Methods of Criminological Research

Victor R Jupp

• Publisher: Routledge
• Year: 2012
• ISBN:
  9781134858309
• Edition: 1
• Topic: Social Sciences
• Subtopic: Sociology
• Index: Social Sciences

Examining the different ways in which data can be collected and analyzed for research on crime and criminal justice, this book deals with social surveys, experimental methods, official statistics, observation and detailed interviews. This practical text includes sections on:

* methods of criminological research
* types of data
* measuring and explaining crime
* studying the criminal justice system
* gaining access
* publishing results.

Throughout, it emphasizes the necessity of examining forms of data collection and analysis within the context of the criminological problems being investigated, the theoretical approaches used to address these problems, and the political and institutional contexts within which research takes place.

1

Theories, Methods, Politics and Problems

---

Introduction

The influence of positivism

Modern criminology has its roots in the mid-nineteenth century and particularly in the challenge to classical thinking represented by positivism. Early classical thinking emphasized free will and therefore portrayed crime as the outcome of voluntary actions based upon rational calculation. It was suggested that individuals committed crimes when they saw the benefits of law-breaking as far outweighing the costs or potential costs. Positivism succeeded in portraying an altogether different conception of crime and also in providing a different basis for its explanation. For example, crime was seen as something into which the individual was propelled by factors largely beyond his or her control and not as an activity into which he or she could freely enter after careful and rational balancing of costs and benefits. Thus positivism involved forms of explanation based upon determinism and the search for causes. Crime and criminality were dependent variables to be explained, and the search was for explanatory or independent variables upon which crime and criminality could be said to be dependent.

There is a further, but related, way in which classicism and positivism differed. With its emphasis upon rational calculation, classical thinking placed the focus upon the means by which the operation of the criminal justice system could increase the costs of criminal activity in relation to the benefits to the criminal. The main thrust of positivism was different. By focusing on deterministic explanations of crime and criminality, early positivists were less concerned with systems of justice and more with locating the ‘propelling forces’ to crime, particularly those assumed to lie within the individual. This is typified in the writings of the Italian biological positivists such as Cesare Lombroso. Lombroso’s concern was to establish, in a systematic and scientific manner, those characteristics of individuals—particularly innate characteristics—which might be deemed to be the causes of criminality. Criminals, he argued, are born like that. His conclusions were reached by the application of theoretical ideas about predispositions towards crime in certain types of individuals, and also by the use of particular methods of research to collect findings which sought to relate criminal behaviour to such predispositions. By comparing the skulls of criminal and non-criminal men he claimed to have evidence with which to assert that criminals shared certain facial and other physical features, including receding forehead, large jaw, handle-shaped ears, dark skin and thick curly hair (Lombroso, 1911). Such physical features and associated criminal propensities were not only innate but represented features which noncriminal groups had outgrown. The criminal was seen as part of a sub-species of humanity.

From today’s standpoint such conclusions seem bizarre and are usually treated with a certain degree of amusement. Nevertheless, forms of positivist methodology have been influential, albeit with different foci and different levels of analysis, and since these early beginnings of positivism there have been successive surges in the changing, and widening of, criminological explanations. For example what may be termed psychological positivism also has a concern with individual propensities to crime but has focused on aspects of personality and how they interact with learning (for example, Eysenck, 1964) whereas other work has placed greater emphasis upon socialization and upbringing (for example, West, 1967, 1969, 1982; West and Farrington, 1973, 1977). The contribution of sociological positivism has been to attempt to shift the focus from a concern with explanations in terms of individual attributes or experiences to an interest in social structure and, for instance, with the way in which crime can be explained in terms of social disorganization and anomie (see, for example, Merton, 1964).

