Review of the Roots of Youth Violence: Literature Reviews

Volume 5, Chapter 6:

Strain Theories5

This section considers four theories that are commonly classified as “strain theories.” These theories include anomie theory (Merton, 1938), institutional anomie theory (Messner and Rosenfeld, 1994), general strain theory (Agnew, 1985 and 1992), and relative deprivation theory (Crosby, 1976; Davis, 1959; Gurr, 1970; Runciman, 1966). Each theory argues that strain creates pressures and incentives to engage in criminal coping as a response to the strain experienced, though each differs with respect to what constitutes the most important sources of strain.

Anomie Theory

Robert Merton published his “Social Structure and Anomie” in 1938. In this article, Merton set forth a theoretical framework for explaining crime rates that differed from the Chicago school criminologists. For example, theorists such as Shaw and McKay (1942) held that urban slum areas foster criminal behaviour through the generational transmission of deviant cultural value. Thus, social disorganization theory assumes that the rejection of conventional middle-class values results in high rates of crime in urban slum communities. Merton, on the other hand, argued that it was the rigid adherence to conventional American values that caused high rates of crime and deviance. In essence, he believed that the widespread conformity to American culture in general, and the American obsession with economic success in particular, produced high levels of serious crime. Understanding why Merton made such a claim requires an understanding of how he viewed American society.

Merton noted that, as opposed to other Western industrialized nations, the United States places an unusual emphasis on economic success. Even more unique is how this emphasis seems to be universal. All members of American society, from the well-to-do to the impoverished, ascribe to the “American dream” that if one were simply willing to work hard enough, one would inevitably reap the economic rewards of such labours. The problem, according to Merton, is that despite the widespread belief in the possibility of upward social mobility, the American social structure limits individuals’ access to the goal of economic success through legitimate means. For example, while the probability of attaining economic success would be enhanced by getting a college education, not all members of American society are able to do so. Those lower on the socio-economic ladder are particularly vulnerable due to their relatively disadvantaged starting point in the race toward affluence.

In essence, Merton’s work contained a discussion of how culture and social structure could cause high crime rates. Merton noted that the American culture, as stated above, places economic success at the pinnacle of social desirability. The emphasis on attaining economic success, however, is not matched by a concurrent normative emphasis on what “means” are legitimate for reaching the desired “goal.” This problem is then exacerbated by the social structural component discussed by Merton, which highlights the structural barriers that limit individuals’ access to the legitimate means for attaining the goal of economic success. This disjunction between culturally ascribed goals (i.e., economic success) and the availability of legitimate means to attain such goals (i.e., social structural limits) in turn puts pressure on the cultural norms that guide what means should be used to achieve the culturally prescribed goal.

Merton referred to this weakening of cultural norms as “anomie.” His adoption of the term “anomie” is based on Durkheim’s (1897) reference to the weakening of the normative order in society, or, put differently, how institutionalized social norms may lose their ability to regulate individuals’ behaviour. In particular, Merton noted that institutionalized norms will weaken, and anomie will set in, in societies that place an intense value on economic success. When this occurs, the pursuit of success is no longer guided by normative standards of right and wrong. Rather, Merton (1968: 189) noted, “the sole significant question becomes: Which of the available procedures is most efficient in netting the culturally approved value?”

Merton was careful to note that there were a number of ways in which individuals may adapt to the “strains” brought on by the inability to secure pecuniary success, and not all of these adaptations are deviant. In his famous typology, Merton proposed that there were a number of adaptations possible in response to social systems that have anomie and blocked opportunities. These adaptations are: innovation, in which the goals are pursued but legitimate means are eliminated and illegitimate means are used; ritualism, in which the goals are abandoned but the legitimate means are pursued; retreatism, in which the goals are abandoned as well as the means; and rebellion, in which the social structure – both goals and means – is rejected and a new structure is advocated. A fifth adaptation is conformity, in which the goals are accepted and pursued, along with the legitimate means. Although Merton failed to articulate what factors determine which deviant adaptations will be adopted (as he acknowledges in his 1938 article), his theory predicts that rates of deviance will be greater when the level of anomie is higher and when the extent of blocked opportunities is greater. Conversely, conformity will be common in social systems when goals and legitimate means are clearly articulated and promoted and when opportunities are equal across individuals and social groups.

