[Summary] Polls or Pols? The Real Driving Force Behind Presidential Nomination
Marty Cohen, David Karol, Hans Noel and John Zaller
The Brookings Review Vol. 21, No. 3 (Summer, 2003), pp. 36 – 39
Published By: Brookings Institutions Press
Background and Introduction
Who or what decides which presidential candidate wins the party primary nomination? The authors argue against the party decline thesis, as well as against the theory that whoever fundraises the most wins. Instead, they say that party insiders play a central role in determining presidential primary outcomes, rather than early polling results or fundraising. During the invisible primary season, candidates attempt to gain the support—both nominal and financial—of party insiders, who judge the candidates based on character, skills, electability, and policy positions that are aligned with the party platform. Although gaining some financial resources matters, most candidates gain the minimum needed to continue the process, so nominal endorsements matter the most by influencing delegates and the public.
The authors used statistical regression analysis to test their argument, though they did not formally show the results via tables or figures. They examined poll results, party insider endorsements, funds raised before the primary, and party primary nomination results. Party insider endorsements were measured by tallying the endorsements found in various media sources by state politicians, celebrities, and presidential incumbents. Polling data was taken from Iowa residents before the primaries, which historically has had high predictive power for who wins the nomination. These data were from 1980 to 2000, and were collected at multiple times before the primary for each presidential election year. For this reason, the authors could also use a time series analysis to determine whether party insider endorsements are adjusting based on previous polling information, or if polling information tends to follow the number of party insider endorsements (as the authors predicted).
Results and Conclusion
The results of the regression analyses indicated that polls and party insider endorsements were equally likely to predict primary nominations for presidential candidates. When the researchers added the amount of funds raised into this regression, they found that this barely predicted nominations, especially compared to polls and endorsements. Therefore, this suggests that funding is not very important during the invisible primary, especially since many rich candidates ran but failed to secure the nomination. Furthermore, the authors found that early changes in endorsements lead to changes in polls at a later time, and vice-a-versa. However, they found that endorsements influenced polling results far more than polling influenced endorsements. Based on this information, the authors concluded that party insider endorsements have more impact on primary presidential nominations than money or polling, thus opposing the implications of the party decline thesis.