Using GLES data we ask whether the differences in the parties’ vote shares between election day voters and postal voters are due to compositional differences (they are simply different subsets of the electorate) or due to associational differences (they apply different logics of decision making). We can reject the second hypothesis and try to make sense of the compositional differences (e.g., pensioners and people from West Germany using postal voting more often) and the parties these groups disproportionately vote for (e.g., CDU/CSU). Maybe a topic that will be of even more (!) significance in the coming months.