps attracts those that are more conservative and authoritarian thus favouring the hypothesis of the dispositional model.
Numerous other studies have used the aforementioned personality assessment instruments (EPQ, 16PF, MMPI) to compare the scores of those that opt for a place in the police corps with the general population. In this way, Gudjonsson & Adlam (1983) using the EPQ and the IVE, found that the police applicants scored higher on Impulsivity, Venturesomeness and Extraversion and lower on Psychoticism (i.e. being more conformist) than the general population. Cooper, Robertson & Sharman (1986) used the 16PF and found that the profile of the British policeman, authorized to carry fire-arms, differs from the general population norms on the factors A (reserved vs. warm), B (concrete vs. abstract reasoning), G (expedient vs. rule-conscientious) and I (utilitarian vs. sensitive). Carpenter & Raza (1987) using the MMPI, collected data from 1981 to 1985. The comparison of their sample with the population norms showed that the police applicants were less depressed and anxious, more assertive and were more interested in developing and maintaining social relationships. Their scores were also more homogenous than those of the general population.
Thus, the results of comparing samples of police candidates with the general population seem to bring evidence in support of the dispositional model. This trend was clearly shown when the instruments used measure relatively stable personality variables, but it could not be found if the variables assessed change as a consequence of experience and daily police practice.
Comparison with other groups
Several American police corps require candidates in the selection process to be assessed by means of intelligence tests. In his book, Spielberger (1979) concluded that intelligence and ability tests are useful predictors for the output at the academy, but not for the occupation as a police officer. Some years later, Aylward (1985) found that I.Q. contributed only 7% to the selection between candidates. Therefore, other factors than the cognitive capacities contribute more to the final decision of who to accept and who not, as e.g. personality variables and attitudes.
Burbeck and Furnham (1984) administered the EPQ to police candidates that were to be selected by means of an interview developed by a group of professional policemen. Once the selection process had finished, they analysed the differences between accepted and refused subjects. They found that the candidates that were accepted were more extraverted and less neurotic than those that were rejected. This strongly indicates that the extraverted as well as the emotionally most stable persons are more valued for the occupation as a police officer. The former as they will spend a substantial part of their time working with people, the latter as they are probably more capable of making decisions in which they do not get emotionally involved. However, the subjects that were turned down scored higher on Extraversion and lower on Neuroticism than the general population as well. Therefore, the data seem to support the dispositional perspective in the sense that the police candidates constitute a self-selected sample that is unusually stable and extraverted, and that from this sample only the most stable and extraverted pass the selection process.
Eber (1991) revised several large-scale studies, collecting data of more than 15,000 police candidates to whom the Clinical Analysis Questionnaire (CAQ; Krug, Cattell & IPAT, 1980) was administered. The CAQ consists of two parts, the first part being the 16PF and the second 12 psychopathological scales. The results indicate a personality profile of the law enforcement candidate in the United States, as evaluated by the 16PF, that is highly controlled, low on anxiety, with a strong character and somewhat independent. Eber (1991) thus established a differential personality profile supporting the dispositional model.
Some years later, Lorr & Strack (1994), administering the same instrument, used cluster analysis to prove the existence of a personality profile among police applicants. Cluster analysis consists of grouping objects based on their similarity on certain characteristics, creating certain subgroups or clusters. The aforementioned study revealed the existence of two personality profiles. The majority group, much like what Eber previously had described as «good policemen» (Eber, 1991), was characterised by a high level of self-control, independence, extraversion and emotional stability. However, one out of every four candidates was allocated to the category denominated «bad policemen». These subjects did not differ from their counterparts regarding independence, but they did manifest a low level of self-control, introversion and a much higher level of anxiety. Thus, even though different methodologies and samples were used in both studies, they both came to the same conclusions. This seems to indicate the existence of a determined personality profile in persons that opt for a place in the police corps.
Hargrave, Hiatt & Gaffney (1986) evaluated, using the MMPI and the CPI, the differential profile of traffic policemen and deputy sheriffs working in court systems. The assumption was that both groups represented two distinct manifestations as the former works independently, patrolling freeways while the latter works more interdependently in court systems, which would be reflected in two aberrant profiles. The results show that both groups do not differ from each other: «traffic officers and deputies are quite similar to each other, and their characteristics are consistent with those reported in other articles which describe a "police personality"» (p. 254). That is, both groups are dominant, independent, flexible, competitive, energetic and are socially ascendent. However, they did differ significantly from the general population.
