Objective: The purpose of this article is to describe the method of item selection tested in the development of the Adolescent Domain Screening Inventory. Method: The convenience sampling frame used for these analyses consisted of 26,781 Communities That Care Youth Surveys. The three item selection methods were used to assess known instrument, convergent construct, and concurrent criterion validities. Results: Method 3, multiple regression on 30-day use, performed the best on the correlations with outcome scales and was indistinguishable from the others on the correlations with the original. Conclusions: This study suggests that the dominant method of short-form item selection, factor analysis loading, might not be only method a researcher can employ and may not be best for all purposes. (Contains 4 tables.)