Merton’s conceptualization of anomie theory rose to a dominant position in the 1950s and 1960s, both theoretically and empirically, and was applied to various deviant behaviours and expanded to include a theory integrated with Cultural Transmission/Differential Association (Cohen, 1955) and one that appreciated that individuals are located in illegitimate as well as legitimate opportunity structures (Cloward, 1959; Cloward and Ohlin, 1960). Anomie theory also endured a period of critique (cf. Hirschi, 1969; Kornhauser, 1978) and a period of benign neglect. Yet anomie has entered a period of resurgence within the last decade. Institutional Anomie Theory (Messner and Rosenfeld, 1997; Messner and Rosenfeld, 2001; Rosenfeld and Messner, 1995) and General Strain Theory (Agnew, 1995; Agnew, 1999) have further refined and revised traditional theoretical contemplations about anomie. As well, anomie has been applied to a host of new areas. For example, anomie has been applied to illuminating the situation of street youth and street criminology (Hagan and McCarthy, 1997); anomie has been examined empirically (Menard, 1995; Menard, 1997); anomie has been utilized to elucidate specific deviant behaviours such as homicide (Messner and Rosenfeld, 1997), transnational crime (Passas, 2000), and white collar crime (Cohen, 1995; Waring, Weisburd and Chayet, 1995); and anomie has been refined by synthesizing it with other theoretical concepts (Passas, 1997).

Institutional Anomie Theory

In Messner and Rosenfeld’s (1994) Crime and the American Dream, Merton’s anomie/strain theory was extended and partially reformulated. Although Messner and Rosenfeld agreed with Merton’s view of American culture, they found his analysis of social structure incomplete. In particular, Merton held that the American system of stratification was responsible for restricting individuals’ access to legitimate opportunities for upward socio-economic mobility, which in turn resulted in high levels of criminogenic anomie in society. What was missing from the anomie tradition, argued Messner and Rosenfeld, was an understanding of how the American dream promotes and sustains an institutional structure in which one institution, the economy, assumes dominance over all others. This apparent “imbalance” in the institutional structure limits the ability of other social institutions, such as the family, education, and/or the political system, to insulate members of society from the criminogenic pressures of the American dream or to impose controls over their behaviour.

As such, Messner and Rosenfeld (1994), like Merton, contend that American culture places a disproportionate emphasis on material success goals. Also consistent with Merton, they maintain that the contradictions implicit in the dominant value system produce strong pressures to employ the most efficient means available to achieve monetary rewards (84–85). However, their conceptualization of the impact of the social structure on the level of anomie and, in turn, on levels of instrumental crime, departs dramatically from Merton's. In particular, Messner and Rosenfeld question Merton’s decision to restrict his analysis of the relationship between social structure and anomie to only one facet the social system, the legitimate opportunity structure (15). More to the point, they argue that an expansion of economic opportunities, rather than lessening the level of anomie in society, may actually intensify culturally induced pressures to use extralegal means to acquire monetary rewards. Insofar as economic vitality reinforces the societal preoccupation with the goal of material success, it is likely to heighten the level of anomie within a collectivity (62, 99–101). Hence, they conclude that the elimination of structural impediments to the legitimate opportunities cannot, in and of themselves, do much to reduce crime rates (108).

Rather than focusing solely on the limitations of the economic structure as the primary source of structural pressure to innovate (i.e., commit crime), Messner and Rosenfeld’s analysis centres on the criminogenic influence of a variety of social institutions in American society. Drawing heavily on Marxist theory, they argue that the cultural penchant for pecuniary rewards is so all-encompassing that the major social institutions (i.e., the polity, religion, education, and the family) lose their ability to regulate passions and behaviour. Instead of promoting other social goals, these institutions primarily support the quest for material success (i.e., the American dream). For example, Messner and Rosenfeld contend that “education is regarded largely as a means to occupational attainment, which in turn is valued primarily insofar as it promises economic rewards” (78). In short, to the extent that social institutions are subservient to the economic structure, they fail to provide alternative definitions of self-worth and achievement that could serve as countervailing forces against the anomic pressures of the American dream. To summarize, Messner and Rosenfeld’s institutional anomie theory holds that culturally produced pressures to secure monetary rewards, coupled with weak controls from non-economic social institutions, promote high rates of instrumental criminal activity (103–108).

The majority of studies that have empirically tested institutional anomie theory have employed property and violent crime as dependent measures. These studies include Chamlin and Cochran, 1995; Maume and Lee, 2003; Messner and Rosenfeld, 1997; Piquero and Piquero, 1998; Pratt and Godsey, 2003 and Savolainen, 2000. Following is an examination of the empirical findings of studies that have investigated institutional anomie theory.

Chamlin and Cochran (1995) were the first to empirically examine institutional anomie theory. They examined the effects of economic (e.g., poverty) and non-economic measures (e.g., family, religion, and polity) on the 1980 property crime rates for the 50 US states. They analyzed the interaction effects of economic and non-economic measures on instrumental crime. For the most part, their findings were consistent with institutional anomie theory predictions. The authors found that higher levels of church membership, higher levels of voting participation, and lower levels of the divorce-marriage ratio reduced the criminogenic effects of poverty on instrumental crime. They concluded that economic measures had no independent effect on crime; rather, it was, as expected, the interplay between economic and non-economic institutions that increased anomie and led to higher levels of crime within a society.

