To investigate the potential interactions among strain, attributions, and anger in influencing the type of criminal response.

Objectives: Several hypotheses are tested in this study.

1. Is it expected that strain will be directly related to crime and substance use net of other predictors? Street youth who experience more labor market strain should report more crime and substance use.
2. Street youth who attribute their labor market problems to external sources and who are angry about these problems will report more crime
3. The revised theory of strain predicts that attributions and emotions will interact with strain to affect the response. It is in particular expected that violent and property crime, but not drug use, will occur more frequently among those street youths, who experience greater labor market strain, attribute it to external sources, and are angry about it.
4. The revised theory predicts that several variables should condition the effect of strain on deviance. It is expected that there are a series of interactions between strain and involvement with delinquent peers, moral inhibitions, conventional commitments, and self-efficacy.

Literature background: This study is building on the strain theory of Agnew (1992). The theory is applicable to a variety of different types of delinquency/crime. There are three major types of negative relations or strain: those that prevent or threaten to prevent the achievement of positively valued goals; those that remove or threaten to remove positively valued stimuli; and those that present or threaten to present noxious or negatively valued stimuli. The first type of strain has been the focus of criminological research, typically measured as the disjunction between aspirations and expectations, with poor empirical results. Strain increases the likelihood that individuals will experience such negative emotions as fear, depression, disappointment, and anger/frustration, with the latter thought to be especially important for delinquency and crime. Agnew (1992) suggests that anger results from strain when people blame their adversity on others. This negative affect creates pressure for some type of corrective action and delinquency/crime may be one response.

Population: Two hundred male respondents were identified using the following inclusion criteria:

- Participants must be male.
- They had to be aged twenty-four and under.
- They must have left or finished school.
- They must spend at least three hours a day, there days a week ‘hanging out” on the street or in a mall.

Convenience and snowball sampling was used.

Methods: Data were collected over a six month period (January to June 1993) in Edmonton, Alberta. The study took place around the core business downtown area. The participants were interviews ; questions focussed on property crime, violent crime, drug use and amount of strain, contact with delinquent peers, delinquent norms, and job commitment, self-efficacy and other external attributes.

Findings: Labor market strain was found to be related to the frequency of the youth’s property, violent, and total crimes. Criminal peers and norms were also related to these crimes, as well as to drug use.

Conclusion: This study proved that the measure of strain is successful in predicting property, violent, and total crime. The findings both support and raise questions about strain theory. It appears that labor market strain is an important independent influence on deviant behaviour. It is important to recognise that strain may be conditioned by the availability of delinquent peers, delinquent peers, delinquent dispositions, and attributions. This study found that perceptions on unfairness over the lack of employment and money increased crime.

Recommendations: Future work should continue to explore the role of strain and its conditioning variables on crime and delinquency, utilizing a broader age range of samples drawn from both street and conventional populations. Additional measures of strain, including the removal of positively valued stimuli, the presentation of negatively valued stimuli, and relative deprivation must still be examined within the street population, and the labor market strain measures should be applied to more conventional populations. Further, it may be important to take into account that opportunity is shaped by broader contexts. It might be useful to include measures such as employment rate, quality of jobs, access to transportation, and travelling time to work to contextualize the area labor market. Further, longitudinal tests need to be undertaken.