Modern psychometrics [44] [45] offers statistically and theoretically well developed methods, such as (genetically informed) latent factor models and Item Response Theory, for addressing the phenotypic measurement issues discussed here, and as such has the potential to contribute considerably to the success of genetic studies.
CFA-OCM appropriately models the categorical nature of the items[ 60] and falls within a larger family of latent variable measurement modeling approaches that includes: CFA for continuous measures, multiple indicator multiple cause (MIMIC) models, and item response theory models.
Model and item fit was evaluated by comparing the observed proportion of responses for each category, with the model predicted values obtained from the item parameters and the estimated latent trait distributions.
Furthermore, 15 out of 24 items presented threshold disorders, which suggests that the response scale is not adequate and therefore contribute to the misfittings found both in model and item levels.
A re-run of the full model and item reduction analysis on the non-imputed dataset (N = 217) resulted in somewhat less favourable but still acceptable fit indices and comparable factor loadings.
First of all, EFA was conducted to determine what items should be included in the models and what items to discard when they did not load on the investigated dimension.
Some even have a selection of floor models and returned items - referred to as Open Box items.
On display were some 300 Stirling drawings, models, and other items.
While a facet represents a model, a package represents a related collection of models and other items.
CFA models were iteratively trimmed by dropping items that were not explained well by the model and regrouping items when less than three items were left in a factor/domain.
Having accepted the 4-dimensional model according to overall model fit and item fit, we then consider the difficulty of the items (Table 2).
