The Prison Preference Inventory: An examination of substantive validity in an Australian prison sample
Faculty of Computing, Health and Science
School of Psychology and Social Science
Toch (1977) developed the 56-item Prison Preference Inventory (PPI) to measure inmates' preferences regarding eight psychosocial aspects of the prison environment. Items on the PPI consist of statements that are supposedly indicative of preferences for one of these eight aspects paired with statements that are indicative of one of the other seven aspects, and respondents indicate their preference. Evidence of construct validity is limited, and no studies have examined substantive validity, that is, whether each statement is reliably indicative of the intended environmental aspect and of only that aspect. The current study examined the substantive validity of the items with Australian prisoners and found that for 21 (37.5%) of the 56 items either or both statements in the pair failed to demonstrate adequate validity. When the more stringent criterion of displaying substantive validity for each of four demographic subgroups was applied, only 8 of the 56 items were deemed valid.