Blais, André. "What Affects Voter Turnout?". Annual Review of Political Science, Vol.9 (2006): 111-125.

Blais, André. "What Affects Voter Turnout?". Annual Review of Political Science, Vol.9 (2006): 111-125.

• Source mine of early materials on factors affecting voter turnout available from page 111 to 116, also page 123.
• The pioneer studies of the political science behind voter turnout conclude that institutional factors are behind producing high levels of voter turnout, with nationally competitive districts -- meaning that parties did not have a clear majority in any voting district -- multiparty structures, unicameral legislatures, and other institutions correlating to voter turnout (111-112).
• Compulsory voting in elections consistently increases voter turnout by around 13%, however this is only true in established democracies. In newer democracies, where the law goes unenforced, there is no difference to turnout. Similar studies conclude that compulsory voting only increases turnout when non-compliance is criminalized (112-113).
• Voter turnout is higher in systems with large voting districts, because ever district because important to the political process. This is also true in systems using proportional representation voting. The reason for these changes, however, remains unexplored (113-114).
• Unicameralism in legislature increases voter turnout, with levels being lowest in systems where the upper house or executive branch is as powerful as the lower house of the legislature. Presumably, this is because people are more likely to vote in contests over powerful institutions which have greater affects on their daily lives. However, some studies have cast doubt on the correlation between institutional power and voter turnout (114-115).
• Age is an important demographic factor in voter turnouts, with older people much more likely to vote. For this reason, voter turnout dropped an average of 5% when states lowered the voting age to 18, as the potential voter population increased by far more than the number of voters (115).
• The admissibility of mail-in ballots, and other factors related to ease of access, are correlated to an increase in voter turnout, likely because people are more likely to do things if those things are easy (115-116).
• Voter turnout tends to increase at the individual level with socioeconomic factors such as age or household income, but it also increases on societal socioeconomic factors, with wealthier countries having a higher turnout overall. However, this relationship is not linear, with the gap being between the poorest countries who lack electoral infrastructure and all other countries (117).
• Poor economic situations also appear to increase voter turnout, although the relationship is not well understood. Additionally, exceptionally small countries have extremely high turnouts, but for an unknown reason not replicate by general trends in community size and voter turnout (117).
• Since the first studies in voter turnout in the 1980s, there have been debates over the effect of party systems on voter turnout, with both the strength of bonds between parties and social groups, and the number of political parties being suggested as variables associated with higher rates of voter turnout (118).
• Many assumed that greater numbers of political parties likely produce higher voter turnout because more parties means more electoral infrastructure to mobilize voters, or that a greater variety of parties will increase the likelihood of voters identifying with a political platform. However, most studies show a negative correlation between the number of political parties and voter turnout (118).
• It is not currently understood why more political parties typically results in a negative association with the voter turnout, and most explanations are without evidence (119).
• Beyond being able to say that neither the bonds between parties and social groups, nor the number of the political parties has a positive effect on voter turnout, very little is understood about the effect of electoral systems on voter turnout (118-119).
• Close elections are consistently correlated with increase in voter turnout (119). However, much of the substructure of this claim is not explored, as the point at which a close election because close enough to be statistically significant is not understood (120).
• The author argues that institutional factors matter less in affecting voter turnout than commonly supposed, and there effects are dependent on other factors such as historical or cultural setting (121).
• It is increasingly important to conduct more analysis of electoral data at regional levels or by parsing out the social and economic context of the elections, as some patterns may be specific to the context of that election (122-123).
• Based on a book by Dr. Mark Franklin, the author suggests that new studies looking at voter turnout should focus on the youngest available voters, who will not have developed a pattern of voting or abstaining. This should although the researcher to look more directly at institutional factors dynamically affecting voter turnout (121).
• The author raises three main objections to this recommendation  by Dr. Franklin, noting that the studies produced from this methodology will be flawed because the behavior of young voters is not generalizable across the entire population, because Dr. Franklin wrongly believes that 'voting culture' is not instilled prior to voting, and because the kind of model that Franklin proposes calls for the use of multilevel analysis (121-122).