1 - Problem behaviour and prevention
by Deklerck Johan, Leuven Institute of Criminology (LINC)

Theme : International Journal on Violence and School, n°10, December 2009

The description of problem behaviour in school education in Flanders and the integral approach to prevention of it are the central themes of this article. After having situated the in Flanders dominant criminological approach of field research and the commonly used concepts and notions of problem behaviour, we make a review of existing empirical quantitative studies in four steps. First we go deeper into (a) figures on ‘general well-being’ within the school system, to make in a next step (b) an analysis of official figures of juvenile delinquency in general and than to concentrate on (c) self-report studies of juvenile delinquency in general and (d) within school education.

Keywords : Violence, delinquency, figures, statistics, school education, Flanders – Belgium, integral prevention, prevention pyramid.
PDF file here.
Click on the title to see the text.

Belgium is a federalized country with a complex political structure . It is divided into three regions for territorial matters as economy, urbanization, employment policies for example (Flanders, Brussels and Wallonia) and three communities for matters relating to the people as culture, youth, education (Flemish Community, the French Community and the German Community). In Flanders, the competencies of the Flemish Region and the Flemish Community are merged in the Flemish government. In contrast with other federal states, regional governments in Belgium have full power concerning their competencies, meaning that these legislations exist at the same hierarchical level as federal laws. The competency over educational matters is a community-based competency . This has resulted in the development of three different educational systems. Despite a lot of similarities between these systems (parents’ “free choice” of school, coexistence of several “educational networks” …), they have been evolving towards bigger differentiation as a result of different policies. Equally, we can speak of different scientific cultures in both areas – Flanders being more influenced by Anglo-Saxon intellectual traditions whereas French-speaking Wallonia has been more influenced by the French debate and research perspectives.
These political and cultural differences affect the way problem behaviour and violence within the school environment is dealt with, resulting in different notions, priorities, and prevention and treatment programs developed within these systems.
In summary, because the name “Belgium” covers two co-existing but very different political, educational and scientific realities, we will present two distinct contributions, dealing respectively with Flanders and with the French Community.

The description of problem behaviour in school education in Flanders and the integral approach to prevention of it are the central themes of this article. After having situated the in Flanders dominant criminological approach of field research and the commonly used concepts and notions of problem behaviour, we make a review of existing empirical quantitative studies in four steps. First we go deeper into (a) figures on ‘general well-being’ within the school system, to make in a next step (b) an analysis of official figures of juvenile delinquency in general and than to concentrate on (c) self-report studies of juvenile delinquency in general and (d) within school education.
This review brings us to a next step. We describe the importance of an integral and positive approach through the description of the prevention pyramid, a model for an integral prevention that is widely spread in the field of prevention in Flanders. In this way, this article wants to give an image of Flanders, as a particular region with an own identity in this field and with an own tradition of a positive and integral prevention. To conclude we make some recommandations for further research.

