The purpose of the current study was to apply and evaluate a procedure to elicit expert judgments about correlations, and to update this information with empirical data. The result is a face-to-face group elicitation procedure with as its central element a trial roulette question that elicits experts' judgments expressed as distributions. During the elicitation procedure, a concordance probability question was used to provide feedback to the experts on their judgments.

We evaluated the elicitation procedure in terms of validity and reliability by means of an application with a small sample of experts. Validity means that the elicited distributions accurately represent the experts' judgments. Reliability concerns the consistency of the elicited judgments over time.

Four behavioral scientists provided their judgments with respect to the correlation between cognitive potential and academic performance for two separate populations enrolled at a specific school in the Netherlands that provides special education to youth with severe behavioral problems: youth with autism spectrum disorder (ASD), and youth with diagnoses other than ASD. Measures of face-validity, feasibility, convergent validity, coherence, and intra-rater reliability showed promising results.

Furthermore, the current study illustrates the use of the elicitation procedure and elicited distributions in a social science application. The elicited distributions were used as a prior for the correlation, and updated with data for both populations collected at the school of interest.

The current study shows that the newly developed elicitation procedure combining the trial roulette method with the elicitation of correlations is a promising tool, and that the results of the procedure are useful as prior information in a Bayesian analysis.

Zondervan-Zwijnenburg, M., Van de Schoot-Hubeek, W., Lek, K., Hoijtink, H. & Van de Schoot, R. (2017). Application and Evaluation of an Expert Judgment Elicitation Procedure for Correlations. Frontiers in Psychology, 8:90.

PhD Student

In her PhD project, Mariëlle focuses on including prior knowledge in statistical analyses (informative Bayesian research) and confronting prior knowledge with new data.

PhD Student

Kimberley works together with Rens on how educational and psychological tests can be improved with new and existing statistical tools. One project focusses, for instance, on how (un)certainty in the test results of individual examinees can be estimated and expressed, to ...

Herbert Hoijtink
Professor Applied Bayesian Statistics
Herbert's main research interest is the evaluation of Informative Hypotheses. These are hypotheses constructed using (in)equality constraints among the parameters of interest.
Visit website