Many countries employ two-tiered proportional electoral systems in which seats are allocated based on both nationwide and constituency-level results. Nonetheless, whether voters are aware of the constituency-level context and how it affects electoral behavior has received limited academic attention. Drawing on individual-level data from the Norwegian Citizen Panel collected before and after the 2021 Norwegian parliamentary election, this study shows that voters are aware of the constituency-level context, as they adjust their expectations about a party's likelihood of winning a constituency seat based on local electoral strength and district magnitude. Moreover, the constituency-level context affects voting behavior: supporters of parties polling around the 4% national threshold are more likely to defect when their preferred party is electorally unviable in low-magnitude districts, whereas local party strength plays a limited role in high-magnitude districts. These findings highlight the importance of accounting for both national- and constituency-level factors when analyzing electoral behavior in districted proportional systems and demonstrate that variation in district magnitude creates unequal strategic incentives across constituencies.
Alexander Verdoes (Thu,) studied this question.