A new multi-university research project found that the debate did nothing to change voter preferences and that Biden is actually doing a better job of holding on to his supporters after the debate than Trump.
A multi-university project called CHIP50, which includes Northeastern University, Harvard University, Rutgers University, and the University of Rochester, issued a new report concluding that the first presidential debate had no impact on voter preferences.
The report found:
The predominant pattern we see in the data is stability in votersâ preferences. Overall, 94% of those who chose Biden and 86% of those who chose Trump in our May survey still preferred the same candidate after the debate. There was a small rate of change from Biden to Trump or vice versaâ less than 3% shifted in either direction. We observed more churn between those who favored either Biden or Trump and the âOtherâ category. About 4% of Bidenâs and 6% of Trumpâs supporters in the earlier wave shifted to âOtherâ post-debate. At the same time, 6% of âOtherâ shifted to Biden, and another 6% shifted to Trump.
The net effect of this churn was still quite small: out of a total sample of 1262 repeat respondents, there was a net shift toward support for Biden of 36 voters, and for Trump of 27 voters. Of course, the presidential debate was not the only consequential event of the last month that might have shifted respondent opinions. To understand the role of the debate, we also examined the shifts in candidate preferences from our May survey wave to the choice of the same respondents reported in June, but before the presidential debate took place. The shifts were similarly small, with 87% of Biden supporters and 85% of Trump supporters retaining the same candidate preference, and less than 2% switching in either direction between the two candidates.
…
The evidence available from CHIP50 does not substantiate the view that the presidential race shifted away from Biden in the immediate aftermath of the recent debate. We also suggest that media reporting on the presidential race should not interpret small apparent changes in voter preferences as reflecting a real change in the state of the race (where âsmallâ should roughly be interpreted as less than twice the margin of error of a survey). For pollsters who use a panel (that is, the same respondents at multiple points in time), we would also recommend reporting the rate at which respondents have shifted among different possible responses.
The Biden Crisis Is A Baseless Media Construct
Democrats who are calling for Biden’s removal are playing into a crisis that the media created. There is no statistical support for the idea that voters have left Biden or that there is an urgent need for Democrats to replace their nominee. The Biden debate crisis appears to be a media construct that panicked Democratic elites have played into. President Biden is doing a better job than Donald Trump of holding on to his supporters.
The dumbest thing that Democrats could do would be to remove the candidate that their voters want.
Indications are growing that Democrats are not in trouble but that the polls and the media are getting this election wrong.
If this trajectory continues, don’t be surprised if Biden and the Democrats pull off another “surprise” win in November.
Awards and Professional Memberships
Member of the Society of Professional Journalists and The American Political Science Association
