Keren Cuervo1 , Lidón Villanueva Badenes1 , Francisco González2 , Cruz Carrión3 , Pilar Busquets3
1Univ. Jaume I, Castellón, España ,2Univ. Valencia, España ,3Juvenile Court of Castellón, España
The aim of this study is to define a profile of juvenile offenders depending on the type of crime (againstproperty or against persons), according to several socio?demographic variables, and a number of indicatorsof juvenile risk. Participants were 395 adolescents between the ages of 14 and 18 with a criminal record inthe juvenile court over a two-year follow-up period. Results showed that in property-related offences theoffender is more likely to be male, from an Eastern European country, and with inconsistent parenting. Onthe other hand, crimes against persons would be committed mostly by girls, Latin American or Africanjuveniles, and with individual factors such as aggressive behaviour, outbursts of anger, poor frustrationtolerance, or little concern for others. These results may be useful in designing crime prevention andoffender intervention programmes.
El objetivo del presente estudio es determinar la existencia de un perfil diferencial de menores infractoresacusados de haber cometido delitos contra la propiedad y contra las personas, según diferentes variablessociodemográficas y diversos indicadores de riesgo juvenil. Los participantes fueron 395 menores entre 14y 18 años, con expediente judicial en el Juzgado de Menores, en un periodo de seguimiento de dos años. Losresultados indican que en los delitos contra la propiedad es más probable encontrar un menor infractor degénero masculino, procedente de los países del Este, con pautas educativas parentales inconsistentes. Porsu parte, los delitos contra las personas serían cometidos en mayor proporción por mujeres, de nacionalidadlatinoamericana o africana, predominando como principales factores de riesgo la agresividad, ataquesde cólera, baja tolerancia hacia la frustración o poca preocupación por los demás. Estos datos pueden serimportantes de cara al diseño de programas de prevención de los delitos e intervención con infractores.
The study of young offenders has frequently been addressed from the paradigm of the protection and risk factors, without also taking into account more criminogenic aspects such as the type of crime and its relationship with a specific youth profile. Perhaps the principle of non-proportionality between the fact itself and the penalty that is followed in juvenile justice has pushed the variable type of crime somewhat into the background. Were this relationship to exist, it could be used to guide the design of intervention programmes that are tailored to meet the needs of youths, since the Spanish Organic Law of Criminal Liability of Minors 5/2000 already includes the principle of flexibility when it comes to adopting and executing the measures seen as the most appropriate under the circumstances of each particular case. Accordingly, this study analyses the possible relation between the type of crime committed (against persons versus against property) and the different socio-demographic variables and risk factors of the minor.
In this work, the classification of crimes against persons and against property was chosen because it is one of the most widely used in the literature on young offenders. A number of studies have shown that the percentage is higher in the case of crimes against property ( Capdevila, Ferrer, & Luque, 2005; Fernández, Bartolomé, Rechea, & Megías, 2009; Garrido, 2009; Iborra, Rodríguez, Serrano, & Martínez, 2011; Núñez, 2012 ), the most common being robbery, robbery with violence and intimidation, and burglary with forced entry ( Bravo, Sierra, & del Valle, 2009; San Juan & Ocáriz, 2009 ). Yet, in recent years there has been an increase in the number of violent crimes and sexual offences ( Capdevila et al., 2005; Fernández et al., 2009 ), or violence that takes place in family or school settings ( Benavente, 2009).
As regards the age of the young offender, the data indicate that the number of offences increases as minors get older ( Instituto Nacional de Estadística, 2011 ). This tendency occurs both in offences against persons and against property, with figures rising from 1,075 at the age of 14 to 2,225 at 17 in the case of offences against persons and from 580 to 1,146 in those committed against property ( Instituto Nacional de Estadística, 2011 ). On analysing the most common course followed by the young offender, that is to say, limited to adolescence with its highpoint at the age of 17, it can be seen that, generally speaking, crimes against property are more common than those against persons ( Moffitt, Caspi, Harrington, & Milne, 2002 ). Nevertheless, no studies have been conducted to analyse particular age groups and their possible relationship with the type of crime.
