In the 2018 midterm elections, voter turnout among New Yorkers rose significantly for the first time since 2002, and voters turned out in record-breaking numbers for one of the most exciting midterm elections in recent memory.
In today’s post, we’ll take a closer look at the demographic characteristics that influenced voter turnout (both positively and negatively) at the neighborhood level. We used Neighborhood Tabulation Areas (NTAs) in our analysis, which are boundaries created by the Department of City Planning. To learn more about NTAs, visit the Department of City Planning's website here.
Race and ethnicity, level of education, and age were the strongest positive predictors of neighborhood turnout in the 2018 midterm elections.
Race and Ethnicity
Race and ethnicity were two of the strongest predictors of voter turnout by neighborhood in the 2018 midterm elections. Neighborhoods with large African American populations saw the greatest positive impact on turnout, followed by neighborhoods with large Latino populations. In fact, the percentage of African American individuals in a given neighborhood had the strongest positive relationship with voter turnout by neighborhood among all the characteristics we studied.
Age group was also a strong predictor of turnout by neighborhood. Neighborhoods with a large percentage of individuals between the ages of 30 and 39 saw the greatest positive impact on voter turnout of any age group, followed by the 70 or older age group and the 50 to 59 age group.
Income had a positive relationship with turnout by neighborhood, but this relationship was much weaker than the relationships for race and ethnicity and age group. Households with a combined income between $35,000 and $49,999 was the only income bracket that had a significant positive relationship with voter turnout in 2018.
Conversely, education, naturalization status, geographic mobility, and gender were the strongest negative predictors of neighborhood turnout in the 2018 midterm elections.
Education was one of the strongest negative predictors of turnout by neighborhood in the 2018 midterm elections. The percentage of individuals with some high school education (but no diploma) in each neighborhood exhibited the strongest negative relationship with neighborhood turnout. Turnout was negatively impacted at a similar rate in neighborhoods with large percentages of individuals with less than a ninth grade education and neighborhoods with large percentages of individuals who have only completed high school. This relationship with neighborhood turnout was still negative for neighborhoods with large populations of individuals with some college but no bachelor’s degree or an associate’s degree, but this relationship was much weaker than it was for the less educated groups described above.
The percentage of naturalized citizens by neighborhood was another negative predictor of voter turnout by neighborhood. In fact, the impact of neighborhoods with large percentages of naturalized citizens on turnout was about as strong as the impact of neighborhoods with very low levels of educational attainment (i.e. no college education).
Neighborhood turnout was also negatively impacted in neighborhoods with a large percentage of individuals who had moved in the last year. While this relationship is a moderate predictor of negative turnout, it is weaker than level of education.
While we expected to see a negative impact on turnout in neighborhoods with low levels of educational attainment, a large immigrant population, and a highly mobile population, it is unusual for the percentage of women in a neighborhood to have a negative impact on turnout. This is because women are generally more likely to vote than men. However, in 2018, large female populations did have a moderately negative impact on voter turnout at the neighborhood level.
To learn more, check out our full report here. And make sure to take a look at our interactive maps showing how turnout through the city looked from the primary to the general elections in 2018, as well as how these demographic factors and others impacted turnout.