22. Social psychologists studying criminal justice issues have long been interested in what influences people’s attitudes about punishment of criminal offenders.

Graham and her colleagues (1997) took advantage of the very public trial of U.S. football star O. J. Simpson to test some basic issues in this area. In the first few days after Simpson was accused of having murdered his ex-wife, the researchers asked people a series of questions about the case. The researchers were mainly interested in the responses of the 177 individuals who believed Simpson was probably guilty, particularly their belief about retribution—how much they agreed or disagreed with the statement, “The punishment should make O. J. suffer as he made others suffer.” The researchers were interested in a number of possible influences on this belief. These included “control” (how much control they believed Simpson had over his actions at the time of the crime), “responsibility” (how much they believed he was responsible for the crime), “anger” they felt toward him, “sympathy” they felt for him, “stability”

(how much they believed his actions represented a stable versus temporary way of behaving), and “expectancy” (if they thought he would commit such a crime again). Graham and her colleagues reported

. . . Table [13] reveals partial support for our hypotheses. As expected, the strongest predictors of the retributive goal of making Simpson suffer were inferences about responsibility and the moral emotions of anger and sympathy.

[S]tability and expectancy . . . were relatively weak [predictors]. (p. 337)

Explain these results as if you were writing to a person who understands bivariate prediction but does not understand multiple regression. (Refer only to the retribution part of the table. You may ignore the t column, which is about the statistical significance of the results.)