Video "podcast" recordings of lectures are popular with students, but are often associated with a decrease in attendance and little increase in performance. Assessment has generally focused on the class as a whole, potentially masking benefits to different subgroups. In this study, conducted in 2 sections of a large active-learning undergraduate introductory biology class with daily podcasts, average attendance remained high (89.5%). More than 50% of the students used podcasts but less than 3% of the variance in actual minus predicted exam performance was correlated with the number of podcasts viewed. Podcast use also varied significantly with gender and ethnicity but even within high use subgroups (females and Asians) less than 6% of the variance in exam performance was correlated with the number of podcasts viewed. These data suggest that lecture capture, even for the students who attend class and use them heavily, do not increase learning gains. Alternative uses for video are discussed.