On Nov. 4, 2020, researchers at the University of Maryland’s Philip Merrill College of Journalism conducted a national survey of 755 registered voters to measure voting behaviors, media use, perceived media bias, and trust in both the news coverage and the electoral process.
Using a 1-5 scale to rate their “faith in the electoral system” (with 5 representing the highest), respondents identifying themselves as Republicans’ averaged 4.0, which was significantly higher than the average 3.7 faith rating by Democrats.
Confirming national voting patterns, 91% of Republicans reported they voted in person versus 64% of the Democrats. Only 9% of the Republicans voted by mail versus 36% of the Democrats. On a 1-5 scale rating how much COVID-19 affected the decision to vote by mail (with 5 indicating the strongest effect), the average rating from all respondents was 2.6. There was no significant differences in the effects of COVID ratings between Republican and Democratic mail-in voters.
Most respondents (45%) reported that social media was their primary source for election news. In second place was television for Republicans (31%) but websites and apps for Democrats (32%). It was the inverse for third place with television the primary source for Democrats (30%) and websites and mobile apps for Republicans (13%). On a 1-5 scale of engagement with 5 the maximum, Republicans reported higher engagement with Facebook (3.0) than Democrats (2.5). Republicans also rated Twitter’s influence for election news significantly higher (3.7) than Democrats (3.2).
In terms of perceived media bias, most participants from both parties indicated panels of pundits on cable-TV news programs — as opposed to traditional newscasts — are, in part, why participants think the news media is biased. By party, Republicans rated cable TV panel programs as more biased (4.2) compared to Democrats (3.7). Voters from both parties said they prefer traditional newscasts over the common pundit-panel news programs. However, more Republicans reported they would view fewer newscasts (3.9) than Democrats would (3.2) if newscasts were more traditional without pundit panels.
The largest proportion of voters (50%) said the reason for voting was their civic duty. Other reasons included a preferred candidate (20% of the voters), key issues (17%), and family and friends (10%). Only 2% said they were motivated to vote by the media coverage. Of the 28 nonvoting participants, only six indicated why. Two didn’t trust the electoral system and two respondents said the candidates were not appealing.
Politically, 47% of the respondents said they were Republican or leaning Republican, 39% indicated Democrat or leaning Democratic, and 13% were Independent. One percent did not report their party.
Of the 755 respondents, 727 reported voting in this election, including 60% men and 40% women. The average age was 37.5 years, and included 70% white, 14% Asian, 7% Black or African American, 4% Hispanic/Latinx and the remainder American Indian, Native Hawaiian or Pacific Islanders.
The survey also measured the extent to which voters engaged in election activities (i.e. registered, donated to a campaign, attended a rally or protest, displayed signage, or posted to social media). Results showed that registering was the most common activity and the second-most popular activity was displaying election signage (24% of Republicans, 22% of Democrats). Only donating to a campaign ranked differently with 3% of Republicans but less than 1% of Democrats donating money to a campaign. Overall, no significant differences in election activities were measured between parties.
Led by Dr. Ronald Yaros, Merrill College associate professor, Merrill director of Ph.D. Studies and Media Engagement Lab director, the research team included doctoral students Keegan Clements, Jodi Friedman, Dinfin Mulupi, Nataliya Rostova and Ho Chun Wong.
For more information, contact:
Josh Land
joshland@umd.edu
301-405-1321