SciELO - Scientific Electronic Library Online

vol.33 issue1The relationship between Gratitude and Life Satisfaction in a sample of Spanish university students: the moderation role of genderCorruption and Sexual Scandal: the Importance of Politician Gender author indexsubject indexarticles search
Home Pagealphabetic serial listing  


Services on Demand




Related links

  • On index processCited by Google
  • Have no similar articlesSimilars in SciELO
  • On index processSimilars in Google


Anales de Psicología

On-line version ISSN 1695-2294Print version ISSN 0212-9728

Anal. Psicol. vol.33 n.1 Murcia Jan. 2017 


Young people belonging to violent groups in the Region of Madrid. Psychosocial process model on the onset and evolution of violent identity behaviour1

Jóvenes pertenecientes a grupos violentos en la Comunidad de Madrid. Modelo psicosocial procesual sobre el inicio y la evolución de la conducta violenta identitaria



María-Jesús Martín2, José-Manuel Martínez2, Rubén García-Sánchez3, Begoña Aramayona2, Carmen Almendros3, and Cristina Jiménez2

2Departamento de Psicología Social y Metodología, Universidad Autónoma de Madrid (Spain).
3Departamento de Psicología Biológica y de la Salud, Universidad Autónoma de Madrid (Spain).

This study has been funded within the actions of the Program "Madrid Joven Integra" within the Operative Program Fight against Discrimination (European Social Fund, objective 3, Region of Madrid) with the collaboration of the Agency for Re-education and Reinsertion of Criminal Minors (ARRMI) of the Region of Madrid. It has been also funded by the Ministry of Science and Innovation through the National Program of Scientific Research, Development and Innovation (Ref. PSI2008-05221/PSIC).





The qualitative research about out-group violence has examined different aspects of this social problem, but the evolution of violent groups has only been approached tangentially. This research analyses the discourses of 124 interviews done with youth from Madrid that belong to violent groups. Describe the interpretation made by young people and identifies variables and psychosocial processes to generate and explore a psychosocial model that attempts to describe the constitution and evolution of this behavior. The strategy of theory generation was applied (first 71 interviews) and afterwards, the method of analytic induction (following 53 interviews) to allow the validation of the identified model. The resulting method proposes that the upsurge of violent groups depends on the previous socialisation focused on three factors: milieu, opportunities and basic socialising processes. The evolution and involution of violent groups would depend on the interaction between the personal and social identities of its members.

Key words: qualitative analysis; Gangs; Interview; Processual Model.


La investigación cualitativa sobre violencia juvenil ha indagado diferentes aspectos de este problema, pero el modo en que los jóvenes se involucran en grupos violentos y cómo evolucionan ha sido abordado solo tangencialmente. Este trabajo analiza cualitativamente 124 entrevistas realizadas a jóvenes de la Comunidad de Madrid que declaran pertenecer a grupos violentos. Describe la interpretación que realizan los jóvenes e identifica variables y procesos psicosociales que permitan generar y explorar un modelo psicosocial que trata de describir el inicio y la evolución de esta conducta. Para ello, se aplicó en un primer momento el método de generación de teoría (primeras 71 entrevistas) y después el método de inducción analítica (segundas 53 entrevistas) para explorar la adecuación a estos nuevos datos. El modelo resultante propone que el inicio de la conducta violenta depende de la socializaáón previa, definida por tres factores: entornos, oportunidades y procesos básicos de socialización; la evolución dependería fundamentalmente de la interacción entre la identidad personal y la identidad social de los jóvenes integrados en grupos violentos.

Palabras clave: estudio cualitativo; grupos violentos; entrevista; modelo procesual.



The Pan American Health Organization (Krug et. al., 2003, p. 5) defines violence as "the intentional use of physical aggression or aggression threats against oneself or another person, group or community that has as main consequence a high probability of causing lesions, death, poor development or deprivation". On the other hand, authors in the United Kingdom talk about "Delinquent youth groups": a group of three or more members, lasting three months or longer, that stays a long time in public spaces, that has participated in criminal acts in the past 12 months and that has at least one structural characteristic, for instance a name, a territory, a leader, some rules, etc. In Spain, the definition of youth gang appears in Instruction 23/2005, of 7 December, of the State Department for Security: its members are young individuals aged from 12 to 32 years, with internal cohesion and discipline structures, that are engaged in violent behaviours and cause social alarm. To standardise criteria, the Eurogang Network2 proposed defining gang as a "street-oriented stable youth group whose involvement in illegal activity is part of its group identity" (Weerman et al., 2009, p. 20).

As shown, we face on the one hand too generic definitions that recommend characterising the type of violence investigated; other times the definition itself considers other criminal acts, which makes difficult to understand the specific events analysed, an essential question (Wood and Allyne, 2013) for the development of interventions aimed specifically at preventing or treating this problem. In the case in question, the characterisation of violence is determined by the fact that, at least in Spain and Europe, many of the criminal acts that are apparently individual are inspired by group norms and habits (Castro, Pérez, García, Gordillo and Gallego, 2012; Martín, Scandroglio, Martínez and López, 2015). For this study, we have rather defined the groups of interest based on the identity violence they exert, defined as a physical aggression performed by one or more persons that, as group members, attempt to intentionally cause physical damage to one or more persons identified as members of a rival or different group. Its main characteristics are a) it allows to distinguish this behaviour from violence against unruly members of the group or those who do not show sufficient commitment with the group; b) it allows to include the violent acts performed indiscriminately against any individual (no matter their number) that belongs to a rival group, thus giving priority to group identity.