The presented review of the literature on the suitability of the dispositional model or the socialization model for the occupation of policeman seems to present data that favour the dispositional perspective, i.e., in a group of certain individuals exist several concrete personality characteristics that make the job of policeman especially attractive to them and therefore they present themselves as candidates.
From another point of view, Gomà-i-Freixanet and colleagues (Gomà-i-Freixanet, Pérez & Torrubia, 1988; Gomà-i-Freixanet, 1991; Gomà-i-Freixanet & Puyané, 1991; Gomà-i-Freixanet, 1995; Gomà-i-Freixanet, 1998) have proposed a classification model in which the physical risk associated with certain behaviours can be categorised as a function of the consequence of the conducted risky behaviour. In this model, the physical risk is defined as the possibility of dying while performing the activity. With this definition in mind we defined a continuum of physically risky activities ranging from antisociality to prosociality, with the risky sports group intermediate on that hypothetical dimension. Thus, the antisocial physically risky activity is one in which the activity can result in harm to others, the prosocial physically risky activity is one in which the behaviour can result in benefit to others, and in the physically risky sports group, the behaviour may only harm oneself. Results obtained with 570 male subjects (Gomà-i-Freixanet, 1995) divided into the three aforementioned risk-taking groups, and a control group for comparative purposes, seem to indicate that a common personality profile for these three types of risky behaviours exist, as well as a differential profile for each of them. These results have received strong support from a replication study conducted with 227 female subjects (Gomà-i-Freixanet, 2001). Following the aforementioned classification, the occupation of police bodyguard is a job that implies a high physical risk but at the same time it can be catalogued as prosocial as another person receives the benefits of the risk taken. Within the police corps, the group of bodyguards is one of the collectives most exposed to physical risk as they are responsible for the security of the highest authorities (VIPs). In this study, we present data of a group of police bodyguards as a prototype of policemen that perform activities that imply a high physical risk from which another person benefits.
The aim of this study is twofold. On the one hand, to provide data on this group and compare it with the general population to determine whether the dispositional model described earlier can be applied to this population. On the other hand, to assess whether our classification model is appropriate to that sample. If so then the bodyguards should share the personality characteristics of the high physical risk groups in general and of the prosocial physical risk groups in particular.
The sample consisted of 20 subjects (M= 30.35 years, SD= 2.91), all male and members of the Mossos d´Esquadra de Catalunya, the Catalan police. At the moment of assessment all subjects were approximately 8 years in the unit and they constituted the complete Bodyguard Unit. To become member of this unit the policemen were selected on several criteria as their length, physical condition, personal characteristics and their possibility of working at irregular hours. Directly after being selected they were trained in different techniques such as personal defence, security-driving, shooting practices and others.
Material and procedure
Personality was assessed by the Eysenck Personality Questionnaire (EPQ; Eysenck & Eysenck, 1975) and the Sensation Seeking Scale Form V (SSS-V; Zuckerman, Eysenck & Eysenck, 1978). The EPQ contains four scales measuring Extraversion (E), Neuroticism (N), Psychoticism (P) and Lie (L). The SSS contains four scales as well, namely Thrill and Adventure Seeking (TAS), Experience Seeking (ES), Disinhibition (Dis) and Boredom Susceptibility (BS), as well as a Total scale (Total) which is the sum of the previous four. Catalan adaptions of these two questionnaires that have shown adequate psychometric properties (e.g. Gomà-i-Freixanet, 1997) were administered.
The bodyguards were contacted when they were attending a continued formation course on stress management at the academy. The questionnaires were administered without any specific instruction as if the material was part of the course, although it was specified that answering the questionnaires was voluntary. Subjects responded in an anonymous form.
Table 1 shows the correlation matrix of all the scales for the sample of the bodyguards. The correlations between the subscales and the Total scale of the SSS and the dimensions of the EPQ are similar to the ones found in other studies (Gomà-i-Freixanet, 1995). In general, the three personality dimensions of the EPQ correlate positively with the Total score on the SSS, as well as with each of its subscales, while the L scale correlates negatively. The absolute values are similar to the findings of other studies although the level of significance is lower, due to the small sample size.
Means, standard deviations and t-test comparisons between the bodyguards and the Catalan norm group are given in Table 2. Bodyguards differ significantly from the Catalan norms (Eysenck et al., 1992) on N, P and L; i.e., they score lower on Neuroticism and Lie and higher on Psychoticism. They do not differ on Extraversion. Regarding the SSS, bodyguards do not differ significantly from the Catalan norms (Pérez & Torrubia, 1986) on any of the subscales, nor on the Total score. It is interesting to note, in spite of the small sample size, the homogeneity (i.e. low SD´s) of the obtained results in the group of the bodyguards compared to the Catalan norms.