The variable gender of the young offender is a classic result that appears persistently in the literature ( Cuervo & Villanueva, 2013; Fernández et al., 2009; Moreira & Mirón, 2013; Ozen, Ece, Remzi, Tirasci, & Goren, 2005 ). Males commit a greater number of crimes and offences than females, 26,527 versus 4,534 respectively ( Instituto Nacional de Estadística, 2011 ), a tendency which remains constant across practically all kinds of crimes. According to these data, the number of offences against persons committed by male minors was 5,274 versus 1,557 committed by females. With regard to crimes against property, males committed 2,879 offences versus 692 in the case of females. In terms of the criminal typology, girls commit a greater number of crimes against persons ( Capdevila et al., 2005 ), while boys are more likely to commit crimes against property ( Vigna, 2012 ). Within crimes against persons, however, those related with family abuse are mostly committed by boys, with figures of 74% versus 26% for girls ( Iborra et al., 2011).
Another important variable in the study of young offenders is nationality. According to Capdevila et al. (2005) , the number of foreign young offenders has grown exponentially in comparison to the number of Spaniards over the last five years, while also displaying higher rates of recidivism and its occurrence at earlier ages. Several studies ( Capdevila et al., 2005; Cuervo & Villanueva, 2013; Iborra et al., 2011 ) yield percentages of Spanish young offenders ranging between 70% and 80%. The remaining 20-30% would be made up of North African, Romanian, or Latin American youths, depending on the province in which the study is conducted. Capdevila et al. (2005) performed a specific analysis of the most numerous groups of foreign minors in Catalonia, namely those from North Africa, who were found to commit more crimes against persons and more serious crimes. Nevertheless, few studies have analysed this relationship between the type of crime and nationality, and it is therefore an issue still to be determined.
On analysing the relationship between recidivism and type of crime, first it should be noted that the rates of recidivism in the Spanish juvenile penal system stand at between 14% and 40.6% ( Acosta, Muñoz de Bustillo, Martín, Aragón, & Betancort, 2012; Bravo et al., 2009; Capdevila et al., 2005; Contreras, Molina, & Cano, 2011; Armas, García, & Castro, 2008; García-España, García, Benítez, & Pérez, 2011; Graña, Garrido, & González, 2006; San Juan & Ocáriz, 2009 ), which are similar percentages to those found by Jennings (2002) in the United Kingdom or by Cain (1997) in Australia. On matching these rates with the type of offence, crimes against property are found to be the ones with slightly higher rates of recidivism than in the case of crimes against persons ( Winner, Lanza-Kaduce, Bishop, & Frazier, 1997 ), although other studies find no significant differences ( Cuervo & Villanueva, 2013; García-España et al., 2011 ).
The minor's risk factors are also related to the crime. The greater the presence of these factors, the more likely the occurrence of a situation involving juvenile delinquency ( Fergusson & Lynskey, 1996; Smith, Thornberry, Rivera, Huizinga, & Stouthamer-Loeber, 2000 ). These juvenile risk factors have been classified for the most part into two groups: factors of a more individual nature and those related to the different settings surrounding the minor, such as family, school, peers, and so forth. The more individual factors include low empathy and little concern for others, impulsiveness, lack of remorse, non-acceptance of responsibilities, scarce self-control, or sensation-seeking ( Aguilar & Godoy, 2013; Bush, Mullis, & Mullis, 2000; Vilariño, Alves, & Mohamed-Mohand, 2013 ). In contrast, the contextual factors include problematic families, negligent or permissive parenting styles, behavioural problems at school, relationships with delinquent peers, inappropriate use of leisure time, substance abuse, migration experiences, and low family income, among others ( Contreras, Molina, & Cano, 2010; Cottle, Lee, & Heilbrun, 2001; Hoeve, Semon, Gerris, van der Laan, & Smeenk, 2011; Kelly, Macy, & Mears, 2005; Ozen et al., 2005 ). Yet, as far as we know, to date no studies have been conducted that relate these individual and contextual risk factors with the type of crime committed. Only Wikström and Loeber (2000) analysed this relationship using the severity of the offence as their criterion. These authors highlight the low relevance of the socio-economic context of the neighbourhood versus other risk factors in the more serious crimes committed by minors.