Messner and Rosenfeld (1997) examined the relationship between the economy, measured as economic inequality, and the polity in relation to criminal homicide rates in 45 modern industrialized nations. They hypothesized that homicide rates and decommodification, a measure of values and resources made available to citizens to reduce their reliance on market forces, would vary inversely. Messner and Rosenfeld found that decommodification had a direct significant negative effect on homicide rates. Nations with greater decommodification scores tended to have lower homicide rates. They also found that the United States had a very low decommodification score and a very high homicide rate.

Building upon Messner and Rosenfeld’s (1997) sample size of 45 nations, Savolainen (2000) added a second cross-national sample that included data from 36 additional countries, increasing the sample size to 81. Savolainen examined the interaction effects between economic inequality and welfare support. He found that economic inequality was a significant predictor of homicide in societies with weak institutions of social protection, thus supporting institutional anomie theory. Savolainen also found that nations offering the most generous welfare programs tended to have the lowest levels of economic inequality.

Pratt and Godsey (2003) highlighted the similarities of the relationships among measures of social support, economic inequality and crime across three theoretical perspectives: institutional anomie theory, social support (Cullen, 1994), and macro-level general strain theory (Agnew, 1999). By examining cross-national homicide rates, they found direct effects for both social support, measured as amount spent on health care, and income inequality, measured as the ratio of median income for the richest and poorest 20 per cent of the population. In addition, they found mediating effects for both social support and income inequality, in that when added together in a model, the main effects for both diminished. Finally, the moderating effect of income inequality and social support suggested that the presence of high levels of social support reduced the effect of economic inequality on rates of homicide.

Maume and Lee (2003) examined institutional anomie theory and used disaggregated homicide rates (e.g., total versus expressive versus instrumental) at the county level as their dependent measure. They suggested that non-economic institutions will not only moderate the influence of the economy on crime rates, but will also mediate the influence of the economy on crime rates. Over all, they found limited support for the commonly used moderating hypothesis; that is, they only found one significant interaction effect that suggested the effect of economic pressure on homicide rates was significantly weaker in counties with low levels of welfare disbursements. More support was found for the mediating hypothesis in that, across all three models, the measure of economy, the Gini coefficient of family income inequality, was reduced when predictors representing the non-economic institutions were added into the model. In support of institutional anomie theory, Maume and Lee conclude that non-economic institutions (i.e., polity, family, and education) mediate the relationship between the economy and instrumental violence.

Piquero and Piquero (1998) examined institutional anomie theory in relation to property and violent crime rates using data from 50 US states and Washington, D.C. Using census data for measures of the independent variables, Piquero and Piquero examined the education component of institutional anomie theory. For the most part, additive effects were significant and in the expected direction for both property and violent crime models. More importantly, the interactive effects revealed that higher percentages of individuals enrolled full-time in college reduced the effect of poverty on both crime types, while the polity-economy interaction was only statistically significant for violent crime rates.

These studies, taken as a whole, suggest partial support for institutional anomie theory. Across a variety of different outcomes as well as aggregated units, and utilizing various measures representing non-economic institutions, there appears to be some support for the presumption of the importance of the economy in explaining instrumental crimes.

General Strain Theory

Agnew’s (1985 and 1992) general strain theory posits that strain leads to negative emotions, which may lead to a number of outcomes, including delinquency. The specific strains discussed in the theory include the failure to achieve positively valued goals (e.g., money or status), the removal of positively valued stimuli (e.g., loss of a valued possession), and the presentation of negatively valued stimuli (e.g., physical abuse). While many specific types of strain may fall into these categories, Agnew has attempted to specify the conditions under which strain may lead to crime. Strains that are 1) seen as unjust, 2) high in magnitude 3) associated with low social control, and 4) create some incentive to engage in criminal coping are most likely to lead to violence and delinquency.

According to general strain theory, individuals experiencing strain may develop negative emotions, including anger, when they see adversity as imposed by others, resentment when they perceive unjust treatment by others, and depression or anxiety when they blame themselves for the stressful consequence. These negative emotions, in turn, necessitate coping responses as a way to relieve internal pressure. Responses to strain may be behavioural, cognitive, or emotional, and not all responses are delinquent. General strain theory, however, is particularly interested in delinquent adaptations. General strain theory identifies various types of delinquent adaptations, including escapist (e.g., drug use), instrumental (e.g., property offences), and retaliatory (e.g., violent offences) outcomes. Coping via illegal behaviour and violence may be especially true for adolescents because of their limited legitimate coping resources, greater influence from peers, and inability to escape many stressful and frustrating environments.

