Exercise handling processes and
individual process characteristics

Lead: DIPF

 

International comparative educational studies like PISA are increasingly switching from paper- to computer-based tests, thus unlocking new possibilities of designing test exercises. A scientific experiment can be simulated on the computer, for instance. Exercises and questions can also be better adapted to the students.

Traditionally, the main focus was whether an exercise was solved or not. Computer testing also enables surveying of processing behaviour. How much time the student spent on an exercise and where they clicked with the mouse can be viewed, for example. This so-called process data permits additional conclusions about processing per se and reveals how the problem was solved or at which point the student failed to solve it, for instance.

Analysing this process data thus produces valuable input for educational practice in order to improve learning processes. The project addresses precisely this aspect and explores which relevant factors can be gleaned from process data and which methods can be applied for an appropriate statistical evaluation of this data.

The extent to which process data provides information on the motivation of the test students in order to then relate it to the test answers is also to be clarified. Conclusions can thus be drawn about actual performance, since less motivated test students do not apply their full capabilities to the test, for example.

The project team will use existing and upcoming data from PISA and PIAAC in tackling the research questions.

The research data can be evaluated in more detail based on questionnaires completed directly on the computer. © iStock.com

© 2012 Zentrum für internationale Bildungsvergleichsstudien

» Webdesign