The Time-on-Task hypothesis asserts that learning is a function of time one allocates to a learning task. Thus, time off-task reduces learning opportunities and is therefore thought to be detrimental to learning. To date, the available research suggests a positive relationship between time on-task and achievement; however, the strength of the correlation fluctuates dramatically. One potential explanation that has been put forth to account for the mixed results is differences in the operational definition of time. The present study tests this hypothesis by examining whether a more stable relationship between on-task behavior and learning can be obtained if time is operationalized in a uniform way. The results of the present study indicate that while on-task behavior was positively correlated with learning outcomes overall, marked variability was still found across classrooms suggesting that the divergent results obtained in previous research are not driven solely by differences in how time is measured.