Of the various types of negative emotions, anger has been identified as playing the key role in mediating the effect of strain on delinquency and violence. This is the case because anger “increases the individual’s level of felt injury, creates a desire for retaliation/revenge, energizes the individual for action, and lowers inhibitions” (Agnew, 1992: 60). Some studies of the mediating model in general strain theory have focused on anger as the sole intervening factor in the relationship between strain and delinquency. Using data from the 1966 Youth in Transition survey, Agnew (1985) conducted the first study of this kind among tenth-grade boys. He found that the strain these boys experienced in school and at home had both a direct effect and an indirect effect (via anger) on property offences, violent offences, and status offences. Agnew also found that anger had the strongest effect on violent offences. Mazerolle and Piquero (1997) focused on how anger mediated the impact of strain on violent responses among college students. They found that exposure to various types of strain affected the students’ assaultive behaviour, both directly and indirectly, through anger. In contrast to the above, Mazerolle et al. (2000) found that strain only affected violent behaviour directly in a sample of high school students and that anger did not have a significant mediating role. Agnew (2004) notes that survey research typically measures trait anger or the disposition of anger, whereas general strain theory argues that strain produces situation-specific or short-term anger, which in turn may lead to crime. Researchers who measure trait anger may find that it does not mediate between strain and crime.

Other studies examined the role of other negative emotions, such as depression and anxiety, but found no mediating effects on delinquent outcomes (violent or non-violent). For example, Aseltine, Gore and Gordon’s (2000) longitudinal study of high school students found that anxiety did not mediate the impact of strain on any kind of delinquent act, although their findings confirmed the mediating role of anger in the relationship between strain and violent/aggressive behaviour (but no such role for non-aggressive acts and drug use). Piquero and Sealock (2000) conducted a study among incarcerated youths on the effects of anger and depression in mediating the impact of strain on both violent and property crime. The results showed that depression failed to predict both types of crime, whereas anger predicted violence but not property crime. There have, however, been a few studies that indicate that the role of other mediating variables, apart from anger, should not be too quickly dismissed. Simons, Yi-Fu, Stewart and Brody (2003), for example, found that strain increased depression, which in turn contributed to crime. Agnew (2004) notes that researchers should attempt to investigate other variables that may mediate between strain and crime. For instance, strain may increase attitudes favourable to aggression, which in turn may lead to crime.

General strain theory has attempted to specify the factors which increase the likelihood that individuals will cope with strain by committing crime. Agnew contends that crime becomes a likely outcome when individuals have a low tolerance for strain, when they have poor coping skills and resources, when they have few conventional social supports, when they perceive that the costs of committing crime is low, and when they are disposed to committing crime because of factors such as low self-control, negative emotionality, or their learning history. Empirical research has offered some support for the above. Agnew et al. (2002), for example, found that individuals with the personality traits of negative emotionality and low constraint were more likely to respond to strain with crime. Such individuals are impulsive, overly active and quick to lose their tempers.

Relative Deprivation Theory

Relative deprivation refers broadly to people’s perceptions of their well-being relative to comparison others. Well-being may be estimated based on a number of dimensions, including wealth, income, power and prestige. “Relative deprivation is [also] used to refer to the emotion one feels when making negatively discrepant comparisons” (Crosby, 1976: 88). Relative deprivation theories argue that when attempting to understand the causes of crime, it is insufficient to examine objective factors such as poverty or inequality, and instead we must try to “delineate the factors that regulate the relationship between objective and subjective status” (Crosby, 1976: 88). According to relative deprivation theory, objective conditions may have little relationship to people’s behaviour, since their perceptions of these conditions may be at odds with the actuality. The research of Agnew et al. (1996) suggests that “dissatisfaction or strain may occur at all class levels, and [this] may help to explain the weak effect of stratification measures on crime. Although one’s ‘objective’ position in the stratification system is important, one’s subjective interpretation of that position may be even more important” (695). The concept of relative deprivation directly measures people’s subjective assessment of their economic position or other dimension of social comparison. Relative deprivation researchers explicitly recognize that people evaluate themselves relative to comparison others, and not all persons may choose the same comparison other. For this reason, relative deprivation is considered by its advocates to be more important when predicting people’s behaviour, compared with more “objective” measures of deprivation such as poverty or inequality.

There are four formal theories of relative deprivation. Davis (1959) argues that people will experience relative deprivation when they lack X, perceive that similar others have X, want X, and feel entitled to have X. Runciman (1966) adds that the individual must think it feasible to obtain X, while Crosby (1976) asserts that individuals must also lack a sense of responsibility for failure to possess X. Runciman distinguishes between egoistic and fraternal relative deprivation. The former occurs at the individual level and the latter when individuals compare their group with other reference groups. Gurr (1970) argued that relative deprivation is the difference between one’s value expectations and value capabilities. Value expectations refer to goods and opportunities that the individual wants and feels entitled to, estimated based on comparisons with others. Value capabilities are the goods and opportunities that individuals already possess. Based on this typology, Gurr distinguishes between aspirational, decremental, and progressive relative deprivation. “All four theorists conceive of relative deprivation as an emotion and not simply as a perception” (Crosby, 1979: 109). This alludes to the distinction between cognitive relative deprivation (the recognition that one is deprived) and affective relative deprivation (the emotions that one attaches to one’s deprivation). Relative deprivation results in feelings of despair, frustration, grievance, and anger, and may be a powerful motivator of crime.

