by Robert A. Gordon
When I wrote "Prevalence: The Rare Datum in Delinquency Measurement and Its Implications for the Theory of Delinquency" ( Gordon, 1976), I was careful not to mention the word "intelligence" in the title. I had two reasons for this. The editor of the book, Malcolm Klein, had requested from me a piece on delinquency measurement, saying, with a confidence in my judgment that I could only hope he would feel was not misplaced, "You'll know what is needed." One of the things I decided was needed was a title that did not affront his conception of what it was I was supposed to do.
The other reason for excluding "intelligence" from the title was that I did not want to risk alienating readers before they had even begun reading the article. After all, a contributor to the Annual Review of Sociology admitted several years later, "If you said 'IQ' you drove most sociologists to the barricades" ( Simpson, 1980). I did not have direct measures of intelligence in my own data, as Hirschi and Hindelang ( 1977) did the next year, but had instead inferred the active role of that variable from what I regarded as strong theory and analyses that had to be devised for the purpose.
I had published two articles earlier, each of which rationalized using the epidemiological term prevalence for the proportion of individuals in an age cohort who become delinquent, according to some operational definition, by age eighteen. Perhaps Klein knew of those articles and invited me because of them, although that is not as likely as one might think, for both had appeared in journals that were not staple reading for criminologists--one in the Journal of the American Statistical Association ( Gordon, 1973), the other in the Journal of Mathematical Sociology ( Gordon & Gleser, 1974). I still smile