The statistics which allow to know the state of the issue in our country are mainly from the Ministry of the Interior and the Annual Reports drawn up by the State Attorney. Therefore, according to the Judicial Technical Police, 3928 members of these groups were identified between 2010 and 2011, 610 of whom were extreme right-wing individuals, 1813 extreme left-wing individuals, 1467 from Latin gangs, and 38 belonged to other types of groups. On the other hand, the State Attorney states that, specifically in the Region of Madrid, in 2006 and 2007 only the presence of Latin Kings was detected; in 2008 the activity of Trinitarios and Dominican's Don't Play was also identified (Pozo, Gallego, Vicente and Pérez, 2013). Accordingly, in 2010 the activity of these groups decreased and in 2011 this falling trend became stable; however, there appears to be (Minister of Justice, 2013, p. 418) "some rise in their activity".

In the past two decades there has been a remarkable increase of qualitative investigations on a high variety of subjects for two main reasons: a) the crisis of the traditional methods of positivist social sciences promoted by the emergence of post-modern epistemology (Tarrés, 2014); b) and by the development and application of accurate quality controls that have allowed to generate and validate theories (Lopez, Blanco, Scandroglio, and Rasskin, 2010). Qualitative studies promote multidimensional understanding of the problem, multidisciplinary design and applications, as well as, in particular, the triangulation of the results obtained by the application of several research techniques for the complex analysis of a behaviour (Denzin and Lincon, 2012).

The application of quantitative methods to the study of group violence has allowed in recent years to identify the psychosocial processes and variables involved in this subject. Mohamed (2011) states as factors contributing to integration of young individuals in gangs the negative experiences at school (academic failure, lack of implication and associated disruptive behaviours) and the family (negative family atmosphere, lack of paternal authority and high number of siblings); also, he mentions peer groups as a place where the young individual builds his/her identity and acquires social recognition by providing a link to the "chosen participation". With this regard, for the author gangs are "positive, selfbuilding spaces" that offer a lifestyle to their members, as well as the necessary material and symbolic resources and, eventually, a social status; this social function of the gang has five dimensions for its young members: physical (consumption), symbolic (power and recognition), psychological (selfesteem), identity (affiliation) and political (social conflict). Hounslea (2011) identified several factors predisposing to participation in youth gangs in the educational and work environments: low educational level, unemployment and lack of a quality employment on the long term; these would be conditions with unique, interactive effects (the low educational level and the lack of training make difficult the access to a quality employment). The author emphasised also the relevance of the lack of opportunities, not only related to employment or education, but also to leisure activities, as well as family factors (low educational level and paternal values consistent with violence as the method for solving conflicts). On the other hand, Shuval et al., (2012) state that youth violence is affected by individual factors (lack of skills to handle interpersonal conflicts), relational (absence of paternal figure at home) and community (troubled neighbourhoods where the presence of gangs is common, socioeconomic disparities based on the place of residence and feeling of insecurity). In a recent literature review about youth gangs, Young, Fitzgibbon and Silverstone (2014), concluded that the understanding of gangs and the reasons of young individuals to join them requires a careful analysis of the interconnection between social institutions (family, school, peer groups and other social networks) and other social-structural factors (deindustrialisation and lack of legitimate opportunities of employment, intimidation, violence, social exclusion and belonging to deprived neighbourhoods, among others).

Maybe the largest study on youth gangs, performed using qualitative methods, is that by Laidler and Hunt (2012) who summarised the results obtained with the application of interviews to young gang members along two decades. According to these authors, participation in youth gangs is related mainly to the search for and maintenance of respect and honour by peers and rivals. Drug trafficking or abuse of drugs, alcohol or other type of substances would act as a "social lubricant" that maintains cohesion and solidarity in the gang, promotes masculinity and comradeship, while acquiring a significant symbolic value in initiation rites and becomes an element that promotes intergroup fights; in this consumption young gang members look for satisfaction on the short term and a method to avoid family conflicts, school problems and lack of job prospects on the long term; drug dealing, on the other hand, becomes an individual way (but not collective) of generating income given the absence of other employment alternatives. With regard to the role of women in gangs, the results obtained by the authors evidenced that, in their respective gangs, women had witnessed or promoted aggressions or directly had a leading role, in particular to defend the honour of their family or of some of the gang members; unlike gang men, a large part of their lives is engaged in prosocial activities (talking, going shopping or looking after children). For women the gang becomes an essential element to attempt to build a family and affective environment.

As shown, the qualitative research on youth group violence has investigated different aspects of the problem, but how young men are involved in violent groups and their evolution has been approached only incidentally; however, this research method is particularly appropriate to approach the complexity and the process of this phenomenon. Based on these assumptions, a qualitative exploratory study was developed, with a double aim: to identify the main psychosocial variables influencing the onset, evolution and, if applicable, the involution of the youth identity violent behaviour; and to propose a model that promotes understanding this process; a model that can be also object of verification using quantitative and qualitative methods.


Material and methods


71 young individuals, 60 men and 11 women, aged from 15 to 29 years, living in the Region of Madrid. Specifically, the sample characteristics were as follows:

- 18 young individuals, all men, who at the time of the interviews were in Centres for the Enforcement of Judicial Measures (CEMEJ) of the Region of Madrid following a firm resolution for a crime of aggression. To be selected they should meet the following criteria: during the year preceding their entrance to the CEMEJ, they had physically assaulted twice or more times, while members of a group, one or more individuals from rival or different groups.

- 53 young individuals, 42 men and 11 women, who at the time of the interview were free and belonged to actively violent youth groups; the selection criteria were: in the past year they have physically assaulted twice or more times, while members of a group, one or more individuals from rival or different groups". To obtain the maximum diversity, they were enrolled in different areas of the Region of Madrid and belonged to different peer groups. These informants were compensated economically with 30 Euros for their participation and, to confirm the truthfulness of the performance of violent behaviours, criteria of sincerity were included in the instrument, as shown in the section below.

Table 1 summarizes the description of the respondents.


