Strategic Voting under Incomplete Information in Approval-Based Committee Elections Jason Tsiaxiras Abstract: In approval-based committee elections, voters vote by submitting an approval ballot, which indicates which candidates the voter approves of, with the purpose of electing a fixed size committee. Recent impossibility results have shown that approval-based committee voting rules that select committees that provide proportional representation to the voters are inherently susceptible to strategic manipulation. The impossibility results obtained in the literature rely on the assumption that voters have complete information about the preferences of their fellow voters. In this thesis, we extend the model of approval-based committee elections in order to account for strategic voting when voters have incomplete information about the preferences of their fellow voters. The purpose of this model is to be able to describe the strategic behavior of real voters more accurately and explore whether there are information restrictions under which voters do not have incentive to misrepresent their preferences. In our analysis we employ formal methods to study voting rules that satisfy a set of normative properties. We also perform simulation experiments to study three well-established proportional voting rules. The results of our experiments show a high prevalence of incentive to manipulate when voters have complete information. Moreover, the experiments show that decreasing the information a voter has does not necessarily decrease the probability that this voter has incentive to manipulate. The negative results of our formal analysis show that, for two natural kinds of incomplete information voters may have, voting rules that satisfy a set of weak normative properties remain susceptible to manipulation. We also present two positive results for the Proportional Approval Voting rule which show that information restrictions exist under which this voting rule is (partially) strategy-proof.