Research of youth’s problematic behaviour in general, and within the school context in particular, has a rather long tradition in Flanders. This tradition is strongly connected to the long history of criminological research in Belgium, in contrast to most other countries in Europe (e.g. the Catholic University of Leuven started with a criminology program in 1928). The three main Flemish universities of Leuven (, Ghent ( and Brussels (, offer a full-fledged criminology program of four years. This study includes a combination of law, psychology and sociology. Different aspects of these scientific disciplines are integrated in order to specifically answer to criminological questions, whereas in other countries criminological questions are subjected to a specialization within another discipline. This means that in the field of youth delinquency, purely sociological or psychological studies only play a marginal role in this country.
We can distinguish three main fields in the criminological research on juvenile delinquency in general. The first is concentrating on mapping problem behaviour and youth delinquency. This is the main subject of this contribution. Below we will describe some figures and give an overview of quantitative research. As in many other countries, except for some smaller local research projects, figures providing information of problem behaviour strictly within the school context, are not available. The figures are about juveniles and their behaviour also in their free time, on the street, at home, etc.. As youngsters mainly spend their time within schools, general figures will also cover what happens within this educational context.
Next to this, different research projects of another type have been conducted in the last years. It concerns evaluation research of programs on bullying, violence, truancy, etc.. These programs are often financed by the Flemish or Federal Government. The results mainly lead to the conclusion of the need for an integral and positive approach. ‘Integral’ means not just attacking the problem child or the symptoms, but taking the larger (school) context into account. ‘Positive’ can be understood as supporting and stimulating a general positive climate in schools as an important part of the answer to problem behaviour.
The last part of research can be classified as action research with the aim to develop answers to this problem behaviour, starting from theoretical and empirical findings (Deklerck, J. & Depuydt, A. & Deboutte, 2001; Depuydt & Deklerck, 2005; Vettenburg, e.a. 2003; Most of this research is realized in an often strong interaction with the field. An example of a result here is the below described model of the prevention pyramid.
In Flemish quantifying criminological research, three main resources are being developed and used. The first resource consists of the official delinquency figures, as registered by the police. Since these figures mainly contain the product of police activities, and therefore do not provide an integral overview of youth delinquency, general self-report studies were developed to complete this picture. For Flanders, two self report studies are to be mentioned. The most recent one is the survey of the Youth Research Platform (JOP or Jeugdonderzoeksplatform) (Vettenburg, Elchardus & Walgrave, 2007; ). In this study, which was accomplished in 2005, 2503 youngsters from 14 to 25 years old were questioned about a variety of aspects of their lives, including involvement in delinquency. In an older (1998-1999) survey-based project called ‘Jongeren in Vlaanderen. Gemeten en geteld’ (JVGG) (De Witte, Hooge, Walgrave; 2000), information on delinquent behaviour of youngsters between 12 and 18 years old was gathered. This research is not limited to the school environment but is oriented towards a larger field, with only some information on what is happening in schools.
Both research projects, Youth Research Platform (Vettenburg et al., 2007) and the research project ‘Jongeren in Vlaanderen. Gemeten en geteld’ (2000) in which general data about youth delinquency were gathered, were recently developed under the impulse of new cooperation between different university departments. This, as a reaction to the existing disparity in Flanders in the field between different local projects with a lack of integrated approaches and problems in the development and validation of fundamental theories (Schillemans, 2004; Van Nuffel, 2003).
Talking about problem behaviour, different notions are used. In a strong connection with the Flemish language tradition they have a particular cultural background and cannot always easily be translated.
First of all, a distinction is made between internalized problem behaviour on the one hand, and externalized problem behaviour on the other. Where the former refers to problems like depression, suicidal ideation, anorexia, etcetera, the latter is used to describe behaviour like bullying, aggression, vandalism, violence, etcetera. Girls usually tend to express feelings of unease and a low self esteem through internalized problem behaviour, whereas their male counterparts in general tend to display externalized problem behaviour (Leer-kracht. Veer-kracht, 2006: 19-20). When talking about externalized problem behaviour different concepts are used . Just like in other countries and regions, Flanders has its own terminology when talking about violence and juvenile delinquency in the practice of education, as well as in the field of scientific research.
“Geweld” or “violence” is a commonly used notion in scientific research. Although everyone “understands” the term, it is interpreted in various ways, from very narrow to very broad. Many definitions are circulating and often the notion of “violence” is mixed up with the concept of “aggression” (Patfoort, 1995; Van Welzenis, 2002; Smith, 2003). Both terms are popular in the practice of education. Another notion that is commonly used is the concept of “antisociaal gedrag” or “antisocial behaviour”’. An often used definition of this kind of behaviour is proposed by Vettenburg, who makes a distinction between four categories: physical violence or threats against people (such as wounding, assault and bullying), other forms of antisocial behaviour (such as theft, vandalism, drug-taking), status-related offences (such as playing truant, going to cafes and dance halls, running away) and deviant school-related behaviour (deliberately being disruptive, arriving late, smoking in school, etcetera) (Vettenburg, 1999:33).
Another commonly used concept is “problematic behaviour” (Peeters, 1995; Deboutte, 2006). It is a very broad concept for all kinds of behaviour that causes problems in social interaction. It covers both interiorised and exteriorised behaviour, also outside the criminological field. It is therefore not easy to make a clear definition of this concept, but the advantage exists in its “interactive”, less stigmatizing quality. In this way, the offender is not pushed into a marginal position, but on the contrary receives possibilities for communication. This concept is preferred to that of “deviancy” of “incivilities”, because of its lower normative character. These last notions are more or less absent in the Flemish educational field.
On a scientific level, they are experienced as very normative and therefore problematic. The frame of reference, e.g. the law, the school regulations, or the normative consensus of what “normal behaviour” should be, and of which one is “deviating”, is very often not questioned, although many consequences can occur at this level (Depuydt, 1998; Deklerck & Depuydt, 2005). As a consequence, “respectless behaviour” is also often used within the educational field. It has been launched by A. Depuydt in her research on the implementation of the approach of “verbondenheid”, translated as “linkedness” (in English) or “reliance” (en Français) as a fundamental prevention of delinquency. Delinquency can be seen e.g. as what is described in the penal law codex, or as a general social appreciation or the norms of society. These norms fluctuate in space and time, for instance physical punishment is not practiced anymore, or in many countries smoking in public spaces is not allowed any longer. A. Depuydt started from a fundamental, existential, ethical approach, with a more absolute character than the permanently evolving rules, laws and regulations, or the continuously changing normativity of society. (Deklerck & Depuydt, 2005: IV, 83-85). Furthermore, this is a more fundamental ethical approach as “experiencing harm or damage”, which has an existential meaning (Deklerck, Depuydt & Deboutte, 2001). The positive element of this concept consists of its possibilities for a more fundamental ethical debate.
A specific notion that has raised a lot of attention is the concept of “pesten” or “bullying”. This term is very widespread in the context of school education as a daily reality in the interaction between pupils and in some cases also teachers or other personnel. It refers to a specific form of problematic behaviour (Deboutte, 1995; Deboutte, 2004; Leonard, 2006; Rigby, Smith & Pepler, 2004) with an own reality, that is recognized by scientists, schools, the policy makers and in different prevention programmes.
At an international level we often forget that concepts are always interpreted in a cultural context. Some countries still accept parents to give their child a light smack to teach them which behaviour is wrong. Within their culture this is not considered a violation of social norms or laws and consequently will not be interpreted as violence, whereas for other countries with a zero tolerance towards this phenomenon, it will be seen as unacceptable and violent. So, even with clear definitions the contextualisation against the local cultural and social background needs to be considered (Smith (Ed.), 2003; Ortega, Sanchez, Van Wassenhoven, Deboutte & Deklerck, 2006).

After having situated the specific background of criminological research tradition and the culturally rooted concepts of problem behaviour, now we go deeper into a review of recent empirical studies. We make four main steps. First, we give an overview of Flemish figures. We start with general figures of ‘well-being’ in school, based on self-report studies and official registrations by the Ministry of Education. Next to this, we go deeper into official registrations by the police, as well as the results of self-report surveys (JOP and JVGG) that specifically apply to youth delinquency in Flanders.
In a last step we go deeper into school settings. As for antisocial behaviour specifically conducted in the school setting, the results of a research project by Vettenburg & Huybregts (2001), and to a lesser extent, the results of self-report studies of Stevens & Van Oost (1994; 1995), the Central Board for Study and Career guidance (Vandersmissen & Thys, 1993) and the Flemish Institute for Scientific and Technological Assessment (Vandenbosch, Van Cleemput, et al., 2006), will be discussed.
The earliest study of antisocial behaviour in school settings, based on a self-report study of 1689 youth, was accomplished in 1988 (Vettenburg, 1988; see also Goris & Walgrave, 2002). However, since this is an older project, these results will not be reviewed in this contribution. Finally, some prevention models for antisocial behaviour in schools and recommendations for future research, will be given.
In Belgium, education is compulsory until the age of eighteen years. Although the majority of the pupils feels rather well in their school, there is only a minority that does not feel satisfied (Mertens & Van Damme, 2000; Huysmans, 2006). The PISA-research (OEDC, 2003) indicates that 33% of the Belgian pupils does not feel at home at school. Stevens and Elchardus (2001) mention that 11% of the pupils, aged 16-18 years, are in general feeling unhappy at school. 14% would like to change schools and at least 18% would prefer to leave school. According to the research of Mertens & Van Damme (2000), this is rather the case for technical and vocational orientations, although these findings are not confirmed in other studies. According to JOP-monitor 1, conducted in 2005 (Vettenburg et al., 2007), secondary pupils are generally satisfied with their education and the school they go to. Girls are more satisfied than boys, and no differences in school satisfaction exist between the three educational tracks (general, technical of vocational secondary education). Pupils repeating a year were less satisfied compared to the other pupils.
Some studies (De Groof, 2003; Smits, 2004; Vettenburg et al, 2007) have questioned the way secondary pupils perceive the relationship with their teachers. These results also suggest that Flemish pupils are satisfied with their teachers and how they are treated.

Despite the conclusion that pupils generally like going to school and perceive the relationship with their teachers as satisfactory, problem behaviour is not absent in Flemish secondary schools. It is generally accepted that chronic problem behaviour at school expresses itself in a combination of learning problems, behavioural problems, school fatigue and antisocial behaviour (Ghesquière & Grietens, 2006; Rombouts, 2000). When looking of the official registrations of ‘truancy’ as an extreme symptom of school fatigue and problem behaviour, there seems to be a small, but significant group of pupils showing this kind of behaviour.

