Aggregation,ratio variables,and measurement problems in criminological research

Issues of measurement error, level of aggregation, and ratio variables have been considered serious problems in criminological research. Although there have been many recent discussions of these issues in sociology and criminology, studies designed to assess the impact of these problems on the results of empirical research have, for the most part, been absent. After reviewing what is known theoretically and conceptually about these issues, an investigation which compares empirical analyses of a particular type of crime, homicide, that use different measurement strategies, different levels of aggregation, and ratio versus nonratio variables is presented. Utilizing homicide data from the mid-1970s and selected independent variables, the results of this investigation indicate that these three problems can interact in an empirical setting such that potential solutions to these problems do not always apply in the manner suggested in previous studies. The results also indicate that there is great risk in ignoring one or more of these problems in empirical research, in that different substantive conclusions can be reached from analyses that ignore these issues compared with analyses that deal directly with them.     

Constructing an index of officially recorded crime: The use of confirmatory factor analysis

Researchers using officially reported crime statistics must frequently combine rates of individual crimes into a single composite. The most popular composite is the FBI Crime Index, which is the sum of the seven Part I crimes; alternatives include forming separate indices for personal and property crimes and weighting the crime categories by their seriousness before summing them. In this paper we construct crime composites through the use of confirmatory factor analysis methods. The results suggest that the measurement assumptions of conventional crime indices are problematic and demonstrate how more satisfactory alternatives can be developed with confirmatory factor analysis techniques.     

Rap sheets in criminological research: Considerations and caveats

The types of errors found in official criminal history records are not completely understood by many researchers, and this lack of understanding can lead to serious misinterpretations. Analyses of a recently developed database of New Orleans offenders indicate that the use of rap sheets with a limited catchment area can lead to gross distortions of the effects of variables related to geographic mobility, such as race and age. Evidence from a number of sources indicates that false-negative error is a serious problem, particularly in fingerprint-based record systems. In addition, arrest records lend themselves to a variety of common misinterpretations by researchers in the coding process, including failing to identify multievent arrests, misclassifying arrests, and treating arrest or custody process events as crimes indicating criminal activity of the individual while free. Solutions to some of these problems are suggested.     

Instrumental variables methods in experimental criminological research: what, why and how

Quantitative criminology focuses on straightforward causal questions that are ideally addressed with randomized experiments. In practice, however, traditional randomized trials are difficult to implement in the untidy world of criminal justice. Even when randomized trials are implemented, not everyone is treated as intended and some control subjects may obtain experimental services. Treatments may also be more complicated than a simple yes/no coding can capture. This paper argues that the instrumental variables methods (IV) used by economists to solve omitted variables bias problems in observational studies also solve the major statistical problems that arise in imperfect criminological experiments. In general, IV methods estimate causal effects on subjects who comply with a randomly assigned treatment. The use of IV in criminology is illustrated through a re-analysis of the Minneapolis domestic violence experiment. The results point to substantial selection bias in estimates using treatment delivered as the causal variable, and IV estimation generates deterrent effects of arrest that are about one-third larger than the corresponding intention-to-treat effects.     

Structural Models of Injustice and Illegalism in Israel

The question of whether a citizen is obliged to obey the law has occupied scholarly thinking since the days of Socrates. Thus, rarely has the question been examined empirically. Based on a survey of 1866 adults, citizens of the state of Israel, this study attempts to construct causal models linking socioeconomic status to subjective measures of injustice, and further to indicators of illegalism and disobedience among Jews and Arabs. LISREL analysis has been found a useful tool to construct the empirical models that measure how members of the two ethnic communities relate to the boundaries of the prima facie obligation to obey the law. Findings show that those who have no trust in one part of the legal system, will also have no trust in the rest of the system which, in turn, serves to legitimize taking the law into one's own hand.     

