The versatility of progress testing assessed in an international context: a start for benchmarking global standardization? B.H. VERHOEVEN 1 , H.A.M. SNELLEN-BALENDONG 2 , I.T. HAY 3 , J.M. BOON 3 , M.J. VAN DER LINDE 3 , J.J. BLITZ-LINDEQUE 3 , R.J.I. HOOGENBOOM 2 , G.M. VERWIJNEN 2 , W.H.F.W. WIJNEN 2 , A.J.J.A. SCHERPBIER 2 , C.P.M. VAN DER VLEUTEN 2 1 Medical Centre Leeuwarden, The Netherlands; 2 Maastricht University, The Netherlands; 3 University of Pretoria, South Africa SUMMARY Sharing and collaboration relating to progress testing already takes place on a national level and allows for quality control and comparisons of the participating institutions. This study explores the possibilities of international sharing of the progress test after correction for cultural bias and translation problems. Three progress tests were reviewed and administered to 3043 Pretoria and 3001 Maastricht medical students. In total, 16% of the items were potentially biased and removed from the test items administered to the Pretoria students (9% due to translation problems; 7% due to cultural differences). Of the three clusters (basic, clinical and social sciences) the social sciences contained most bias (32%), basic sciences least (11%). The differences that were found, comparing the student results of both schools, seem a reflection of the deliberate accentuations that both curricula pursue. The results suggest that the progress test methodology provides a versatile instrument that can be used to assess medical schools across the world. Sharing of test material is a viable strategy and test outcomes are interesting and can be used in international quality control. Introduction Progress testing is a longitudinal form of testing knowledge in a comprehensive way (Van der Vleuten et al., 1996b). Periodically, i.e. 4 times a year, a newly written but parallel test is administered to all the students or residents from a particular programme at the same time. First year candidates will not be able to answer many questions, second year candidates somewhat more and so on until a final level is reached. The test consists of a large number of multiple choice or true/false questions covering all disciplines within the curriculum. The progress test (PT) was originally developed to assess knowledge in a problem based learning context to avoid test directed studying and to reinforce self directed learning (Van der Vleuten et al., 1996b). All individual study activities will be rewarded by the test. Extensive experience and empirical research provided evidence that the progress test works in that way (Blake et al., 1996; Blake et al., 1995; Van Berkel et al., 1995; Van der Vleuten et al., 1996b; Van Til, Van der Vleuten & Van Berkel, 1997; Verhoeven et al., submitted). Progress testing is a valuable assessment tool in any context emphasizing self directed learning. PTs are currently used in a number of undergraduate medical schools and some postgraduate training programmes (Blake et al., 1996; Fourie et al., 2002; Pollemans, 1994; Ram, 1998; Schuwirth et al., 1993; Shen, 2000; Van Leeuwen, 1995). Another advantage is that the test procedure is not course specific but comprehensive and intended to reflect the final objectives of a curriculum. Progress tests in a medical undergraduate domain, for example, are stratified by agreed standards and tables of specification (Verhoeven et al., 1998b; Verwijnen et al., 1982). These specifications are not confined to one specific medical curriculum. Therefore the test can easily be applied in other medical schools (Verhoeven et al., 1998a). This allows not only compensation of a disadvantage of progress testing, but also delivers an advan- tage. This potential expanded application of the PT would underline its value as a comparative assessment instrument and help to overcome the resource intensity drawback. The disadvantage of progress testing is the resources that are required to develop and maintain an item bank with sufficient new and high quality questions (Verhoeven et al., 1998b). So why not share and collaboratively develop and maintain such a bank across institutions? Three medical schools in The Netherlands collaboratively develop progress test ques- tions (Bulte & Ket, 2000). A fourth school participates by buying the test. The same test is administered in all four schools at the same time. Thereby the progress testing procedure allows for quality control. It is possible to compare subjects and schools and evaluate growth of knowledge (Van Hessen & Verwijnen, 1990; Verhoeven et al., 1998b; Verhoeven et al., 2002; Verwijnen et al., 1990). This versatile characteristic of the progress test might be a good step on the way to more global standardization of high quality tests (Cohen, 2000; Prideaux & Gordon, 2002; The Executive Counsel, 1998). The items in the PT are the materialized objectives of a (core) curriculum and the performance of the group of candidates are the realized objectives. Cross institutional sharing and collaboration, and quality control are easily achieved with this instrument. Correspondence: B.H. Verhoeven, Theodorus de Leeuwlaan 13, NL-1391 CB Abcoude, The Netherlands. Tel: þ31 650247557; Fax: þ31 43 3618612; email: basverhoeven@mac.com Medical Teacher, Vol. 27, No. 6, 2005, pp. 514–520 514 ISSN 0142–159X print/ISSN 1466–187X online/05/060514-7 ß 2005 Taylor & Francis DOI: 10.1080/01421590500136238