A number of researchers have employed relative deprivation as a predictor of crime.6 Stiles et al. (2000) investigate the impact of relative deprivation (called perceived economic deprivation) on property crime, violent crime and drug use. These authors include drug use as a dependent measure and suggest that relative deprivation may result in various responses other than hostility displaced on others. Relative deprivation was specified with respect to three reference groups: friends, neighbours, and perceived national norms. Accordingly, relative deprivation was assessed by asking respondents to compare their total family income to these three groups. Stiles et al. argue that once people recognize that they are deprived, they may develop feelings of envy, injustice and low self-worth. They therefore hypothesize that the relative deprivation–crime relationship is mediated by negative self-feeling. Survey data from 6,074 adults are employed in their study. Using nine logistic regression models, Stiles et al. initially examine the relationship of the three relative deprivation predictors to the three crime measures. Poverty and four controls are employed in each model. Relative deprivation (friends) significantly predicted violent and property crime; relative deprivation (neighbours) predicted property crime and drug use; and relative deprivation (national norms) predicted all three crimes. To examine the mediating effect of negative self-feeling, Stiles et al. then include this variable in all models and assess the change in the coefficient of each relative deprivation measure. A significant decrease in the predictive ability of relative deprivation indicates a mediated effect (Baron and Kenny, 1986). In seven of the nine models, the explanatory power of relative deprivation decreased significantly with the addition of negative self-feeling. Poverty was a significant predictor of property crime in all nine models, both without and with the inclusion of negative self-feeling. Stiles et al. conclude that relative deprivation increases negative self-feeling, which in turn leads to crime.

Baron (2004) examines the effects of strain on property crime, violent crime and drug use. In so doing, he includes two operationalizations of relative deprivation as predictors. The first measure assessed the level of respondents’ satisfaction/dissatisfaction with their monetary status. This assessment implies a comparison with others and represents a measure of relative deprivation that relates specifically to economic status. The second measure was broader in nature and asked respondents to give an overall ranking of themselves relative to others in Canadian society. Crime was measured using self report. The sample was 400 homeless youths from Vancouver. Regression analysis indicated that the monetary dissatisfaction operationalization of relative deprivation predicted property crime, while the second more inclusive operationalization predicted property and violent crime. Baron (2004) uses the same data set as Baron (2003) and has almost identical findings. Baron represents a test of Agnew’s (1985, 1992 and 2001) revised strain theory.

The findings of Baron are consistent with Agnew et al. (1996), which represents a test of classic strain theory. In this respect, it should be noted that one of Agnew’s operationalizations of strain, “failure to achieve positively valued goals” very closely resembles the operationalization of relative deprivation used by a number of researchers.

Grant and Brown (1995) conducted a double-blind experiment in which they manipulate collective relative deprivation and investigated its effects on participants’ collective protest actions, counter-normative actions, derogatory out-group remarks, and hostility/dislike for the out-group. Such dependent measures acknowledge that the responses to relative deprivation may be cognitive, affective or behavioural. The respondents were 259 students who were organized into groups for the experiment. These students were selected by pre-test designed to select only those students who believed that women should be given more encouragement to apply for high-status jobs. In each experimental session, two groups were placed in adjoining rooms to facilitate the belief that an inter-group interaction was taking place. They were told that both groups’ task was to discuss this issue and to develop a position statement for use by their university. Each group’s statement would be evaluated by the other group, who would make recommendations for a payment between $3 to $13 per student. Students were further told that such payments are usually around $10, but rarely less than $8, thus creating an expectation that could be violated. The two groups never met directly; all communication was through a confederate. The “evaluation,” which in fact came from the experimenters and not from the other group, was used to manipulate feelings of collective relative deprivation. Groups were randomly given positive (or negative) feedback and a payment of $10 (or $4) was recommended. The condition with negative feedback and a payment of $4 was supposed to induce feelings of relative deprivation. Just after this manipulation, collective relative deprivation was measured to verify if the manipulation was successful. The dependent variables were also measured. Group differences in the dependent measures for the relatively deprived and the non-deprived groups were compared. There were significant differences in all four dependent measures supporting the influence of relative deprivation.

Rosenfeld (1982) is one of the few studies which explores the relationship of relative deprivation to crime using structural rather than individual measures. The dependent variables are property and violent crimes. Controls include population size and per cent black. Relative deprivation is defined as the product of the intensity of deprivation, the scope of deprivation, and the level of economic aspirations among poor families in the Standard Metropolitan Statistical Area. Intensity refers to the degree of discrepancy or difference between economic capabilities and expectations, and is operationalized as the difference between the mean income of families below the poverty level and the mean income of all families in the SMSA. The scope of deprivation refers to the proportion of the population sharing some specified level of deprivation, and is operationalized as the percentage of families with incomes below the federal poverty level. Aspirations are measured by the ratio of median years of school completed by heads of poor families to median years of school completed by all family heads. This measure assumes that the economic aspirations of low-income people will vary directly with their educational attainment. This measure of relative deprivation has merit, since it assesses economic inequality within the context of level of economic aspirations. This measure successfully predicted homicide, rape, assault, burglary, larceny and auto theft, but not robbery. Notably, in this model, unemployment predicted burglary and larceny, but not the other crimes. Rosenfeld (1986) uses the same data set as Rosenfeld (1982), but employs a slight model re-specification (the southern region control variable is replaced by per cent black) and derives almost identical findings. The only differences are that relative deprivation no longer predicts auto theft, while unemployment does not predict any of the seven crimes.

