The outcomes of studies on the effectiveness of special education (SE) remain unclear. There are only a few studies on the effects of SE that have used advanced methodology to minimise the influence of potential selection bias. This study examined the plausible effects of SE on students' affective and motivational outcomes using a quasi-experimental method of propensity score matching with longitudinal data within the context of the Finnish multi-tiered support system. The participants of this study were fifth- and sixth-grade students from 30 primary schools who took part in a larger study of school inclusion in Eastern Finland (ISKE). Data from 553 students included information from questionnaires on students' emotional engagement with school, self-concepts, and goal orientations from themselves, academic achievement from their teachers, and socioeconomic status from their parents. We examined the effects of receiving SE service by following students from fifth and sixth grade using three different regression models in ANCOVA. Results revealed not only the influence of selection bias on outcomes but also that when comparing matched groups of students, SE did not have effects on students' affective or motivational outcomes.