This paper studies how political competition can lead candidates to strategically increase the salience of specific issues, in order to influence voting decisions of marginal groups, with non trivial consequences for turnout rates. In my setup issues differ in their divisiveness, to be defined as the extent to which members within a coalition disagree. Political candidates face a trade-off; they can choose to increase the salience of cohesive issues which energize their own (but also their opponent) constituency or divisive issues which alienate the opponent (but also their own) supporters. Using a model of probabilistic voting, I derive equilibrium campaign strategies of the candidates and predictions on turnout. The results are relevant for the literature on political participation, pointing out that it is crucial to analyze campaign effects on different groups of voters, to explain turnout rates actually observed. We also show descriptive evidence on US Presidential campaigns, supporting the model.