Orientation Observation In-depth interviews Document analysis and semiology Conversation and discourse analysis Secondary Data Surveys Experiments Ethics Research outcomes



Social Research Glossary

About Researching the Real World



© Lee Harvey 2012–2020

Page updated 29 April, 2020

Citation reference: Harvey, L., 2012–2020, Researching the Real World, available at
All rights belong to author.


A Guide to Methodology

CASE STUDY Religion and Delinquency

Travis Hirschi and Rodney Stark (1969) undertook a study of the effect of religion on delinquency. They argued that relations between religiousness and delinquency offer a critical test of the relevance of religion. If religion is immaterial in an area where it has concentrated its efforts then its failure seems to be acute.


Despite the assumptions of law enforcers, the judiciary and the clergy in the United States that religion has a beneficial effect on delinquency, empirical studies are less conclusive. Unfortunately, there is a

relation between the findings and the religiosity of the researcher. While most studies conducted by criminologists suggest that religion has little or no effect, research by religionists tends to indicate that religion is just what it has always been thought to be, a powerful “aid to the sword” in the maintenance of conformity, a factor in delinquency at least equal to the variables traditionally considered important by criminologists. (Hirschi and Stark, 1969, p. 203)


The aim of the study was to see if religious belief and affiliation had any effect on delinquency.


Hirschi and Stark assessed functionalist analyses of religion. Functionalists argue that religion functions to maintain social control. Religious sanctioning systems play an important role in ensuring conformity to social norms (Durkheim, 1912; Davis, 1948). The Durkheimian view is that religion plays this role:

  • by legitimating social and individual values;
  • through rituals that reinforce commitment to these values;
  • through a system of eternal reward and punishment that helps to ensure the embodiment of these values in actual behaviour.

Religious ritual has a central place in functionalist analyses. According to Hirschi and Stark, involvement in ritual promotes:

      • the internalisation or acceptance of moral values including the belief that people deserve fair and just treatment;
      • acceptance of the legitimacy of legal authority;
      • belief in the literal existence of a supernatural world and therefore the belief that one may be punished in the world to come for violations in this world.

Belief in supernatural sanctions is presumed by sociologists to promote and maintain conformity. For example, Edward Ross (1920) has argued that Christianity has made the doctrine of future life a strong deterrent influence.

Hirschi and Stark accept that people probably do have internal values and ethics that sometimes govern their actions. What they questioned was whether religion has anything to do with sustaining these personal values. Does belief in hell-fire and heavenly glory deter people from unlawful acts?


Hirschi and Stark had three main hypotheses:

1. Religious training prevents delinquency by promoting the development of moral values.

2. Religious training prevents delinquency by promoting the acceptance of conventional authority.

3. Religious training prevents delinquency because it promotes belief in the existence of supernatural sanctions.


These hypotheses involve five concepts: delinquency, morality, conventional authority, belief in supernatural sanctions, and religiosity.


Delinquency was operationalised using the following self-report questions:

1. Have you ever taken little things (worth less than $2) that did not belong to you?

2. Have you ever taken things of some value (between $2 and $50) that did not belong to you?

3. Have you ever taken things of large value (worth over $50) that did not belong to you?

4. Have you ever taken a car for a ride without the owner's permission?

5. Have you ever banged up [destroyed] something that did not belong to you on purpose?

6. Not counting fights that you may have had with a brother or sister, have you ever beaten up on [sic] anyone or hurt anyone on purpose?

These represented operationalisations of petty larceny [theft], grand larceny, auto theft, vandalism and assault. Respondents were asked to date any acts that fell into these categories. Those who admitted to two or more offences in the year prior to the survey are referred to below as the 'delinquent' group.

Morality was operationalised through two statements that the respondents were asked to agree or disagree with:

7. To get ahead, you have to do some things that are not right.

8. Suckers deserve to be taken advantage of.

Wordly authority was operationalised using the following two statements:

9. It is alright to get around the law if you can get away with it.

10. I have a lot of respect for the local police.

Belief in supernatural sanctions was operationalised through the following two statements:

11. There is life beyond death.

12. The devil actually exists.

'High believers' were those people who agreed with both statements.