Competing ethical paradigms in forensic psychiatry and forensic psychology: Commentary for a special section of legal and criminological psychology

In her paper Gwen Adshead (2013) identifies the principle of respect for justice as the foundation for ethical practice in forensic psychiatry. This commentary discusses how forensic psychologists have approached a key aspect of the justice principle, the ethical imperative to promote the welfare of the individual client. The approaches of the psychiatric and psychological professions are compared and the role of psychologists as allied health professionals working in the forensic setting considered.     

Linking self- and social control with deviance: Illuminating the structure underlying a general theory of crime and its relation to deviant activity

The present paper operationalizes and empirically tests the most recent theoretical speculations of Hirschi and Gottfredson regarding an individual level characteristic of self-control and its relation to earlier specifications of control theory as well as the literature on personality. Linkages are drawn between their broad delineation of self-control and personal disorders of hyperactivity, impulsivity, attention deficits, and minor conduct problems. Psychologists disagree about whether such disorders represent single or multiple traits and whether both behavioral and cognitive measures can appropriately depict personality characteristics. Employing structural equation techniques, support for several propositions derived from Gottfredson and Hirschi's thesis is found: Self-control subsumes several personality disorders and is significantly comprised by early behavioral indicators of aggression and fighting, is inversely related to other elements of the social bond, is moderately stable over a short period of time, and significantly predicts criminal convictions. However, questions remain regarding the ubiquity of self-control, the magnitude and meaning of stability, and the power of this perspective to explain all forms of self-reported delinquency.     

Calibration procedure and the robustness of CGE models: Simulations with a model for Poland

Calibration is commonly used in order to specify a numerical computable general equilibrium model. As it involves the evaluation of model parameters on the basis of a single observation, there is a concern about the reliability of the results obtained from the model. A one-sector model for the Polish economy is specified and calibrated using various years as a benchmark equilibrium. Different assumptions are also made about some elasticity parameters. The model response is examined by simulating an increase in government expenditure. The results are quite robust to the choice of the base year but very sensitive to some elasticity parameters such as trade substitution elasticities.     

Micro-models of criminal careers: A synthesis of the criminal careers and life course approaches via semiparametric mixed poisson regression models,with empirical applications

Much recent research and debate in criminology have centered around how to conceptualize and model longitudinal sequences of delinquent and criminal acts committed by individuals. Two approaches dominate this controversy. One originates in thecriminal careers paradigm, which emphasizes a potentialheterogeneity of offending groups in the general population—thus leading to a distinction between incidence and prevalence of criminal offending, a focus on the onset, persistence, and desistence of criminal careers, and the possibility that criminals are a distinctive group with constant high rates of offending. Another approach places criminal events within a broader context ofstudies of the life course by explicitly substituting the conceptualization of “social events” for that of “criminal careers”. With respect to analytical models, this approach emphasizes a potentialheterogeneity of offenders with respect to order of criminal events from first to second to higher orders and thus suggests an analysis of the “risks” or “hazards” of offending by order of offense. Some extant commentaries on the criminal careers and life course approaches to conceptualizing and modeling longitudinal sequences of delinquent and criminal events committed by individuals have emphasized their differences and incompatibilities. In contrast, we apply recently developed semiparametric mixed Poisson regression techniques to develop conditions under which the two conceptual/modeling approaches are formally equivalent. We also modify the semiparametric mixed Poisson regression model of criminal careers to incorporate information on order of the delinquent/criminal event and develop an empirical application. This modification demonstrates the complementarity of the criminal careers and life course approaches, even though they have somewhat different foci.     

Unemployment,imprisonment, and the stability of punishment hypothesis: Some results using cointegration and error correction models

A version of the stability of punishment hypothesis is used to illustrate the concept of cointegration and its relationship to error correction models. The hypothesis is then tested and rejected using data from England and Wales. Finally, a dynamic time-series model relating imprisonment to convictions, crime, and Unemployment is developed and tested.The empirical results in this work are based upon those originally prepared for a paper to be given by the author and Steve Box at the meeting of the American Society of Criminology held in Montreal in November 1987. Steve Box died in September 1987 before it could be completed.The paper is dedicated to his memory and it is hoped that he would have approved its content, if not, perhaps, the dryness of its form.     

