Do party conventions and running mates move the polls?
The short answer
- A party's national convention usually lifts its presidential nominee in the polls; in U.S. races from 1964 to 1988, the typical gain was about 6 percentage points. (Campbell, Cherry and Wink, 1992)
- The bounce fades, but not overnight: in that study the direct effect wore off over about two months, so much of it carried into the fall campaign. (Campbell, Cherry and Wink, 1992)
- Whether a running mate wins extra votes at home is disputed; a 2019 study of individual voters from 1952 to 2016 finds no clear evidence of such a home-state boost. (Devine and Kopko, 2019)
What the studies found
- In Gallup and Harris polls from 1964 to 1988, a presidential nominee typically gained about 6 percentage points after the party's convention, and 4 of the 14 conventions gave double-digit gains. (Campbell, Cherry and Wink, 1992)
- Analyzing 185 polls from 1964 to 1988, the same study finds a convention's direct effect on a nominee's poll standing wore off over about two months: 54 days for Democrats and 59 days for Republicans. (Campbell, Cherry and Wink, 1992)
- The party holding the first convention of the summer usually got an extra boost, but the authors find that extra support was very short-lived, unlike the normal bounce. (Campbell, Cherry and Wink, 1992)
- Not every nominee gets a bounce: in that study, Lyndon Johnson's standing did not change after the 1964 Democratic convention, and George McGovern dropped two points after the 1972 one. (Campbell, Cherry and Wink, 1992)
- A later study of U.S. presidential elections from 1948 to 2000 estimates that about a third of the net convention bounce is temporary, while much of a convention's effect on voters lasts. (Campbell, 2001)
- In Gallup polls of registered voters from 1964 to 2008, the median nominee gained five percentage points after the party's convention; Bill Clinton's 16-point gain in 1992 was the record. (Gallup (Jones), 2012a)
- In Gallup's polls from 1964 to 2008, the only nominees who did not gain after their convention were George McGovern in 1972 and John Kerry in 2004. (Gallup (Jones), 2012a)
- Gallup finds bounces have shrunk: they averaged 3.8 points per candidate from 1996 to 2012, down from 6.2 points from 1964 to 1992. (Gallup (Jones), 2012b)
- In 2012, Gallup measured a three-point bounce for Barack Obama, from 47% to 50% of registered voters, and no gain for Mitt Romney after the Republican convention. (Gallup (Jones), 2012b)
- In U.S. presidential elections from 1884 to 2008, presidential candidates usually got a large boost in their home states, but running mates' home-state boosts were statistically negligible, appearing only with certain experience and state size. (Devine and Kopko, 2013)
- The same study finds that no home-state advantage, for either a presidential or a vice presidential candidate, changed the outcome of any presidential election since 1884. (Devine and Kopko, 2013)
- In their book, the same authors conclude a running mate's home-state advantage is highly conditional and rarely decisive in the Electoral College, though it could decide an election under narrow conditions. (Devine and Kopko, 2016)
- Using national surveys of voters from 1952 to 2016, a 2019 study finds no evidence of a statistically significant home-state boost for running mates. (Devine and Kopko, 2019)
- Another study of elections from 1884 to 2012 disagrees: comparing each home state with a blend of similar states, it finds running mates do raise their ticket's vote at home, by more than earlier studies found. (Heersink and Peterson, 2016)
- In U.S. presidential elections, feelings about running mates typically play little role in how people vote, but they matter much more when a running mate draws heavy media attention, as Sarah Palin did in 2008. (Ulbig, 2010)
- In 2008, voters' views of Sarah Palin strongly predicted their vote, the largest effect of any running mate in the elections studied back to 1980, and she may have cost her ticket support among swing voters. (Knuckey, 2012)
What the evidence can't tell us
- The detailed study of bounces covers only 14 conventions from 1964 to 1988, a set its authors call small in number. (Campbell, Cherry and Wink, 1992)
- Gallup measures a bounce by comparing its last poll before a convention with its first poll after, so the number shows a short-term change, not the effect on election day. (Gallup (Jones), 2012a)
- Researchers have not settled the running-mate question: the 2019 study says its results conflict with the synthetic-control study and with its own authors' earlier finding of a boost in some states. (Devine and Kopko, 2019)
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.
- James E. Campbell, Lynna L. Cherry and Kenneth A. Wink (1992). The Convention Bump. American Politics Quarterly 20(3), now American Politics Research; read as the published version posted on the first author's University at Buffalo site. Free copy · Publisher's page
- James E. Campbell (2001). When Have Presidential Campaigns Decided Election Outcomes?. American Politics Research 29(5); read as the published version posted on the author's University at Buffalo site. Free copy · Publisher's page
- Jeffrey M. Jones, Gallup (2012). Candidate Support Typically Up Five Points After Convention (August 23, 2012), with its page 2 table 'Change in Candidate Vote Support Before and After Conventions, 1964-2008 Gallup Polls'. Gallup News (polling organization's report of its own survey data). Free copy
- Jeffrey M. Jones, Gallup (2012). Obama Gets Three-Point Convention Bounce (September 11, 2012). Gallup News (polling organization's report of its own survey data). Free copy
- Christopher J. Devine and Kyle C. Kopko (2013). Presidential Versus Vice Presidential Home State Advantage: A Comparative Analysis of Electoral Significance, Causes, and Processes, 1884-2008. Presidential Studies Quarterly 43(4); abstract read on the University of Dayton eCommons record (the authors' accepted manuscript is attached there). Free copy · Publisher's page (summary only: no figures taken from it)
- Christopher J. Devine and Kyle C. Kopko (2016). The VP Advantage: How Running Mates Influence Home State Voting in Presidential Elections. Manchester University Press (book); abstract read on the University of Dayton eCommons record. Free copy (summary only: no figures taken from it)
- Christopher J. Devine and Kyle C. Kopko (2019). Bringing Voters into the Equation: An Individual-Level Analysis of the Vice Presidential Home State Advantage. Presidential Studies Quarterly 49(4); abstract read in its OpenAlex record. Free copy · Publisher's page (summary only: no figures taken from it)
- Boris Heersink and Brenton D. Peterson (2016). Measuring the Vice-Presidential Home State Advantage With Synthetic Controls. American Politics Research 44(4); abstract read in its OpenAlex record. Free copy · Publisher's page (summary only: no figures taken from it)
- Stacy G. Ulbig (2010). The Appeal of Second Bananas: The Impact of Vice Presidential Candidates on Presidential Vote Choice, Yesterday and Today. American Politics Research 38(2); abstract read in its OpenAlex record. Free copy · Publisher's page (summary only: no figures taken from it)
- Jonathan Knuckey (2012). The "Palin Effect" in the 2008 U.S. Presidential Election. Political Research Quarterly 65(2); abstract read on the University of Central Florida STARS record. Free copy · Publisher's page (summary only: no figures taken from it)