Poor Well-Being Associated with Shift to Trump in 2016


  • Counties with low life ratings more likely to shift to Trump in 2016
  • Improved life satisfaction since 2012 linked to Clinton support
  • Daily anger and worry show no relationship to voting shifts

WASHINGTON, D.C. — U.S. counties in 2016 that took the biggest jump in voting Republican for president between 2012 and 2016 had residents who reported below-average satisfaction with their current lives and below-average optimism for the future. Accordingly, counties that shifted most strongly to the Democratic candidate reported above-average life satisfaction and higher optimism for the future.

Poor Well-Being Associated with Shift to Trump in 2016

These county-level findings are based on more than 176,000 interviews with U.S. adults across all 50 states and the District of Columbia, conducted from January to December 2016. Over 3,000 counties nationwide were aggregated into one of six clusters based on the change in their vote for president from 2012 to 2016, ranging from more than 10-percentage-point Republican gains in 2016 to more than 10-point Democratic gains. Examining 2016 data that were restricted to just January through October, thus excluding results captured after the election itself, resulted in very similar results to those found from the entire year.

This analysis was done in partnership with researchers at the Yale School of Medicine and the University of Cincinnati College of Medicine and was originally published in PLOS ONE.

Positive Daily Emotions, but Not Negative, Linked to Shift

People’s assessments of how satisfied they are with their life at present and their forecast for how satisfied they will be in the future are measures of evaluative well-being, or how individuals assess their lives generally. People’s reports of their daily emotions, such as happiness, sadness and anger, are, in turn, measures of experiential well-being, or emotions experienced on a day-to-day basis.

Unlike the analysis of presidential voting trends and evaluative well-being, which points to a strong association between satisfaction levels and party preferences, the results pertaining to presidential vote and experiential well-being are mixed. Residents of counties that voted more heavily Democratic for president in 2016 than in 2012 had residents who were more likely to report happiness and enjoyment on any given day than residents of those counties that voted more Republican.

Reports of negative emotions, in contrast, did not show a clear pattern across county clusters based on presidential vote change. Residents of each cluster were similarly likely to report significant anger or worry on any given day. And the cluster that shifted most strongly to the Democratic candidate had slightly higher worry than any of the remaining groups.

Poor Well-Being Associated with Shift to Trump in 2016

Red-Shifting Counties Had Smaller Improvement in Well-Being

Residents of counties that shifted more Republican in 2016 not only rated their current and future lives worse than their left-shifting counterparts, but also reported lower levels of improvement in these measures since 2012, a period of time when both metrics were steadily improving nationally among all adults. Those who lived in counties that shifted at least five percentage points more strongly to Donald Trump in 2016 than Mitt Romney in 2012 had roughly half the improvement over this four-year span as did the residents of counties that voted for Hillary Clinton by a similarly large margin compared with Barack Obama four years earlier.

Poor Well-Being Associated with Shift to Trump in 2016


The 2016 U.S. presidential election has been analyzed through many means, but not explicitly as a function of population well-being at the county level. The results of this analysis strongly suggest that current life satisfaction, future life optimism and experiencing positive daily emotions are all tied to shifts in voter preferences, with stronger or more greatly improving levels of each associated with shifting toward the political party that was currently in the White House (e.g., voting more strongly Democratic in 2016 than in 2012). Lower or worsening well-being, in contrast, is associated with shifting away from the status quo (e.g., voting more strongly Republican in 2016 than in 2012).

The absence of a relationship between specific negative daily emotions and these same shifts belies some general observations pertaining to the high levels of anger, stress and worry observed among supporters of Trump at his campaign rallies. Anecdotal evidence such as this may be legitimate, but the research shows that counties that shifted to the right in 2016 were no more likely to be populated with residents experiencing anger on any given day than those counties that shifted to the left.

For leaders, these results underscore the relevance of the well-being of their constituents to major electoral outcomes. Those who are better able to pursue and realize a life well-lived are more likely to prefer maintaining ideological continuity in the White House, while their counterparts with lower — or worsening — well-being are more likely to seek out change.

Much thanks to Jeph Herin, Brita Roy, Carley Riley and Harlan Krumholz for their collaboration and partnership in this research.


Results are based on 176,622 telephone interviews conducted Jan. 2-Dec. 30, 2016, and 352,653 interviews conducted Jan. 2-Dec. 30, 2012, as a part of the Gallup-Sharecare Well-Being Index, with a random sample of adults, aged 18 and older, living in all 50 U.S. states and the District of Columbia.

The margin of error for each reported county cluster varies based on the total number of respondents in each cluster. See table below for typical margins of error for each group based on the 2016 sampling. Margins of error for 2012 results would be smaller. All reported margins of sampling error include computed design effects due to weighting.

Poor Well-Being Associated with Shift to Trump in 2016

Each sample of national adults in 2016 included a minimum quota of 50% cellphone respondents and 50% landline respondents, with additional minimum quotas by time zone within region. Each sample of national adults in 2012 required a minimum of 40% cellphone respondents. Landline and cellular telephone numbers were selected using random-digit-dial methods.

Learn more about how the Gallup-Sharecare Well-Being Index works.