In a large-scale field experiment, we use text message reminders to increase student attendance and attainment within UK government-sponsored literacy and numeracy programs for adults. High dropout rates present a major challenge in this context – in our data, approximately 25% of students stop attending within the first ten weeks. However, we find that our text intervention has a large and persistent affect on student attendance as well as literacy and numeracy performance on exams. Text messages lead to a 4% point increase in average attendance relative to the control group in the initial weeks after implementation. Using data from the full academic year, we find that these effects persist, showing a 7% point increase over the full trial period, and an 8% point increase in the likelihood of passing exams.