Developing new Test description indices, funding by Strukturfond at the Universität Tübingen, Kap. 1415


Actual situation in test usage

Psychological measurement keeps a central function in scientific research as well as in daily practical usage through practitioner. Both sides are interested in high quality measurement, what should be indicated also through a formal test description. The development and improvement of standards and guidelines to describe psychological and educational tests are part of a continuous strategy (APA, 1985). But practitioners to often neglect standards (Schorr, 1995; Steck, 1997 for Germany speaking countries; Archer, Maruish, Imhof & Piotrowski, 1991). The author had developed criteria, to evaluate the usability of a formal test description and to show, that there is a lack of easy-to-understand and widely applicable formal test criteria. Innovations within the Item-Response-Theories (Linden & Hambleton, 1997) offers seldom used opportunities to describe a measurement. To evaluate the validity and reliability of an assessment variance based approaches are heavily used (Multi-Trait-Multi-Method, Campbell und Fiske, 1979), that leads often to measures of association.

Own approach

The author had defined in several works (Müller, 2000d, 2001g, 2002a,b) five new indices, which are distinguished by their wide applicability, especially in different test theories.