Not all researchers have discovered significant relationships between relative deprivation and crime. Burton et al. (1994) examine the relationship of three operationalizations of strain to self-reported crime using a sample of 555 adults. Crime here is classified as utilitarian and non-utilitarian, which broadly overlap with property and violent crimes respectively. The operationalizations of strain are: economic aspiration/expectation disparities, blocked opportunities and relative deprivation. Relative deprivation measures respondents’ assessment of their economic situation relative to comparison others in their reference group. It should be noted that the aspiration/expectation measure of strain is considered to be an alternative operationalization of relative deprivation (Gurr, 1970). Initial analysis, using age, sex, and income as controls, reveal that blocked opportunities and relative deprivation, but not the aspiration/expectation measure, were significant predictors of crime. Burton et al. then include measures from competing theories (low self-control, differential association and social bond) in this model. Relative deprivation and blocked opportunities became non-significant for both utilitarian and non-utilitarian crime. Baron and Kenny (1986) point out, however, that if a mediating variable is entered into a regression model, the predictive power of the independent variable (here relative deprivation) decreases in the case of partial mediation, and becomes non-significant in the case of full mediation. Researchers have argued that relative deprivation may have an effect on crime through its effect of weakening social bonds (Kennedy et al. 1998). That is, relative deprivation weakens social bonds and decreases interpersonal trust, which in turn encourages crime. It may not be, as Burton et al. (1994) conclude, that relative deprivation is not an important predictor of crime. Failure to consider the potential mediating role of social trust could have resulted in an erroneous conclusion by these authors.

Policy Implications

Anomie theory, general strain theory, and relative deprivation theory have identified various types of strain which may induce delinquency and youth violence. The basic principle common to all three theories is that strain creates pressures that necessitate coping behaviours. Under some conditions, these coping behaviours may be deviant. Institutional anomie theory adds to these theories by indicating that strain conditions may be perpetuated by a wide range of institutions in society, apart from the economy. These theories indicate that policy interventions need to address the various types and sources of strain in order to address the issue of youth violence. It should be borne in mind, however, that while these theories point to a wide range of sources and types of strain, policy interventions should focus primarily on those that have the strongest relationships to crime and delinquency and on those that are most cost-effective. Following is an examination of some of the types of strain identified by general strain theory and the policy implications that follow from them. Similar arguments hold for the other theories.

Agnew (2001) has identified the conditions which must be present for strain to result in crime and delinquency. He argues that straining events can lead to crime when they are 1) seen as unjust, 2) high in magnitude, 3) associated with low social control, and 4) create pressure or incentive for criminal coping. He has identified a number of strains which have such characteristics. Policy interventions may focus on these types of strains with the aim of reducing them and hence reducing deviant adaptations. In addition, policy interventions may focus on improving youths’ ability to cope with the various types of strain. Some of the more important types of strain outlined by Agnew are examined next.

“The failure to achieve core goals that are not the result of conventional socialization and that are easily achieved through crime” can encourage delinquency (Agnew, 2001: 343). Such goals may include the desire for money, thrills/excitement, high levels of autonomy and masculine status. If barriers to achieving these goals are related to ascribed status, such as race or religion, people may see their inability to achieve such goals as unjust. This may encourage the adoption of illegitimate means to achieve these goals, since legitimate means are perceived to be blocked. Parental rejection is another strain likely to lead to deviant behaviour. Parental rejection creates strain “because it may seriously threaten many of the child’s goals, values, needs, activities, and/or identities” (343). Parents who reject their children are more likely to model deviant and aggressive behaviours which may be learnt by children. Research has found that parental rejection is strongly related to delinquency (Sampson and Laub, 1993). Educating children about coping, and about what may be expected of them when they become parents in the future, may address this issue. Parental education also represents another area of focus. Another important type of strain identified by Agnew is supervision or discipline that is very strict, harsh, erratic or excessive given the infraction. Such discipline is likely to be seen as high in magnitude and unjust. Such sanctions do not function as effective controls and, importantly, may serve to undermine attachments and commitments to conventional values, others, and institutions. Additionally, the sanctioning agent models aggressive behaviour to the youth involved. “Data indicate that parents, school officials, and possibly criminal justice officials who employ this type of discipline/supervision increase the likelihood of crime” (Agnew, 2001: 344). Other forms of strain that induce deviance include child neglect and abuse, negative secondary school experiences, work in the secondary labour market, homelessness, abusive peer relations, criminal victimization, and experiences with prejudice and discrimination based on ascribed characteristics such as race. Policy interventions may focus on reducing such strains and in improving youths’ abilities to cope with such strains in an attempt to reduce youth violence.