Individual interviews focusing on the problem, performed on the basis of a planned semi-structured outline including 47 main questions (see Table 2), structured as follows: 1) Socialisation environments, 2) Variables and processes related to identity violence 3) Variables and processes related to violence in general, 4) Variables and processes related to personal identity, and 5) Cross-check questions repeated twice at different times of the interview, to test consistency and sincerity of the discourse. In the case of young individuals at liberty, the interview was applied at two different times, separated by a mean of 44 days (SD = 7 days), to check for any changes in the frequency of behaviours or in the personal or group situations of the respondent; young individuals placed in institutions were interviewed only once as the behaviour or the situation of the respondent could not change.


The enrolment of informants at liberty was performed by a group of people from youth associations, non-profit organisations and social intervention professionals; all of them received individualised training and were paid for their work. In the case of young individuals placed in institutions, the selection was performed by the managers (generally directors or programme coordinators) of the CEMEJ. In a second stage a pilot application of the interview was performed with the first three young individuals enrolled; the analysis generated changes of adaptation of the language to the type of population and cuts in the questions to avoid exceeding two hours of duration. Finally, two of the principal investigators conducted the interviews; in the case of young individuals at liberty, they were held in different places agreed between the recruiters and the participants: 19 in offices at the University; 13 at youth social or cultural centres; 10 at premises of youth or political associations; 8 at meeting premises of the group, and 3 at personal rooms of the young individuals at "squatted" houses. All young individuals placed in institutions (18) were interviewed at offices of the CEMEJ. The only conditions to the places where the interviews were conducted, in both cases, were that the acoustic conditions enabled recording the talk and that privacy was ensured.

Each participant was informed of the objectives of the investigation and the guarantee of anonymity of their identity and confidentiality of their statements; also they were asked for consent to record their statements. For this, an Informed Consent Protocol was drawn up that was approved by the Ethics Committee of Autonomous University of Madrid.

Analysis of results

An analysis of the discourse was made, divided into two stages. In the first, an attempt was made to generate a theoretical psychosocial model consistent with the statements of all individuals in the first interview (71) about the onset and evolution of the violent behaviour. In the second, this model was compared to the analysis of the transcripts of the second interview to the 53 young individuals from violent groups that were at liberty.

Coding of the information. A primary intra-category coding was performed based on the contents of the interview (Table 2) and a secondary axial or inter-category coding in two main stages: a) origin or onset of the identity violent behaviour; b) evolution or involution of the identity violent behaviour.

Content analysis. The statements of all individuals interviewed were included in each section of the interview. For the analysis, a syncretic approach was used, based on the "grounded theory" for the discovery of processes or variables and relationships between them and the "analytical induction" for checking and "partial" or "universal" generalisation (Flick, 2014). There were two axes of main analyses: transversal or conceptual (referred to the analysis of each component or category) and longitudinal or processual (referred to the process of onset, evolution and, in case, involution of the violent behaviour).

Operatively, "the constant comparative method" was applied (Hernández Carrera, 2014) by which provisional inferences were generated, consistent with the opinions expressed by the interviewed individuals; each new statement about the same matter (from the same interviewed individual or the rest of informants) was compared to previous hypotheses to value their theoretical consistency. The observed disagreements required adopting two measures a) the priority was to examine the possibility of reformulating the hypothesis to allow inclusion of this new approach; b) in case this was not feasible, a classification or structuring of the universe of contents of the interviewed individuals took place, assuming the existence of different perceptions about the matter. "Invariant" or final hypotheses were postulated, that met two conditions: a) finding explicit empirical support in the statements of all subjects; b) there is no statement that disagrees with them totally or in part.

Saturation. There was no previous formalised theory, but there was an express theory formed by the selection of contents of the interview. Based on the application of the constant comparative method, it was considered that a hypothesis would have been saturated when it could not be reformulated in part comparing it to the relevant statements of seven different interviewed individuals.

Quality controls. The analysis was fully performed by an investigator. Then, the conclusions of the study (invariant hypotheses) and the original material (statements of the interviewed individuals structured by thematic blocks) were subject to re-analysis by other three investigators by a non-face-to-face Delphi panel to refute empirically and argumentatively the conclusions of the former. An evaluation form was drawn up in a table divided into three sections (columns): exact statements of the respondents in each section, hypothesis formulated in a seven-point Likert type scale of agreement, and proposed reformulation. After the three consultation rounds, the inter-judge saturation conditions could be met: a) That the scores in the scales of agreement was above 5.5 points; b) That no judge had expressed a degree of agreement under 4.5 points.

The structuring of the declarative material and its subsequent analysis were performed with the support of the software QDA-Miner 3.03.



The following introduction adopts a processual format that intends to investigate the onset, evolution and, if appropriate, the involution followed by young individuals belonging to violent groups. This study discusses only the "ultimate invariant hypotheses" resulting of the analysis of "constant comparison" and the subsequent interjudge evaluation. In general, each statement or group of statements is exemplified with the literal statement of some participant; said statements can be seen in the file of Complementary Material, the corresponding quote is mentioned in the text in format [D- x].

The onset

Previous experience andfirst aggressions

The informants can be divided into two groups in relation to their previous experience with violence: with and without experience in carrying out interpersonal aggressions. A minority had not carried out this type of aggressions during the pre-adolescent period, though they have suffered them; the aggressions suffered by this type of young individuals led them to adopting two main types of measures (not exclusive) to achieve a perception of greater security; passive self-defense (avoiding places, persons or situations of risk or threatening); active self-defense, oriented to the practice of aggressive sports (karate, aikido, boxing, full-contact) or to have relationships with powerful violent groups; this powerful nature may be translated into two aspects that are mentioned together or separately: prestige in the neighbourhood and ability to generate respect or fear in the rivals. For all interviewed individuals, the main or added value of joining these groups is their efficacy in personal protection [D 1-2].

Young individuals previously carrying out aggressions joined the violent group for one or two of these three reasons: following a relative that was already a member, feeling more powerful and to achieve some economic resource [D 3-5].

