This article examines the relative size of gaps in knowledge and participation between the more and less educated as they vary by the quantity and type of news media use. We predicted that the gap between high and low education groups would be smaller among heavy television news users than among light users, whereas the gap between high and low education groups would be larger among heavy newspaper users than among light users. We also predicted that the gap in general political participation — but not voting—would be greater among both heavy television news users and heavy newspaper users than among light news users. These predictions were based on logic derived from the communication effects gap hypothesis, the cognitive psychology of learning, and research on political behavior. Analyzing data collected as part of the American National Election Study during the 1996 U.S. presidential campaign, we found that gaps in knowledge between higher and lower education groups were greater among light than heavy users of television news. A similar pattern was found for knowledge gaps across levels of newspaper use, but this pattern was weaker and may possibly be attributed to ceilings imposed by the nature of the test or a natural ceiling in the information domain. By contrast, neither television news use nor newspaper use was related to gaps in voting; however, newspaper use, but not television news use, was related to gaps in general political participation.