The second aim of this study consisted of whether the bodyguards share the personality profile of high physical risk groups in general and/or that of prosocial physical risk groups in particular. In order to assess this, we compared the bodyguards with three groups of male subjects similar in age from another study we conducted (Gomà-i-Freixanet, 1995): two risky groups –a high physical risk prosocial group and a high physical risk sports group- and a control group selected by not doing any risky activity. Table 3 shows means, standard deviations and t-test comparisons between the bodyguards and these three groups.
An analysis of the results indicates that bodyguards differ significantly from the risky sports group on 6 out of the 9 administered scales, from the controls on 5 out of the 9 scales and from the prosocials on only 2 out of the 9 scales. More specifically, bodyguards differ significantly from the control group and the risky sports group on the N, P and L scales of the EPQ in the sense that the former are more stable, conformist and sincere. In relation to the SSS, bodyguards differ significantly from the risky sports group on the Total score and on the ES and BS subscales, and from the controls on the TAS and BS subscales. Finally, the bodyguards only differ significantly from the prosocial group on the L scale of the EPQ and the BS subscale of the SSS in the sense that bodyguards are more sincere and less susceptible to boredom.
With regard to our first aim, to compare the bodyguards to the general population, the results from the Eysenck Personality Questionnaire indicate that they scored significantly lower on Neuroticism and on the Lie scale, i.e. they are more emotionally stable subjects, capable of responding in an adequate form to unforeseen and critical situations. Their high sincerity could be explained by the fact that this subset of policemen has probably been submitted to a rigorous selection process, where for security reasons only the most sincere were selected. In relation to the Psychoticism dimension, the bodyguards scored significantly higher than the Catalan norms, which seems paradoxical with regard to the expected results. However, if we compare the Psychoticism scores of the bodyguards with those of the other comparison groups (see Table 3) the tendency shows to be inverse, i.e., their Psychoticism scores are significantly lower compared to those from the risky sports group and controls. The unexpected high Psychoticism scores compared to the Catalan norms, could therefore probably be explained by these norms themselves. In the normative group, and specifically in the males, a correlation of -.32 between P and L scales was found. The high correlation between these two scales along with the fact that the Catalan norms were significantly higher than the English norms on the L scale, made the authors of the Catalan version (Eysenck et al., 1992) state that «the Catalan norms of the Psychoticism scale have to be interpreted with caution» (p. 228). Thus, if we take the risky sports group and control group as a reference instead of the Catalan norms, the bodyguards score significantly lower on Psychoticism. Regarding Extraversion, bodyguards do not differ significantly from the general population, i.e. they are ambiverted with average scores on Extraversion.
To summarize, the obtained results from the EPQ indicate that police bodyguards, compared to the general population, are emotionally stable subjects, sincere and ambiverted and, compared to our own reference groups, they are lower on Psychoticism as well. Indeed, this appears to be the appropriate profile in regard to the content of their work: persons that are capable of reacting rationally, not emotionally, to a situation that implies physical risk and to which you can trust the protection of someone else’s life.
Regarding the Sensation Seeking Scale, the results were in the predicted direction. The scientific literature (Zuckerman, Eysenck & Eysenck, 1978), as well as our own results (Gomà-i-Freixanet, 1995; Gomà-i-Freixanet, 2001) show that Extraversion and SSS display a high positive correlation between them. If the bodyguards do not differ significantly from the general population on Extraversion, as shown earlier, they should neither do so on the SSS as the present findings confirmed.
Gudjonsson & Adlam (1983) found that recruits that were at the academy at the time of the research scored equal to the general population on Neuroticism and Lie, lower on Psychoticism and higher on Extraversion. At their turn, Burbeck & Furnham (1984), comparing aspirants that were accepted with the general population, found that the former were more extraverted, less neurotic, lower on Psychoticism and higher on Lie. However, in both studies the subjects were assessed when they did not yet form part of the police corps and this could cause the subjects responding in a desired direction according to the social stereotype of a policeman, i.e.: stable, flexible, and eager to interact with people.