The highest percentage is identified in part-time education (28.23%). In the full-time educational system, 3.45% of pupils in the vocational educational track are identified as truants, while the mean percentage of the full-time system only reaches 0.57% . These numbers suggest that there are some important differences between the different educational tracks. They also seem to suggest that this kind of behaviour is on the rise; an alternative explanation is that schools have become more aware of the problem, resulting in more adequate registrations. It thus seems that a general feeling of satisfaction goes with a small, but significant group of pupils not satisfied with school and showing problem behaviour. In the next part of this paper will be focused on the concepts and notions that are used in the study and description of this behaviour in Flanders.
The reflection about concepts and definitions leads us to the question about the prevalence of this kind of behaviour in the Flemish educational system. Although this may seem to be a simple question of registration and interpretation, answering this question is not that easy. In this case, the level of registration is an important element: are the numbers the result of official records, self-report or victim surveys? Are these official records a consequence of police intervention? Is it about suspected persons or convicted offenders at the court level?

This figure demonstrates that the crime that is known by the official authorities provides by no means a real and reliable reflection of the total amount of criminal activity (see e.g. Linch & Addington, 2006). A lot of considerations can determine the non-reporting to the police . This is especially true regarding the so called “petty crime”. As this is the case with crime in general, it also holds for school delinquency. Violence, vandalism, verbal aggression, most cases of bullying, etcetera can be solved within the school context, in a formal or informal matter, e.g. by the intervention of the class teachers, the principal, or other responsible personnel, through peer mediation or group conferencing, by a cooperation between school and parents. For different reasons, e.g. the image of the school, to give the offender a ‘new’ or ‘last’ chance, etcetera, schools can decide not to report to the police. Self report studies and victim surveys (Debarbieux, 2004) can offer a solution to this dark number. Although for very serious cases as murder or rape self report will not be possible as offenders are not willing to report this, it is a very useful instrument for daily problem behaviour and petty delinquency. Comparison between self report studies is not always evident, because methodological differences (e.g. sample size, age group, by mail, classical or face-to-face) can influence the results (Eliaerts, 2007; Van Kerckvoorde, 1995; Vettenburg, Elchardus & Walgrave, 2007) and present very different pictures of the same social reality . Comparing figures between different surveys and between countries should take these basic elements into account. Considering these suggestions, it remains important to explore the problems in the field of violence and delinquency within the educational system. Gathering data is therefore very useful and important. Together with the data, enough information about the methodology, limits and possibilities of these data should be added. In Belgium only few statistics that are able to present a general overview of delinquency, bullying and problem behaviour in school are available. However, figures on juvenile delinquency in general can be given.
A recent study (Goedseels, Detry & Vanneste, 2007) of the number of criminal acts with a minor suspect being brought to the public prosecutor in 2005 revealed that at a nation-wide level in sum 45 722 cases were declared, of which 38 747 were cases declared with suspected minors. This implies that 55 minors on 1000 minors between 12 and 18 years old have committed a criminal act that was brought to the prosecutor’s level. Turned around this suggests that 95% of all minors have not been involved. Looking at the prevalence of the most serious forms of crime, these numbers state that only 1 on 100 000 minors was accused of homicide or manslaughter, 3 on 10 000 of robbery and only 1 on 10 000 of armed robbery.

These tables show the prevalence of criminal acts committed by minors that were discovered and brought to the public prosecutor in 2005. When the variable ‘gender’ is taken into account it could be easily stated that boys commit much more crimes than their female counterparts. Generally speaking, only 20% of the different forms of criminal acts are committed by girls. When the variable ‘gender’ is integrated, the study suggests that an increase of the prevalence of committing delinquent behaviour starts occurring at the age of 12, with the onset of puberty. This prevalence increases until the age of 16-17, but starts diminishing afterwards. Most juveniles gradually stop offending after the age of 17-18, a finding which is firmly grounded in numerous self report studies. Another important limitation of these official numbers is that they do not allow determining if these criminal acts are committed in the school context.
The figures of two self report studies, the first survey of the Youth Research Platform (JOP) and the survey that was accomplished for the research ‘Jongeren in Vlaanderen. Gemeten en geteld’ will be presented here. Because of methodological differences of both surveys, we should be careful in comparing these figures, which you find both in the diagram below.

These figures show that “severe” youth delinquency is a rather marginal phenomenon. Fare dodging (25.5 % according to JVGG, 32.7% according to JOP), theft (23.4% according to JVGG, 15.5% according to JOP) and vandalism (20.7% according to JVGG and 14.7% according to JOP) are being committed the most. Violence (12.6% according to JVGG, 6.1% according to JOP) and carrying a weapon (12.7% according to JVGG and 4.0% according to JOP) are rather marginal phenomena. The large difference between JVGG and JOP considering the rates of violence and carrying a weapon can be explained by a difference in formulation of the questions. In JVGG was asked if the pupils had been involved in a fight. In JOP was asked if they had beaten up someone so badly that this person was injured. If the criterion of causing injury in the JOP-research would have been left out, the violence rates in this database might have been higher. Furthermore, in JOP, the question about carrying weapons very distinctly excluded the possession of a pocked knife. If pocket knifes were included in the question, the rates of weapon carrying might have been higher.
Except for truancy, these figures do not apply specifically for the school setting. They could have taken place outside schools, during leisure time, on the road, at home, etcetera. As well, the information about violence is limited, since no details on the kind of violence are provided.

This diagram shows that boys are generally more involved in delinquency than girls (a difference that is significant, ?=0.01). However, gender differences vary according to the type of delinquency. Again, this gender aspect is not to be reduced to the school environment. These figures give an overall view of the prevalence of delinquency in Flanders. They particularly show that the “problem” of youth deviance should not be overestimated. However, as noted earlier, the way in which the questions were formulated in the questionnaires has an influence on the results. We should thus be careful to make general conclusions about crime rates and developments in Flanders. Finally, we can not generalize these figures to the French speaking part of Belgium.
In 2001, Vettenburg & Huybregts reported the results of their research of antisocial behaviour in schools, based on self-report of 4829 pupils from 95 different Flemish schools. The following data are based on their report.

Most antisocial acts are reported by 10 to 15 % of the sample, except for vandalism, which seems to occur less often (Vettenburg & Huybregts, 2001:127). This table also shows that boys report more antisocial behaviour in school than girls, a difference that is significant (?=0.01). The sex difference is specifically high for vandalism at school, boys do this three times more often than girls (Vettenburg & Huybregts, 2001:128).

