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.

Relationship between the retirement, disability, and unemployment insurance programs: the U.S. experience

This article was prepared initially for an international conference of social security program administrators and researchers. They examined the reasons for, and implications of, a recent trend in several European countries toward making it easier to qualify for retirement or disability benefits as a way of alleviating long-term unemployment. The article notes that the United States has not followed this trend. Instead, this country has continued to use temporary extensions of unemployment insurance benefits as a way to help the long-term unemployed during recessionary periods. Since the mid-1970's, the emphasis in retirement and disability insurance programs has been to strengthen the financial integrity of these programs rather than to expand eligibility. Described here are the progression of extended benefit provisions of unemployment insurance through the most recent recession, the historical development of early retirement features in the social security program, and the more recent attention that has been paid to the financing issues that have played a central role in legislation during the late 1970's and early 1980's. Unemployment experience and trends toward early retirement are examined, along with the role of public and private employee pension plans that supplement social security retirement benefits. Preliminary data from the Social Security Administration's New Beneficiary Survey show the prevalence of such pension coverage for recent retirees and the extent to which these pension benefits were claimed before normal retirement age.     

Studying the reach of deterrence: Can deterrence theory help explain police misconduct?

This article reports the first perceptual deterrence study of a sample of police officers. The study investigated the influence of traditional deterrence considerations, extralegal sanctions, and impulsivity on the intention to commit several hypothesized acts of police misconduct. The results were largely consistent with perceptual deterrence findings from samples of college students, experienced offenders, and corporate managers. In particular, this study found that both legal and extralegal sanction threats potentially deter police misconduct. Further, it found that impulsivity diminished the deterrent influence of both sanction forms. The study also found that some of the effects of the explanatory variables depended on whether officers had prior punishment experience. The article discusses the implications of its findings for combating police misconduct and for deterrence research generally.     

Cash benefits for short-term sickness, 1948-81

At least three-fifths of all workers are under some kind of formal plan offering income replacement during temporary periods of illness, even though there is no national sick-pay program. In 1981, income loss due to short-term non-work-related disabilities totaled +41.3 billion. Sick-pay benefits were +15.6 billion, representing 38 percent of the income that otherwise would have been lost. The replacement rate has been fairly stable since 1974, following earlier periods of uneven increases from the 17 percent recorded for 1948, the first year of this series. This article includes estimates for 1980 and 1981 and revisions for 1978 and 1979 of the income loss and benefits arising from short-term sickness. Also included are the losses and benefits for the first 6 months of long-term disability.     

Theoretical issues and social indicators: A societal process approach

Well-being involves various levels: the individual, the institutional-distributive, and the societal. These levels are interrelated. Social indicators of well-being must be theoretically based so as to take into account these levels. A societal process model is proposed to describe the levels of society and the nature of well-being at each level. Social indicator types regarding the output and distribution of well-being, the effect of policy manipulatable and nonmanipulatable inputs, and the secondary consequences of inputs are suggested and examples are provided.The ideas presented herein have evolved from the authors' participation in a research project conducted for the U.S. Agency for International Development (AID/csd—3642), but do not necessarily reflect the views of that agency. Thanks are extended to J. Steven Picou, James Copp, Jon Alston, Bill Howard, and Ben Crouch, as well as to the then editor of this journal, Garry Brewer, and several anonymous reviewers who were kind enough to provide both cogent critique and encouragement. Journal Paper No. J-8808 of the Iowa Agriculture and Home Economics Experiment station, Ames, Iowa. Project No. 1837.