Applications and extensions of subpopulation theory: a caseworkers guide

This paper extends the calculation of conditional probabilities from those given by Balding and Nichols to casework situations where a series of possible DNA types are possible. Such situations may occur when a sample is identified containing a mixture of DNA from two or more people or where extra information can be determined about the subpopulation under consideration by analysis of additional samples. Using this approach, the error in the estimated likelihood ratios is expected to reduce as the number of additional individuals typed from the subpopulation increases.     

Visualization of latent fingermark corrosion of brass, climatic influence in a comparison between the U.K. and Iraq

Through a comparison of fingermark sweat corrosion of α phase brass in both the U.K. and Iraq, we show how samples from Iraq have improved fingermark corrosion over U.K. samples that require no additional enhancement prior to visualization. Over 50% of Iraqi samples produced fingermark corrosion with full ridge detail compared with 0% from the U.K. X-ray photoelectron spectroscopy analysis of the fingermark corrosion products showed that Iraqi samples exhibit more dezincification with the Zn:Cu ratio averaging 1:1.82 compared with 1:3.07 for U.K. samples. Auger spectroscopy showed the presence of both zinc oxide and copper (I) oxide. No copper (II) was observed on the surface of the corroded brass. Opportunities to exploit the optical properties of these thin film oxides to enhance the visualization of fingermark corrosion are considered, and the potential to use fingermark corrosion of metal as a means of visualizing fingerprints in war zones is discussed.     

中国城市化进程与犯罪率之间关系的实证研究——基于结构突变的协整分析

在我国城市化发展进程中,伴随着犯罪率的显著上升,二者具有显著的长期趋势的一致性和短期波动的差异性。＂严打＂运动的外部冲击对协整模型具有显著的结构影响,＂严打＂运动在短期也起到了遏制犯罪的震慑效果。     

Predicting juvenile delinquency: A review of correlates and a confirmation by recent research based on an integrated theoretical model

This study is an exploration of the effects of setting cases for trial in a court that permits jury trials for juveniles. Although few cases actually go to trial, cases set for trial have an impact on juveniles and the court. The study reported here is based on data from a study of 710 youths upon whom delinquency petitions were filed in a suburban court in 1980. Of these youths, 94 had cases set for trial and 7 actually went to trial.

The first part of the article gives a brief historical perspective on the juvenile right to jury trial, discusses why defense attorneys set cases to trial, and describes some of the difficulties in measuring the impact of jury trials. The second part reports on the impact of trial setting in “Suburban Court” on case outcomes and case processing time. A partial correlation analysis shows that there is no significant association between setting a case for trial and either adjudication or final disposition, even when other factors are held constant. Setting a case for trial has a significant association with long case processing times, however. Cases set for trial take almost twice as long as other cases to move from filing to adjudication.     

Assessing Stability and Change in Criminal Offending: A Comparison of Random Effects, Semiparametric, and Fixed Effects Modeling Strategies

