Toward a Model-Based Approach to the Clinical Assessment of Personality Psychopathology - Eaton - 2014 - Article

The goal of psychological assessment is to characterize an individual's standing on individual difference cnstructs of clinical relevance. Assessment of personality disorders, on the other hand, is typically reflective of official nosological systems such as the DSM. 

Structural assumptions about personality variables are inextricably linked to personality assessment. When assessing the personality traits, the assessor needs to measure the full range of the trait dimension to determine where an individual falls on it. The purpose of the assessment is to determine which criteria are present, calculate the number of present criteria, and note whether this sum meets or exceeds a diagnostic threshold. 

Many assumptions about the distribution of data reflecting personality constructs resulted from expert opinion or theory. Both 'type' theories (introvert vs. extravert) and the dimensional theories (introvert-extravert dimension) of personality have been proposed. When you use the 'type' theory, you will only need questions assessing the introvert or extravert type of the scale, no 'inbetween questions'. There have been multiple comparisons between categorical and dimensional theories, but none about a categorical-dimensional hybrid possibility. 

The problem with for instance the DSM and personality psychopathology is that they are highly comorbid and elements of these disorders tend to vary continuously. Fortunately, data-driven methodologies have emerged as an important tool in adjudicating these issues. 

Model-based tests of distributional assumptions

It is critical for personality theory and assessment that underlying distributional assumptions of symptomatology be correct and justifiable, because different distributions impact the way clinical and research constructs are conceptualiszed, measured and applied to individuals. 

The latent trait model posit the presence of one or more underlying continuous distributions. Individuas are dispersed along this trait dimension in a particular way such that there are no zones of rarity. The latent trait is thus continuous and not segmented. All items are associated with a particular location along the span of the latent trait. 

The latent class models are based on the supposition of a latent group (class) structure for a personality construct's distribution. These classes are defined by patterns of item endorsement across individuals. This assumption is referred to as conditonal independence. 

The hybrid models are newcomers to the collection of possible models to test distributional assumptions. These models are often grouped together as factor mixture models, combine the continuous aspects of latent trait models with the discrete aspects of latent class models. For instance, you can show that not everyone within each of those two general groups has the same degree of narcissistic pathology. 

An illustration by application to data on magical thinking

You can show the application of model-based comparisons of latent traits, class and hybrid distributions for illustrative purpose by focusing on data relating to magical thinking. Magical thinking can involve beliefs in extrasensory perception, telepathy, mind-reading, or fortune telling. 

In this paper there were 301 participants with schizophrenia, schizoaffective disorder, or bipolar disorder. Then all participants completed the Schizotypical Personality Questionnaire (SPQ), a 74-item true-false questionnaire that measures interpersonal, cognitive-perceptual, and disorganized schizotypical characteristics with nine subscales. 

This paper shows that distributional assumptions about personality psychopathology are critical t how it is conceptualized and assessed. Measures of personality psychopathology that reflect trait, class and hybrid disorder distributions will necessarily differ markedly, reflecting the fundamentally different ways that these competing models characterize individual differences.