Development of a valid psychological scale for youth police selection among young unemployed and excluded persons: The Nemesis scale

1.1 Youth, unemployment and poverty

Youth, unemployment, and poverty conform a set of situations that are often inseparable. In many countries, young people have become a lost generation because they have very little access to the labour market, causing the burden on their families to impoverish the family income. This phenomenon, which has become a daily aspect of industrialized countries [1], is increasingly more acute in developing countries. In such countries, the only access to the labour market for certain sectors of the population is through informal economy (i.e., outside the official figures), and—what is more worrisome—by means of criminal activities that affect the public health. An example of this is the case of young people from the Andean coca areas, who produce and collect the coca leaf, which will subsequently be processed and turned into cocaine in other countries, flooding the international markets.

It is difficult to convince the youth of these regions to stop engaging in illegal business when the legal alternatives provide a very low remuneration for them. Only through a profession of high social prestige, which involves higher wages than those offered by the agricultural tasks of the area, it is possible to convince them.

The program set out in this work aims to remove from social exclusion youth who, if they are not offered an alternative, would normally end up cultivating the coca leaf. The alternative was to prepare them for entry into law enforcement. To this end, appropriate training was designed so that candidates could be selected in the future to the Peruvian police force. The PNP/NAS Pre-Police Schools Project “Sowing opportunities” was created. The project was supported by the United States Embassy in Peru, through the NAS (U.S. Office of Anti-Narcotics Affairs), and with a Peruvian organization, Centre for Prevention Studies, Treatment, Research and Health (CEPTIS) as an executive body. The PNP/NAS Pre-Police Schools Project aims to create opportunities to incorporate and train young people from areas at high risk of drug trafficking and terrorism and offer them a non-criminal life alternative. In addition, the training they will receive is intended to provide them with at least a minimal education, which will have an impact on the development of their cognitive, emotional, and social abilities. In this way, young people who participate in this program may develop personally, achieving a more appropriate and comprehensive quality of life.

This research will centre on the creation of an instrument to predict which young people are best suited to receive such pre-police training.

Unemployment and precariousness, on the one hand, create a picture of social and psychological dismemberment. A jobless person is more likely to suffer health problems and, as time goes by and he or she does not enter into the labour market, such health problems increase. On the other hand, for some populations, it is practically impossible to achieve a socially “normalized” job, as society does not offer many job opportunities to the disadvantaged classes, generating an informal economy among persons who are unemployed. In the case at hand, young people from many areas of Andean jungle areas find their informal occupation in the cultivation and collection of the coca leaf, causing not only damage to themselves, but to humanity as a whole. As mentioned, it is difficult to offer a legal and socially regular job to these young people, as they often perceive much higher pay by engaging in marginal activities.

This social reality leads us to recall the case of Marienthal (Austria, a small municipality near Vienna), which was analyzed at the beginning of the 20th century. In this town, the unemployment rate had skyrocketed; 75%of its working-age inhabitants were unemployed, and 30%of the families had severe material deficiencies. Practically all the population was inactive, as they had a single activity, textiles, and were suffering the effects of the Great Depression (the Crack of 1929). Three renowned social scientists, Marie Jahoda, Paul Lazarsfeld, and Hans Zeisel, analysed the situation of that municipality to determine how unemployment affected daily life, health, and the attitudes of the people [2]. Their results showed that, in general, the precarious situation of collective unemployment, far from favouring the workers’ solidarity, increased their individualism. Usually, if there is a lack of jobs in societies where work is the centre of people’s lives, this affects not only income (i.e., access to necessary goods and services), but also people’s emotional and mental well-being.

In this new scenario, unemployment and precarious employment are again extending to become structural. Marienthal was used to describe social and personal disorganization arising from unemployment: other Marienthals emerged, and cities such as Detroit (USA), which had been a major industrial centre, rapidly degraded, dragging down entire social sectors. The example of Marienthal continues to produce a rich literature in this regard [3–5].

In the 21st century, technological development has created a new horizon. Machine learning and robotics are a paradigm shift that threatens to produce a “useless class” (Yuval Noah Harari), a huge mass of people with no chance of ever obtaining paid employment [6–8]. Harari posits that a significant percentage of humanity could be expelled from the labour market. This situation raises many issues, especially regarding the economic and political power of the more fragile social sectors. Two questions are unavoidable: What means of subsistence will there be for this huge mass of unemployed people with no possibility of entering the labour market? What will protect all these people from the physical and mental deterioration associated with the lack of work activity? Like other scholars such as the British academician Selina Todd, in the face of this bleak future, Harari proposes a radical change in the social worldview of work: an educational, political and economic transformation [9, 10].

As Harari and Todd advocate, a change of mindset is necessary. People’s lives should assume other priorities. Working to make others rich—the essence of the current free market system—should no longer be the central axis of our existence, and economic security and strong community ties should be offered to all people. In fact, there are already some communities of poor people who manifest their happiness in environments characterized by solidarity and cooperation. Technological development, which threatens to leave half of humanity unemployed, could help shape a society in which human relations come before industrial relations. Lower wages and the need to provide aid to large masses of unemployed people have generated the label of “the working poor” (i.e., workers whose incomes fall below the poverty line). Hence, the concept that having a job implies leaving poverty behind is shattered [11].

However, in the case at hand, a solution must be found to provide socially accepted work to a population of young people who will turn to illegal work if no alternative employment is found. This illegal work undermines the whole of society and is the first step in the worldwide production, distribution and sale of cocaine.

Moreover, the different analyses carried out (usually by Eurofund) show that these difficulties have not only an immediate impact, but also significant long-term effects on the development of the young people involved. In this sense, the experience of previous periods of recession highlights the long-term consequences of early unemployment and precariousness. At the individual level, these situations lead to a lower lifetime income, more frequent transitions between employment and unemployment, higher rates of poverty even after retirement, and worse physical and mental health [12]. Similarly, at the collective level, the exclusion of young people from the labour market can undermine the growth potential of economies and is a clear threat to social cohesion [13]. In addition, the diminishment of these young people’s capacity for political and social participation exacerbates previously existing class differences. In short, the social and economic processes of recent years lead to an inter-generational interpretation of the crisis and oblige us to address its potential future conflict and to treat this scenario as an emergency.

Research shows that this everyday framework fuels hopelessness about the future. As the unemployment situation worsens and becomes chronic, hope for a plausible future for oneself or one’s children vanishes. The process follows a pattern: initial cautious optimism turns into resignation and ends in despair. In the meantime, the sociocultural fabric of the working-class community is disintegrating. This is noted in the decrease in reading rates, the paralysis of cultural activities and previous associative life, and the progressive loss of trust in former co-workers [14]. After this introduction about unemployment, youth and social marginalization, we will refer specifically to studies that focus on police training.

1.2 Police studies and police training

Police officers are very important to ensure safety on a regional and national level. The police profession can be classified as a risk, and some studies have indicated that the rate of suicide in this group, which carries out public protection activities, is elevated [15]. This is despite candidate selection focusing on recruiting suitable police officers, in order to have suitable police officers, an appropriate selection of candidates is crucial [16–18]. Many studies have tried to determine the main skills—psychological, cognitive and physical—that a good police officer should possess. For example, earlier instruments have addressed aspects like: psychological processes, characteristics of the selection procedure (the selector vs. the selection of different groups of people: gender or racial differences) as well guidelines for the selection interview. However, earlier research did not address in detail the psychological processes involved in selecting the best functioning police officer. Many studies have highlighted a good police officer’s physical skills, an aspect that can be consulted in the meta-analysis carried out by Hunter and Hunter [19], as well as subsequent works [20, 21].

Different studies have focused on determining the socio-demographic variables that could adequately predict the selection of a good police officer [22]. The most studied variables are age, origin and sex, but not all studies agree about this [23, 24], as this study finds that demographic variables are poor predictors of success within the academy or subsequently in police performance. Furthermore, the use of variables such as sex in the selection process leads to questioning the social ethics of the process. Gender issues have been debated in these studies, as police work has classically been a man’s job, implying a clear discrimination against women in this field [25], and also both female and minority applicants have faced challenges in recruitment, hiring and selection [26]. Bogićević-Milikić [16] has suggested that interventions aimed at managing gender-sensitive talent should be carried out to make police bodies sensitive to gender issues.

In addition, some studies attempt to determine whether police force candidates’ characteristics—including school behaviour and prior history of criminal or antisocial behaviour—can predict good future police officers. Thus, for example, the work of Gray [27] analyzed the problematic behaviours of boys who want to become policemen, concluding that they had more alcohol-related problems (according to self-report questionnaires) but a smaller percentage had been arrested and they were less likely to consume drugs (except marihuana) than a control group of individuals who did not want to become police officers.

