Test MultiDimensionnel de Personnalité (TMDP): Norms in T-scores and Percentile Rank order scores for the general and clinical population

This is the authors' abstract. We don't add key points for this paper.

Converting raw scores to T-scores simplifies the interpretation of test results by providing a standardized metric.A simple linear conversion of raw scores to T-scores can produce misleading T-scores if the raw score distribution deviates significantly from normality.The aim of the present research was to provide normalized T-scores and percentile rank scores for the Test MultiDimensionnel de Personnalit e (TMDP).T-scores were established by first transforming raw scale scores to a normal distribution with either an approach based on the Item Response Theory or with an approach based on Rankit percentile-based normalization.The three resulting T-scores (Linear-, IRT-and Rankit-based T-scores) were compared.Results revealed that both normalization approaches yielded similar normalized T-scores, which deviated substantially from linear T-scores for most scales, especially in the higher range of the scores.Benefits of normalized T-scores and the need for gender-and age-based norms are discussed.