Psychological and sociological strands of positivism have each had a major influence on social science in general and the study of crime in particular. They have played an important role in the development of criminology and still retain prominence in what has been termed conventional or mainstream criminology (Cohen, 1981). Historically, positivist forms of analysis and of explanation have sought to plant ‘science’, scientific thinking and systematic empirical investigation firmly at the heart of the study of crime. The emphasis upon empirical investigation has meant that positivism has been a major influence on the development of methods of social research, particularly those methods which collect and use ‘hard’, quantitative data. (The term positivism is used here in a very general sense and without reference to the different nuances which can be attached to its meaning. For a detailed discussion of these see Halfpenny, 1982; Bryman, 1988.)

The positivist position has also had secondary influences in terms of the theoretical critiques which have been mounted against it, critiques which have themselves subsequently influenced criminological research: for example, in the late 1960s and early 1970s a number of specific theoretical approaches to the study of crime collected together under the general banner heading of ‘new deviancy’. The influence of these various expressions of new deviancy has been to minimize the importance of deterministic causal explanations of crime in favour of an interest in the role of social meanings and interactions in the social construction of crime. A radical and critical approach to the study of crime subsequently developed out of the new deviancy school. This radical or critical strand within criminology shares with new deviancy the dislike of causal thinking but at the same time contributes a greater concern with social control, the role of the state in crime control and the historical development of social structure. In doing this it has also contributed an interest in the role of empirical criminological investigation in such crime control. What is more, feminist theory and research—which include elements of a radical approach—have sought to rewrite the criminological agenda to give greater emphasis to the reconceptualization of the nature of crime in terms of the structural position of women in society and also to the experiences of women as victims of crime. The various theoretical critiques of positivist explanations have been mirrored by debates about the appropriateness of particular methods of social research (especially those which collect ‘hard’ quantitative data); by the development of methods geared to the subjective and humanistic aspects of crime; by critical analyses of the political uses of criminological research; and by formulation of methods appropriate to feminist research. (In relation to the latter see, for example, Roberts, 1981; Bowles and Duelli Klein, 1983.)

In short, the development of criminology has not been characterized by theoretical or methodological unity. There have been variations in the degree of emphasis which has been given to positivist explanations in causal terms and also in the degree of emphasis which has been given to different units and level of analysis. The latter has seen shifts from a concern with individual crime and criminality to an interest in social structures within which crime is committed and including the historical development of such structures. However, whatever the corners that have been turned, criminology has developed by interactions between theory and the data generated by methods of investigation; by a self-reflective consideration of the respective contributions of theory and method; and via issues, debates and disputes about the way in which they should relate to each other. What is more, in more recent decades there has been critical discussion of the role of criminological investigation and its potential contribution as a mechanism of social control.

The value of considering ‘methods’

Here we are not explicitly concerned with a discussion of what is, or should be, the appropriate territory of the criminological enterprise, nor are we explicitly concerned with the range of broad theoretical approaches and specific theoretical positions which populate this territory. Rather, the aim is to focus on some of the main methods by which criminological research is carried out. However, in this endeavour it would be folly to ignore the central problems of criminology and the theories which are brought to bear on them. On the one hand, the methods that come to be used have implications for the way in which problems are conceptualized and for the type of explanations employed. On the other, the problems and theories of criminology have implications for the kinds of methods that are used. Some methods of research are more useful and appropriate to the investigation of certain aspects of crime than others. For example, detailed insights into the way in which crime is experienced are unlikely to be captured by the short and highly structured format required for self-completion mail questionnaires. What is more, different theoretical positions seem to have preference as to method because of the types of data which can be collected, the level or unit of analysis which is used and the degree of primacy which is to be given to the search for causes.

