[OSF] Personality inventories are predominantly curated using factor analytic psychology/personality approaches. Indicators capturing common and thus redundant variance are preferentially selected, whereas indicators capturing a large proportion of unique variance outside the broad trait domains are omitted from further research. Even recent research dealing with lower-level personality traits such as facets or nuances has invariably relied on inventories founded on this factor analytic approach. However, items can also be selected to ensure low instead of high communality amongst them. The expected predictive power of such item sets is higher compared to those compiled to capitalize on the indicators’ redundancy.
To investigate this, we applied Ant Colony Optimization (ACO) to select personality-descriptive adjectives with minimal inter-item correlations. When used to predict the frequency of everyday life behaviors, this ‘crude-grit’ set outperformed a traditional big 5 personality item set and sets of randomly selected adjectives. The size of the predictive advantage of the crude-grit set was generally higher for those behaviors that could also be predicted better by the big-five item set.
This study provides a proof-of-concept for an alternative procedure for compiling personality scales, and serves as a starting point for future studies using broader item sets.
…Prediction and How it is Affected by Item Selection: Predictions improve, when the dimensionality of the predictor set increases. This relationship is supported by studies showing the incremental predictive power of personality facets over factors ( et al 2020; 1998) or nuances over facets (Seeboth & Mõttus, 2018; et al 2021). The latter findings suggest that relations between criteria and personality traits (or facets) are driven by narrow personality nuances because they contain unique information. However, as nuances have so far been conceptualized within an implicit higher-order model of personality, empirical results do not entail how explained variance changes with the selection strategy for indicators.
A recent study by et al 2020 provides first insights about how different item selection strategies affect predictive power of personality scales. They found higher predictive performance for personality measures developed to measure many relatively independent dimensions compared to narrowly arranged big few measures. Nevertheless, these high-dimensional item sets have still been extracted based on factor analytic approaches. Therefore, in this manuscript we aim to select a broad set of indicators via an item sampling algorithm, moving beyond the factor analytical approaches on personality.