Measuring Crime and Criminality

John MacDonald

• Publisher: Taylor & Francis
• Year: 2017
• eBook ISBN: 9781351506403
• Edition: 1
• Topic: Sozialwissenschaften
• Subtopic: Kriminologie

Measuring Crime and Criminality focuses on how different approaches to measuring crime and criminality are used to test existing criminological theories. Each chapter reviews a key approach for measuring criminal behaviour and discusses its strengths or weaknesses for explaining the facts of crime or answers to central issues of criminological inquiry. The book describes the state of the field on different approaches for measuring crime and criminality as seen by prominent scholars in the field. Among the featured contributions are: The Use of Official Reports and Victimization Data for Testing Criminological Theories; The Design and Analysis of Experiments in Criminology; and Growth Curve/Mixture Models for Measuring Criminal Careers. Also included are papers titled: Counterfactual Methods of Causal Inference and Their Application to Criminology; Measuring Gene-Environment Interactions in the Cause of Antisocial Behaviour and What Has Been Gained and Lost through Longitudinal Research and Advanced Statistical Models? This volume of Advances in Criminological Theory illustrates how understanding the various ways criminal behaviour is measured is useful for developing theoretical insights on the causes of crime.

1

The Self-Report Method and the Development of Criminological Theory

Terence P. Thornberry and Marvin D. Krohn

The core purpose of etiological theories of crime is to explain why some people engage in criminal behavior while others do not. There are many secondary purposes, for example, to explain why some people engage in more or less serious criminal behavior than others and why they do so at different rates and periodicities over their life course. There can, of course, be more refined or specialized questions; for example, why do some people commit certain types of offenses like violence or sex offenses, or how do certain contexts facilitate criminal behavior? But these specialized issues do not really alter the core issue—explaining why some people engage in criminal behaviors.

Criminologists have adopted a wide range of theoretical perspectives—from genetic models to broad macro-social perspectives—to address this core objective. They have also used a variety of research designs including the examination of official records, cross-sectional and longitudinal surveys, experimental designs, and, more recently, counterfactual approaches to causality. Indeed, many of these approaches are discussed in this volume. But whatever theory is adopted or research approach is taken, one central ingredient is required if we are to make progress in our scientific understanding of criminal behaviors: a valid measure of actual involvement in criminal behavior. That is, we need to know as accurately as possible whether each person in a study sample actually engaged in criminal behavior. Absent valid information on the behavior of interest, it is hard to see how science can progress.

Measuring actual involvement in crime is a difficult task. Crime, by its very nature, is a secretive and sensitive topic. Those who engage in crime are liable for punishment and not likely to broadcast their activities. Thus, in criminology, the scientific need for full and accurate information to both describe and explain the phenomenon is hampered by the very characteristics of the behavior to be explained. This difficulty does not obviate the scientific necessity of generating valid and accurate measurement, however. It is a challenge that must be overcome if scientific knowledge is to advance.

Criminologists have adopted several approaches to the measurement of criminal behavior. They include the use of direct observation of criminal acts, life histories of criminals, the examination of official records such as arrest reports, self-report surveys, victimization surveys, biochemical assays such as urinalysis, and third-party reports, for example, from parents. The present chapter focuses on the self-report method and assesses its advantages for both informing and empirically assessing theories of criminal behavior. Although far from perfect, our core contention is that self-report data are currently the best single source of data to use in developing and testing etiological theories of crime.

We begin with a brief history of the self-report method and key requirements to generate a psychometrically acceptable measure. Following that, we return to an assessment of the advantages of the self-report measure for theoretical criminology.

Historical Overview

The development and widespread use of the self-report method of collecting data on delinquent and criminal behavior is one of the most important innovations in criminological research in the twentieth century. Initially, there was great skepticism about whether respondents would be willing to tell researchers about their participation in illegal behaviors. The earliest studies by Porterfield (1943) and by Wallerstein and Wylie (1947) found that not only were respondents willing to self-report their delinquency and criminal behavior, but also they did so in surprising numbers. Since those very early studies, the self-report methodology has become much more sophisticated in design, making it more reliable and valid, and extending its applicability to many criminological issues.

Although the self-report method began with the contributions of Porterfield (1943, 1946) and Wallerstein and Wylie (1947), Short and Nye (1957, 1958) “revolutionized ideas about the feasibility of using survey procedures with a hitherto taboo topic” and changed how the discipline thought about delinquent behavior itself (Hindelang et al., 1981: 23). Short and Nye’s research is distinguished from previous self-report measures in its attention to methodological issues, such as scale construction, reliability and validity, and sampling, and their explicit focus on the substantive relationship between social class and delinquent behavior. Short and Nye did not find social class to be inversely related to delinquency based on their self-report scale. Other studies in the late 1950s and early 1960s also used self-reports to examine the relationship between social status and delinquent behavior (e.g., Reiss and Rhodes, 1959; Clark and Wenninger, 1962; Akers, 1964; Empey and Erickson, 1966; Gold, 1966; Voss, 1966). With few exceptions, these studies supported the general conclusion that if there were any statistically significant relationships between measures of social status and self-report delinquent behavior, they were weak and clearly did not mirror the findings of studies using official data sources. This failure to find a relationship between social status and delinquency served to question extant theories such as anomie theory (Merton, 1938) and social disorganization theory (Shaw and McKay, 1942) that were built on the assumption that crime was most prevalent among the lower class. It is the first clear example of how the self-report method changed the theoretical landscape.