An important theoretical problem for criminologists is an explanation forthe robust positive correlation between prior and future criminaloffending. Nagin and Paternoster (1991) have suggested that the correlationcould be due to time-stable population differences in the underlyingproneness to commit crimes (population heterogeneity) and/or thecriminogenic effect that crime has on social bonds, conventionalattachments, and the like (state dependence). Because of data andmeasurement limitations, the disentangling of population heterogeneityand state dependence requires that researchers control for unmeasuredpersistent heterogeneity. Frequently, random effects probit models havebeen employed, which, while user-friendly, make a strong parametricassumption that the unobserved heterogeneity in the population follows anormal distribution. Although semiparametric alternatives to the randomeffects probit model have recently appeared in the literature to avoid thisproblem, in this paper we return to reconsider the fully parametric model. Viasimulation evidence, we first show that the random effects probit modelproduces biased estimates as the departure of heterogeneity from normalitybecomes more substantial. Using the 1958 Philadelphia cohort data, we thencompare the results from a random effects probit model with a semiparametricprobit model and a fixed effects logit model that makes no assumptions aboutthe distribution of unobserved heterogeneity. We found that with this dataset all three models converged on the same substantive result—evenafter controlling for unobserved persistent heterogeneity, with models thattreat the unobserved heterogeneity very differently, prior conduct had apronounced effect on subsequent offending. These results are inconsistentwith a model that attributes all of the positive correlation between priorand future offending to differences in criminal propensity. Sinceresearchers will often be completely blind with respect to the tenabilityof the normality assumption, we conclude that different estimationstrategies should be brought to bear on the data.     

Opportunities for property victimization: Fixed and random effects models in a cross-national scope

Cross-national research on victimization often does not consider the cultural effects on opportunities leading to victimization. The current study uses a routine activities/lifestyles theoretical framework to examine the opportunities leading to two types of property victimization across forty-five countries. I employ fixed effect and random effect models in efforts to find the best explanation of theft and burglary victimization. The results indicate going out in the evening for leisure activities is an important source of opportunity for both types of property victimization. While a fixed effect model offers the most appropriate explanation for burglary victimization, theft is best explained by a random effects model. For theft victimization, variation in whether or not respondents work or go to school is explained, in part, by the level of development of the country. The findings provide evidence that there are different explanations of opportunity for burglary and theft victimization in a cross-national scope.     

Estimating Mean Length of Stay in Prison: Methods and Applications

This paper examines various approaches to studying the mean length of stay in prison. The literature contains a wide range of estimates of this quantity. The discrepancies that appear in these estimates and in the conclusions reached from them have been the subject of several reviews. We build upon that work, using the life table as the gold standard, to demonstrate the inaccuracy of common measures such as the ratio of the population size to the annual number of entrances or the mean length of time served by those exiting in a particular period. This demonstration is conducted in two parts. One part uses model populations with constant growth rates; the second part relies upon simulated prison populations with time-varying rates of entrance and exit. In addition, we introduce two new indirect measures that are more accurate than several existing indirect measures and that are relatively easy to use. The new measures are based on the entrance rate or the exit rate and adjust for the growth rate of the prison population.

Citation analysis of randomized experiments in criminology and criminal justice: a research note

In recent years multiple studies have used citation analysis as a way to examine the most cited scholars and works in criminology and criminal justice and its subfields. This study is the first to focus on the most cited randomized experiments in criminological research. Using citation counts from both Social Science Citation Index and Google Scholar, and multiple sources of randomized experiments, I present the most cited experiments in six areas: policing, courts, corrections, schools, community, and early prevention. Experiments published in non-criminological (e.g., medical) journals tend to be the most cited, and early prevention research is generally cited most frequently of the six categories. I conclude by considering possible reasons why certain experiments are more cited than others.     

Modeling the Deviant Y in Criminology: An Examination of the Assumptions of Censored Normal Regression and Potential Alternatives

Many dependent variables of criminological interest have censored distributions. Investigations that use such variables increasingly have turned to the Tobit model, a censored regression technique that is specified based on a latent dependent variable. When used under suitable circumstances, this model provides appropriate estimates. This paper discusses key assumptions of the Tobit model. It then highlights the risk of violating these assumptions and reviews alternative flexible parametric and semiparametric modeling techniques, currently used sparingly in criminology, which researchers may find helpful when assumptions regarding the error terms are untenable. By using an empirical example focused on sentencing outcomes and comparing estimates across analytic methods, this study illustrates the potential utility of simultaneously estimating the Tobit model along with some alternatives.