Joining or forming a group with violent behaviours is related to the convergence of behaviour of both types of young individuals (previous aggressors and those suffering violence) in carrying out the aggressions. It had two main effects; a) perceived positively: a better self-perception based on a greater feeling of security or control, or of personal courage or the belief of having the support of the group members; b) perceived negatively: they gained a high number of enemies and in parallel increased threats [D 6-7].

In the first stages of joining violent groups social mobility is likely: most remember they changed of group themselves or report that a peer did it. Informants declaring to have moved to other groups mentioned one of these three reasons: disagreements with peers or leaders, lack of efficacy to defend the group members and problems to combine the activities of the original violent group with the demands of other groups, friends, partners or relatives [D 8-10].

Socialisation environments

The description and assessment made by young individuals of their relatives allow to classify them in four types, ordered by frequency: anomic, bipolar or discontinuous, authoritarian and inductive or democratic. The anomic family is characterised by a limited capacity of influence and coercion, a reduced effort to participate in the life of their children, that tries to avoid conflicts, that has little or no control on their leisure activities or their friends and that when external signs of problems appear (bruises or wounds) prefers not to approach them or do it indirectly for two main reasons: they feel they are no prepared and/or move their socialising responsibility to other entities (in particular the educational); these are families that "do not see", that "do not want to see" or that "diminish the importance of what they see". The evidences of blows and bruises in their children, leisure late at night in pre-adolescents, possession of knives, brass knuckles or other artefacts, can cause alarm, but infrequently lead to an evident conflict with their children [D 11-12]. The authoritarian family is characterised by its desire of a permanent, close control and supervision of all aspects of their lives [D 13-14]. The bipolar or discontinuous family is characterised by not following a stable educational or upbringing pattern, oscillating suddenly and cyclically from carelessness to rigidity; this type is in principle a basically anomic family, that reacts initially to the evidence of violence in their children as an authoritarian family, with punishments, leisure restrictions and threats, but the relative failure of their attempts on the short or middle term or the internal conflicts between the parents resulting of the lack of agreement in approaching the problem, leads them to look for new coping methods (denial, avoiding the problem) similar to those used by anomic families [D 15-16]. Only one of the young individuals interviewed had a democratic or inductive parental supervision, characterised by the use of reasoning as educational standard, highlighting the consequences of the behaviour of children on other persons and on themselves.

It has been shown that, in general, parents maintain expressly a negative attitude towards youth violence, but their children remember relevant nuances on this general opinion. Therefore, it appears to be frequent that parents also talk about positive effects of violence, linked in particular to the defense of personal integrity, in response to an attack and even that they consider desirable an aggression in the first term to avoid greater damages [D 17-18].

With regard to academic experience, all interviewed young individuals rate it negatively. The most highly valued elements are peers (that in many cases are also part of the violent group). Teachers are generally poorly valued for their affective coldness and their lack of personal involvement; classes are considered boring in all cases, unrelated to the rest of the community or their personal interests [D 19-22].

With regard to work experience, in most cases it consisted of eventual, unskilled, low-paid jobs [D 23-24].

All interviewed individual declared to have or have had some affective partner or couple; in some cases they are identity partners (belonging to the violent group) and in others non-identity partners (not linked to the violent group). In the first case, the partner knows and supports clearly both that the young individual belongs to the group and the violent behaviour he/she conducts; in the case of non-identity partners, young individuals act in two ways: a) when the partner is not aware of the group membership and the violent behaviour practices, young individuals hide them, anticipating rejection of the situation by the partner; b) if the partner knows the activities performed by the young individual, they divide their time between the group activities and those carried out with the partner, while trying to avoid conflicts with the latter [D 25-27].

The results described above allow to conclude that the interviewed young individuals do not appear to have developed a positive personal and social identity linked to the traditionally normalising socialisation environments (family, academic or work institutions), but have looked for social support in social groups which are against the rules or antisystem, formed by individuals with vocational or attitude affinity (with the same hobbies or attitudes) and geographically close (from the same neighbourhood or school). In all cases the young interviewed individuals joined the group looking for a stable satisfactory source of social support which allows to generate a good self-image.


The expectations of young individuals about the duration of their violent acts are strongly conditioned by social identity: most of the interviewed individuals have a "diverse social identity", that is, state to belong to several social groups; these young people and their groups occasionally carry out criminal activities (low scale drug dealing or consumption); in these cases they declare that their violent behaviour will disappear when between 23 and 25 years of age and usually anticipate this discontinuation for the emergence of other responsibilities (occupational) and other people (partners, work peers, new friends) [D 28-29]. Another group of young individuals show a "unique social identity" or "merged with their personal identity" and show other peculiarities: they link their social activity to the violent group, that appears to meet all their needs; these are young individuals who frequently carry out criminal activities related to abuse drug traffic or extortion; there appears not to be expectations or wishes to change and they think it is likely that all or a large part of their life takes place in this group; any potential change requires a change in the group or in a large part of it; selfperception appears to depend on staying in the group (with this regard, it can be stated that they have merged identities - individual and social-); they participated in more or less elaborate initiation rites including activities with a strong personal implication, with a high cost for the protagonists (insulting, injuring, stealing or even raping a member of the rival group); carrying out "voluntarily" these behaviours promotes identification with the group and internalisation of its rules, habits and attitudes [D 30-31].

Powerful, active groups, directly or indirectly known, are models and motivate grouping; they value in these groups the social support they provide to their members and their capacity to generate respect or fear in enemies [D 32-33].

Once the group is formed, it is essential to check its efficacy to respond to a previous aggression (in a first stage) and to prevent future aggressions (later). The first violent actions are key elements in the increase of personal and group security; the success associated with these aggressions strengthens group cohesion. The high efficacy of these first aggressions is mainly due to the fact that they are planned and are usually directed at isolated members of rival groups [D 34-35].