Hence, the aim of this work is to determine whether the profile of young offenders charged with committing crimes against property and against persons varies according to different socio-demographic variables (gender, age, nationality) and different indicators of juvenile risk. It is expected that crimes against persons will be committed in a higher proportion by older adolescents, girls, and foreign minors (mainly North African). Individual factors will be more related to crimes against persons. The study addresses the issue by combining data from interviews and court records, together with an objective, standardised instrument: the YLS/CMI Inventory, built on evidence-based practice ( Guy, Packer, & Warnken, 2012; Henggeler, 2004 ). Likewise, it should be noted that this study also draws on a two-year prospective follow-up period with official data on recidivism.Method Participants
The participants were all minors from a province in the Valencian Region in Spain, who had committed a crime or offence for which they had appeared in the Juvenile Court within the period 2008-2010. The sample included a total 395 minors altogether, 74 of whom were females and 321 males (18.7% and 81.3%), with ages ranging between 14.3 and 19.3 years, the mean age being 16.08. In 203 cases (51.4%) the type of crime committed was against persons and in 192 cases (48.6%) they involved property. In terms of nationality, 303 were Spanish (76%), 36 were from countries in Eastern Europe (9.1%), 31 were Latin Americans (7.8%) and 25 came from African countries (6.3%).Instrument
The instrument that was used in the study was the Youth Level of Case Management Inventory (YLS/CMI) of Hoge and Andrews (2002), which was translated by Garrido, López, Silva, López, and Molina (2006) as the Inventario de Gestión e Intervención para Jóvenes (IGI-J). The purpose of the inventory is to assess the minor's risk of recidivism based on information provided from different sources.
The 42 items, or risk factors, of the inventory are grouped into 8 areas: 1) prior and current offences/adjudications; 2) family circumstances/parenting; 3) education/employment; 4) peer relations; 5) substance abuse; 6) leisure/recreation; 7) personality/behaviour; and 8) attitudes, values, beliefs. The sum of all the items present in the minor provides us with a level of risk of recidivism, both by area and overall.
In order to determine the internal consistency of the inventory, the Cronbach's alpha statistic was used. The Cronbach's alpha for all the items on the Inventory was .87, and therefore it can be concluded that it presents very high internal consistency.Procedure
The data for this study were obtained from the analysis of the records of minors in the Juvenile Court of a Spanish province. The analysis included the number of offences for each minor in a 2-year follow-up, the types of crimes, demographic data, and the risk of recidivism.
As regards the type of crime, a distinction was made between crimes committed against persons and those against property. The following were considered crimes against property: theft either as a crime or an offence, damage to property, fraud, attempted robbery, vehicle robbery/theft, burglary, and burglary with forced entry. Crimes against persons included actual bodily harm, threatening behaviour, reckless driving, robbery with intimidation, perjury, inhuman treatment, resisting arrest, assault, negligent homicide, crimes against public health, and domestic abuse.
The variable criminal recidivism refers to charges filed after the date of the first assessment carried out on the minor by the Youth Offending Team, and which will hereinafter be referred to as the “baseline”. Therefore, each minor has a different baseline, considered as of March 2008. For this study all the charges after this “baseline” have been taken into account. In this regard, recidivists were taken as being those minors who were involved in an offence or crime within the two years following the baseline.Statistical Analyses
To compare the two groups, contingency tables were drawn up and χ2 tests were performed for the categorical variables. In the cases of low frequency in any of the categories, Monte Carlo correction and Fisher's exact test were used. The power of the significance was analysed using Phi and Cramer's V, the intensity of the significance being considered low when it is below .150, medium between .150 and .300, and high when it is above .300. Analysis of variance and Student's t -tests for independent samples were conducted when comparing the means of different variables under study. In these cases, the effect size was calculated in accordance with Cohen (1988) and the confidence interval was 95% in all the analyses.Results Type of Crime and Age
Taking into account the ages of the subjects included in our sample, there are no statistically significant differences depending on whether the crime was committed against property or against persons, with a value of F (1, 39) = 0.119, p = .73. The mean age for crimes against persons is 16.10 and SD = 1.12, and in crimes against property the values are 16.06 and SD = 1.27. Dividing the minors into two age brackets ( Table 1 ) also failed to reveal any significant differences, χ2(1) = 0 .255, p = .61.