This table shows that antisocial behaviour in school also differs by age. Most antisocial acts are being committed until the age of 16. From the age of 17, the prevalence of all antisocial acts (except truancy) declines (Vettenburg & Huybregts, 2001:128).
Except for sex and age, Vettenburg and Huybregts (2001) found in their research that the most important predictors of antisocial behaviour in schools are school fatigue, earlier problems in school, problematic behaviour outside of the school, a tolerant attitude towards violence, the bond with parents (a strong bond with prosocial parents prevents antisocial behaviour in school, a strong bond with antisocial parents encourages it), the bond with peers (a strong bond with prosocial peers prevents antisocial behaviour in school, a strong bond with antisocial peers prevents it) and the connection between the school and the student (Hay, I., & Ashman, A. F., 2003; Hirschi, 1969).
Other, older, self-report studies focus specifically on the prevalence of bullying in Flemish schools. In 1993, the Central Board for Study and Career guidance questioned 1054 pupils from 12 tot 18 years old (high school students). The results of this study show that 18 per cent of high school students are victim to bullies, of which 2.6 per cent said to be bullied very badly (Vandersmissen & Thys, 1993). In 1994, Stevens & Van Oost questioned 10 000 students between 10 and 16 years old about bullying and being bullied. In primary schools, 15.9 per cent of the pupils reported to bully others regularly, 23 per cent reported to have been the victim of bullying. In secondary schools, these statistics are respectively 12.3 per cent and 15.2 per cent (Stevens & Van Oost, 1994, 1995). In accordance with the research of Vettenburg & Huybregts (2001), both these studies found a decrease in bullying behaviour with age, and boys bullying more often than girls.
Finally, in 2006, the Flemish institute for Scientific and Technological Assessment (viTWA) questioned 636 primary school children and 1416 high school students about cyber bullying. This questionnaire also included a couple of questions about “traditional” bullying, which showed much higher rates of bullying than the older studies. 49.3 % admits they have bullied others over the last three months, 78.6% report having witnessed bullying. 10.4% reports having been bullied over the last three months. A tiny minority said they have been both a victim and a bully (Vandenbosch, Van Cleemput et al., 2006).

The review above brings us to a next step. In Flanders there is a growing tradition of an integral and positive approach (see a.o. to tackle problem behaviour. Policymakers and schools, as well as many organisations working in the field of school education, are aware that problem behaviour is a complex phenomenon, and that not only life course events and the social context of the children or youngsters (Harter, 1999) but that also the context of the school plays an important role (Debarbieux, 2006). Punitive schools can e.g. provoke or strenghten problem behaviour, whereas democratic, pupil oriented schools can realize the opposite. To map this contextual approach, the prevention pyramid, a widely spread model in Flanders, is an instrument for analysis and policy making. It will be presented below.
Interaction between research, policy makers, school education and the criminological field in general, has lead to the largely accepted opinion that a simple punitive reaction to “the problem child” is not the best answer. A larger perspective is necessary. In order to reduce deviant behaviour in school, symptoms of the general functioning of the system, such as the societal pressure on adolescents; educational models and methods; the quality of communication in, and the democratic functioning of, a school; pedagogical styles of the teachers, etcetera, should be addressed as a way of a positive prevention (Deklerck, 1996; Galloway & Roland, 2004). With this term is meant that, instead of reacting defensively against problematic behaviour with, for example, punitive measures, (De Cauter, 1990; Munthe, Solli, Ytre-Arne & Roland (Eds.), 2004; Cornell, 2006) problematic behaviour should be prevented by creating challenging opportunities for youth.
Against this background, the “prevention pyramid” was developed (Deklerck, 1996; Deklerck, 2006). This preventive model answers to the actual complex variety of prevention measures, going from “repressive” and “punitive” to “oriented towards general well-being and social support”. Diverse prevention measures to problem behaviour in general, and violence in particular, are therefore included in this model, which is used in Flanders as an assessment instrument for schools (see e.g. the province of Limburg (Dienst Welzijn en Gezondheid van de Provincie Limburg, 2006). The model puts focus on the preventive quality of general well-being and organizes preventive measures into four levels. This categorization is based on the degree of problem orientation: How strongly does the problem manifest itself and how clearly do the measures highlight the problem? The levels are complementary, they all contribute more or less to a positive social climate and they are all necessary for an integrated prevention policy.