Except for the interviewed individuals that have a "unique social identity" or a merged personal and social identity, young individuals usually perceive the initial aggressions as defensive, justified by previous attacks; later, they start to carry out aggressions with no immediate provocation, that are perceived as "preventive" or "educational actions" to gain respect or fear in the enemies and to avoid future intergroup disputes, as they show clearly the readiness of the group to defend itself strongly. In addition, it increases interdependence, trust and friendship among the group members that, in turn, reinforces their perception of security associated with the power of the group to defend its members; and in parallel self-esteem grows [D 36-37].

As the group's success in confrontations increases, individual and group self-efficacy and self-esteem grow and the tolerance level to extra-group provocations (real or presumed) decreases. The group emerges as a daily reality and its rules are used cognitively and affectively to face problems and interpret the daily reality. At this point, the actionrepression dynamics and the strong internal cohesion facilitate maintaining the group [D 38-39]. Even when these conditions start to be dysfunctional, young individuals usually bias the interpretation of reality in subtle forms (probably unconsciously) that help maintain a positive self-esteem linked to the violent group; for instance, they tend to generalise the violence they exercise in all or a large part of the community, systematically to out-groups and interpret with hostility the behaviour of any of its members [D 40-43].

These biases have a high instrumental value in maintaining and potentiating self-esteem (individual and social) linked to the group identity. However, most of these young individuals show discourses where they recognise some individual responsibility in the aggressions, with four conditions: the already mentioned generalisation of violence to the community; mentions of the negative effects of the capitalist socioeconomic system; the impossibility to change the aggressive behaviour for fear of the social vacuum that would appear if they left the in-group: or highlighting the reprisal they could suffer for leaving the group or for moving to another [D 4447].


Involution occurs when they manage to leave the group or when its rupture is evidenced. Mobility appears to develop ritually, though not necessarily consciously and it is developed in four stages: 1) Phase of personal instability linked to a reduced self-esteem dependent on belonging to the violent group and in particular to the immediate circle of friends; 2) Stage of unstable sensitivity, where the group member is more open to new leisure proposals from other groups, without adopting any decision yet; 3) Diversification and trial stage, where the young individuals carry out different activities with both groups, the violent and the alternative, that help them value the degree of improvement in their selfesteem; 4) Stage of rupture where the young individual progressively or suddenly stops going to the meetings and activities of the violent group and focuses his/her interest on the new group. Sometimes the alternative group can be replaced by the partner, provided this meets the fundamental condition of providing a better individual and social self-esteem. Finally, after assuming the new group identity, new habits and rules (for instance, prosocial ones) are assumed very easily and rapidly [D 48].

A number of conditions facilitate this process:

1. The appearance of new persons or relevant groups for the subject that maintain habits and rules contradictory with those of the in-group. The induction of a higher complexity and divergence in their social identity by joining the labour market or the emergence of stable affective relations usually have strong effects in the violent young individual, whenever it is difficult for him/her to integrate persons, activities and groups in his/her previous social structure. In both cases (or in similar situations) a marked imbalance or cognitive dissonance occur due to the emergence of different, opposing social identities [D 49-50].

2. An internal crisis in the in-group caused by serious setbacks in the fight against other groups, for some members which are part of the immediate circle of the subject leaving the group, or for the express dissatisfaction of this part of the group [D 51].

3. A crisis of confidence in the in-group, for the absence of evident reciprocity to the young individual or his/her close friends [D 52].

4. The impairment of personal identity in the group for the difficulties to perceive empathy, but also for the inability of behaving in an altruistic way, of asserting themselves [D 53].

5. Death or serious wounds in a group member as a result of a beating, arrest or imprisonment. In these cases, the crisis increases if these facts lead to a loss or reduction of the perception of personal and group power [D 54].

6. Persons who do not show commitment with the group suffer serious problems [D 55].

7. A conflict with the family arising for the infeasibility to continue hiding violent activities (police report, evident wounds, etc.) or because the young individual has an established perception that these activities can directly affect his/her family (phone threats or direct threats to parents or siblings) [D 56].

8. The conflict arising with the extra-identity partner, provided at least two conditions are met: a) The relationship must be affectively stable and bring strong positive feelings in the violent young individual, and b) They must express an identity conflict, that is, they must clearly oppose to the violent activities of the young individual or that he/she belongs to the violent group [D 57-58].


Discussion and conclusions

The results of the qualitative analysis allow to develop a model including psychosocial variables involved in the onset of the violent behaviour, its evolution and involution. A model based on six "invariant" or "universal" hypotheses is proposed, that influence the building and interpretation of the reality of the young individuals interviewed.

1) Out-group violence must be defined on the basis of an identity criterion, the recognition of a rival or different group identity. The violence of a single individual against another isolated individual belonging to a rival or different group is considered to be group violence. This hypothesis is consistent with previous studies (Martín et al., 2011; Scandroglio and López, 2013) and with the principles of the social identity's theory of the minimal group paradigm (Tajfel and Turner, 1986) and the "Self-categorisation theory" (Turner, 1995).

2) The peer groups of these young individuals are perceived as daily socialisation environments, based on a face-to-face interaction and on a strong personal involvement, in a way similar to that discovered by Laidler and Hunt (2012) in US gangs or in violent Spanish groups (Pérez and Castro, 2013; Scandroglio and López, 2013).

3) In-groups are perceived as a source of social support with a high impact in self-esteem and personal development. They comply with instrumental (economic support) and expressive functions (affective support, defence), according to Lin and Ensel (1989) identified in several studies on violent and antisocial groups (Alleyne and Wood, 2010; Wood and Alleyne, 2013). The feeling of insecurity suffered by these young individuals for confrontations or previous threats and the confidence in the ability of the group to defend them is key for them to join the violent group (Martín et al., 2011). Once part of the group, reciprocity and social responsibility are the main basic principles of internal performance of groups and promote internalisation of most rules and behaviours performed in them.

