The election of the French representative to the European Parliament is based on a proportional representation system with a 5% threshold. Due to this threshold, votes for parties that do not reach this threshold are lost, and the corresponding voters are not represented, and not taken into account for the allocation of seats. In this context, we wanted to test alternative voting methods that could mitigate this issue by allowing voters to rank more than one list. We conducted an online voting experiment during the election of the French representative to the European Parliament in June 2024. In this experiment, participants were asked to vote with the alternative voting methods, and give their opinion. We ran the experiment on two sample of participants, a self-selected one in which every voter could participate, and a representative one, collected through a poll institute. If you are using this dataset in a scientific paper, please cite our paper Reallocating Wasted Votes in Proportional Parliamentary Elections with Thresholds (Théo Delemazure, Rupert Freeman, Jérôme Lang, Jean-François Laslier, Dominik Peters). V2.0: Added the representative dataset and updated the documentation. V1.2: same bug but in "participants.csv" for voting intentions. V1.1: there was a bug in the data of the "vote_two_choices.csv" file => NaN values were replaced by "ESPERANTO LANGUE COMMUNE".