The prevention pyramid consists of five levels. These are numbered 0 to 4. Four of them are to be considered as “action levels”, in which a number of initiatives are developed and accomplished. The zero level refers to the general political, ecological, sociological and economical context that is influencing the different action levels and the policy people and schools can work out. It is not an action level. The top four levels (levels 1 to 4) provide a general overview of all possible crime prevention measures. It here concerns the level of curative measures (level 4), the level of specific prevention measures (level 3), the level of the general (level 2) and the level of fundamental prevention (level 1). The levels differ from each other in their degree of problem orientation. The higher we climb the greater becomes this problem orientation. This problem orientation constitutes the basic criterion for categorising crime prevention measures into different categories, and the point of departure for organising these measures.
Level 1: fundamental crime prevention
Level 1 represents ‘fundamental crime prevention’ or a policy that promotes the general quality of life. It is often forgotten in discussions of crime prevention. Nevertheless, it is basic to achieving a good result from any other prevention measure. It in general concerns areas such as social policy, emancipation policy or welfare work in the broad sense of the word, good school organisation, possibilities for recreation and leisure, ombudsman services, minimum wage, measures to combat environmental pollution and degeneration, medical care, procedures political involvement, the protection of nature, social subsidy policy, providing sufficient recreational zones, and others. The links with the problem of crime are more and more indirect. Measures at this level are oriented positively. Their purpose is an improvement in the basic climate of society. At this level of crime prevention we depart from a direct causal link between measures proposed and criminal facts. Applied to school education, the first level consists of the improvement of the general climate within schools. This level is oriented towards the improvement of the general well-being of everyone in the school and the democratic functioning of the school. These measures, that are usually part of the basic policy of the school, have an important preventive effect although they can be considered as general school politics.
The point of departure for level 1 is thus not a formulation of any problem, nor does it provide a problem-oriented answer. We can view it as an offensive, positive strategy in the sense of creating a pleasant, peaceful environment, good working conditions, smooth communication and decision making. It is a framework in which the higher levels can better thrive. In the example of vandalism in the bicycle shed in a school, it means lessening the need for aggression because people feel better, and less need for formal control due to less aggression. The focus is thus no longer exclusively or primarily directed to problem groups (e.g. hooligans, drug addicts) or problem situations (e.g. football stadiums before or after the match, nightlife districts), but to the societal framework as bearer of problem areas. Measures at this level have an indirect preventative effect. They, however, influence directly the general quality of life of the society and in this way provide a basis for more directed crime prevention policy.
Level 2: general crime prevention
At this level of general crime prevention we think for example of limiting violence and aggression in the media, projects around after-school reception for high-risk groups, organising interesting leisure activities or creating special work opportunities for high-risk youth, youth work in underprivileged neighbourhoods, projects around neighbourhood redevelopment, around poverty, community work, training in non-violent solutions to conflicts, limiting the speed of motor vehicles, fan-coaching hooligans, payment with credit cards, decreasing the production of materials that damage the environment, the sale of low alcohol beer or forbidding glass and sharp objects at large gatherings, re-directing the energy that currently goes into the cocaine harvest in Latin America, relaxation training for addicts to tranquillisers (yoga, relaxation, autogenous training), etc.
In level 2, when a problem occurs, prevention measures are taken that are oriented towards social well-being, such as amelioration of the infrastructure, non violent communication, gender and interculturality, etc.. The answer is positive and reduces the problem orientation. The problem orientation at level 2 (general crime prevention) is thus clearly different than level three (specific crime prevention) and level 4 (curative measures). Level 2 has as point of departure the problem but it formulates non-problem oriented answers that indeed have a preventative effect. The same example of vandalism in a bicycle shed (problem) then receives an answer in the form of a closer location, in public view, better lighting and better maintenance (non-problem oriented answers with a preventative effect). In this way it becomes more difficult to steal or vandalise bicycles (preventative effect), and the bicycle shed at the same becomes a more pleasant location for the users. The risk of crime is no longer accented and thus feelings of fear and insecurity disappear or lessen. Measures at this level have a directly preventative effect, yet this is not their only or primary purpose. In addition to pursuing goals of prevention (especially in the sphere of primary prevention), we can also note some more societal goals, dependant upon their position in the helping, emancipating sphere. Some developments in the financial-economic sphere (such as payment with credit cards) also have a preventative effect and as such may also be promoted within a policy of crime prevention. The example of "improved coordination of public transport with pub closing times" also illustrates this. This also has a preventative value and moreover it represents a service to the public. Although problem analysis remains the point of departure, answers are sought that are not directly related to the problem and thus do not contain the risk of highlighting, confirming and reinforcing the problem as such and certain groups, zones and situations. The point of departure is that the constructive intervention into, or the dealing with, environmental factors creates a climate in which criminal behaviour becomes impossible, is less able or unable to thrive. Measures on level 1 and 2 can be the same, but the point of departure is different. Level 2 starts from the analysis of a preventative need, and level 1 doesn’t.
Level 3: specific crime prevention
Level 3 (specific crime prevention measures) concerns measures directly intended to prevent specific crimes. The analysis as well as the answers formulated remain problem oriented. It concerns a direct approach to crime. These measures can be technical-preventative in nature (e.g. electronic surveillance), targeted intervention in the infra(-structure) (e.g. safes), or of an organisational nature (e.g. drawing up a safety plan).
Within this level fall various measures such as information concerning drugs, camera surveillance of high-risk zones, police patrols, security audits in firms, alarm systems, campaigns concerning theft, police surveillance and other measures that increase the chance of getting caught, locks, burglary protection, security agents, alcohol checks for forbidden weapons or drug smuggling, targeted sensitisation of high-risk groups, access management and control, security services, guarding money transports, projects around burglary prevention, around vandalism or sexual aggression, or providing problem-directed information to educational institutions by the police concerning drugs or weekend traffic accidents after the use of alcohol. In crime prevention, this group of measures spontaneously receives the most attention. Intervention at this level deals directly with a specific problem that one wishes to prevent or limit. The point of departure is a certain problem. The intent is to tackle this problem. It here concerns a countermove. These measures are thus also problem oriented, both in their analysis and their approach. If they succeed, they have a specific and directly preventative effect. And this is also the primary purpose of such intervention.
The directly preventative level 3 has the same problem orientation as level 4. Its point of departure is the problem and in addition it formulates problem-oriented answers. The example of vandalism at a bicycle shed (problem) leads e.g. to the answer to this in the form of using locks and cameras. The problem analysis (bicycle theft) thus here receives a problem-directed answer. The presence of locks and cameras does not cause users to forget the problem, but rather emphasises it. They are always reminded of the risk attached to this place. This can place a burden on the social climate and reinforce feelings of insecurity.
Level 4: curative measures
Level 4 (curative measures) is the most specific. It is the last (top) step in the crime prevention pyramid. It also forms the last series of measures that always are situated on the border between the preventative and the curative. It concerns repairing and limiting damage once something bad has happened and learning from it for the future. With respect to drug related crimes it concerns arresting and sentencing the makers of synthetic drugs, the drug dealers, cleaning up the drug network and possibly the detention and detoxification (ambulant and in centres) of the addicts. This level also includes the registration, reporting, investigation and prosecution of crimes. In addition, here are located reparation and restitution by financial and other means to the victim such as in forms of victim-offender mediation. Therapy and guidance for the offender are also here (either within or outside the prison setting). Also help for the victim needs to be offered here. Local and state police and sometimes also the office of the public prosecutor have an important role here. Yet many other institutions or persons can play a smaller role such as the initial reception of or discussion with the victim, support in reporting to the police, looking into the possibilities for the intervention of insurance.
Quick intervention in acute situations (e.g. arresting a hooligan, rounding up a band of thieves) can prevent the situation from becoming worse and thus have a preventative function. Detaining offenders can prevent a relapse. Dealing with a victim’s fear or therapy for the offenders, the restoration of societal relations and networks through forms of restorative law, etc. can have preventative significance. The degree of problem orientation is the greatest at this curative level because it is here that the problem takes place. The analysis of the situation and the approach are problem oriented. At this level we must search for a good solution and the prevention of a worsening by quick and proper action.
Level 0: the societal context
This level should be seen as the context. Within the pyramid it is not seen as an action level. This is the wider societal framework, society in its different dimensions, the societal-organisational dimension and ordinary life. The higher levels are firmly anchored in the wider societal context. Crime prevention policy (implemented at levels 1 to 4) cannot be seen apart from the societal, social and ecological context. This ‘level 0’ is the bedding. With the means and limitations that arise from this context one must initiate preventative campaigns, develop a crime prevention policy. Prevention policy occurs within this broader societal context and thus can never be seen apart from this. The political and economic policy implemented, the presence of natural and cultural riches, the situation in this world, current conceptions of morals and ethics concerning what is and is not of value, the degree of institutionalisation influence the criminal and crime prevention policy (and vice versa). This context conditions the possibilities of implementing a crime prevention policy in a positive or a negative way. The collective will to collaboration on the part of the citizen, the subsidy policy on the part of authorities, technological developments, etc. are examples here. The appeal to crime prevention can be an important signal for the quality of this societal environment. Crime prevention measures at the higher levels will not be effective if serious problems are present at this level. Policy makers, in other words, should implement a policy that supports the targeted crime prevention measures. And likewise crime prevention measures must be implemented that have a connection within this societal context and which influence this favourably.
Thus problem orientation forms the central criterion for categorising crime prevention measures.
This can be schematically represented as follows:

