Learning potential and anxious tendency: Test anxiety as a bias factor in educational testing

Contrary to conventional educational testing, in so-called dynamic assessme nt subjects are allowed to consult help during testing or are offered prior training. The differential results of both testing procedures are sometime s ascribed to the idea that dynamic tests reflect the breadth of the zone o f proximal development on top of independent achievement. Alternative expla nations claim that conventional tests are more strongly biased towards vari ous characteristics of persons, which have a negative influence on performa nce, when compared to dynamic tests. In this study, it was hypothesised tha t static as well as dynamic assessment is biased towards anxious tendencies of subjects, but the former more strongly than the latter. In order to inv estigate this supposition, the performance of subjects on dynamic and stati c tests was systematically compared and related to measures of test anxiety in a longitudinal experiment. In the experiment, repeated measures of inde pendent mathematics achievement as well as mathematics learning potential w ere gathered among students of secondary education in the Netherlands. Prio r to every mathematics test, subjects filled out a test anxiety questionnai re. After every mathematics test, subjects filled out a general state anxie ty questionnaire. The participating subjects were students from secondary e ducation, either preparing for higher vocational training or University, ag ed approximately 15 years on average. The results of the experiment showed that lack of self-confidence is an imp ortant constituent factor of test anxiety, apart from worry and emotionalit y. The data supported the assumption that such testing procedures are less biased towards anxiety than conventional tests, but it was not established that dynamic testing procedures render results that are not biased by test anxious tendencies.