Religiosity was operationalised through church attendance. This was in accord with the theoretical view about involvement in religious ritual.

There were five categories of church attendance:

A. Once a week
B. 2–3 times a month
C. Once a month and/or religious holidays
D. Hardly ever
E. Never.


The sample was drawn from all students entering the public junior and senior high schools in Western Contra Costa County, California, in the autumn of 1964. The original sample was 5,545 students, 4,077 (74%) completed the questionnaire. This total sample consisted of 1,588 white boys, 1,001 black boys, 675 white girls and 813 black girls.


Hirschi and Stark controlled for gender and 'race' in their study and presented data separately for black and white, boys and girls. For illustrative purposes only the data for the white boys will be used.


'Delinquents' (white boys who admitted committing two or more delinquent acts) were more likely to agree than disagree with amoral statements (No. 6 and No. 7).

They were also more likely to feel it was 'all right to break the law if they could get away with it' (44% of delinquents agreed with this compared to 13% who disagreed).

Delinquents were also three times more likely to disagree than agree that they had respect for the local police.

For each of these statements the gamma coefficient yyyyyyyyyy is reasonably high. The gamma coefficient is a measure of association. The nearer gamma is to 1 the higher the relationship between delinquency and attitude. The nearer gamma is to 0 the lower the relationship.

These results suggest that delinquency is fairly strongly related to attitudes about morality and authority. This is what you might expect. The question is, to what extent does religion effect this relationship?

Hypothesis 1 suggests that religion prevents delinquency by promoting the development of moral values. However, high church attenders are virtually no different from low church attenders in agreeing with the statements about morality (Nos. 7 and 8).

Hypothesis 2 suggests that religious training prevents delinquency by promoting the acceptance of conventional authority. Again, high church attenders are not much different from low attenders in their attitude towards law and the police.

So church attendance does not seem to have any effect on attitudes and hypotheses 1 and 2 are rejected.

Hypothesis 3 suggests that religious training prevents delinquency because it promotes belief in the existence of supernatural sanctions. The results suggest church attendance is related to belief in supernatural sanctions (49% of high attenders are high believers while only 12% of low attenders are high believers). However, there is no relationship between delinquency and belief in supernatural sanctions. So hypothesis 3 is also rejected.

Overall, then, it appears that religiosity has no effect on delinquency. This is further confirmed by the very small difference in percentage delinquents (those admitting two or more offences) for different church attendance groups.

Reflective Activity
1. Are you convinced by the logic of the analysis? If not, what do you think is wrong with it?
2. Do you think that it is reasonable to do the analysis using as delinquents those who have admitted two or more offences in the last year?
3. Would it have been preferable to have simply compared church attenders with church non-attenders?

Hirschi and Stark argued that the data for all four groups supported the findings that religiosity had no effect on delinquency. This raises doubts about the functionalist view of religion. As religion appears to have no impact on delinquency, the idea that religion serves to maintain social control is undermined. This is reinforced by the apparent lack of effect that belief in religious sanctions has on ensuring conformity to social norms.

Note that the sample questions above are numbered sequentially but this does not correspond to the numbering in the original (where there were more questions). In addition, as a check, Hirschi and Stark examined police records of reported delinquency and there were considerable discrepancies between the self-reported delinquency and police records. However, the overall results were very similar when the authors used the police data as when they used the self-reported data.

Functionalists provide only one view of the role of religion. Marxists argue that religion has traditionally been used to oppress people. It has been used as a direct instrument of social control (for example, the inquisition in Spain, fundamentalist religious courts in Iran and so on). Religion also provides legitimacy for unequal and oppressive social structures. It thus has an ideological role. However, it has been argued that, in Western capitalist societies, the once-dominant role of religion as an ideological tool has been replaced by education (Althusser, 1971).


Return to Falsificationism and middle-range theorising (Section 8.2.3)

Return to Operationalisation (Section 8.3.5)