This approach makes possible the enlarging of crime prevention approaches that are too narrow to positively oriented measures, to the promotion of the quality of life and the societal context in general. It concerns here not a choice between the different levels (not ‘either-or’, but ‘both’). It concerns a balance between the different levels with as point of reference the total societal quality. The levels are linked and require each other. It concerns an integral approach (see below).
The model contains a hierarchy. It is a continuum of continuously more specific crime prevention measures. The construction of more basis-oriented levels of prevention always precedes the more specific prevention levels. In the movement through prevention levels from 0 to 4 there is increasing attention for continuously more specific, targeted crime prevention measures. Or the other way around, in moving through levels 4 to 0, there is a decreasing attention for specific preventative interventions, in favour of an increasing attention for the general quality of life. Each level must be strong enough to build another on top of it. The base is the global societal context. If this is functioning sufficiently, it operates in a fundamentally preventative manner. It is a first requirement before moving to a more specific crime prevention policy. Specific crime prevention measures can thus become more directed and still remain in proportion. Thus specific crime prevention measures (level 3, e.g. frequent police patrols, sensitisation around burglary prevention) in a neighbourhood that has significantly deteriorated and that has high unemployment makes little sense if work is also not done to improve the basic quality of life in this neighbourhood (levels 0, 1 and 2, e.g. using improvement programs, welfare work, social and economic neighbourhood development) that in turn cannot be seen apart from the functioning of the wider society to which this neighbourhood belongs (level 0, e.g. unemployment as a result of economic recession).
Positive versus negative orientation
The general and fundamental levels 1 and 2 are other in nature than the specific prevention and curative levels of 3 and 4. The first levels strive for a better quality of life, a healthy social climate (positive orientation). They are oriented to a positive goal such as democratic decision making, minority rights, improved living conditions, etc. The latter levels are problem oriented (negative orientation). They are directed to crime, the protection against such and tackling it. Here a given problem is at the centre, a negative fact the needs to be avoided or combated (e.g. camera surveillance to protect against theft, private security services to protect against burglary or damage, police patrols to protect against vandalism, etc.) We can speak of a negative orientation. This brings us to the general, basic tension in the field of prevention. A balance must be sought between these negative and positive orientations, with special attention to the positive orientation. A too one-sided problem-oriented approach to crime prevention can be counterproductive.
Policy feedback
Systematically maintaining crime data can provide much information concerning the implementation of crime prevention policy. This can take the form of figures and statistics, but also that of experiential data. Figures and statistics are provided mostly at level 4. These can provide information about the different levels where crime prevention measures need to be taken. In this way feedback from the different levels can be fed back into policy.
A wave of burglaries in a certain neighbourhood can lead to measures at the different levels. These can range from, for example, a temporary intensification of police patrols, an information campaign about burglary protection (level 3), to the installation of better street lighting and better methods of dealing with roaming youth in the evening (levels 2 and 1).
Forming attitudes and structural measures
The horizontal division of the crime prevention pyramid into different levels ranging from a general to a specific prevention approach to problem areas must be supplemented by a vertical division. Here we find considerations concerning forming attitudes, changing the mentalities among the groups concerned to complement the structural measures (cf. diagram “horizontal and vertical division”). These two groups of measures must be implemented as complements to each other at the different levels (from 1 to 4).
With attitude formation we think of, among others, information and sensitisation campaigns via the media, advertising campaigns, lectures, talks, meetings, involving the citizen in political decision making via an open policy with possibilities for involvement and discussion, a personal approach to the target groups involved, increasing involvement, ombudsman services, etc.
Under structural measures we can make a distinction between organisational (e.g. uniform crime prevention regulations, alarm procedures, surveillance regulations, disaster plans, etc.) and technical measures (this includes a broad range of measures such as physical and architectural protection in high-risk zones by cameras, electronic security, access control, adequate lighting, etc.). The installation of extra lighting in a dark parking lot can be a structural measure but so can the improvement of the road infrastructure for bicyclists and the installation of solid, safe bicycle sheds, as well as major investments in green parks, the drastic reduction of disposable packaging, forbidding the sale and possession of weapons, the creation of more work opportunities, training programmes in education concerning learning to deal with conflicts and assistance in reconverting the cocaine farmer’s industry.
Crime prevention policy should promote the synergy between measures in the area of attitude formation and structural measures. A one-sided accent on one of the groups of measures usually ends in the failure of the crime prevention measures taken. For example it makes no sense to forbid youth to come into contact with drugs (attitude) if there is no meaningful alternative offered.
Integral approach
If a society is undergoing an increase in unrest, fear, insecurity and crime, policy makers must investigate the problem fields at the different levels. This can range from political and economic policy (level 0), welfare policy (levels 1 and 2) to the organisation and operation of the police and the courts (levels 2, 3 and 4). Measures at the specific prevention levels (3 and 4) are here perhaps insufficient.
Without a thoroughgoing analysis of the underlying levels, the threat is real that one is only treating symptoms while the macro-environment that creates, maintains and reinforces these problems is left untouched. Such a policy quickly becomes pernicious. A policy that wishes to decrease the level of these problems must thus look to a broad spectrum of measures at the different levels, with an eye to a positive orientation and to a balance between structural measures and attitude formation (cf. above). It concerns here an integral approach.
The crime prevention pyramid can contribute to the criminological discussion concerning where crime prevention begins and ends. It is not necessary here to look for possible conflicts between the different levels, but rather to the linking of the different levels back to level 0. In this way very diverse crime prevention measures can make sense if they are meaningful within a general policy of promoting the quality of life. The pyramid shows that prevention measures in school should not strictly focus on level three and four by, for example, camera surveillance or sanctioning difficult students. The perspective of prevention can be broadened to a more integral approach (all levels) with an accent on positive measures (level one and two). In this way, direct and indirect answers to violence can be employed more appropriately (O’Moore & Minton, 2006; Blaya, 2006; Galloway & Roland, 2004).

