Gender Differences in Delinquency and Situational Action Theory: A Partial Test

Abstract

This paper investigates to what extent a recent perspective in criminology, Situational Action Theory, is valid for both males and females and to what extent elements from the theory explain gender differences in delinquency. Data are used from the first (N = 843) and second (N = 616) wave of the Study of Peers, Activities, and Neighborhoods, which included detailed data about core elements of Situational Action Theory (morality, self-control, unsupervised peer activity and peer deviancy), proposed indirect causes (bonds with parents and school, parental monitoring) and self-reported delinquency. Cross-sectional and lagged Tobit regression analyses show that the core relations with delinquency are not significantly different between boys and girls and that the elements of Situational Action Theory are able to explain gender differences in delinquency for a substantial part. However, there is still a remaining independent effect of gender on current and lagged delinquency.

Keywords:

• juvenile delinquency
• Situational Action Theory
• gender differences
• gender gap
• sex differences

Acknowledgments

We want to thank Per-Olof Wikström of Cambridge University for sharing the questionnaire and the space-time budget interview that were developed in the Peterborough Adolescent and Young Adult Development Study (PADS+); Beth Hardie and Caroline Moul for training our interview staff, and Kirsten Grandia, Evelien Hoeben and Lieneke Spel for coordinating the data collection. We also thank various anonymous reviewers for their constructive and detailed comments.

Disclosure statement

No potential conflict of interest was reported by the authors.

Notes

1 Researchers disagree whether one should use the word “gender” or “sex” in reference to differences in delinquency between males and females. Those who use “sex” prefer a neutral term, indicating that the variable itself does not explain anything; those who use “gender” refer to cultural prescriptions that are adhered to being a male or female. We do not have a preference but chose to use “gender” in this paper because this word seems to be most commonly used in studies in this area, and because we include variables on which boys and girls may differ because of cultural prescriptions (e.g. parental monitoring).

2 Apart from studies based on theoretical perspectives, there are also studies that try to explain gender differences in crime, without adhering to a particular theory. Wong (2012) for example, attempted to explain the “gender gap” in crime by analyzing differential exposure and vulnerability to a set of risk and protective factors, and by analyzing potential differential thresholds for delinquency. She found support that girls are differentially exposed to a set of risk factors, but not for differences in vulnerability or thresholds.

3 The perception choice process may have two forms according to Wikström: habitual (when a person routinely either takes the opportunity to commit an offense or refrains from it) or deliberately (when a person actively considers the pros and cons of an offense). When people choose deliberately, their choices depend on the extent to which they are able to exercise self-control, and on the external controls they perceive in a setting (deterrence).

4 It is also possible that girls are less exposed to criminogenic settings because of their lower propensity towards criminal behavior. Because of their higher levels of self-control and morality, girls may be less inclined to spend time in risky settings and/or with risky or delinquent friends.

5 We did not have measures for other important potential causes of the causes that are mentioned in the theory, such as socioeconomic status and family adversity.

6 In total, there were 942 adolescents in the 1st and 4th grades of the participating schools. From this population, 27 adolescents were not approached because the school was too busy at the moment of data collection; 15 did not participate in the study because parents withdrew them from the study; 13 did not show up at the moment of data collection; 6 appeared to have moved to another school; and 3 were ill during the data collection period. From the remaining 878 adolescents, 35 did not complete all research instruments of the study.

7 The reasons why 227 respondents exited the study varied: 33 respondents could not be contacted at all; 147 respondents refused participation; 16 respondents repeatedly did not show up at an appointment; 12 parents of respondents refused permission; and 18 respondents reported other reasons for not participating (e.g. stress or problems at home). One respondent was left out of the sample because the answering pattern revealed strange answering tendencies and internal inconsistencies.

8 In 2008, about 10% of the total Dutch population had one or two parents who were not born in the Netherlands (source: www.statline.cbs.nl). No specific figures were available for the 12/13 and 15/16 year olds however, and the percentage will probably higher among these categories of the population.

9 These figures depart from the total Dutch youth population. The website of the Dutch national statistics office CBS (statline.cbs.nl) provides figures for students of 15/16 year old (for the younger cohort, school levels are often combined and no exact information is available). This reveals that in the SPAN sample students from practice education are very strongly overrepresented (only 3% of the total population follows this type of education), vocational education is slightly underrepresented (about two thirds in the total population), medium level is underrepresented (about a quarter in the total population) but high level education is slightly overrepresented (about 20% in the total population).

