Development and Rasch Analysis of the 18-Item Health Resilience Profile (HRP)

Abstract

Existing resilience measures have psychometric shortcomings, and there is no current gold-standard resilience measure. Previous work indicates adults enrolled in a health coaching program may benefit from a resilience measure that is tailored and contextualized to this sample. This two-part study aimed to develop and evaluate a resilience instrument focused on health behavior change in adults in a health coaching program. Two studies were conducted to (1) create a resilience instrument (Health Resilience Profile; HRP) specific to adults attempting health behavior change (n = 427; female = 83.8%; age = 44.5 ± 11.9 years) and to (2) optimize the instrument performance using Rasch analysis (n = 493; female = 62.1%; age = 49.8 ± 12.5 years). Study 1 identified two issues: (1) four unacceptable misfit items and (2) inappropriate rating scale functioning. Study 2 evaluated an improved instrument based on the outcome of study 1 resulting in one more misfit item, and unidimensionality was supported. The new four-category rating scale functioned well. The item-person map indicated that item difficulty distribution was well matched to participants’ resilience level, and items were free from measurement error. Finally, items did not show differential item functioning across age, sex, alcohol use, and obesity status. The 18-item HRP is optimized for adults in a health coaching program.