Can’t We Make It Any Shorter?

Abstract

Psychological constructs are becoming increasingly important in social surveys. Scales for the assessment of these constructs are usually developed primarily for individual assessment and decision-making. Hence, in order to guarantee high levels of reliability, measurement precision, and validity, these scales are in most cases much too long to be applied in surveys. Such settings call for extremely short measures validated for the population as a whole. However, despite the unquestionable demand, appropriate measures are still lacking. There are several reasons for this. In particular, short scales have often been criticized for their potential psychometric shortcomings with regard to reliability and validity. In this article, the authors discuss the advantages of short scales as alternative measures in large-scale surveys. Possible reasons for the assumed limited psychometric qualities of short scales will be highlighted. The authors show that commonly used reliability estimators are not always appropriate for judging the quality of scales with a minimal number of items, and they offer recommendations for alternative estimation methods and suggestions for the construction of a thorough short scale.