The absence of a relationship between class and delinquency, especially when compared to the strength of the relationship using official data, raises fundamental questions about the operations of the juvenile and criminal justice systems. The contradictory findings from these two data sources were one of the primary factors in the development and prominence of labeling theory (e.g., Lemert, 1951; Becker, 1963) during the middle part of the twentieth century.

During this same period, researchers began to recognize the true potential of the self-report methodology. By including questions concerning other aspects of an adolescent’s life, as well as a self-report delinquency scale on the same questionnaire, researchers could explore a host of etiological issues. Theoretically interesting issues concerning the family (Nye and Olson, 1958; Dentler and Monroe, 1961; Voss, 1964; Stanfield, 1966; Gold, 1970), peers (Short, 1957; Erickson and Empey, 1963; Reiss and Rhodes, 1964; Voss, 1964; Matthews, 1968; Gold, 1970) and the school (Reiss and Rhodes, 1963; Elliott, 1966; Polk, 1969; Gold, 1970; Kelly, 1974) emerged as the central focus of self-report studies. The results of these studies led to a shift in emphasis in theoretical criminology, away from macrolevel explanations for crime to more individual and family-based theories. The self-report method was used both to develop and to test social control theory (Hirschi, 1969; Hindelang, 1973; Conger, 1976), social learning theory (Akers et al., 1979), self-concept theory (Kaplan, 1972; Jensen, 1973), strain theory (Elliott and Voss, 1974; Johnson, 1979), and deterrence theory (Waldo and Chiricos, 1972; Anderson et al., 1977; Jensen et al., 1978).

Most of these self-report studies were cross-sectional in design. The next major step in the development of this measurement approach, and its next major impact on theoretical criminology, came with the introduction of longitudinal self-report studies. The first major longitudinal self-report study is the National Youth Survey (NYS), conducted by Elliott et al. (1979). The NYS began in 1976 surveying a national probability sample of 1,725 youth aged eleven through seventeen and has followed them well into their thirties. The survey design was sensitive to a number of methodological deficiencies of prior self-report studies and has been greatly instrumental in improving the self-report method. Since that time, an increasing number of panel studies have collected self-report data from sample members over long portions of the life course. They include the three studies of the Program of Research on the Causes and Correlates of Delinquency conducted in Denver, Pittsburgh, and Rochester (Thornberry and Krohn, 2003) as well as similar studies conducted in Seattle (Hawkins et al., 2003), Montréal (Tremblay et al., 2003), and other cities.

Along with criminal career research based on official data (Wolfgang et al., 1972; Blumstein et al., 1986), these longitudinal self-report studies have changed our understanding of patterns of delinquency and reshaped theoretical criminology. Instead of focusing on static attributes of delinquency—for example, the level, variety, or seriousness of offending—longitudinal studies using the self-report method added a more dynamic orientation to the study of delinquency and crime. New descriptions of crime emerged including issues such as age of onset, the distribution of the age–crime curve, length of careers, trajectories or patterns of offending, intermittency, desistance, and so forth. Descriptions of these new and different aspects of offending careers required a shift in theoretical attention.

The application of the self-report method within longitudinal designs coupled with the conceptual framework developed by Glen Elder (1997) facilitated the emergence of life-course criminology as the dominant theoretical perspective in the latter part of the twentieth century (see, e.g., the summaries in Thornberry, 1997, and Farrington, 2005). Traditional theories had focused almost exclusively on between-individual explanations for offending, attempting to explain why delinquents differed from nondelinquents. Developmental theories, while interested in that issue, became increasingly focused on within-individual differences in attempting to explain changing patterns of offending over the life course. For example, there was an increased focus on explaining the link between early-onset offending and later criminal careers, explaining the notable level of continuity in offending over the life course, explaining different patterns of offending such as desistance and late bloomers, explaining the developmental consequences of involvement in delinquency, and so forth. Rather than focusing primarily on adolescent influences as traditional theories had, life-course explanations examine the origins and consequences of delinquency and crime over the full life course, from childhood through adulthood.

In sum, the advent and development of the self-report method has influenced the shape and development of theoretical criminology in several ways. The influence that the method had on changes in criminological theories was intimately linked to improvements in the scope and quality of self-report measures. We turn now to a discussion of how the self-report method itself developed over time.

