Abstract
The current study examined processes relating problem-solving skills training to the
performance of adolescent educationally at-risk students. This study began with a model based on
the belief that both educational-risk level and problem solving skills training would influence
behavioral efficacy and attempted to investigate processes related to locus of control and self-
efficacy that underlie this influence. A quasi experiment first varied the problem solving skills
training (trained, not trained) at different levels of student educationally at-risk students (at-risk,
non-risk) and then measured problem-solving cognitive skill, locus of control, self-efficacy, and
behavioral outcome efficacy on a sample of 88 students. Separate models for at-risk and non-risk
students evolved. Models for both groups represent a process in which problem solving skills
training influences locus of control (positively for at-risk but not positively for non-risk students)
which then increases self-efficacy and subsequent behavioral outcome efficacy.