Research and development of measures to document ongoing learning within the content areas are in the beginning stages. As such, the current study adds to efforts in the development of the vocabulary-matching measure. Using a modified format of the previously studied vocabulary-matching measure, 63 middle school students completed alternate forms of an administrator-read science measure weekly across 24 data collection periods. Results of hierarchical linear modeling analysis revealed a significant mean growth rate across time (0.26 items per week), significant individual variation in learning rates, and growth rate differences that were significantly predicted by student disability status. General education students had a higher rate of growth than those with disabilities included for science. Results are analyzed and limitations as well as directions for future research are discussed. It is concluded that current findings lend additional validity support to the interpretation of vocabulary-matching scores as indices of learning in the content areas. (Contains 5 tables and 1 figure.)