Using the cumulative deviation method for cross-institutional benchmarking in the Berlin progress test.

The Berlin Progress Test has grown to a cooperation of 13 universities. Recently, comparisons between the participating schools became an area of high interest. Muijtjens et al. [Muijtjens AM, Schuwirth LWT, Cohen-Schotanus J, Thoben AJNM, van der Vleuten CPM. 2008a. Benchmarking by cross-institutional comparison of student achievement in a progress test. Med Educ 41(1):82-88; Muijtjens AM, Schuwirth LWT, Cohen-Schotanus J, van der Vleuten CPM. 2008b. Differences in knowledge development exposed by multi-curricular progress test data. Adv Health Sci Educ 13:593-605] proposed a method for cross-institutional benchmarking based on progress test data. Progress testing has some major advantages as it delivers longitudinal information about student's growth of knowledge. By adopting the procedure of Muijtjens et al. ( 2008a , b), we were able to replicate the basic characteristics of the cumulative deviation method. Besides the advantages of the method, there are some difficulties as errors of measurement are not independent, which violates the premises of testing statistical differences.