Just 45% of voters are able to give an accurate description of the Electoral College.
Given that lack of knowledge, any polling data on changing the Electoral College should be reviewed with caution.
For example, 59% of voters initially say they favor replacing the Electoral College with a direct popular vote. Seventy-two (72%) percent of Democrats favor replacing the Electoral College with nearly half (49%) strongly favoring moving to a popular vote.
However, when even a little additional information is added, the numbers shift dramatically. We highlighted this with a question informing voters that eliminating the Electoral College could lead to many candidates in the race and a new president receiving just 30% of the popular vote. With that information, support for replacing the Electoral College dropped to 45%.
Many people assume you can make a major change like replacing the Electoral College without changing anything else. In this case, the standard assumption might be that we would have just two major candidates for president.
However, history suggests a major change will lead to other changes. In the 1970s, the system of nominating presidents changed to the current model of primaries in virtually every state. Prior to that change, there were landslide elections on a regular basis. Since then, the only landslide election was in 1984 when Reagan won by nearly 20 points and captured the Electoral College votes from 49 states.
Going further back, when the Constitution was was written the delegates assumed there would be a 50% turnover in Congress every election. That's the way it was in state legislatures at the time. And, that assumption held true for over a century. However, in 1968, Congressional turnover fell to single digits for the first time and we have had extremely low levels of turnover since then.
This Napolitan News Service survey of 1,000 Registered Voters was conducted online by Scott Rasmussen on October 2-3, 2024. Field work for the survey was conducted by RMG Research, Inc. and has a margin of error of +/- 3.1.
