Prior performance is an important factor that directly predicted performance at posttest for
modelling (β=0.43) and intra-mathematical performance (β=0.28). However, students’ performance
in modelling or in solving intra-mathematical problems at pretest did not influence
the number of solutions they developed during the teaching unit. We also did not find
significant indirect effects of prior performance on students’ performance at posttest. Although
students who showed better performance at pretest outperformed low achievers at posttest, this
did not result from the indirect effects of prior performance on the number of solutions they
developed during the teaching unit