Computerized Adaptive Testing of Personality Traits

Abstract

A computerized adaptive testing (CAT) procedure was simulated with ordinal polytomous personality data collected using a conventional paper-and-pencil testing format. An adapted Dutch version of the dominance scale of Gough and Heilbrun’s Adjective Check List (ACL) was used. This version contained Likert response scales with five categories. Item parameters were estimated using Samejima’s graded response model from the responses of 1,925 subjects. The CAT procedure was simulated using the responses of 1,517 other subjects. The value of the required standard error in the stopping rule of the CAT was manipulated. The relationship between CAT latent trait estimates and estimates based on all dominance items was studied. Additionally, the pattern of relationships between the CAT latent trait estimates and the other ACL scales was compared to that between latent trait estimates based on the entire item pool and the other ACL scales. The CAT procedure resulted in latent trait estimates qualitatively equivalent to latent trait estimates based on all items, while a substantial reduction of the number of used items could be realized (at the stopping rule of 0.4 about 33% of the 36 items was used).

Influence of Context on Item Parameters in Forced-Choice Personality Assessments

“…Personality assessments nowadays have not integrated CAT methodology as widely or deeply as modern cognitive ability tests, but the prospect is promising. For example, Hol, Vorst and Mellenbergh (2008) conducted a realdata simulation study using SS personality items, and found that CAT only requires as few as 33% of the original items to reach similar levels of measurement accuracy.…”

Validation of the Intelligent Cognitive Assessment System (ICAS) for stroke survivors

“…Advances in computer technology and the Rasch model provide possible alternatives in test development including the use of computer-adaptive testing (CAT) [8]. CAT uses an item bank stored in the computer.…”

“…In other words, CAT selects only questions that are appropriate to the test taker's ability level based on previous responses [11]. In this way, the same number of questions can yield a more precise estimation of cognitive ability in CAT than in a conventional paper and pencil test [8,9].…”

The Development and Validation of a Multidimensional Measure of Moral Orientation

“…In addition to IRT, another area of test development that is receiving attention is Computer Adaptive Testing (CAT; Forbey, Ben-Porath, 2007;Hol, Vorst, Mellenbergh, 2008). With the majority of personality tests being administered online and over the computer, researchers have begun studying the benefits of CAT for personality scales.…”

“…This is perhaps the greatest benefit to CAT -the ability to make more refined measurements. Not only can CAT offer more refined measurement, it can considerably shorten the time of test administration (Hol et al, 2008), which in turn may limit fatigue effects, boredom, and careless responding. It seems that the use of CAT for moral orientation research would be particularly helpful, because the scenario-based nature of these assessments would benefit from a shorter administration time.…”