This study evaluated the construct validity and reliability of a newly devised 32-item problem quality rating scale intended to measure the quality of problems in problem-based learning. The rating scale measured the following five characteristics of problems: the extent to which the problem (1) leads to learning objectives, (2) is familiar, (3) interests students, (4) stimulates critical reasoning, and (5) promotes collaborative learning. The rating scale was administered to 517 polytechnic students enrolled in problem-based curricula and the data collected were subjected to confirmatory factor analysis. The results revealed a good fit of the data with the hypothesized five-factor model. The coefficient H values of the five factors suggested acceptable factor reliability. Overall, the psychometric characteristics of the rating scale indicated adequacy of the instrument to measure the quality of problems in problem-based learning. Although there are other ways to assess problem quality, the ease of use and means to measure multiple indicators makes the problem quality rating scale useful.