A Psychometric Evaluation of Single‐Item Measures

Single‐item measures are increasingly used in human resource management research due to practical constraints such as survey length, anticipated respondent burden, or repeated measurement. Information concerning which psychometric properties single‐item measures can and cannot meaningfully demonstrate is fragmented and incomplete in the HRM literature. This manuscript provides a comprehensive psychometric evaluation of single‐item measures, drawing on classical test theory, item response theory, and generalizability theory. Logic, math, and previous psychometric research inform the assumptions, practical considerations, and tradeoffs involved in assessing the factor structure, reliability, and validity of single‐item measures. Single‐item measures are incapable of empirically demonstrating latent factor structure or internal consistency, and common workarounds for these issues involve relying on untestable and untenable assumptions. In contrast, some forms of reliability (e.g., temporal consistency, inter‐rater reliability) and validity (e.g., criterion‐related validity, aspects of construct validity) are more conceptually and mathematically appropriate for evaluating single‐item measures. Content validity, including content coverage and clarity of item wording, is a critical consideration when developing or using a single‐item measure. This manuscript concludes by offering practical guidance for researchers and practitioners, including a checklist to aid decisions about when the use of single‐item measures is justified.