paperKB
coga / coga-kb
Help
Sign in

Chunk #31 — Results — Appropriateness of Item Response Theory to harmonize Neuroticism and Extraversion scores

Source
Harmonization of Neuroticism and Extraversion phenotypes across inventories and cohorts in the Genetics of Personality Consortium: an application of Item Response Theory.
Embedded
yes

Text

Figures 2 and 3 display item parameter values for the NEO-FFI and EPQ-R Neuroticism and Extraversion items for all cohorts in which these inventories were assessed. These parameters are based on a Bayesian hierarchical analysis (Verhagen and Fox 2013a, b) which takes into account any potential mean and variance differences across cohorts. All Bayes factors were smaller than 0.3. However, the item parameters were largely the same across cohorts for most items, with few striking differences. Item parameters tend to be more similar when cohorts have the same language. An example is NEO-FFI Neuroticism item 9 (‘At times I have been so ashamed I just wanted to hide’) for which the two Finnish cohorts show somewhat different item parameter values compared to the other cohorts. Examples from the NEO-FFI Extraversion scale are items 10 (‘I don’t consider myself especially “light-hearted” (R)) and 11 (‘I am a cheerful, high-spirited person’) that show differences across English speaking (red lines) and Dutch speaking cohorts (green lines). Similarly for the EPQ-R items, where item parameters for the Croatian cohorts (black lines) are very similar,