Age Brackets and Type of Crime.
|14 and 15 years (34.2%)||16 to 18 years (65.8%)|
|Crimes against persons||49.6% (67)||52.3% (136)|
|Crimes against property||50.4% (68)||47.7% (124)|
|Total||100% (135)||100% (260)|
Regarding the type of crime and gender, statistically significant differences are observed with a value of χ2(1) = 4.22, p = .040, Φ= -.103. On examining only the crimes committed by the group of girls in Table 2 , it can be seen that most of them are crimes against persons (62.2%). In contrast, on analysing the percentages for the group of boys, there is hardly any difference between crimes committed against persons or against property, the percentages being very similar in both cases (48.9% and 51.1%, respectively).
Gender of the Minors and Type of Crime.
|Male (81.3%)||Female (18.7%)|
|Crimes against persons||48.9% (157)||62.2% (46)|
|Crimes against property||51.1% (164)||37.8% (28)|
|Total||100% (321)||100% (74)|
If we turn to look at the relationship between type of crime and nationality of the minors, statistically significant differences are found, with a value of χ2(3) = 13.53, p = .004, Φ= .24. An analysis of the proportion of crimes committed by Spanish minors does not reveal any big differences in terms of crimes against property or those against persons (48.5 and 51.5%, respectively). Nevertheless, most of the crimes committed by the group from Eastern European countries are against property (72.2%), while the Latin Americans and Africans commit more crimes against persons. Biggest differences are found in Eastern European and Latin American nationalities (see Table 3).
Nationality of the Minors and Type of Crime.
|Spanish||Eastern European||Latin American||African|
|Crimes against persons||51.5% (156)||27.8% (10)||71% (22)||60% (15)|
|Crimes against property||48.5% (147)||72.2% (26)||29% (9)||40% (10)|
On analysing the type of crime and recidivism within the 1-year span (re-offends/does not re-offend), no statistically significant differences are observed: χ2(1) = 0.265, p = .607, Φ= .02. As can be seen in Table 4 , proportionally in the group of crimes against persons, there are no big differences between re-offending and non re-of-fending minors (48.3% and 51.9%). The same occurs in the group of crimes committed against property, where there are no differences in terms of whether they were committed by re-offending or non re-offending minors (51.7% and 48.1%).
Type of Crime and Recidivism.
|Crimes against persons||51.9% (174)||48.3% (29)|
|Crimes against property||48.1% (161)||51.7% (31)|
|Total||84.8%/100 (335)||15.2%/100 (60)|
On analysing recidivism according to the number of offences in one year's follow-up, no statistically significant differences are found between the crimes committed against persons and against property, with a value for F(1, 39) = 0.26, p = .608.Type of Crime and Risk Factors for Recidivism
On analysing what risk factors may be associated to what type of crime, statistically significant differences are found in only three areas of the YLS Inventory: parenting, personality/behaviour and attitudes/ values/beliefs. The minors involved in crimes against property present higher risk scores in inconsistent parenting. Minors with crimes against persons show higher risk scores in physical aggressiveness, outbursts of anger, poor frustration tolerance and insensitivity (see Table 5 ). In accordance with the evaluation carried out by Cohen (1988) , the effect size would range between low and medium in all cases, the highest magnitude being the one corresponding to inconsistent parenting ( d = -0.34). The other risk factors in the inventory do not show any significant differences between the two groups of crimes.
Mean Scores According to Type of Crime for Each Risk Factor.
