The data described above allow us to calculate the number of years of a given student’s early childhood in which television sig- nals were available. In order to make the magnitudes we measure in the analysis below more easily interpretable, we will also use data on the rate at which television ownership actually diffused among households in each county. We will use the term televi- sion exposure to refer to the expected number of years a child’s household owned a television during the child’s p