Other studies have focused on the advantages and disadvantages of becoming a police officer. The main advantage is that most candidates who are admitted to the police academy are guaranteed a job before starting their formation in the academy. Furthermore, the costs of the police academy are mostly assumed by the public institution that selects the candidates. On the other hand, the disadvantage of becoming a police officer is the high personal risk involved in this profession. Many studies have shown that a police career may be a way to integrate minority groups into society, granting them easy access to education and job security. In this regard, one of the most frequently discussed topics is how officers from each minority group react when dealing with people from their own group, and whether their adequate professional performance can be predicted in this situation. White [28] and White, Cooper, Saunders and Raganella [29] have addressed this issue.

A different type of research has attempted to verify whether candidates’ school behaviour during the police academy training period, as well as past job performance prior to selection, are good predictors of future performance. An example of this approach is the work of Wright, Dai and Greenbeck [30], who focus on determining whether past job performance is a good predictor of admittance into the police academy. Another work in this regard is the aforementioned study by Henson, Reyns, Klahm and Frank [23], who found that previous work experience (type of job) was a poor predictor of success within the academy or subsequently as police officers. According to White [28], having good previous academic training is not a predictor of better police performance.

Some scholars have focused on the period after selection and training to determine whether service history in the daily course of police officers’ work can adequately predict future performance. White, Cooper, Saunders, and Raganella [29] analyzed the stability of individuals’ motivations to become police officers over time, especially among women and minority groups. Their results showed high stability. A strange result in this study was that white males tended to report low job satisfaction, and a relationship between low satisfaction and concealed motivations was found.

Finally, in the selection process, one of the main variables to consider (besides cognitive variables and the physical examination) is the psychological examination of the candidate. The fundamental contribution of psychologists to police officer selection has been to conduct processes that rule out candidates with some kind of mental illness. This is the perspective we focused on in our study.

The prospect of selection through psychological procedures (which does not preclude other types of tests, especially those of physical ability) has increased since the National Advisory Commission on Criminal Justice Standards and Goals recommended in 1973 that all law enforcement agencies follow a formal selection process that includes (a) a written test of mental ability or aptitude, (b) an oral interview, (c) a psychological examination, and (d) a background check. Subsequently, in 1998, the “International Association of Chiefs of Police” developed several guidelines to follow when carrying out the psychological evaluations prior to granting a job [18].

Out of all above perspectives, we are interested in this last one, because it is the perspective of the Legal Psychology and because it is essential to verify the adequate mental health of future police officers in order to prevent misfortunes in the performance of their duties, as they are armed.

In this paper, we focus on determining the psychological characteristics that may contribute to predicting the adequate performance of a police officer in the future. However, we do not verify that the selected subjects have adequate mental health. In short, we are interested in predicting which subjects will become good officers, through the use of psychological testing. Barrick and Mount [31], and Tett, Jackson and Rothstein [32] carried out two meta-analyses that provided an in-depth review of the psychological tests used.

The studies that have focused on determining candidates’ adequate mental health have obtained unpromising results [33]. In general, these studies show that personality variables are not good predictors of future police performance. This does not mean they should be discarded, as it would be unthinkable for future police officers to have inadequate mental health. However, it does show that other psychological tests focusing on future police performance are needed.

Perhaps the work of Lough and Von Treuer [33] has gone furthest in determining the validity of psychological mental health tests for the prediction of adequate performance of future police officers. These authors focus in particular on well-known instruments within the field of clinical psychology, which, interestingly, are also the most predictive. Essentially, the following instruments (among others) were examined: the AIFP (Australian Institute of Forensic Psychology test battery), the CPI (California Personality Inventory), the IPI (Inwald Personality Inventory) and the MMPI (Minnesota Multiphasic Personality Inventory). The results indicated that none of the tests obtained adequate validity parameters but, within this general conclusion, the most valid instruments were the CPI and the AIFP. We therefore aim to further develop this area by introducing and thoroughly examining a new selection instrument that is based on mental health and that possesses the psychometric properties of reliability and validity.

There is considerable agreement that a future police officer should have the following traits [34]:

Adequate emotional maturity, which includes not only behavioural self-control but also emotional self-control in the face of victims who approach the police not only with physical injuries but also often with a great deal of emotional distress.

In any case [35], selection processes can be quite unpredictable depending on the circumstances, but at the same time, there is an increasing demand for good police officers. Henson et al. [23] have indicated how psychological selection of future officers may be used. In first place, this involves screenings that will prevent individuals with mental health problems from gaining access to the job [36, 37]. The problem arises in the identification of the traits required of a good police officer, as well as the identification of good police officers, which can therefore be used as criteria [28, 38]. As Lough and Von Treuer [33] stated, normally, the psychological tests used have a limited degree of validity. This was already stated by Burkhart [39] regarding intelligence testing. In general, researchers agree that a good future police officer must not have negative psychological traits (low intelligence or a high level of psychoticism or paranoia, for example), but the failure to manifest these deficiencies does not mean that these people will be good police officers.