|YLS area||Risk indicators||1Crime||M||SD||t||p||Cohen's d||r|
|Prior and current offences / dispositions||Three or more prior convictions||(1)||.04||.19||-1.63||.104||-0.17||-.08|
|Two or more failures to comply||(1)||.03||.17||-0.649||.517||-0.05||-.02|
|Three o more current convictions||(1)||.05||.22||-0.352||.725||-0.04||-.02|
|Difficulty incontrolling behavior||(1)||.19||.39||0.125||.901||0.02||.01|
|Poor relations (father-youth)||(1)||.09||.28||0.378||.706||0.03||.01|
|Poor relations (mother-youth)||(1)||.05||.27||0.331||.741||0||0|
|Education / Employement||Disruptive classroom behavior||(1)||.37||.48||0.750||.454||0.08||.04|
|Disruptive behavior on school property||(1)||.05||.21||-0.355||.723||-0.04||-.02|
|Problems with peers||(1)||.14||.34||1.533||.121||-0.16||-.08|
|Problems with teachers||(1)||.24||.42||0.797||.426||-0.06||-.03|
|Unemployed/not seeking employment||(1)||.25||.43||-0.317||.751||0.12||.06|
|Peer relations||Some delinquent acquaintances||(1)||.58||.49||1.310||.191||0.10||.05|
|Some delinquent Friends||(1)||.41||.49||1.103||.268||-0.06||-.03|
|No/few positive acquaintances||(1)||.37||.48||-.437||.662||0.02||.01|
|No/few positive Friends||(1)||.39||.49||.282||.778||-0.36||-.18|
|Substance abuse||Occasional drug use||(1)||.30||.45||-.807||.420||-0.69||-.32|
|Chronic drug use||(1)||.18||.38||.143||.887||0.06||.03|
|Chronic alcohol use||(1)||.03||.17||.550||.582||0.03||.01|
|Substance abuse interferes with life||(1)||.09||.29||.358||.721||-0.08||-.04|
|Substance use linked to offence(s)||(1)||.01||.09||-.884||.377||-0.06||-.03|
|Leisure / Recreation||Limited organized activities||(1)||.66||.47||-.794||.428||0.02||.01|
|Could make better use the time||(1)||.62||.48||.125||.901||0.02||.01|
|No personal interests||(1)||.53||.50||.635||.526||0.06||.02|
|Personality / Behaviour||Inflated self-esteem||(1)||.06||.24||1.258||.209||0.09||.04|
|Outbursts of anger||(1)||.28||.45||2.88||.004*||0.29||.14|
|Short attention span||(1)||.10||.30||-0.515||.607||-0.06||-.03|
|Poor frustration tolerance||(1)||.33||.47||2.35||.019*||0.22||.11|
|Inadequate guilt feelings||(1)||.22||.41||0.578||.564||0.04||.02|
|Verbally aggressive / Impudent||(1)||.09||.29||0.926||.355||0.07||.03|
|Attitudes, values and beliefs||Antisocial/procriminal attitudes||(1)||.30||.45||1.016||.310||0.11||.05|
|Not seeking help||(1)||.25||.43||-0.085||.932||0||0|
|Actively rejecting help||(1)||.09||.28||-1.318||.188||-0.13||-.06|
|Callous, little concern for others||(1)||.05||.22||2.49||.013*||0.23||.11|
Note. M = Mean, SD = standard deviation; size effect values: Cohen's d and r. 1Crime = (1) person, (2) property.
* p <.05
The aim of the study was to determine whether the profile of young offenders charged with crimes against property and against persons varies according to different socio-demographic variables (gender, age, nationality) and different juvenile risk indicators. In this sense, some characteristic profiles associated to one type of crime or another were found and are discussed in the following.
Whether the minor is older or younger does not seem to be related to the type of crime committed, regardless of whether the age is taken into account as a continuous or a categorical variable (younger and older age groups). Nevertheless, the variable gender of the minor does appear to be related to the type of crime. Although crimes against property are more frequent among young offenders as a whole ( Fernández et al., 2009; Graña et al., 2006 ), girls commit more crimes against persons, as expected, in line with conclusions from earlier studies ( Capdevila et al., 2005; Iborra et al., 2011 ). This finding could be related with a greater general social orientation among young females than in the case of young males ( Calvo, Gonzalez, & Martorell, 2001; Moreira & Mirón, 2013; Sánchez-Queija, Oliva, & Parra, 2006; Velásquez, Martínez, & Cumsille, 2004 ), the more negative side of which is reflected in a greater number of interpersonal conflicts. Another possible explanation that is compatible with the foregoing could have something to do with the scant perception of self-efficacy young females have when it comes to committing specific crimes, such as robbery or vandalism (both against property) as compared with their male counterparts ( Garrido-Martín, Masip, & Herrero, 2009 ).
