Testing levels of competencies in biological experimentation
Currently, efforts are being made to improve the quality of biology education in Germany and develop tests to assess student achievement. With the introduction of national science education standards in 2005, there has been a growing need for criterion-oriented tests, i.e. tests that assess whether or not a specific level of competency has been reached. Here we present an approach to measuring students levels of competencies in experimentation. In particular, the focus lies on the competencies of forming hypotheses, planning experiments and analysing data. These competencies were selected because they are crucial to experimentation as problem-solving, according to the SDDS model (Klahr 2000), and because they are central to the newly introduced Biology standards. One specific reference point of this study is the international scientific literacy test for PISA 2000. Five levels of competencies were proposed for PISA 2000. Items were mapped onto the levels of competencies by dividing up the maximum total sum score into five segments and by assigning items with a low/high difficulty to a low/high level of competency. The approach chosen in the present study differs, as closed-end test items were developed with response categories that can be directly related to a specific level of competency. Each item could be answered in different ways that are indicative of a particular level of competency. Item development, thus, took into consideration the rich research literature on qualitative differences among levels of competencies in experimentation.