4) Intergroup conflicts promote the process of emergence of group rules, without requiring a previous or reasoned analysis. Group members react according to the theory of emergent norms (Turner and Killian, 1987): the stress of the conflict promotes the salience of the group norms leading individuals to react fast. This union of the group allows to increase the perception of group efficacy to face them, an impression that is supported by a higher probability of real success against enemies.

5) Social, group and personal identities are strongly linked: a substantial change in any of them leads to equilibrating changes in the rest. Some violent groups socialise their members similarly to Spanish gnostic sects (Rodríguez, 2013) or fundamentalist religious groups (Ysseldyk, Mathe-son and Anisman, 2010); in these cases, group and personal identity appear to be joined (Swann, Gómez, Dovidio, Hart and Jetten., 2010) and the self-esteem of the member is only due to the violent in-group. In the rest of the cases, the anticipation of "social vacuum" (for rejection or discontinuation of group support) is a strongly unwanted situation.

6) Young individuals from violent groups build the social reality with bias and then usually reinterpret it based on these previous beliefs (Vasquez, Lickel and Hennigan, 2013; Wood and Allyne, 2013). It is frequent to recognise in their discourse bias and socio-cognitive heuristics, extensively reported in psychosocial literature: "false consensus" "confirmatory trends", "vicarious personalism", and "ultimate attribution error" (Vaughan and Hoog, 2014).

Based on these assumptions, the model given in Figure 1



Onset of the violent behaviour

The onset of violent activities in young individuals depends on the previous socialisation that, as shown in Figure 2, is delimited by three linked factors: socialisation environments, opportunities of socialisation and basic socialisation processes. Given the processual nature of the model, they are presented as influences of greater to lesser generality; therefore, socialisation environments delimit the opportunities and these in turn frame the processes that may occur. It is proposed that the influence of these socialisation environments is essential for the onset of the behaviour and, though it is evident that they still exert effects later, once the young individual has joined the group, its norms and habits are the most influential factors. The socialisation environments make reference to the most important interaction niches where violent young individuals have grown: family, school and neighbourhood. These places limit the opportunities of socialisation that can be classified as formal and informal; the former are represented by the family (understood in terms of organisation and extensive) and school institutions. The informal opportunities are represented by forms of affiliation based on the shared geographical environment (the community, the neighbourhood) and by two psychosocial characteristics which are emergent forms of affiliation: skills (abilities) and attitudes (affections, trends of action).



Within these socialisation opportunities, the basic socialisation processes represent the processes of social influence that determine specific norms and behaviours in specific areas of performance. Two cognitive-social processes and three educational styles have been identified that appear to influence the development of violence and the preference for violent groups. With regard to the cognitive-social processes, the modeling of identity violence (in particular from older siblings and close childhood and pre-adolescence friends) and perceived self-efficacy (understood as the perceived ability to gain respect and social success and solve conflicts or problems by exercising violence) are the two central elements identified. Both factors, modelling and perceived selfefficacy, are part of the cognitive-social theory (Bandura, 1987) and have been identified in Spanish youth groups in previous investigations (Martin et al., 2011; Scandroglio and López, 2013). In addition, three educational styles have been identified (applied by the family, at work and/or at school) characterised by their inability to achieve internalisation of prosocial rules and generalisation to different situations and environments, which predispose to joining a violent group: the anomic, the authoritarian and the bipolar. Axiological and regulatory rigidity (autocratic families), its absence or ambiguity (anomic families) or the variation of the educational pattern applied (bipolar families) have made difficult the development of a positive personal and social identity and have induced indirectly that young individuals acquire it by belonging to alternative social groups. The anomic and bipolar styles were already related (Martin et al., 2011) to violent and antisocial behaviours. The negative impact of the lenient family with the aggressive, disturbing behaviours of young individuals has been extensively studied (Aroca-Montoliu, Lorenzo-Moledo and Miró-Pérez, 2012; Cerezo, Méndez and Ato, 2013). Both the anomic and the authoritarian styles are also strongly associated with violent and antisocial behaviours (Garcia and Gracia, 2010; Gracia, Fuentes and Garcia, 2010). In the specific case of violent youth groups, some studies have found that the socioeconomic level of the family (for instance, poverty or economic disadvantage), the structural characteristics of the family (for instance, single-parent families), the children upbringing styles and participation of family members in gangs are associated with belonging to this type of groups, though the results obtained in the studies are not fully consistent (Young et al, 2014).

The key element when making the decision of joining or forming a violent group is the process of "convergent predispositions" (see Figure 3).



Convergent predispositions are a concept theoretically linked to the "convergence theory" that originally postulated some rationality in crowd behaviour (Berk, 1974). Although it was criticized for its excessive simplicity (Perry and Pugh, 1978), in the case of violent groups (a smaller, more homogeneous group than the crowd) it is appropriate to explain the basic process of constitution or joining. Individual predispositions (attitudes, skills or abilities, behaviours or habits and personal norms), fruit of previous socialisation, are key elements in the onset of the violent behaviour. A favourable attitude to violence, skills and motivation for physical aggression, experience in exercising it and norms favourable to the use of violence to solve conflicts, are highly distinctive elements in the selection of the members of a group or its initial formation.

Evolution and involution

Once the behaviour has begun, the interaction between personal identity and social identity is the critical element which allows to postulate an explanation of the evolution and involution of young individuals, and has shown a complete adjustment to the discourse of all interviewed individuals. Figure 4 shows the primary variables involved in this interaction.



The association between personal and social identity is well established in the modern literature both in the Theory of Social Identity and in the modern hypotheses of the SelfCategorisation Theory (Vaughan and Hoog, 2014). It is a hypothesis consistent with the current investigations applied to youth group violence (Laidler and Hunt, 2012; Scandroglio and López, 2013; Tendayi Vicki and Abrams, 2013). The conditions that promote maintaining the violent identity behaviour are mainly two: a) how the self-esteem and selfconcept included in personal identity are linked to the violent group; b) the balance between group identity and the rest of identities (individual, partner, family, school).


