Cruise Scientific   ¨   Visual Statistics Studio   ¨   Measurement and Scaling   ¨   Visual Statistics Illustrated

Krus, D.J., Sherman, J.L., & Kennedy, P.H. (1977) Changing values over the last half-century: the story of Thurstone's crime scales. Psychological Reports,  40, 207-211.

CHANGING VALUES OVER THE LAST HALF-CENTURY:
THE STORY OF THURSTONE’S CRIME SCALES

David J. Krus
University of Southern California
Jay L. Sherman, and Patricia H. Kennedy
Arizona State University 

Summary.— A follow-up of Thurstone’s study, published 50 yr. ago, concerning the perceived seriousness of crimes. The trend toward the judgment of sex offenses as less serious and offenses against the person as more serious was observed. The obtained scale also showed heightened sensitivity of subjects to the burglary, as the study took place during the Watergate impeachment proceedings against President Nixon

In the late twenties, Thurstone (192 7b) had subjects choose between all possible pairs of 19 crimes and offenses. Using the method of paired comparisons, he built a scale of crimes from the more serious as rape and homicide to the less serious as bootlegging and vagrancy. This study was one of the earliest quantitative studies of attitudes. The present study is a 50-yr-later attempt to continue Thurstone’s legacy. The analysis of the obtained data was complemented by the reanalysis of the original Thurstone’s study by using a more powerful computational model.

METHOD

The list of 171 pairs of nineteen offenses was administered to 209 undergraduate students at the University of Southern California. The instructions were identical to those used by Thurstone (1927b, p. 386), however, the list of crime definitions was slightly modified to conform to the contemporary legal code. Although the original Thurstone’s study was carried on at the University of Chicago, the home background of the University of Southern California students was varied enough to compensate for possible regional differences. The student responses are tabulated (Table 1).

 TABLE 1

Proportion of Subjects Who Judged Column Offense More Serious Than Row Offense

 

RESULTS AND DISCUSSION

Both data sets were analyzed by the domain-referenced method for scaling dominance matrices in the paired-comparisons mode (Krus, 1977) and transformed into the T scores to facilitate the inter-distributional comparisons, as shown in Table 2.

TABLE 2
Normalized Case III values of the judged seriousness of
criminal offenses over the 1926–1966 time interval.

 

Scales shown in Fig. 1 were selected from the Thurstone’s original categories of sex offences, offences causing personal injury, and offences against the property.

Fig. 1.   Comparison of 1926–1966 changes in scale values
of judged seriousness of selected criminal offences.

 

 

As can be observed in the above figure, over a half-of-a-century, the sex offences were seen relatively as less serious, while the offences likely to cause a personal harm were perceived as more serious. The elevated perceived seriousness of burglary is likely to the fact that this study was taking place during the Watergate-related impeachment proceedings against President Nixon This finding is significant as it suggests that the method of pair comparisons, enhanced by the domain-referenced method of analysis, can be used as a powerful tool for the analysis of natural experiments, as conceptualized by Auguste Comte.

REFERENCES

• Comte, A. (1848) A General View of Positivism. New York: Robert Speller & Sons (1957)
• Krus, D. J. & Kennedy, P. H. (1977) Normal scaling of dominance matrices: The domain-referenced model. Educational and Psychological Measurement, 37, 189-193. (Request Reprint)
• Thurstone, L. L. (1927a) A law of comparative judgment. Psychological Review, 34, 273-286..
• Thurstone, L. L. (1927b).The method of paired comparisons for social values. Journal of Abnormal and Social Psychology, 21, 384-400.

 

 

Cruise Scientific   ¨   Visual Statistics Studio   ¨   Measurement and Scaling   ¨   Visual Statistics Illustrated