Bearing this in mind, what is the value of a consideration of method? How one addresses a question such as this depends upon how widely or narrowly method is interpreted. Matters of method can be interpreted rather narrowly as being about the types of data collected by criminological researchers, about the methods by which they collect them and about the ways in which such data are analysed. The technicalities of such matters often remain buried in the main text of criminological writings or are treated cursorily in a brief methodological appendix. Either way they are invariably viewed by all except those with a specialist interest, as matters to be passed over in the rush to get to the central assertions and conclusions. After all, in his classic exposition on the sociological imagination C. Wright Mills enthusiastically and persuasively implored us to avoid the ‘fetishism of method and technique’ (Mills, 1970, p. 246). This edict, however, was not intended to warn us against all method but to warn against an obsession with the matters of method to the exclusion of all others. Indeed, a consideration of social science research methods and the data they collect is important at a number of levels.

In the first place, at a technical level, it facilitates an appreciation of the ways in which empirical investigation is, and can be, carried out. But secondly, and more importantly, it provides a platform from which to generate clues as to the credence which can be placed upon criminological findings and the conclusions which can be erected upon them. For example, official statistics on crime are collected and processed by the Home Office and are published each year in Criminal Statistics. Such statistics provide one means by which researchers can seek to measure the extent of crime in society What is more, they can also provide the basis for explanations of crime by relating crime levels to the features of social areas, such as types of housing tenure (see, for example, Baldwin and Bottoms, 1976) or to other trends in society such as changes in levels of unemployment (see, for example, Tarling, 1982). However, such statistics should not be taken for granted. An understanding of the nature of official statistics on crime and of the way in which they are collected and processed leads to questions about whether or not they can be treated as objective indicators of the level of society’s criminality. Part of the reason for this is that many criminal acts are unknown to the police. What is more, even when known, many criminal acts are not officially recorded. Also, as Kitsuse and Cicourel (1963) and others have argued, crime statistics could be viewed more appropriately as indicators of those organizational processes at work in the criminal justice system which result in the recording of some criminal acts and the non-recording of others. In short, questioning the ways in which data are collected, processed and analysed provides a base for evaluating findings of criminological research. In asking such questions we are asking about the validity of particular research designs and the data they generate and use. We can call this method validity. An assessment of method validity is not an end in itself but a contribution to the overall evaluation of how well social research methods can, and do, capture the social world as it is, or as people think it is.

Methodological validity

Closely related to the above is a third and much more fundamental justification for inspection of the methods of criminological research, one that goes much further than asking whether a particular method of collecting or using data is the most appropriate to the task at hand. This involves a consideration of the theoretical and methodological assumptions implicit in the use of particular methods and designs. All such assumptions are inextricably bound up with issues about the way in which theory and method connect. Two issues have already been identified as being important in the development of criminology and these can be made more explicit here. One of these concerns questions about what should be the appropriate focus and unit of analysis in any explanation of crime—the individual, the social group or the social structure. Whatever the level or unit of analysis chosen there are implications for the type of data which should be collected, for how it can and should be collected, and for the extent to which we can validly jump from data collected at one level (say, from individuals) to the making of assertions about another level (say, collectivities of individuals).

A second issue concerns the degree of credence which should be placed upon explanations founded upon determinism and causal thinking, irrespective of the unit of analysis employed. This is related to the fundamental debate about the appropriateness of the positivist paradigm to the social sciences in general and to criminological research in particular. The positivist paradigm brings with it key assumptions about the nature of social reality and about the way in which it can be investigated. For example, it is assumed that social phenomena such as crime can be treated as objective facts which can be apprehended, equally objectively, by the researcher. What is more, this is linked to the (often taken for granted) assertion that ‘crime-asobjective-fact’ can be quantified by the application of the basic principles of measurement. One typical use of quantification has been in relation to questions about the extent of crime and its social and historical distribution. On the surface, such questions would seem capable of easy resolution by reference to officially recorded crime statistics but, as has already been pointed out, such official statistics record the number of crimes reported to the police but exclude crimes about which the police know nothing or about which the police wish to know nothing. Such statistics might even be better seen as the outcomes of policing practices rather than indices of the extent of crime (see, for example, Box, 1981; Bottomley and Pease, 1986). More fundamentally, the use of official statistics to measure the extent of crime is based upon official definitions of criminal acts and completely skirts the question of what, in the first place, should be treated as crime. For example, it can be argued that criminological research should extend itself beyond legal definitions of crime to encompass the violations of basic human rights implicit in racism and sexism. (For a discussion of differing conceptions of what is crime, including a human-rights viewpoint, see Bottomley, 1979.) It is because of major question marks against the quantification of aspects of crime, and particularly the often unthinking and indiscriminate application of numbers and their subsequent manipulation, that many eschew quantitative strategies in favour of so-called ‘qualitative’ strategies and so-called ‘qualitative’ data. Both the inbuilt quantitative assumptions of positivism and the challenges which have been made to them have had important implications for the range of methods which has been used to conduct criminological research.

