Longitudinal Research in Criminology

by

Steven Zane

• LAST REVIEWED: 24 September 2020
• LAST MODIFIED: 24 September 2020
• DOI: 10.1093/obo/9780195396607-0287

Introduction

Broadly, longitudinal research is research that involves longitudinal data, that is, data with a time dimension. This can be contrasted with cross-sectional data, which records information about the units of analysis at a particular point in time. Longitudinal research in criminology can be used for a variety of purposes, including quantifying trends in human behavior over time, describing the progression of life events, identifying patterns of behavior change, testing theories of crime causation, and evaluating the impact of interventions including criminal justice policy. Different types of longitudinal studies can be used for each of these aims. Panel studies look at multiple waves of data collection using same measures and sample. Cohort studies involve a particular group of individuals that are studied over time, such as a “birth cohort” or a “prison cohort.” Time-series studies involve a series of measurements taken at periodic time intervals in order to measure the impact of a change (such as a policy intervention) by comparing before- and after- measurements of the phenomena of interest. Trend studies look at change over time in a total population or sample that is generalizable to that population. In what follows, longitudinal research in criminology is reviewed with an emphasis on prospective cohort designs. First, an overview of longitudinal research and the methods for analyzing longitudinal data is first provided. Next, major longitudinal cohort studies are discussed, including early (pre-1970) and later (post-1970) cohort studies, long-term follow-ups of prospective cohort studies, and major longitudinal-experimental designs. These studies have allowed researchers to control for possible cohort effects, i.e., similarities within the group, when examining patterns in offending over time. This is especially useful for “life course” researchers who are interested in how a sequence of socially defined events affects individuals over time (i.e., age and period effects). Applications of longitudinal research are then discussed. Most notably, this involves developmental and life-course criminology, which requires longitudinal research to examine criminal careers, early risk factors for offending, offending trajectories, and adult transitions and desistance. Additionally, longitudinal research has been utilized to examine intergenerational transmission of crime, to test major theories of crime, to assess the impact of criminal justice policy, and to examine aggregate trends in crime and punishment. Each of these is briefly discussed.

General Overviews

Longitudinal research is an expansive topic, with many important distinctions such as type of longitudinal design (e.g., trend, cohort, panel, time-series) as well as the different uses of longitudinal research for crime. There are several helpful introductions to longitudinal research in the social sciences, such as Elliot, et al. 2008. Deschenes 1990 provides a good introduction in the particular context of criminology. Menard 2002 provides a more technical discussion of longitudinal research designs in criminology, and Menard 2008 is a more comprehensive account of longitudinal research in criminology in the form of a handbook. Liberman 2008 is a useful reference for major topics in criminology that have relied on longitudinal research.

• Deschenes, Elizabeth P. 1990. Longitudinal research designs. In Measurement issues in criminology. Edited by Kimberly L. Kempf, 152–166. New York: Springer-Verlag.

  Provides a nice overview of longitudinal research in criminology, including types of longitudinal research (trend studies, panel designs, cohort studies, time-series designs), and uses of longitudinal research for examining trends, identifying patterns of change, testing theories, and evaluating interventions. Also includes the history of the debate regarding the use of longitudinal versus cross-sectional research in criminology.

• Elliot, Jane, Janet Holland, and Rachel Thompson. 2008. Longitudinal and panel studies. In The SAGE handbook of social research methods. Edited by Pertti Alasuutari, Leonard Bickman, and Julia Brannen, 228–248. Thousand Oaks, CA: SAGE.

  Chapter that describes longitudinal studies in social science generally, including prospective versus retrospective designs, cohort studies, quantitative versus qualitative longitudinal research, and different analytical approaches for quantitative longitudinal data.

• Liberman, Akiva M. 2008. The long view of crime: A synthesis of longitudinal research. New York: Springer-Verlag.

  This book serves as an excellent reference for longitudinal research in criminology. It includes six chapters, each devoted to a special topic: trajectories of criminal activity over the life course; early childhood and development of delinquency; adolescent street gangs; transitions to adult roles; work and crime; the effects of justice system sanctions; and using longitudinal research to address knowledge about causes. In total, chapters cover sixty longitudinal studies in criminology.

• Menard, Scott. 2002. Longitudinal research. 2d ed. Quantitative Applications in the Social Sciences 76. Thousand Oaks, CA: SAGE.

  Part of the SAGE series in quantitative application in the social sciences, this is an excellent and short graduate-level introduction to various aspects of longitudinal research, including an overview of the purposes of longitudinal research, an explanation of different kinds of designs for longitudinal data collection, major issues in design and measurement in longitudinal research, and overview of different analytical approaches for longitudinal data.

• Menard, Scott, ed. 2008. Handbook of longitudinal research: Design, measurement, and analysis. New York: Elsevier.

  Authoritative guide for issues of design, measurement, and analysis in longitudinal research, written by a criminologist. Chapters on measurement issues cover topics such as panel conditioning, panel attrition, and non-ignorable non-response. Chapters on analytical methods cover such topics as survival analysis, latent class analysis, and various approaches that utilize structural equation modeling and multilevel modeling, such as panel analysis and latent growth curve models.