Assessing Personality With Multi-Descriptor Items

Abstract

The need for efficient personality inventories has led to the wide use of short instruments. The corresponding items often contain multiple, potentially conflicting descriptors within one item. In Study 1 (N = 198 university students), the reliability and validity of the TIPI (Ten-Item Personality Inventory) was compared with the reliability and validity of a modified TIPI based on items that rephrased each two-descriptor item into two single-descriptor items. In Study 2 (N = 268 university students), we administered the BFI-10 (Big Five Inventory short version) and a similarly modified version of the BFI-10 without two-descriptor items. In both studies, reliability and construct validity values occasionally improved for separated multi-descriptor items. The inventories with multi-descriptor items showed shortcomings in some factors of the TIPI and the BFI-10. However, the other scales worked comparably well in the original and modified inventories. The limitations of short personality inventories with multi-descriptor items are discussed. (PsycINFO Database Record (c) 2019 APA, all rights reserved)