Quantification is often closely related to the goal of developing and testing theories of the social world with wide applicability and strong explanatory power. Typically, this goal is founded upon a viewpoint which sees research as a means by which hypotheses and models of the social world can be tested. Hypotheses are formulated (say, asserting that there is a relation between levels of crime in communities and various indicators of deprivation in these communities), and data are collected, perhaps from official statistics, to see how well such hypotheses ‘fit’ the quantified empirical reality. For example, in Chapter 3 we refer to studies summarized by Tarling (1982) which seek to establish statistical relationships between levels of unemployment and levels of crime with which to make inferences about causal connections between these phenomena. The methodological significance of such attempts to provide statistically verified models of the social world lies not just in the range of data and methods which are used but also in the way in which theory and method are seen to relate to each other. Method is typically seen as a handmaiden of theory, that is, as something to assist in the verification of theory.

An alternative strategy is to treat theoretical generalizations as the product of empirical investigation, or at the very least as the outcome of a flexible and continuous interchange between theory and data. This is often, but not always, linked to a belief that ‘qualitative’ data and not quantitative data provide more valid representations of the social world. It is also typically associated with the assertion that social science theories should be grounded in the everyday theories of the people they are studying. Within such a strategy of inquiry the researcher starts by getting immersed in the social world of those whose actions and ideas he or she wants to understand and explain. It is on the basis of such first-hand experience and as a result of data collection and analysis that generalizations about actions, ideas and experiences can be formulated. This is characterized in what Glaser and Strauss (1967) refer to as the discovery-based approach. Basically, this approach eschews any notion of an obligatory and unilinear transition from problem through theory to method. Rather, it advocates moving backwards and forwards between three broad phases of inquiry within each of which there is a constant exchange between problem, theory and method. In Chapter 3 the contrasts between discovery-based, qualitative research and the more formal protocols of quantitative research are illustrated by reference to ethnographic studies of school and youth culture. For example, in his study of kids in Sunderland Corrigan (1979) immersed himself in first-hand collection of data by the use of observational methods and detailed informal interviews in the schoolground and on the streets. One of his central conclusions is that the actions of the kids are only understandable in terms of the kids’ perceptions of differential and unequal power relations between themselves and teachers and police officers. ‘Messing about’ at school and ‘doing nothing’ on the streets are means by which the kids can seek to subvert the power of others. For Corrigan, this conclusion has validity because it emerged spontaneously and naturally from the data and because it was founded upon the perceptions of the subjects themselves rather than imposed from above in the form of some preordained and rigid hypothesis. This grounded and flexible form of analysis implies a more interactive relationship between theory and method than is found in the stringent strategies of hypothesis testing.

The distinctions between positivist and non-positivist research and also between quantitative and qualitative data are easily exag...

Table of contents

• Cover
• Half Title
• Social Research Today edited by Martin Bulmer
• Full Title
• Copyright
• Dedication
• Contents
• Preface
• 1 Theories, Methods, Politics and Problems
• 2 Methods of criminological research
• 3 Measuring and explaining crime
• 4 Studying the criminal justice system
• 5 Conclusion
• References
• Index