This result is consistent with other studies showing that effects were stronger when the
measuring instruments were more closely related to the intervention (cf. the meta-analyses
by Hattie et al., 1996; Seidel & Shavelson, 2007). On the other hand, related constructs
(see, for example, studies about self-concept by Fennema, 1989) may profit from the
intervention as well, as affect related to other types of problems did in the present study.