CHAPTER 1 Criteria for Scale Selection and Evaluation John P. Robinson, Phillip R. Shaver, and Lawrence S. Wrightsman The original idea for this handbook of attitude and personality measures came from Robert Lane, a political scientist at Yale University. Like most social scientists, Lane found it difficultto keep up with the proliferation of social attitude measures. In the summer of 1958, he attempted to pull together a broad range of scales that would be of interest to researchers in the field of political behavior. Subsequently, this work was continued and expanded at the Survey Research Center of the University of Michigan under the general direction of Philip Converse, with support from a grant by the National Institute of Mental Health. The result was a three-volume series, the most popular of which was the last, Measures of Social Psychological Attitudes. That is the focus of our first update of the original volumes. Readers will note several differences between this work and its predecessors. Most important, we have given responsibility for each topic to experienced and well-known researchers in each field rather than choosing and evaluating items by ourselves. These experts were also limited to identifying the 10 or 20 most interesting or promising measures in their area, rather than covering all available instruments. This new structure has resulted in more knowledgeable review essays, but at the expense of less standardized evaluations of individual instruments. There are many reasons for creating a volume such as this. Attitude and personality measures are likely to appear under thousands of book titles, in dozens of social science journals, in seldom circulated dissertations, and in the catalogues of commercial pub- lishers, as well as in undisturbed piles of manuscripts in the offices of social scientists. This is a rather inefficient grapevine for the interested researcher. Too few scholars stay in the same area of study on a continuing basis for several years, so it is difficult to keep up with all of the empirical literature and instruments available. Often, the interdisciplinary investigator is interested in the relation of some new variable, which has come to attention casually, to a favorite area of interest. The job of combing the literature to pick a proper instrument consumes needless hours and often ends in a frustrating decision to forego measuring that characteristic, or worse, it results in a rapid and incomplete attempt to devise a new measure. Our search of ihe literature has revealed unfortunate replications of previous discoveries as well as lack of attention to better research done in a particular area. Measures of Personalily and Social Psychological Alrirrrdes Copyright O 1991 by Academic Press. Inc. All rights ofrepmduction in any form reserved.