Violent Spanish groups can increase their influence in young individuals with convergent predispositions for the skills they are attributed to generate respect and fear in rival groups and to defend their members (Martin et al., 2011). When in- or out-group conflicts cause a reduction of the positive group identity of its members, maintaining the group integrity depends on two factors: the efficacy of the strategies of social creativity and the increase of the intergroup conflict or the social competition (Vaughan and Hoog, 2014). Social creativity makes reference to the set of strategies which allow to reinterpret the reality favourably for the in-group by biased comparison of characteristics of the in-group to those of the out-group. Social competition involves an increase in the frequency and seriousness of the violent behaviour, which usually causes a reduction of dissents and drop-outs from the group and promotes a greater control of its members, increasingly unable to feel secure without the support of their peers; similar results have been found by Hajek (2012) in the case of sexual orientation and have been mentioned by Wood and Alleyne (2013) in the case of the formation of gangs.


The identity conflict is the condition necessary for the involution of violent behaviour. When the violent group is not able to be the fundamental origin of the self-esteem and self-concept of its members, and/or other persons (partner) or groups (family) or institutions (work) involve expressly a cognitive-affective-behavioural conflict, the probabilities of involution increase. In these cases, social mobility based on the presence of social support alternatives (family, partner, another group of friends, satisfactory employment) replaces the social creativity in the effort to keep or improve selfesteem. Without this external support, the probability of leaving or changing of group, even when group self-esteem is negative, becomes null or reaches very low values.



The authors would like to express their gratitude also to all young persons participating in this investigation.


1 This article contains Complementary Material available at

2 Eurogang is a thematic network aimed at developing a common framework for comparative research in Europe about youth gangs, using for it a research design and standardised methodological instruments.



1. Alleyne, E., & Wood, J. L. (2010). Gang involvement: psychological and behavioral characteristics of gang members, peripheral youth, and nongang youth. Aggressive Behavior, 36, 423-436.         [ Links ]

2. Aroca-Montolíu, C., Lorenzo-Moledo, P., & Miró-Pérez, J. L. (2012). Características de las familias que sufren violencia filio-parental: un estudio de revisión. (Characteristics families suffering filio-parental violence: a review study). Educatio Siglo XXI, 30(2), 231-254.         [ Links ]

3. Bandura, A. (1987). Pensamiento y acción. Fundamentos sociales. (Thought and action. Social foundations). Barcelona: Martínez Roca.         [ Links ]

4. Berk, R. A. (1974). Collective behavior. Dubuque, W.C: Brown.         [ Links ]

5. Castro, F. V., Pérez, M. L., García, V., Gordillo, M., & Gallego, P. A. (2012). Bandas Juveniles violentas en España (Violent youth gang in Spain). INFAD Revista de Psicología, 1(1), 383-392.         [ Links ]

6. Cerezo, F., Méndez, I., & Ato, M. (2013). Moderating role of family and friends' factors between disocial behavior and consumption in adolescents. International Journal of Clinical and Health Psychology, 13, 171-180.         [ Links ]

7. Denzin, N. K., & Lincoln, Y. S. (2012). Manual de investigación cualitativa. El campo de la investigaáón cualitativa. (Handbook of qualitative research. The field of qualitative research). Barcelona: Gedisa.         [ Links ]

8. Flick, U. (2014). La gestión de la calidad en Investigación Cualitativa. (Managing uality in Qualitative Research). Madrid: Morata.         [ Links ]

9. García, F., & Gracia, E. (2010). ¿Qué estilo de socialización parental es el idóneo en España? Un estudio con niños y adolescentes de 10 a 14 años. (What is the optimum parental socialization style in Spain? A study with children and adolescents aged 10-14 years). Infancia y Aprendizaje, 33(3), 365-384.         [ Links ]

10. Gracia, E., Fuentes, M. C., & García F. (2010). Barrios de Riesgo, Estilos de Socialización Parental y Problemas de Conducta en Adolescentes. (Neighborhood Risk, Parental Socialization Styles, and Adolescent Conduct Problems). Intervención Psicosocial, 19(3), 265-278. doi: 10.5093/in2010v19n3a7.         [ Links ]

11. Hajek, Ch. (2012). Communications and Identities Charazterized by Male Sexual Orientation. In H. Giles (Ed.), The Handbook of Intergroup Comunication (pp. 211-222). Nueva York: Routledge.         [ Links ]

12. Hernández-Carrera, R. (2014). La investigación cualitativa a través de entrevistas: su análisis mediante la Teoria Fundamentada. (The qualitative research through interviews: analysis by Grounded Theory). Cuestiones Pedagógicas, 23, 187-210.         [ Links ]

13. Hounslea, D. S. (2011). Youth Gang Membership: An investigation of Young People joining and leaving Gangs. Retrived from:         [ Links ]

14. Krug, E. G., Dahlberg, L. L, Mercy, J. A., Zwi, A. B., & Lozano, R. (Eds.) (2003). Informe mundial sobre la violencia y la salud. (World report on violence and health). Washington: Organización Panamericana de la Salud.         [ Links ]

15. Laidler, K. J., & Hunt, G. P. (2012). Moving beyond the gang-drug-violence connection. Drugs (Abingdon England) 19 (6), 442-452. doi: 10.3109/09687637.2012.702144.         [ Links ]

16. López, J., Blanco, F., Scandroglio, B., & Rasskin, I. (2010). Una aproximación a las prácticas cualitativas en Psicologia desde una perspectiva integradora. (A qualitative approach to the practices in psychology from an integrative perspective). Papeles del Psicólogo, 31(1), 131-142.         [ Links ]