This leads us to a last step in this contribution after the the description of the research, the review of empirical data and the model of the prevention pyramid. As mentioned before, Belgium is a small but complex federalized country, with different policies and research traditions in the northern (Flemish speaking) and southern (French speaking) part of the country. Although Flanders has a long history of criminological research in the field of juvenile problem behaviour and youth delinquency, this research used to be scattered all over the region. Research projects that were set up to tackle this disparity are not limited to the school environment but oriented towards a larger field, with only some information on what is happening in schools.
Therefore, the collection of specific data about violence and deviant behaviour in schools at a regional level should be encouraged. In doing so, European standards for data gathering should be followed, in order to allow comparison with other countries and regions.
Furthermore, investments in longitudinal research at a European level should be made. These data should not only be used for knowledge, but also lead to scientific and practical answers to the following questions. “Which programs will effectively tackle violence and bullying?” “How can we contribute to the implementation and evaluation of best and promising practices in our region and at a European level?”
Secondly, as shown in the paragraph about well-being in schools, the Ministry of Education is registering truancy figures. Although this is a positive initiative, the registration should become more specified (different forms of truancy should be identified) and these figures should be used to construct an integral education- and school policy that includes prevention, problem-solving and the promotion of well-being in schools.
Finally, anthropological-sociological issues considering definitions should also receive some attention. International comparison shows differences in local terminology that cannot be translated. Concepts of violence should be described and contextualized against the cultural background in which they are used. Clarification of definitions and concepts is important in order to construct a European scientific language and to interpret country bound figures in a correct way.


