Two experiments tested the idea that the means by which children acquire efficacy information can produce different levels of task motivation and self-perception of competence. In Experiment 1, children periodically received either ability attributional feedback, effort feedback, ability plus effort feedback, or no attributional feedback. Although the three feedback conditions did not differ in rate of problem solving, ability feedback alone led to the highest self-efficacy and achievement. In Experiment 2, children pursued either difficult or easier performance goals, and received either direct or comparative attainment information. Difficult goals enhanced rate of problem solving; combining difficult goals with direct information resulted in the highest self-efficacy and skill. Collectively, these results support the idea that self-efficacy is an important variable in understanding students' achievement behavior. (Author/PN)