In a previous study, Phillips (1998) observed that accounting students possess several dimensions of beliefs about the nature of knowledge, and provided evidence that one of the belief dimensions (i.e., that knowledge is uncertain) was related to a component of unstructured problem‐solving performance (i.e., evaluating the relevance of case facts). Phillips (1998) also proposed that the relationship between students' beliefs and unstructured problem solving was mediated by their study strategies, but did not test this proposition. The current study replicates the belief dimensions observed by Phillips (1998) and examines the empirical relationship among students' beliefs, study strategies, GPAs, and unstructured problemsolving performance. Results indicate that one dimension of beliefs (i.e., that knowledge is complex) was associated with a dimension of study strategies (i.e., consolidating knowledge) and that these two dimensions were related to cumulative GPA and, after controlling for GPA, with a component of unstructured problem‐solving performance (i.e., consolidating analyses). These findings, in conjunction with the results reported by Phillips (1998), are consistent with the theory that performance on an unstructured problem depends, in part, on the degree to which student beliefs and study strategies match the features of an “ideal” solution for the problem. This theory helps to explain how two equally knowledgeable students can differ in how they cope with unstructured problem solving, with one insisting on simple answers and the other remaining open to complex and integrative solutions.