Although it is not the object of this work, we want to mention that we must not only select good police officers, but that it is essential to take care of their occupational health, as many researches demonstrate how we are dealing with a profession with a high rate of occupational diseases [40, 41], its origin frequently being job stress [42]. In this sense, all kinds of treatment techniques have been used to avoid stress, including yoga [43], as well as recently those from the Job Demand Control Support model (JDCS), including Formal mentoring [44, 45].

The objective of this research is to determine the psychological variables that reflect the above-mentioned characteristics, designing a psychological test that detects these traits. The aim of this work was to integrate all the specified aspects within a battery that is not too lengthy and to verify that the resulting test has adequate reliability and validity. Therefore, the objective of this research is to create a psychological test that can be considered a good predictor of adequate police performance.

2.1 Participants

A sample of 713 unemployed persons with no initial qualifications was used. Participants were selected to join a pre-police training program at Peruvian schools in Ayacucho, Tingo María, Huánuco, Mazamari and Santa Lucía. All were selected to be future students for the PNP/NAS “Sowing Opportunities” pre-police school project (see section on Procedure). Once the data were coded, eight subjects were eliminated because they had not completed the entire questionnaire, and so the total number of participants in the sample was 713. The percentage of men (59.5%) was somewhat higher than of women (40.5%) in the final sample. Mean age was 18.56 years (SD = 1.062, range 15 –21 years).

2.2 Instruments

We created an instrument called Nemesis to measure suitability to become a good future police officer. This scale should have the least number of items possible.

Validity will also be tested, in this case, using a clinical test, Derogatis’ Symptom Checklist (SCL-90). This 90-item scale detects psychosomatic symptoms within the period of the last two months and is rated on a 5-point Likert-type response format, ranging from 0 (I have never displayed the symptom) to 4 (I display it continuously). The scale provides information on the following symptoms: Somatization, Obsessive-Compulsive Disorder, Interpersonal Sensitivity, Depression, Anxiety, Hostility, Phobic Anxiety, Paranoid Ideation, and Psychoticism. It also provides global health indices. If the Nemesis is valid, it should correlate negatively with the indicators of this instrument.

2.3 Procedure

This research collected the data from the PNP/NAS “Sowing Opportunities” Pre-police Schools Project. The U.S. Embassy in Peru, through NAS (Narcotics Affairs Section of the United States), with the help of the Peruvian organization CEPTIS (Centro de Estudios de Prevención, Tratamiento, Investigación y de Salud, Centre for Prevention, Treatment, Research, and Health) as executing agency, has promoted the Pre-Police Schools in Peru with the aim of contributing to the comprehensive training of young people to prepare them to enter the Superior Technical School of the Peruvian National Police (PNP).

The Pre-Police School is an opportunity for young people and is framed within a strategy of cultural integration that seeks to create opportunities to incorporate youth into society. It provides them with free comprehensive training based on the principle of equal opportunities, so that the young beneficiaries can participate in the entrance examination for the PNP Superior Technical School in better conditions in terms of academic and psychological preparation, the personal interview, physical training, and general culture.

This project has been active since 2007 in areas of Peru at a high risk for drug trafficking and terrorism, particularly in Huamanga (Ayacucho), Mazamari (Junín), Santa Lucía (San Martín), and Tingo María (Huánuco), with aim of providing comprehensive training to selected youth. The PNP/NAS “Sowing Opportunities” Pre-Police Schools Project was developed within the framework of the objectives of reducing the supply of drugs and a strategy of cultural integration of young people. It aims to create opportunities to incorporate and train youth from areas at a high risk for drug trafficking and terrorism, in order to offer an alternative, non-criminal life. It thus seeks to improve their cognitive, emotional, and social skills so as to enable them to achieve better personal development and to strengthen ties to their communities. The goal is to promote a culture of legality and to give these youth a better quality of life.

To achieve the objectives of the above-mentioned Pre-Police Schools Project, the youth are prepared and given a solid education to “catch up” with their peers in academic and psychological factors, physical training and general culture. The experience is a model of comprehensive training. So far, a total of 2,230 youths have benefited from the Pre-Police Schools in the project’s intervention zones.