There are two possible explanations for the discrepancy regarding the results on Extraversion between these studies and the present. In the first place, the bodyguards in this sample were already for eight years in the corps which makes it plausible to assume that their responses were not affected by social desirability. Therefore, our results may be more adjusted to the reality than those described earlier. Another possible explanation could be that to become member of the Bodyguard Unit those are selected from the entire, compared to the general population, relatively more extraverted police sample that are less extraverted, causing them not being different from the general population. We cannot confirm the latter assumption as our sample of bodyguards has been selected based on subjective criteria by the superior commanders and not by means of personality questionnaires which makes a systematical comparison impossible. Thus, regarding our first aim we can conclude that they do not differ on the traits measured by the SSS but they do differ on Neuroticism, Psychoticism and Lie as measured by the EPQ. Taking into account that the personality dimensions as evaluated by the EPQ are relatively stable and enduring traits (Eysenck & Eysenck, 1985), our results favour the dispositional model, i.e. the bodyguards display a determined set of personality traits that distinguish them from the general population.
Our second objective consisted in whether the police bodyguards share the personality profile of the high physical risk groups in general and/or that of the prosocial physical risk groups in particular. In our work with risky men (Gomà-i-Freixanet, 1995) the only variable that discriminated the physical risk groups (antisocial, prosocial and sports) from the control group was the Thrill and Adventure Seeking scale from the SSS in the sense that the risky groups scored higher than the control group. The present study showed that the bodyguards also differed significantly from the control group on that scale. It follows that police bodyguards seem to pertain to this group of physically risky activities attracted by the seeking of thrill and adventure. If we subsequently compare police bodyguards with the prosocial group, the results indicate that they only differ significantly on two out of the nine compared scales. Therefore, our police bodyguards are more similar to groups that practice activities that imply physical risk of a prosocial kind (e.g. firemen, security guards, prison warders or life-savers from the Red Cross) than to those that practice activities that imply physical risk as well but of a sports leisure type (e.g. alpinism, diving, water-skiing, parachuting, hang gliding or ultralight flying). The only two scales on which the bodyguards differ from the prosocials are the L scale from the EPQ and the BS scale from the SSS. As we have mentioned earlier, the possible explanation of this elevated sincerity could be the subjects have probably undergone an exhaustive selection process within the police corps itself, as they are responsible for the protection of the physical integrity of the highest authorities. With respect to Boredom Susceptibility, bodyguards score significantly lower, meaning they are less prone to boredom. At first sight this may seem paradoxical if we limit ourselves to the social stereotype of this profession: permanently alert individuals working under high pressure due to constant possible danger. If however the bodyguards themselves are asked to give an accurate description of their work they respond that the most habitual is the routine, and not the unexpected. As a consequence, a low susceptibility to boredom or, similarly, a high tolerance for monotony allows the subject with this profile to maintain a high level of attention even in monotonous conditions.
Summing up, the data provided in this paper favour the hypothesis of the dispositional model in the sense that the police bodyguards of this sample differ from the general population on two big personality dimensions: Neuroticism and Psychoticism, but not on Extraversion. On the other hand, the same data bring empirical evidence about the appropriateness of the classification model of Gomà-i-Freixanet being presented elsewhere (Gomà-i-Freixanet, 1995; Gomà-i-Freixanet, 2001). Police bodyguards display a personality profile similar to persons that practice activities with high physical risk in general, i.e. danger and adventure seeking attracts to him, and to persons that practice high physical risk activities of a prosocial kind in particular: they are ambiverted, emotionally stable, flexible, not attracted by the seeking of experiences nor disinhibited and demonstrate the distinctive characteristic of a high sincerity and a low susceptibility to boredom.
These results have implications in a practical and theoretical sense. The former as the personality profile found in this study could be used as a criteria in the selection process of future applicants, the latter as it gives empirical evidence supporting the aforementioned classification model. We have to mention that although these results are quite interesting from both points of view, a replication study with a larger sample is needed before any generalization can be done.
Austin, T.L., Hale, D.C. & Ramsey, L.J. (1987). The effect of layoff on police authoritarianism. Criminal Justice and Behavior, 14, 194-210.
Aylward, J. (1985). Psychological testing and police selection. Journal of Police Science and Administration, 13, 201-210.
Burbeck, E. & Furnham, A. (1984). Personality and police selection: Trait differences in successful and non-successful applicants to the metropolitan police. Personality and Individual Differences, 5, 257-263.
Burbeck, E. & Furnham, A. (1985). Police officer selection: A critical review of the literature. Journal of Police Science and Administration, 13, 58-69.
Carpenter, B.N. & Raza, S.M. (1987). Personality characteristics of police applicants: comparisons across subgroups and with other populations. Journal of Police Science and Administration, 15, 10-17.
Cochrane, R. & Butler, A.J.P. (1980). The values of police officers, recruits and civilians in England. Journal of Police Science and Administration, 8, 205-211.
