Toward a second generation of statistical prediction rules in psychodiagnosis and personality assessment

«Guidelines» zur Primärprävention kardiovaskulärer Erkrankungen empfehlen die Bestimmung des kardiovaskulären Gesamtrisikos mittels Risikokalkulatoren (sog. «Prediction rules»). Während die «Prediction rules» als Messinstrumente eingesetzt werden, geben die «Guidelines» Handlungsanleitungen, welche Konsequenzen aus der Berechnung gezogen werden sollten. An einer Tagung wurden Fragen zur Anwendung von «Prediction rules» in der Grundversorgung diskutiert. Die Anwendung scheint in der Praxis einfach zu sein, doch sind einige Barrieren vorhanden, diese Instrumente einzusetzen. Teilweise ist nicht bekannt, wie zuverlässig das Risiko überhaupt berechnet werden kann, teilweise fehlen in den «Prediction rules» Risikofaktoren (z.B. Adipositas), die intuitiv doch wichtig sind und drittens muss zwischen «Prediction rules» und «Guidelines» differenziert werden. In diesem Artikel werden, sofern vorhanden, Antworten auf Fragen, die an dieser Tagung aufgekommen sind, beschrieben.

Download Full-text

Handwriting Analysis and Personality Assessment

European Psychologist ◽

10.1027//1016-9040.5.1.44 ◽

2000 ◽

Vol 5 (1) ◽

pp. 44-51 ◽

Cited By ~ 5

Author(s):

Peter Greasley

Keyword(s):

Personality Traits ◽

Personality Assessment ◽

Personnel Selection ◽

Empirical Studies ◽

Handwriting Analysis ◽

Recruitment Process ◽

Academic Literature ◽

Personnel Recruitment ◽

The Subject ◽

Better Than

It has been estimated that graphology is used by over 80% of European companies as part of their personnel recruitment process. And yet, after over three decades of research into the validity of graphology as a means of assessing personality, we are left with a legacy of equivocal results. For every experiment that has provided evidence to show that graphologists are able to identify personality traits from features of handwriting, there are just as many to show that, under rigorously controlled conditions, graphologists perform no better than chance expectations. In light of this confusion, this paper takes a different approach to the subject by focusing on the rationale and modus operandi of graphology. When we take a closer look at the academic literature, we note that there is no discussion of the actual rules by which graphologists make their assessments of personality from handwriting samples. Examination of these rules reveals a practice founded upon analogy, symbolism, and metaphor in the absence of empirical studies that have established the associations between particular features of handwriting and personality traits proposed by graphologists. These rules guide both popular graphology and that practiced by professional graphologists in personnel selection.

Download Full-text

The Persnickety Pervasiveness of Rating Enhancement in Personality Assessment

European Journal of Psychological Assessment ◽

10.1027/1015-5759/a000610 ◽

2020 ◽

pp. 1-13

Author(s):

Alicia A. Stachowski ◽

John T. Kulas

Keyword(s):

Response Bias ◽

Personality Assessment ◽

Current Paper ◽

Psychological Measurement ◽

Shared Sets ◽

Self Reports ◽

Original Target ◽

Target Profile ◽

Do So ◽

Normative Influences

Abstract. The current paper explores whether self and observer reports of personality are properly viewed through a contrasting lens (as opposed to a more consonant framework). Specifically, we challenge the assumption that self-reports are more susceptible to certain forms of response bias than are informant reports. We do so by examining whether selves and observers are similarly or differently drawn to socially desirable and/or normative influences in personality assessment. Targets rated their own personalities and recommended another person to also do so along shared sets of items diversely contaminated with socially desirable content. The recommended informant then invited a third individual to additionally make ratings of the original target. Profile correlations, analysis of variances (ANOVAs), and simple patterns of agreement/disagreement consistently converged on a strong normative effect paralleling item desirability, with all three rater types exhibiting a tendency to reject socially undesirable descriptors while also endorsing desirable indicators. These tendencies were, in fact, more prominent for informants than they were for self-raters. In their entirety, our results provide a note of caution regarding the strategy of using non-self informants as a comforting comparative benchmark within psychological measurement applications.

Download Full-text