The variable nationality of the young offender has been shown to be statistically related with the type of crime. That is to say, whereas in the case of Spanish minors the distribution between crimes against persons and against property is similar, among foreign minors each type of crime is significantly associated with a particular nationality. Thus, most of the crimes committed by minors from Eastern European countries are against property, while among Latin Americans and those from African countries crimes against persons predominate. These results support and extend previous studies in which Latin American minors were more frequently involved in crimes related with persons ( Iborra et al., 2011 ), in the same way as those from Northern African countries ( Capdevila et al., 2005).
In order to understand this situation, the circumstances surrounding these groups must be taken into consideration, since they can act as risk or protection factors. The Latin American minors have been living in Spain for several years and live with their families, so it can be expected that most of their basic needs are already covered. Therefore, we will have to analyse how other variables such as acculturation or whether or not they belong to Hispanic street gangs influence the fact that they commit a greater number of crimes against persons. On the other hand, minors from Eastern European countries, mainly from Romania and of Gypsy extraction, have been used by mafias or adults – who may or may not be their parents ( Lázaro, 2007 ) – in different countries of Europe to commit robberies or theft, which would explain their greater presence in crimes against property.
Another of the variables studied in this work is the higher or lower rate of juvenile recidivism depending on the type of offence. The type of crime does not appear to determine a higher rate of recidivism in young offenders, which supports the results of previous works that did not find any significant differences ( Cuervo & Villanueva, 2013; García-España et al., 2011 ). This ambiguity or contradiction in the results from earlier studies may be due to the wide range of methodologies used. Such differences make it difficult to compare them directly and include different kinds of participants (confined versus low-risk minors), different recidivism follow-up periods, and so forth. What seems clear, however, is that there continues to be a higher rate of recidivism in crimes against property, although the studies in the literature do not contain any solid empirical evidence to confirm the fact.
When it comes to analysing the risk factors of the minor (evaluated by the YLS Inventory) present in one type of offence or the other, significant differences appear. These differences are discussed in the following. On the one hand, minors who commit crimes against persons present more individual risk factors, as it was hypothesised, such as physical aggressiveness, outbursts of anger, and insensitivity towards others, all of which are factors that have been related with psychopathic traits in juvenile delinquents or aspects of personality ( Aguilar & Godoy, 2013; Barudy, 2000; Contreras et al., 2011; DeLisi et al., 2010; Herrero, 2002; Vilariño et al., 2013 ). On the other hand, in crimes against property, the minors that were studied were characterised by presenting a greater degree of inconsistent parenting, a factor that has also been related to delinquency in earlier studies ( Kelly et al., 2005).
In this respect, some authors find similar traits associated to a predominant type of crime. For example, there are impulsive young offenders with a lack of empathy, which leads them to fail to recognise the needs and feelings of others, who find it difficult to establish interpersonal links in time and who commit crimes mostly against persons ( Barudy, 2000 ). Hence, it is possible that the minors who commit crimes against property are rational actors who choose to carry out this action because they perform a classical cost-benefit assessment and assume beforehand that the benefits will be greater than the costs ( Kessler, 2004 ). This internal evaluation could be encouraged by an absence of adequate external regulation, given the inconsistence of the parenting. This rational assessment would not be so present in the case of crimes against persons, which would be of a more “expressive” nature, displaying aggressiveness and anger towards others in a less rational way.
Although the results from this study are relevant for designing prevention and intervention programmes, the following limitations should also be taken into account. First, the data come from a single province and therefore cannot be generalised to the whole Spanish population, despite the fact that some of the results found run in the same direction as those from previous studies. Likewise, in future research it could be interesting to take into account not only the initial crime, but to add the crime committed by the minor on re-offending, since this would allow more accurate predictions to be made about the recidivism of young offenders according to the type of crime and the risk factors. Despite these limitations, this study provides a differential profile of the minors depending on the type of crime. Thus, in crimes against property the young offender is more likely to be male, Spanish or from an Eastern European country, and with inconsistent parenting. In contrast, crimes against persons would be committed predominantly by Spanish, Latin American or Africa female minors with individual risk factors such as aggressiveness or poor frustration tolerance.Conflict of Interest
The authors of this article declare no conflict of interest.Financial Support
We are grateful to Fundación Caixa-Castelló Bancaixa (P1.1B2010-16) and the Spanish Ministry of Education (EDU2010-21791) for funding this study.
Copyright © 2018. Colegio Oficial de Psicólogos de Madrid