Do endorsements change votes?
The short answer
- Endorsements can change some votes, but how many depends on surprise: in the 2000 and 2004 U.S. presidential races, unexpected newspaper endorsements moved readers far more than expected ones. (Chiang and Knight, 2008 (published 2011))
- In U.S. House and state legislative races, which draw less attention, a newspaper endorsement reprinted on Election Day appears to give a small boost, but it showed no clear effect in governor's races. (de Leon, 2013)
- A rare celebrity endorsement can matter in a primary: Oprah Winfrey's backing of Barack Obama in the 2008 Democratic primary is estimated to have won him about 1 million extra votes in the states studied. (Garthwaite and Moore, 2013)
What the studies found
- A study of the 2000 and 2004 U.S. presidential races used surveys of 32,014 newspaper readers, comparing those interviewed before their paper published its endorsement with those interviewed after. (Chiang and Knight, 2008 (published 2011))
- In 2000, surprising presidential endorsements by the Denver Post and the Chicago Sun-Times convinced about 3 percent of their readers to switch to the endorsed candidate, by the authors' estimates. (Chiang and Knight, 2008 (published 2011))
- Expected endorsements, such as the New York Times backing Al Gore and the Dallas Morning News backing George W. Bush in 2000, convinced less than one percent of their readers to switch. (Chiang and Knight, 2008 (published 2011))
- Treating every newspaper endorsement alike, the same study finds only a small average effect that cannot be told apart from zero, suggesting only credible, surprising endorsements move readers. (Chiang and Knight, 2008 (published 2011))
- The effect of credible endorsements was much stronger among moderate readers than among readers as a whole. (Chiang and Knight, 2008 (published 2011))
- In 2002 and 2006 U.S. House and state legislative races in eight states, candidates whose newspaper endorsement was reprinted on Election Day won a slightly larger share of the vote than other endorsed candidates. (de Leon, 2013)
- The same study finds no statistically significant Election Day endorsement effect in governor's races, where the author says voters are better informed. (de Leon, 2013)
- The Election Day gain came from newspapers backing a candidate whose politics differed from the paper's own, which the author reads as a sign that newspapers can sometimes give voters unbiased information. (de Leon, 2013)
- Comparing counties with more and fewer Oprah magazine subscribers, a study estimates Oprah Winfrey's 2008 endorsement gave Barack Obama about 1 million more primary votes in 45 states and Washington, D.C. (Garthwaite and Moore, 2013)
- The same study finds the endorsement also raised overall turnout in the 2008 Democratic primaries and increased donations to Obama's campaign. (Garthwaite and Moore, 2013)
- In 2002 U.S. congressional primaries, a candidate's share of the endorsements from party-linked people and groups affected the candidate's vote share, even accounting for money and candidate quality. (Dominguez, 2011)
- In a 2006 field experiment in two close Pennsylvania state house races, a well-known liberal group endorsed Democrats, and Republicans used that endorsement as a cue to vote against them. (Arceneaux and Kolodny, 2009)
- Survey experiments on local elections find that party endorsements, endorsements from public officials or a nonpartisan voter guide can help voters pick candidates whose policy views match their own. (Boudreau, Elmendorf and MacKenzie, 2015)
- In multiethnic local elections, endorsements from ethnic groups made voters weigh candidates' ideology less, and among white voters this reaction was driven by stereotypes. (Boudreau, Elmendorf and MacKenzie, 2019)
- Outside the U.S., when several prominent British newspapers switched to backing the Labour Party before the 1997 election, they persuaded a considerable share of their readers to vote Labour. (Ladd and Lenz, 2009)
What the evidence can't tell us
- Because Oprah Winfrey is so unusually popular, the authors say their estimate is likely the most a celebrity endorsement could do; since it covers one party's primary, they say it is unclear whether a general election would work the same. (Garthwaite and Moore, 2013)
- Endorsement effects are hard to measure, because it is difficult to find a way to gauge an endorsement's reach that is not tied to how much voters already liked the candidate. (Garthwaite and Moore, 2013)
- Even where the Election Day endorsement effect shows up, its author notes it is small and unlikely to decide who wins. (de Leon, 2013)
Sources
Every study here was read in a free copy, linked below. A number on this page comes only from a study read in full; one read only in its summary is described in words.
- Brian G. Knight and Chun-Fang Chiang (2008). Media Bias and Influence: Evidence from Newspaper Endorsements. NBER Working Paper 14445 (October 2008); later published in Review of Economic Studies 78(3), 2011, doi 10.1093/restud/rdq037. Free copy · Publisher's page
- Fernanda Leite Lopez de Leon (2013). The Tuesday Advantage of Politicians Endorsed by American Newspapers. B.E. Journal of Economic Analysis and Policy 13(2); read as the published version in the University of Kent's open repository. Free copy · Publisher's page
- Craig Garthwaite and Timothy J. Moore (2013). Can Celebrity Endorsements Affect Political Outcomes? Evidence from the 2008 US Democratic Presidential Primary. Journal of Law, Economics, and Organization 29(2); read as the published version posted on the first author's Northwestern University faculty page. Free copy · Publisher's page
- Casey B. K. Dominguez (2011). Does the Party Matter? Endorsements in Congressional Primaries. Political Research Quarterly 64(3); abstract read in its Crossref record. Free copy · Publisher's page (summary only: no figures taken from it)
- Kevin Arceneaux and Robin Kolodny (2009). Educating the Least Informed: Group Endorsements in a Grassroots Campaign. American Journal of Political Science 53(4); abstract read on its HAL (Sciences Po) record. Free copy · Publisher's page (summary only: no figures taken from it)
- Jonathan McDonald Ladd and Gabriel S. Lenz (2009). Exploiting a Rare Communication Shift to Document the Persuasive Power of the News Media. American Journal of Political Science 53(2); abstract read in its Crossref record. Free copy · Publisher's page (summary only: no figures taken from it)
- Cheryl Boudreau, Christopher S. Elmendorf and Scott A. MacKenzie (2015). Informing Electorates via Election Law: An Experimental Study of Partisan Endorsements and Nonpartisan Voter Guides in Local Elections. Election Law Journal 14(1); abstract read in its OpenAlex record. Free copy · Publisher's page (summary only: no figures taken from it)
- Cheryl Boudreau, Christopher S. Elmendorf and Scott A. MacKenzie (2019). Racial or Spatial Voting? The Effects of Candidate Ethnicity and Ethnic Group Endorsements in Local Elections. American Journal of Political Science 63(1); abstract read in its Crossref record. Free copy · Publisher's page (summary only: no figures taken from it)