Agnew, Robert, Francis T. Cullen and Velmer S. Burton Jr. (1996). A new test of classic strain theory. Justice Quarterly, 13(4), 681−704.

Agnew, R. (1985). A revised strain theory of delinquency. Social Forces, 64, 151−167.

Agnew, R. (1992). Foundation for a general strain theory of crime and delinquency. Criminology, 30, 47−87.

Agnew, R. (1995). The contribution of social psychological strain theory to the explanation of crime and delinquency. In Freda Adler and William S. Laufer (Eds.), The legacy of anomie theory (pp. 113−138). New Brunswick, NJ: Transaction.

Agnew, R. (1999). Why Do Criminals Offend: A general strain theory of crime and delinquency. In Francis T. Cullen and Robert Agnew (Eds.), Criminological Theory: Past to Present. Los Angeles: Roxbury Press.

Agnew, R. (2001). Building on the foundation of general strain theory: Specifying the types of strain most likely to lead to crime and delinquency. Journal of Research in Crime and Delinquency, 38, 319−361.

Agnew, R. (2004). A general strain theory approach to violence. In M.A. Zahn, H.H. Brownstein and S.L. Jackson (Eds.), Violence: From Theory to Research. Newark, N.J.: Lexis Nexis.

Agnew, R., T. Brezina, J.P. Wright and E.T. Cullen. (2002). Strain, personality traits, and delinquency: Extending general strain theory. Criminology, 40, 43−72.

Aseltine, Robert H., Susan Gore and Jennifer Gordon. (2000). Life stress, anger and anxiety, and delinquency: An empirical test of general strain theory. Journal of Health and Social Behaviour, 41, 256−275.

Baron, R. and D. Kenny. (1986). The moderator-mediator distinction in psychological research: Conceptual, strategic and statistical considerations. Journal of Personality and Social Psychology, 51, 1173–1182.

Baron, S. W. (2003). Self-control, social consequences, and criminal behavior: Street youth and the general theory of crime. Journal of Research in Crime and Delinquency, 40(4), 403−425.

Baron, S. W. (2004). General strain, street youth and crime: A test of Agnew's revised theory. Criminology, 42(2), 457−483.

Burton, Velmer S. Jr., Francis T. Cullen and David T. Evans. (1994). Reconsidering strain theory: Operationalization, rival theories, and adult criminality. Journal of Quantitative Criminology, 10(3), 213−239.

Burton, V.S. and F.T. Cullen. (1992). The empirical status of strain theory. Journal of Crime and Justice, 15, 1−30.

Chamlin, M. B. and J.K. Cochran. (1995). Assessing Messner and Rosenfeld's institutional anomie theory: A partial test. Criminology, 33, 411–429.

Cloward, Richard A. (1959). Illegitimate means, anomie, and deviant behavior. American Sociological Review, 24, 164–176.

Cloward, Richard A. and Lloyd E. Ohlin. (1960). Delinquency and Opportunity: A Theory of Delinquent Gangs. New York: The Free Press.

Cohen, Albert K. (1955). Delinquent Boys: The Culture of the Gang. Glencoe, IL: The Free Press.

Cohen, Deborah Vidaver. (1995). Ethics and crime in business firms: Organizational culture and the impact of anomie. In Freda Adler and William S. Laufer (Eds.), The Legacy of Anomie Theory (pp. 183−206). New Brunswick, NJ: Transaction Publishers.

Crosby, Faye. (1976). A model of egoistic relative deprivation. Psychological Review, 83(2), 85−113.

Crosby, Faye. (1979). Relative deprivation revisited: A response to Miller, Bolce and Halligan. The American Political Science Review, 73(1), 103−112.

Davis, James. (1959). A formal interpretation of the theory of relative deprivation. Sociometry, 22(4), 280−296.

Durkheim, E. (1897). Suicide: A Study in Sociology. New York: Free Press.

Grant, Peter and Rupert Brown. (1995). From ethnocentrism to collective protest: Responses to relative deprivation and threats to social identity. Social Psychology Quarterly, 58(3), 195−212.

Gurr, T. R. (1970). Why Men Rebel. Princeton: Princeton University Press.

Hagan, John and Bill McCarthy. (1997). Anomie, social capital, and street criminology. In Nikos Passas and Robert Agnew (Eds.), The future of Anomie Theory (pp. 124−141). Boston: Northeastern University Press.

Hirschi, Travis. (1969). Causes of Delinquency. Berkeley: University of California Press.

Kennedy, Bruce, Ichiro Kawachi and Deborah Prothrow-Stith. (1998). Social capital, income inequality, and firearm violent crime. Social Science and Medicine, 47(1), 7−17.