Development of the Self-Report Method

Since the introduction of the self-report method by Short and Nye (1957), considerable attention has been paid to the development and improvement of the psychometric properties of self-report measures of delinquent behavior. The most sophisticated and influential work was done by Elliott and his colleagues (Elliott and Ageton, 1980; Elliott et al., 1985; Huizinga and Elliott, 1986) and by Hindelang et al. (1979, 1981). From their work, a set of characteristics for acceptable, that is, reasonably valid and reliable, self-report scales has emerged. Five of the most salient of these characteristics are the inclusion of (1) a wide array of offenses, including serious offenses; (2) frequency response sets; (3) screening for trivial behaviors; (4) application to a wider age range; and (5) the use of longitudinal designs. Each of these design features will be briefly discussed here; a fuller discussion can be found in Thornberry and Krohn (2001).

Inclusion of a Wide Array of Delinquency Items

The domain of crime covers a wide range of behaviors, from petty theft to aggravated assault and homicide. If the general domain of delinquent and criminal behavior is to be represented in a self-report scale, it is necessary for the scale to cover that same wide array of human activity. Simply asking about a handful of these behaviors does not accurately represent the theoretical construct of crime. In particular, it is essential that a general self-report delinquency index tap serious as well as less serious behaviors. Early self-report scales tended to ignore serious delinquent acts and concentrated almost exclusively on minor forms of delinquency. In part, this was due to concern about having a sufficient number of serious offenders in the sample to allow for reliable conclusions. With the increased concern for surveying adolescents at risk for serious and violent delinquency (see Thornberry and Krohn, 2003), more serious offenders could be included. Failure to include serious offenses misrepresents the full domain of delinquency, contaminates comparisons with other data sources, and impedes the testing of many delinquency theories that set out to explain serious, chronic delinquency.

Inclusion of Frequency Response Sets

Many early self-report studies relied on response sets with a relatively small number of categories that tend to censor high-frequency responses. For example, Short and Nye (1957) used a four-point response with the highest category being “often.” Aggregated over many items, the use of limited response sets had the consequence of lumping together occasional and high-rate delinquents, rather than discriminating between these behaviorally different groups. Taking into account the full range of responses had implications for the relationship between demographic factors and delinquency (Elliott and Ageton, 1980) and, in turn, theories in which such factors play a significant role.

Inclusion of Screening for Trivial Behaviors

Self-report questions have a tendency to elicit reports of trivial acts that are very unlikely to elicit official reactions and even acts that are not violations of the law. This not only occurs more frequently with less serious offenses but also affects responses to serious offenses. For example, respondents have included such pranks like hiding a classmate’s books in the respondent’s locker between classes as theft, or events that are merely roughhousing between siblings as serious assault. Two strategies are generally available to address this issue. First, one can ask a series of follow-up questions designed to elicit more information about the event, such as the value of the property stolen, the extent of injury to the victim, and the like. Second, one can use an open-ended question asking the respondent to describe the event and then probe to obtain information necessary to classify the act. Both strategies have been used with some success.

Application to a Wider Age Range

With increasing emphasis on the study of crime across the entire life course, self-report surveys have had to be developed to take into account both the deviant behavior of very young children and the criminal behavior of older adults. The behavioral manifestations of illegal and antisocial behaviors can change, depending on the stage in the life course at which the assessment is taking place. For the very young child, measures have been developed that are administered to parents to assess antisocial behavior such as noncompliance, disobedience, and aggression (Achenbach, 1992). For the school-age child, Loeber et al. (1993) have developed a checklist that expands the range of antisocial behavior to include such behaviors as stubbornness, lying, bullying, and other externalizing problems. Unfortunately, there has not been as much development of instruments targeted specifically at adults (Weitekamp, 1989), and many studies simply use an upward extension of adolescent instruments.

Use of Longitudinal Designs

Perhaps the most significant development in the application of the self-report methodology is its use in following the same subjects over time in order to account for changes in their criminal behavior. This has...

Table of contents

• Cover
• Half Title
• Title Page
• Copyright Page
• Table of Contents
• Introduction: The Measurement of Crime and Estimating Treatment Effects in Criminology
• 1. The Self-Report Method and the Development of Criminological Theory
• 2. Socially Desirable Response Bias in Criminology: An Example of Its Effect in Testing the Effects of Self-Control
• 3. How Do We Measure the Severity of Crimes? New Estimates of the Cost of Criminal Victimization
• 4. Communities and Crime Theories: Construct Validation Deficits, a Paucity of Comparisons, and a Boudon–Coleman Metamodel Resolution
• 5. The Coming of a Networked Criminology?
• 6. What Can Genetically Informed Research Tell Us about the Causes of Crime?
• 7. Bounding Disagreements about Treatment Effects with an Application to Criminology
• 8. Randomized Experiments and the Advancement of Criminological Theory
• 9. Causal Inference via Natural Experiments and Instrumental Variables: The Effect of “Knifing Off” from the Past
• 10. Criminal Career Research: A Statistical and Substantive Comparison of Growth Modeling Approaches
• 11. Understanding Desistance: Theory Testing with Formal Empirical Models
• 12. Meta-Analysis and the Relative Support for Various Criminological Theories
• Contributors
• Index