Prof. Dr. Arendasy
Theory and practice are not at odds with each other – but a challenge that we are addressing.
Brief biography:
Martin Arendasy is Professor of Psychological Assessment and Methodology at Karl Franzens University in Graz.
03/1998: Mag.rer.nat. (psychology: focusing on differential psychology, applied psychology); 1999-2001: Head of department (applied psychometrics) Manpower-Austria; May 2000: Dr.phil. (psychology, focusing on applied psychometrics); April 2004: Post-doctoral lecturing qualification (psychology, Vienna); June 2007: Awarded a patent for automatic item generation; February 2009: Professor of Psychological Methodology and Computer-based Modeling, Graz; September 2011: DGP Prize for Computerized Psychological Assessment; September 2013: Professor of Psychological Assessment and Methodology, Graz. Since 2009 advisor (psychological assessment, methodology) to various professional bodies; since 2012 chair of the MedAT expert group (admission procedures for degree courses in human medicine and dental medicine at public medical universities in Austria); since 2015 in charge of test development for the talent center of the Styrian chamber of commerce.
Interview:
What was the aim behind development of the tests?
Cognitive abilities and specific facets of personality are among the most important success factors in different careers and types of training. Assessment decisions often have far-reaching practical consequences. We were therefore particularly concerned that the tests should not only be reliable and practically relevant: they also needed to be fair and not discriminate against anyone.
A key challenge was to meet the steadily growing need for items of good psychometric quality. Because of the increasing prevalence of test repetition and coaching to prepare people for tests in HR and education, test items reach the end of their useful life relatively quickly. The increasingly multicultural nature of our society also poses new challenges for assessment. It is therefore becoming more important to develop tests simultaneously in various languages in such a way that comparison of their results is both fair and valid. Using the min-max approach to automatic item construction that we have developed we have managed to find a way of coping with this challenge.
For what target group were the tests developed?
BFSI and INSBAT can be used quite widely. We were particularly interested in personnel recruitment and development, the selection of trainees and students, and occupational and career counseling in the civil and military sectors.
What are the special features of the tests?
INSBAT and BFSI are theory-led and modular in design, which means that they are easy to adapt to the particular assessment query. They have also been proved to be fair and valid. In addition, because the INSBAT items are presented adaptively the test is particularly secure. This enables the need for test repetition, and applicants’ need for information about the test and opportunities to practice, to be combined without difficulty with the need for the tests to be of high psychometric quality. Practice and psychometric quality are not at odds with each other but a challenge that we are addressing and will continue to address. I believe that in Schuhfried we have chosen the right partner for this.
How to you see tests developing in the future? What aspects of this are you particularly interested in?
We are currently working on the further development of INSBAT. New practically relevant subtests of linguistic and spatial abilities have been created for INSBAT 2 and the item pools of the existing subtests have been significantly enlarged. This enables us to achieve greater differentiation in the upper and lower parts of the ability range where this is relevant for assessment purposes. In addition there will in future be exclusive item pools for an online version and customer-specific item pools. Finally, the new developments include a new item selection algorithm. Here we are trying to strike a balance between reasonableness and acceptance on the one hand and psychometric quality and test security on the other. These new developments are based on ongoing customer feedback on practical use of the tests, combined with research by the Department of Psychological Assessment and Methodology at the University of Graz.
In due course there will also be new test forms for BFSI that will help reduce the problem of deliberate faking. The relevant studies are already in progress.
The authors of INSBAT and BFSI regard their tests as living tools that need to be constantly updated on the basis of practice and research in order to meet the practical requirements of everyday assessment in the best way possible.
For further information visit the website of Karl Franzens University, Graz.