All the subjects who participated were informed of the purpose of this study and signed an informed consent form. They were specifically informed that their data would anonymous and handled statistically, and that they could see the results and withdraw permission for the use of their data at any time. None of them declined to participate. Data were collected between February and December 2017.

The psychological evaluation was conducted by senior officer psychologists from the PNP. The evaluation consisted of an assessment of intellectual abilities, a personality assessment, an assessment of skills through a personal interview, and an evaluation of candidates’ vocation for police service (also through a personal interview).

The selected subjects from one of the police preparation classes were administered a wide range of items that might predict good future police behaviour (200 items) to determine the adequacy of the items to be included in the final scale. The 200 items were selected on the basis of the consensus of a group of people about the characteristics of good police performance. This discussion group included the following participants: two Peruvian police commanders, an expert on the battle against drugs, an expert in social and labour insertion of young unemployed people, and two psychologists from Ceptis (the organisation that trained the young people). The SCL-90-R was also administered [46–48] in its validated Spanish version [49].

Firstly, the adequacy of the items for the future questionnaire was determined through factor analysis, and reliability of the proposed scale was deter-mined using Cronbach’s alpha. Secondly, the validity was determined by analyzing the Pearson correlation of the scores on the proposed scale with the SCL-90 score. Finally, logistic regression was used to determine whether the items on the Nemesis scale can predict possible admission to police training. The PASW of the SPSS, version 20.0, was used for the statistical analyses.

This research was approved by and Ethical Committee in Spain, ref. 15/2018.

3.1 Instrument reliability

A factor analysis was performed with the initial 200 items of the Nemesis scale. The factor extraction method was maximum likelihood, with varimax rotation and Kaiser normalization. The rotation converged after 11 interactions. The results indicated that the first factor explained 18.82%of the total variance, whereas the remaining factors explained a very low amount (the second factor accounted for 7.55%, the third for 7.27%, etc.). When adding factors, the increase in explained variance was very small, indicating that the proposed scale should have one factor.

After confirming the items that loaded on that factor (i.e., that did not load on the other factors), 14 items were selected. Table 1 shows the factor loadings of each item on the factor.

Table 2 presents the communalities of the selected items. Goodness of fit was calculated with chi-square: χ²(49) = 50.393, p = 0.418 (nonsignificant, so the scale was adjusted to a normal distribution).

After determining the factorial structure of the instrument, we confirmed the homogeneity of the 14 items comprising the scale. For this purpose, we calculated the reliability with the alpha coefficient, α= .817, for the standardised items, which is acceptable. We also examined the intercorrelations of all the scale items, obtaining adequate results. These correlations are shown in Table 3. The average score of the scale items was 52.36 (SD = 7.294; max. score = 14, min. score = 70).

We then proceeded to calculate the correlation between the global score on the Nemesis scale with each of the individual factors and global scores of the SCL-90. The Pearson correlations were significant with the following scales: Somatization, Interpersonal Sensitivity, Anxiety, Hostility, Phobic Anxiety, and Psychoticism. In addition, the correlations with the three global indexes of the test were significant: GSI (Global Severity Index, also called the general symptomatic index; this is a generalized and indiscriminate measure of intensity of overall mental and psychosomatic distress); PST (breadth and diversity of psychopathology); and PSDI (global distress, an indicator of average symptomatic intensity). All the correlations were negative, indicating that the scale is valid: that is, the higher the score on the Nemesis scale, the lower the manifestation of psychosomatic symptoms. The results are shown in Table 4.

Finally, we applied a binary logistic regression to the data to determine whether the proposed test could determine higher or lower degree of successful admittance (i.e., who will or will not pass the entrance exams) to the police academy. As some subjects did not complete the entire training process, we could only work with 597 candidates. Omnibus tests applied to the model coefficients showed high significance, χ²(14) = 79.141, p = .001. The estimate concluded on iteration 4 because subsequent parameter estimates changed by less than .001. The Hosmer and Lemeshow test showed high significance, χ²(14) = 18.988, p = .015. The classification table (see Table 5) shows that the mean correct prediction is 64.8%: that is, higher than 50%, which would indicate of lack of predictiveness. The increase in predictiveness over 50%is 14.8%. We also calculated sensitivity (the test’s capacity to correctly identify candidates likely to pass future police training) and specificity (the test’s capacity to correctly identify as negative cases the candidates who were not likely the academy entrance exam). These indices obtained the following values: sensitivity = 56.98%and specificity = 71.38%.