Confirmatory Factor Analysis for the first-order three-factor model of OMS-HC15 showed acceptable model fit indices (χ2/df = 2.11, CFI = 0.943, TLI = 0.932, RMSEA = 0.082, SRMR = 0.086), but one item demonstrated inadequate factor loading (< 0.3). Considering the poor factor loading and concerns raised on the relevance of this item by previous researchers, this item was removed. However, as removing this item resulted in a slightly worse model fit, modification indices were examined, and constraints on three error covariances were relaxed. This led to a good model fit for the 14-item scale (CFI = 0.962, TLI = 0.951, χ2/df = 1.88, RMSEA = 0.074, SRMR = 0.079). Higher-order, bifactor, and unidimensional models were not supported. The 15-item scale showed acceptable reliability (ω = 0.80; α = 0.69), which improved after removing Item 11 (ω = 0.81; α = 0.70). Subscale reliabilities were lower (Social Distance: ω = 0.70, α = 0.69; Disclosure/Help-Seeking: ω = 0.62, α = 0.61; Attitudes: ω = 0.58, α = 0.55).
No takes yet. Share an insight, caveat, or question.
Samaranayake et al. (2025) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: