The Perfect Design: Optimization Between Reliability, Validity, Redundancy in Scale Items and Response Rates

Abstract

There has been a continuing emphasis on reporting high reliability estimates as a measure of survey quality. In lieu of increasing the reliability of a survey exi>eriment, researchers invariably increase the number of scale items thus critically affecting the response rates. Further, in doing so the researchers overlook the fact that highly reliable scales may or may not be construct valid. The paper deals with a critical yet overlooked issue pertaining to measurement of latent constructs. We attempt to outline the complex and intertwining relationship between the construct validity, reliability, redundancy (in terms of number of items) and response rate for a multiitem scale.