BLAYA C., (2006), Violences et maltraitances en milieu scolaire, Paris, Armand Colin.
BURSSENS D., (2007), Onder, op en over de schreef: jongeren en delinquentie, in VETTENBURG N., ELCHARDUS M., WALGRAVE L., (ed.)., Jongeren in cijfers en letters: bevindingen uit de JOP-monitor 1, Leuven, Uitgeverij Lannoo.
CORNELL D.G., (2006), School Violence. Fears Versus Facts. London, Lawrence Erlbaum Associates, Publishers.
DEBARBIEUX E., (2004), Les enquêtes de victimation en milieu scolaire : leçons critiques et innovations méthodologiques, in Déviance et Société, 3, 28, 317-333.
DEBARBIEUX E., (2006), Violence à l'école : un défi mondial ?, Paris, Armand Colin.
DEBOUTTE G., (2006), Ruwe leerlingen en warme scholen: een (on)mogelijk samengaan?, Unpublished paper for the conference ‘Agressie op school’, Eekhoutcentrum, KULeuven Campus Kortrijk on 6 december 2006.
DE CAUTER F., (1990), Methodiek van de preventieve projectwerking, Leuven, Acco.
DE GROOF S., (2003), Leerlingenparticipatie naderbij bekeken. Een kwantitatief en kwalitatief onderzoek naar de betekenis en het belang van participatie op school, Vrije Universiteit Brussel, Vakgroep Sociologie, Onderzoeksgroep TOR.
DE GROOF S., STEVENS F., (2004), Over onderwijs. Gezocht: effectieve school waar het goed toeven is, met plaats voor inspraak en aandacht voor sociale ongelijkheid, in BURSSENS D., DE GROOF S., HUYSMANS H., SINNAEVE I., STEVENS F., VAN NUFFEL K., VETTENBURG N., ELCHARDUS M., WALGRAVE L., DE BIE M., (Eds), Jeugdonderzoek belicht. Voorlopig syntheserapport van wetenschappelijk onderzoek naar Vlaamse kinderen en jongeren (2000-2004), 31-58, Unpublished report, K.U.Leuven, VUB & UGent.
DEKLERCK J., (1996) De preventiepiramide, een oriëntatiekader voor geïntegreerde preventie, Panopticon, 4, 347-363.
DEKLERCK J., DEPUYDT, A., DEBOUTTE G., (2001), Verbondenheid als antwoord op ‘de-link-wentie’. Preventie op een nieuw spoor, Leuven, Leusden, Acco.
DEKLERCK J., DEPUYDT A., (2001), The prevention pyramid: a model for fundamental crime prevention, Acta Criminologica, 14, 1, 73-80.
DEKLERCK J., (2001), Naar een fundamentele criminaliteitspreventie, Welzijnsgids, 40, Kluwer, 41-98.
DEKLERCK J., (2005), Delinquentie en haar context bekeken vanuit een cyclisch model van ‘integratie-desintegratie, Diss. doct. KULeuven, Afd. Strafrecht, Strafvordering en Criminologie, as part of DEKLERCK J., DEPUYDT A., “‘Re-ligare’ als antwoord op ‘de-linquentie’. Aanzet tot een ethische, contextuele en ecologische criminologie”, Diss. doct. KULeuven, Afd. Strafrecht, Strafvordering en Criminologie.
DEKLERCK J., (2006), Onveiligheid integraal aanpakken: de preventiepiramide, Tijdschrift voor Veiligheid, 3, 19-37.
DEPUYDT A., (1998), Verbondenheid als onderzoeksthema en maatschappelijk project in het spoor van ‘Mens, Maatschappij en Marginaliteit’, 137-168, in DE BATSELIER S., (1998), Liber Amoricum S. De Batselier. Gent, Druk in de Weer.
DE WITTE H., HOOGE J., WALGRAVE L., Ed., (2000), Jongeren in Vlaanderen: gemeten en geteld. 12- tot 18- jarigen over hun leefwereld en toekomst, Leuven, Universitaire Pers.
DIENST WELZIJN EN GEZONDHEID VAN DE PROVINCIE LIMBURG, (2006). Werkmap Welbevinden op school (WOS). Samen werk maken van meer welbevinden in de klas en op school, Leuven, CEGO Publishers.
ELIAERTS E., e.a., (Ed.), (2006), Criminologie: tussen kritiek en realisme, Liber Amica/orum Christian Eliaerts, Brussel, VUB Press,
GALLOWAY D., ROLAND E. (2004), Is the direct approach to tackling bullying always the best? in RIGBY K., SMITH P.K., PEPLER D., (Eds.), (2004), Bullying in schools. How successful can interventions be?, 37-53, Cambridge, Cambridge University Press.
GOEDSEELS E., VETTENBURG N., WALGRAVE L., (2000), Delinquentie, in DE WITTE H., HOOGE J., WALGRAVE L., (Ed.), Jongeren in Vlaanderen: gemeten en geteld. 12- tot 18-jarigen over hun leefwereld en toekomst, Leuven, Universitaire Pers Leuven..
GOEDSEELS E., DETRY I., VANNESTE C., Dir., (2005) Onderzoek met betrekking tot de productie en wetenschappelijke exploitatie van cijfergegevens inzake jeugddelinquentie en jeugdbescherming, Eerste onderzoeksrapport, Analyse van de instroom op de jeugdparketten voor het jaar, Collectie van onderzoeksrapporten & onderzoeksnota's n°20b, Nationaal Instituut voor Criminalistiek en Criminologie, Hoofdafdeling Criminologie, Brussel.
GORIS P., WALGRAVE L., Red., (2001), Van kattenkwaad en erger. Actuele thema’s uit de jeugdcriminologie, Leuven, Garant.
HARTER, S., (1999), The construction of the self: A developmental perspective, New York, Guilford.
HAY I., ASHMAN A.F., (2003), The development of adolescents’ emotional stability and general self-concept: The interplay of parents, peers and gender. International Journal of Disability, Development and Education, 50, 79-93.
HUYSMANS H., (2006), De jeugd van tegenwoordig…doet het zo slecht niet, in VETTENBURG N., ELCHARDUS M., WALGRAVE L., (Ed.), Jongeren van nu en straks. Overzicht en synthese van recent jeugdonderzoek in Vlaanderen, 221-252, Leuven, Lannoocampus.
JIMERSON S.R., FURLONG M.J., (2006), Handbook of School Violence and School Safety: From Research to Practice, London, Routledge.
LYNCH J.P. & ADDINGTON L.A., (2006), Understanding Crime Statistics: Revisiting the Divergence of the NCVS and the UCR, Cambridge, Cambridge Studies in Criminology, Cambridge University Press.
MINISTERIE VAN ONDERWIJS EN VORMING, (2007), Leerplicht. Wie is er niet als de schoolbel rinkelt? Evaluatie 2006-2007, Brussel, Ministerie van Onderwijs en Vorming.
MUNTHE E. , SOLLI E., YTRE-ARNE E., ROLAND E. (Eds.), (2004), A report from the conference ‘Taking Fear out of Schools’. Organized by the OECD, the Norwegian Ministry of Education and Research, Directorate for Primary and Secundary Education and the University of Stavanger. Stavanger, Norway, 5-8 September 2004. Stavanger, University of Stavanger, Centre for Behavioural Research.
OEDC, (2003), Literacy skills for the World of Tomorrow. Further results from Pisa 2000, Paris, OECD.
O'MOORE, M. & MINTON, S.J., (2006), Protecting Children: Making the school environment safe, in GITTINS C. (Ed.) Violence reduction in schools - how to make a difference, 79-94, Strasbourg, Council Of Europe.
ORTEGA R., SANCHEZ V., VAN WASSENHOVEN L., DEBOUTTE G., DEKLERCK J., (2006), Vista Unit A1: Understanding the Definition and Context of School Violence. VISTA: School Bullying and Violence: Taking Action., English/A1f.pdf
PEETERS J., (1995), Moeilijke adolescenten. Antwerpen, Garant uitgevers.
GALLOWAY D., ROLAND E. (2004), Is the direct approach to tackling bullying always the best? In: Rigby 37-53
RIGBY K., SMITH P.K., PEPLER D., (Eds.), (2004), Bullying in schools. How successful can interventions be? Cambridge, Cambridge University Press.
SCHILLEMANS L., (2004), Jeugdonderzoek in Vlaanderen. Het waarom, wat, wie en hoe van het JOP alias Jeugdonderzoeksplatform, Retrieved August 6, 2008, from website Jeugdonderzoeksplatform,
SMITH P.K. (Ed.), (2003), Violence in Schools. The response in Europe, London, Routledge/Falmer.
SMITS W., (2004), Maatschappelijke participatie van jongeren. Bewegen in de sociale, vrijetijds- en culturele ruimte, Brussel, V.U.B., Vakgroep Sociologie, Onderzoeksgroep TOR.
STEVENS V., VAN OOST P., (1995). Pesten op school: Een eerste deelonderzoek naar het optreden van pesten en gepest worden bij kinderen tussen 10 en 14 jaar. Tijdschrift voor Klinische Psychologie, 3, 239-259.
STEVENS V., VAN OOST P., (1995), Pesten op school: een actieprogramma. Handboek leerlingenbegeleiding. Probleemgedrag aanpakken, 15, 93-100.
VANDENBOSCH H., VAN CLEEMPUT K., MORTELMANS D., WALRAVE M., (2006), Cyperbesten bij jongeren in Vlaanderen, Brussel, viWTA.
VANDERSMISSEN V., THYS L., (1993), Onderzoek naar de schoolbeleving in Vlaanderen. Omgang met medeleerlingen, Caleidoscoop, 4, 4-9.
VAN NUFFEL K., (2003), Jeugdonderzoek in Vlaanderen. Het JOP stelt zich voor, Retrieved August 6, 2008, from website Jeugdonderzoeksplatform,
VAN KERCKVOORDE J., (1995), Een maat voor het kwaad? Over de meting van criminaliteit met behulp van officiële statistiek en door middel van enquêtering, Leuven, Universitaire Pers.
VAN RIJSSELBERGHE L., PORTZKY G., VAN HEERINGEN C., (2002), Zelfbeschadigend gedrag bij adolescenten in Vlaanderen. Een onderzoek naar de verspreiding van zelfbeschadigend gedrag en de beïnvloedende factoren door middel van vragenlijstonderzoek in Vlaamse scholen in het kader van het CASE-onderzoek, Gent, Eenheid voor Zelfmoord onderzoek, UGent.
VETTENBURG N., ELCHARDUS M., WALGRAVE L., (Ed.) (2007) Jongeren in cijfers en letters: bevindingen uit de JOP-monitor 1, Leuven, Lannoo,.
VETTENBURG N., HUYBREGTS I., (2001), Onveiligheidsgevoelens en antisociaal gedrag. Een onderzoek naar de samenhang tussen de onveiligheidsgevoelens bij leerkrachten en het antisociaal gedrag van leerlingen, Onuitgegeven onderzoeksrapport, Leuven, K.U.Leuven, Afdeling Strafrecht, Strafvordering en Criminologie, Onderzoeksgroep Jeugdcriminologie.
VETTENBURG N., MELIS B., VAN GILS J., GORIS P., VERDONCK D., BURSSENS D., (2003), Preventie gespiegeld, Leuven, Lannoo.
X (2007), Leerplicht. Wie is er niet als de schoolbel rinkelt? Evaluatie 2006-2007, Brussels, Ministery of education, An abstract of this document can also be consulted at:


Read also

> Summary
> 2 - Schools being tested by violence
> 3 - Deviant behaviour and violence in Luxembourg schools
> 4 - School violence in Spain
> 5 - European trends in research into violence and deviance in schools
> 6 - Extra - Stakes of violence in education in Africa

<< Back