Kornhauser, R. R. (1978). Social Sources of Delinquency: An Appraisal of Analytic Models. Chicago: University of Chicago Press.

Maume, M. O. and M.R. Lee. (2003). Social institutions and violence: A sub-national test of institutional anomie theory. Criminology, 41, 1137–1172.

Mazerolle, Paul and Alex Piquero. (1997). Violent responses to strain: An examination of conditioning influences. Violence and Victims, 12, 323−343.

Menard, Scott. (1995). A developmental test of Mertonian anomie theory. Journal of Research in Crime and Delinquency, 32, 136–174.

Menard, Scott. (1997). A developmental test of Cloward’s differential-opportunity theory. In Nikos Passas and Robert Agnew (Eds.), The Future of Anomie Theory (pp. 142−186). Boston: Northeastern University Press.

Merton, R.K. (1938). Social structure and anomie. American Sociological Review, 3, 672−682.

Merton, R.K. (1968). Social Theory and Social Structure. New York: Free Press.

Messner, S.F. and R. Rosenfeld. (1994). Crime and the American Dream. Belmont, CA: Wadsworth.

Messner, S.F. and R. Rosenfeld. (1997). Crime and the American Dream (2nd ed.). Belmont, CA: Wadsworth.

Messner, S.F. and R. Rosenfeld. (2001). Crime and the American dream (3rd ed.) Belmont, CA: Wadsworth.

Messner, Stephen F. and Richard Rosenfeld. (1997). Political restraint of the market and levels of criminal homicide: A cross-national application of institutional-anomie theory. Social Forces, 75, 1393–1416.

Passas, Nikos. (1997). Anomie, reference groups, and relative deprivation. In Nikos Passas and Robert Agnew (Eds.), The future of Anomie Theory (pp. 62−94). Boston: Northeastern University Press.

Passas, Nikos. (2000). Global anomie, dynomie, and economic crime: Hidden consequences of neoliberalism and globalization in Russia and around the world. Social Justice, 27, 16–44.

Piquero, A. and N.L. Piquero. (1998). On testing institutional anomie theory with varying specifications. Studies on Crime and Crime Prevention, 7, 61–84.

Piquero, Nicole and Miriam Sealock. (2000). Generalizing general strain theory: An examination of an offending population. Justice Quarterly, 17, 449−484.

Pratt, T. C., and T.W. Godsey. (2003). Social support, inequality, and homicide: A cross-national test of an integrated theoretical model. Criminology, 41, 611–643.

Rosenfeld, Richard. (1982). Inequality, relative deprivation and crime: Explaining some discrepant findings. Paper presented at the 1982 annual meeting of the American Society of Criminology, Toronto, Canada.

Rosenfeld, Richard. (1986). Urban Crime Rates: Effects of Inequality, Welfare Dependency, Region, and Race. In James M. Byrne and Robert J. Sampson (Eds.), The Social Ecology of Crime (pp. 116−130). New York: Springer-Verlag.

Rosenfeld, Richard and Stephen F. Messner. (1995). Crime and the American dream: An institutional analysis. In Freda Adler and William S. Laufer (Eds.), The Legacy of Anomie Theory (pp. 159−182). New Brunswick, NJ: Transaction.

Runciman, W. G. (1966). Relative Deprivation and Social Justice: A Study of Attitudes to Social Inequality in Twentieth-Century England. Berkeley: University of California Press.

Sampson, Robert and John Laub. (1993). Crime in the Making. Cambridge, MA: Harvard University Press.

Savolainen, J. (2000). Inequality, welfare state, and homicide: Further support for the institutional anomie theory. Criminology, 38, 1021–1038.

Shaw, C. R. and H.D. McKay. (1942). Juvenile Delinquency and Urban Area: A Study of Rates of Delinquents in Relation to Differential Characteristics of Local Communities in American Cities. Chicago: University of Chicago Press.

Simons, R.L., C. Yi-Fu, E.A. Stewart and G.H. Brody. (2003). Incidents of discrimination and risk for delinquency: A longitudinal test for strain theory with an African American sample. Justice Quarterly, 20, 827−854.

Stiles, B. L., Liu, X., and H.B. Kaplan. (2000). Relative deprivation and deviant adaptations: The mediating effects of negative self-feelings. Journal of Research in Crime and Delinquency, 37(1), 64−90.

Waring, Elin, David Weisburd and Ellen Chayet. (1995). White collar crime and anomie. In Freda Adler and William S. Laufer (Eds.), The legacy of Anomie Theory (pp. 207−226). New Brunswick, NJ: Transaction Publishers.

5 This section was prepared with the assistance of Randy Seepersad, PhD candidate, Centre of Criminology, University of Toronto.

6 Many researchers use the terms “inequality” and “relative deprivation” synonymously. Both terms, however, refer to distinct constructs.


Volume 1. Findings, Analysis and Conclusions

Volume 2. Executive Summary

Volume 3. Community Perspectives Report

Volume 4. Research Papers

Volume 5. Literature Reviews