There is widespread belief that exposure to television has harmful effects on children's cognitive development. Most studies that point to a negative correlation between hours of television watching and cognitive outcomes, fail to establish causality. Using the National Longitudinal Survey of Youth (NLSY) we study young children between 5 and 10 years of age during late 1990s and early 2000s. We find strong evidence of negative correlations between hours of television watched and cognitive test scores. However, once parent's characteristics and unobserved child characteristics are taken into account these correlations go away. We find that hours of television viewed "per se" do not have any measurable impact on children's test scores. Our results are robust to different model specifications and instrumental variable estimates. We conclude that despite the conventional wisdom and the ongoing populist movement, proactive policies to reduce children's television exposure are not likely to improve children's cognitive development and academic performance.