Krus, D. J., & Hoehl, L .S. (1994) Issues associated with international incarceration rates. Psychological Reports, 75, 1491-1495.

Issues associated with international incarceration rates

David J. Krus and Lenore S. Hoehl
Arizona State University

Abstract. Issues associated with international rates of incarceration were discussed within the framework of a computer-aided search of over 50,000 records on 30 nations, followed by multiple regression analysis and indicating that, jointly, unequal distributions of national income and disintegration of family are two principal determinants of incarceration rates.

The search for causal determinants of criminal behavior and their concomitant variables is usually conducted within a specific cultural context. Comparison of international crime and incarceration rates are less often attempted and show many difficulties such as disparate methods of data collection and diverse laws and efficiency of enforcement of these laws within different countries. The present study is concerned with one aspect of criminal behavior or rather its eventual outcome, the incarceration rates as observed and compared among countries. The purpose of the present article is to describe major factors that influence incarceration rates as viewed from the international perspective.

Method

A multiple regression was used for analysis, with the international incarceration rates for the 1989 (Mauer, 1991, p.5) and 1990-1991 (Mauer, 1992, p.5) serving as a criterion variable. The incarceration rates for 1989 and 1990-1991 were merged into a single variable, with the most recent rates taking precedence over the rates reported earlier. Predictor variables which might explain the incarceration rates were located by a computer-aided search of the compendium of world rankings, compiled by the Facts on File Corporation and the World Model Group (Kurian, 1991). Based on the latest available data, Kurian's compendium contains over 50,000 records on more than 200 countries of the world, extracted from government records, the United Nations statistical yearbooks, the World Bank, and the International Monetary Fund files. After a potential predictor variable was identified, it was reconfirmed by searching the primary source. Nations with corresponding entries found in both Mauer’s reports and Kurian’s compendium are listed in Fig. 1, together with their incarceration rates per 100,000 population.

Fig. 1.  International incarceration rates per 100,000 population, 1989-1992.

The predictor variable with the largest correlation with the international incarceration rates is the index of unequal distribution of wealth, computed as a ratio of the percentage of national income received by the richest 10% to that received by the poorest 20% of households. (Kurian, 1991, pp. 72-73). The predictor variable with the second largest correlation with the criterion variable, the family disintegration index, is based on divorce rates per 1,000 population (Kurian, 1991, pp. 31-32). 

Results

The Pearson correlation between index of incarceration rates and index of unequal distribution of wealth was .48 (p<.01), that between incarceration rates and index of family disintegration was .43 (p<.05). The correlation between indices of unequal distribution of wealth and family disintegration as -.41 (p<.05). The beta weights for simultaneous inclusion of both predictor variables into the multiple regression analysis were .79 for the unequal distribution of wealth index and .74 for the disintegration of family index.
A multiple regression analysis returned a multiple R equal to .83, showed that jointly, the best predictors of international incarceration rates were the indices of unequal distribution of wealth and family disintegration. Inspection of the pattern of cross-correlations between the predictor variables and the criterion variable, their corresponding beta weights, and their intercorrelations showed a pattern typical of positive cooperative suppression, characterized by positive beta weights exceeding the correlation coefficients between the predictor variables and the criterion (cf., Krus and Wilkinson, 1986). This cooperative suppression is easy to understand. Impoverishment or breakdown of family by themselves does not always precipitate crime, but poverty and a lack of family support often do. Thus the combined indices of unequal distribution of wealth and family disintegration accounted for 69% of variance in international incarceration rates analyzed, as shown in Figure 2.

Discussion

The results of the regression analysis suggest the joint role of the unequal distribution of a nation's wealth (the explanation favored by liberals) and the role of family disintegration (the explanation favored by conservatives) as determinants of incarceration rates and, indirectly, crime. An uneven distribution of wealth of the society was a primary predictor of incarceration rates. This explanatory construct is clearly defined and open to the interpretation that by reducing the distance between the rich and the poor, crime rates seem likely to decrease, diminishing incarceration rates. The degree of disintegration of families within a society also shows a strong relationship with incarceration rates. The significance of this determinant is obvious and its contribution to the explanation of the variance of the criterion variable is substantial.

References

Kurian, G.T. (1991) The New Book of World Rankings. New York: Facts on File, Inc.
Krus, D. J., & Wilkinson, S. M. (1986) Demonstration of properties of a suppressor variable. Behavior Research Methods, Instruments, and Computers, 18, 21-24.
Mauer, M. (1991) American Behind Bars: A Comparison of International Rates of Incarceration. Washington, D.C.: The Sentencing Project.
Mauer, M. (1992) American Behind Bars: One Year Later. Washington, D.C.: The Sentencing Project.