Colman, A.M. & Gorman, L.P. (1982). Conservatism, dogmatism and authoritarianism in British police officers. Sociology, 16, 1-11.
Cooper, C.L., Robertson, I.T. & Sharman, P. (1986). A psychometric profile of British police officers authorized to carry firearms: A pilot study. International Review of Applied Psychology, 35, 539-546.
Eber, H.W. (1991). Good cop includes bad cop: A supplementary concept of police brutality. Paper presented at the annual meeting of the Society of Multivariate Experimental Psychology, Albuquerque, NM. (cited in Lorr & Strack).
Eysenck, H.J. & Eysenck, M.W. (1985). Personality and individual differences: A natural science approach. New York: Plenum.
Eysenck, H.J. & Eysenck, S.B.G. (1975). Manual of the Eysenck Personality Questionnaire. London: Hodder & Stoughton.
Eysenck, S.B.G., García-Sevilla, L., Torrubia, R., Avila, C. & Ortet, G. (1992). Versió catalana de l’EPQ per a adults: Un instrument per a la mesura de la personalitat. Annals de Medicina (Barc), 9, 223-230.
Genz, J.L. & Lester, D. (1976). Authoritarianism in policemen as a function of experience. Journal of Police Science and Administration, 4, 9-13.
Gomà-i-Freixanet, M. (1991). Personality profile of subjects engaged in high physical risk sports. Personality and Individual Differences, 12, 1087-1093.
Gomà-i-Freixanet, M. (1995). Prosocial and antisocial aspects of personality. Personality and Individual Differences, 19, 125-134.
Gomà-i-Freixanet, M. (1997). Consensual validity of the EPQ: Self-reports and Spouse reports. European Journal of Psychological Assessment, 13, 179-185.
Gomà-i-Freixanet, M. (1998). Personalidad y conducta arriesgada en mujeres: resultados preliminares. En P. Sánchez López y A. Quiroga Estévez (Eds.), Perspectivas actuales en la investigación psicológica de las diferencias individuales. Madrid: Centro de Estudios Ramón Areces.
Gomà-i-Freixanet, M. (2001). Prosocial and antisocial aspects of personality in women: a replication study. Personality and Individual Differences, 30, 1401-1411.
Gomà-i-Freixanet, M., Pérez, J. & Torrubia, R. (1988). Personality variables in antisocial and prosocial disinhibitory behavior. In T.E. Moffit & S.A. Mednick (Eds.), Biological contributions to crime causation, (211-222). Dordrecht: Martinus Nijhoff Publishers.
Gomà-i-Freixanet, M. & Puyané, P. (1991). Personalidad en alpinistas vs. otros grupos que practican actividades relacionadas con la montaña. Psicothema, 3, 73-78.
Griffeth, R.W. & Cafferty, T.P. (1977). Police and citizen value systems: Some cross sectional comparisons. Journal of Applied Social Psychology, 3,191-204.
Gudjonsson, G.H. & Adlam, K.R.C. (1983). Personality patterns of British police officers. Personality and Individual Differences, 4, 507-512.
Hargrave, G.E., Hiatt, D. & Gaffney, T.W. (1986). A comparison of MMPI and CPI test profiles for traffic officers and deputy sheriffs. Journal of Police Science and Administration, 14, 250-258.
Johnson, E.E. (1990). Psychological tests used in assessing a sample of police and firefighter candidates: An update. American Journal of Police, 4, 85-92.
Krug, S.E., Cattell, R.B. & IPAT (1980). Clinical Analysis Questionnaire manual. Champaign, IL: Institute for Personality and Ability Testing.
Lorr, M. & Strack, S. (1994). Personality profiles of police candidates. Journal of Clinical Psychology, 50, 200-207.
Pérez, J. & Torrubia, R. (1986). Fiabilidad y validez de la versión española de la escala de Búsqueda de Sensaciones (forma V). Revista Latinoamericana de Psicología, 18, 7-22.
Rokeach, M., Miller, M.G. & Snyder, J.A. (1971). The value gap between police and policed. Journal of Social Issues, 27, 155-177.
Spielberger, C.D. (1979). Police selection and evaluation: Issues and techniques. New York: Praeger Publishers.
Wright, B.S., Doerner, W.G. & Speir, J.C. (1990). Pre-employment psychological testing as a predictor of police performance during an FTO program. American Journal of Police, 4, 65-83.
Zuckerman, M., Eysenck, S.B.G. & Eysenck, H.J. (1978). Sensation Seeking in England and America: Cross-cultural, age and sex comparisons. Journal of Consulting and Clinical Psychology, 46, 139-149.