This paper describes an experiment that investigated the effects of goals and self-evaluation on self-regulation processes and achievement outcomes of (n=44) fourth grade students who received instruction and practice on fraction operations. Half of the students were provided with the goal of learning how to solve problems (learning goal) and the other half were given a goal of solving problems (performance goal). Within each goal condition, half of the students regularly assessed their problem-solving capabilities. Providing a learning goal with or without opportunities for self-evaluation or a performance goal with self-evaluation led to higher self-regulated performance, self-efficacy, skill, and task orientation, as well as lower ego orientation, compared with providing a performance goal without self-evaluation. The learning goal with self-evaluation led to greater persistence than the performance goal without self-evaluation. Task orientation correlated positively with self-efficacy and skill, and ego orientation related negatively to these measures. Implications of the results for educational practice are discussed. Contains 34 references. (Author/MKR)