17. Martín, M. J., Martínez, J. M., Espinosa, J., Blanco, A., De la Corte, L., Giménez, A. ... González, J. L. (2011). Grupos juveniles violentos (Violent Youth Groups). Murcia: Fundación Diagrama Intervención Psicosocial.         [ Links ]

18. Martín, M. J., Scandroglio, B., Martínez, J. M., & López, J. (2015). Caracterización actitudinal e intencional de la violencia juvenil exogrupal en la Comunidad de Madrid (Attitudinal and intentional characterisation of juvenil group violence in the Madrid Area). Anales de Psicología, 31, 207-216. doi: 10.6018/analesps.31.1.163701.         [ Links ]

19. Ministerio de Justicia. (2013). Memoria de la Fiscalía General del Estado. (Report of the Attorney General of the State). Madrid: Ministerio de Justicia.         [ Links ]

20. Mohamed, M. (2011). La formation des bandes. Entre la famille, l'école et la rue. (Gangs Training. Between family, school and the street). Paris: Presses Universitaires de France.         [ Links ]

21. Pérez, M. y Castro, F. V. (2013). Dimensiones simbólicas entorno a los grupos juveniles violentos en España. (The symbolic dimensions surrounding violent youth groups in Spain). INFAD Revista de Psicología, 1(1), 549-556.         [ Links ]

22. Perry, J. B., & Pugh, M. D. (1978). Collective behavior: Response to social strees. St Paul: West.         [ Links ]

23. Pozo, A., Gallego, P., Vicente, F., & Pérez, M. L. (2013). Spain, National Analysis. In S. Brutto y A. Minesso (Eds.) Interaction of different subjects. Towards a strategic Common answer concerning juvenile gangs (pp. 142-96). Verona: Tipolitografía don Calabria.         [ Links ]

24. Rodríguez, P. (2013). Características de la socialización de los hijos/as en contextos familiares sectarios: el rol de las figuras parentales según su estilo básico de relación/educación (Features of socialization of children/as in familiar contexts sectarian: the role of parental figures according to their basic style of relationship/education). Ajayu 11(1), 58-90.         [ Links ]

25. Scandroglio, B., & López, J. S. (2013). Violencia grupal juvenil. (Violent youth groups). Psicología Política, 46, 95-115.         [ Links ]

26. Shuval, K., Massey, Z., O' Caughy, M., Cavanaugh, B., Pillsbury, C. H., & Groce, N. (2012). I Live by Shooting Hill. A Qualitative Exploration of Conflict and Violence among Urban Youth in New Haven, Connecticut. Journal Health Care Poor Underserved, 23(1), 132-143. doi: 10.1353/hpu.2012.0024.         [ Links ]

27. Swann, W. B., Gómez, A., Dovidio, J. F., Hart, S., & Jetten, J. (2010). Dying and Killing for One's Group: Identity Fusion Moderates Responses to Intergroup Versions of the Trolley Problem. Psychological Science, 21(8), 1176-1183. doi: 10.1177/0956797610376656.         [ Links ]

28. Tajfel, H., & Turner, J. C. (1986). The social identity theory of intergroup behavior. In S. Worchel & W. Austin (Eds.). Psychology of intergroup relations (pp. 7-24). Chicago: Nelson-Hall.         [ Links ]

29. Tarrés, M. L. (2014). Observar, escuchar y comprender sobre la tradición cualitativa (Look, listen and understand: about the qualitative tradition in social research). México: Colegio de México-Flacso.         [ Links ]

30. Tendayi Viki, G., & Abrams, D. (2013). The Social influence of groups on individuals. In J. L. Wodd & T. Gannon (Eds.): Crime and Crime Reduction. The importance of group processes (pp. 3-33). New York (NY): Routledge.         [ Links ]

31. Turner, J. C. (1995). Self-categorization theory. In A. S. R. Manstead & M. Hewstone (Eds.): The Blackwell Encyclopedia of Social Psychology (pp. 562-567). Oxford: Blackwell.         [ Links ]

32. Turner, R. H., & Killian, L. M. (1987) Collective behavior. Englewood Cliffs: Prentice Hall.         [ Links ]

33. Vasquez, E., Lickel, B., & Henningan, K. (2013). Applying socio-psychological models to understandig displaced and group-based agression in street gangs. In J. L. Wodd & T. Gannon (Eds.): Crime and Crime Reduction. The importance of group processes (pp. 56-74). New York (NY): Routledge.         [ Links ]

34. Vaughan, G. M., & Hoog, M. A. (2014). Intergroup Behaviour. In G. M. Vaughan & M. A. Hoog, Social Psychology (pp. 363-411). French Forest: Pearson Australia.         [ Links ]

35. Wood, J. L., & Alleyne, E. (2013). Street Gangs: Group processes and theoretical explanations. In J. L. Wodd & T. Gannon (Eds.): Crime and Crime Reduction. The importance of group processes (pp. 34-55). New York (NY): Routledge.         [ Links ]

36. Ysseldyk, R., Matheson, K., & Anisman, H. (2010). Religiosity as identity: toward an understanding of religion from a social identity perspective. Personality and Social Psychology Review, 14(1), 60-71. doi: 10.1177/1088868309349693.         [ Links ]

37. Young, T., Fitzgibbon, W., & Silverstone, D. (2014). A Question of Family? Youth and Gangs. Youth Justice, 14, 171-185. doi: 10.1177/1473225414537569.         [ Links ]



María Jesús Martín.
Departamento de Psicología Social y Metodología.
Universidad Autónoma de Madrid.
Ciudad Universitaria Cantoblanco.
CP 28049 Madrid (Spain).

Article received: 25-05-2015
revised: 04-03-2016
accepted: 04-02-2016

Creative Commons License All the contents of this journal, except where otherwise noted, is licensed under a Creative Commons Attribution License