10 The 20 offense types are: scratch on objects with paint, pen or spray paint; destroy or damage something; set fire; steal something worth less than five euro from a shop; steal something worth more than five euro from a shop; buy stolen goods; bicycle theft; moped theft; theft from a house; theft from a car; theft from elsewhere; robbery; steal from someone covertly; threaten someone; kick or hit someone on the street; injuring someone by kicking or hitting; sell soft drugs; sell hard drugs; carry a weapon; use a weapon.

11 In the SPAN study, we chose to use a frequency scale, in line with previous studies on Situational Action Theory. An alternative scale for delinquency is a variety scale, consisting of the number of different offense types. It appears that in our data, the frequency and variety scale are highly correlated (r = .923 at T1 and r = .913 at T2), and analyses using a variety scale results in similar findings as those that are presented here.

12 Some previous studies of Situational Action Theory employed more extensive scales for morality that also included items about drug use and non-criminal forms of rule breaking (e.g. Wikström et al., 2012). The scale that is used in the current study is a subscale explicitly focused on the breaking of moral rules that are laid down in the law. It was constructed on the basis of factor analysis on all items of the more extensive measure.

13 A different strategy was followed in the studies by Pauwels (2012a, 2012b). These studies focused on the interaction effects, employed OLS regression models but also did robustness checks by investigating nonlinear effects and negative binomial models. These studies reported that the interaction terms as predicted by the theory were related to offending among boys as well as girls.

14 This relates to the score on the delinquency frequency scale. Because the delinquency frequency score is based on ordinal item responses, the difference in the actual number of offenses committed by boys and girls is probably higher.

15 Additional analyses indicate that also the prevalence of any offending is significantly higher among boys (80% in the first wave and 73% in the second) than among girls (62% in the first wave, 56% in the second). With regard to offense types, it appears that boys and girls differ most apparently in their involvement in more serious and violent offenses. For example, in the first wave, 5% of the boys reported a burglary, compared to only .3% of the girls. Similarly, 10% of the boys used a weapon compared to 3% of the girls. The differences between boys and girls are much smaller for minor property crimes, and even non-significant for minor shoplifting and graffiti.

16 Translated to actual values on the response categories, the difference in parental monitoring between boys and girls is .5 per item. This actual value difference per item is less impressive for bond with parents (.06) and bond with school (.12).

17 Translated to actual values on the response categories, the difference between boys and girls in morality is most impressive: .68 per item. For self-control, this actual value difference per item is .27; for unsupervised activities with peers .16; and for rule-breaking peers .22.

Additional information

Notes on contributors

Frank M. Weerman

Frank M. Weerman is a senior researcher at the Netherlands Institute for the Study of Crime and Law Enforcement (NSCR). His publications have focused on the explanation of juvenile delinquency, on co-offending and youth gangs, and on the role of delinquent peers and peer-related activities.

Wim Bernasco

Wim Bernasco is a senior researcher at the Netherlands Institute for the Study of Crime and Law Enforcement (NSCR) and a professor in the Department of Spatial Economics at VU University Amsterdam. His research interests focus on spatial aspects of crime, including geographic variations in crime and delinquency, offender travel behavior, and target selection. More recent work addresses situational causes of offending and victimization.

Gerben J.N. Bruinsma

Gerben J.N. Bruinsma was the director of the Netherlands Institute for the Study of Crime and Law Enforcement (NSCR), a national research institute of the Netherlands Organization for Scientific Research (NWO), and he is professor of environmental criminology at VU University Amsterdam. His research interests are social disorganization, crime places, history of criminology, and criminological theory.

Lieven J.R. Pauwels

Lieven J.R. Pauwels is a professor of criminology in the Department of Penal Law and Criminology at Ghent University (Belgium). His research interests focus on neighborhood effects and urban crime, empirical tests of integrated theories of offending and victimization, and survey measurement issues. Recent work has focused on the relationship between lifestyles and violent youth group involvement. Correspondence to: Frank M. Weerman, Netherlands Institute for the Study of Crime and Law Enforcement (NSCR), P.O. Box 71304, 1008 BH Amsterdam, The Netherlands. E-mail: [email protected]