Rationalizing Delinquency: Understanding the Person-situation Interaction through Item Response Theory

• Author: Kyle J. Thomas
• DOI: 10.1177/0022427818789752
• Published date: 01 February 2019
• Date: 01 February 2019
• Subject Matter: Articles

Article

Rationalizing

Delinquency:

Understanding

the Person-situation

Interaction through

Item Response Theory

Kyle J. Thomas

1

Abstract

Objectives: I argue that a person-situation complex of delinquent rationaliza-

tions can be conceptualized by relatingrationalizations to item responsethe-

ory (IRT), where approval of delinquency is predominately a function of the

individual willingness to rationalize (yj) and situational difficulty o f applying a

rationalization (b

i

). This framework offers testable predictions and addresses

extant criticisms. Method: Adolescents from a public high school (N¼223)

and subjects from the National Youth Survey (N¼1,436) were asked their

degree of approval for delinquency under various circumstances. Graded

response models assessed the joint effects of individual and situational

characteristics on approval of delinquency. I test whether diffe rences in

self-reported offending (SRO) and willingness to offend (WTO) are con-

sistent with predictions derived from IRT models. Results: Approval of

1

Department of Criminology and Criminal Justice, University of Missouri–St. Louis, St. Louis,

MO, USA

Corresponding Author:

Kyle J. Thomas, Department of Criminology and Criminal Justice, University of Missouri–

St. Louis, 331 Lucas Hall, St. Louis, MO 63121, USA.

Email: thomaskj@umsl.edu

Journal of Research in Crime and

Delinquency

2019, Vol. 56(1) 3-41

ªThe Author(s) 2018

Article reuse guidelines:

sagepub.com/journals-permissions

DOI: 10.1177/0022427818789752

journals.sagepub.com/home/jrc

delinquency is a joint function of individual and situational characteristics.

Some situations are so “easy” to rationalize that most everyone is predicted

to approve of delinquency, and others are so “difficult” that only those very

high in yare predicted to express approval. SRO and WTO differences

between individuals and situations are consistent with the IRT predictions.

Conclusion: The findings demonstrate the utility of IRT for understanding

delinquent rationalizations. The implications of the findings for theory and

person-situation explanations are discussed.

Keywords

rationalizations, item response theory, delinquency

Attitudes favorable to delinquency are not typically “oppositional values

that ...make crimes morally correct” but situational rationalizations that

allow individuals to justify behavior in specific circumstances (Matsueda

1988:296; Sykes and Matza 1957). Sutherland (1973), for example, argued

that delinquency is the result of a person-situation interaction comprised of

(1) one’s ability to rationalize delinquency and (2) the situational circum-

stances where rationalizations can be applied. Cressey (1954) stated that

situational rationalizations have advantages over other explanations of

crime because it can explain both between-individual variation in offending

and within-person variation across situations (see also Wikstro¨m 2006).

Indeed, the concept of “rationalizations” has been incorporated into many

prominent explanations of crime (Akers 1998; Laub and Sampson 2003;

Matza 1964; Sutherland 1973; Sykes and Matza 1957; see also Maruna and

Copes 2005). But beyond acknowledging that situations are important for

reasons beyond the objective opportunities provided (Sutherland 1947),

rationalization theorists have not linked individual and situation character-

istics in a framework that describes how the two jointly affect the approval

of delinquency.

The failure to clearly incorporate situational elements into the concep-

tualization of delinquent rationalizations has had both theoretical and meth-

odological consequences. Theoretically, the inclusion of situational

elements into rationalization theories has been described as “not particu-

larly convincing” (Costello 1 997:425), as having “low genera lizability”

(Birbeck and LaFree 1993:122), and as an ad hoc attempt to account for

attitude-behavior incongruence (Glueck 1956; Hirschi 1996). Methodolo-

gically, the lack of clarity as to how individuals and situations interact to

4Journal of Research in Crime and Delinquency 56(1)

affect the approval of behavior has led to inconsistencies in measurement.

While some scholars have called for measures that ask respondents to

evaluate the appropriateness of behavior under various situational circum-

stances (Thomas2018; see also Fishbein and Ajzen 1975), “rationalizations”

have been captured using moral belief items that ask about one’s abstract

attitudes toward a behavior and incorporate no situational information

(Rebellon et al. 2014). Moreover, some scholars have assumed that rationa-

lizations are directly observable and operationalized the construct using a

summated scale,while others have argued that itis a latent constructreflected

in responses to items (Matsueda 1988). These two issues are related, as a

clearer conceptualization of the person-situation complex of delinquent ratio-

nalizations can lead to greater consistency in its operationalization.

The difficulty in linking individual attitudes and situational circumstances

lies in the fact that no two situations are completely identical, and one must

identify general characteristics that vary across situations and then link these

to the individual evaluation of behavior. I suggest that this can be accom-

plished by drawing on Matza (1964), who argued that the excuses used to

rationalize delinquency are embedded in the society and, thus, most everyone

has the potential to rationalize crime. There are, however, individual differ-

ences in the underlying willingness to apply these excuses, where some may

rationalize only in extreme situations (defending oneself from assault) and

others rationalize in a broader range of circumstances (challenging one’s

honor). It follows that there is situational variation in the ease at which a

rationalization can be applied: It is easier to justify hitting someone who hits

you first than hitting someone who gives you a dirty look. Taken together,

there are between-individual and between-situation differences in the ten-

dency to rationalize because there are differences in the (1) individu al will-

ingness to rationalize delinquency and (2) situational difficulty of applying

a rationalization. This leads to another possibility: That individuals’

approval of delinquency in some situations is more indicative of their

underlying willingness to rationalize that others. There are some situations

which may be so easy or difficult to rationalize delinquency that virtually

everyone holds similar views on the behaviors appropriateness. Other

situations, however, are better at discriminating between individuals high

and low in the underlying willingness to rationalize delinquency, suggest-

ing that one’s underlying willingness to rationalize is likely to be a stronger

predictor of behavior in some situations rather than others.

Based on this conceptualization, the joint effects of the individual will-

ingness to rationalize and the situational difficulty of rationalizing can be

mathematically linked and operationalized through item response theory

Thomas 5