Suchergebnisse

Researchers in psychology have long known that preferences are constructed in the decision-making process, influenced by choice environments that trigger unconscious biases and heuristics. As a result, choices, including those of voters, can be manipulated by political information. Personalised political messages, designed to influence based on detailed personal profiles, can undermine voter autonomy. We suggest that these practices should therefore be regulated, and discuss policy options and approaches, specifically the appropriate balance between freedom of political speech and privacy rights and interests, the implications of voter analytics for the electoral process, and how and by whom sophisticated voter analytics practices should be regulated.

AbstractAre ordinary citizens better at predicting election results than conventional voter intention polls? The authors address this question by comparing eight forecasting models for British general elections: one based on voters' expectations of who will win and seven based on who voters themselves intend to vote for (including 'uniform national swing model' and 'cube rule' models). The data come from ComRes and Gallup polls as well as the Essex Continuous Monitoring Surveys, 1950–2017, yielding 449 months with both expectation and intention polls. The large sample size permits comparisons of the models' prediction accuracy not just in the months prior to the election, but in the years leading up to it. Vote expectation models outperform vote intention models in predicting both the winning party and parties' seat shares.

Low and uneven turnout is a serious problem for local democracy. Fortunately, one simple reform—shifting the timing of local elections so they are held on the same day as national contests—can substantially increase participation. Considerable research shows that on-cycle November elections generally double local voter turnout compared with stand-alone local contests. But does higher turnout mean a more representative electorate? On that critical question, the evidence is slim and mixed. We combine information on election timing with detailed microtargeting data that includes voter demographic information to examine how election timing influences voter composition in city elections. We find that moving to on-cycle elections in California leads to an electorate that is considerably more representative in terms of race, age, and partisanship—especially when these local elections coincide with a presidential election. Our results suggest that on-cycle elections can improve local democracy.