Did election results trump frames of newspaper endorsements?

With the endorsement of only two of the top 100 circulation newspapers in the US, Republican Donald J. Trump stunned the country by becoming the 45th president of the United States on November 8, 2016. Never before in the history of US politics had a presidential candidate received so few major newspaper endorsements.

Democrat Hilary Clinton was endorsed by 57 newspapers while Libertarian candidate Gary Johnson was endorsed by 4, and 3 newspapers recommend ‘Not Trump.’ The other 31 either did not endorse as a matter of principle or chose none of the candidates.

In comparison, in 2012 President Obama was endorsed by 41 of the top 100 circulation newspapers and Republican Mitt Romney 35; the other 24 newspapers did not endorse.

Was Trump’s victory as stunning as a rebuke to the influence of newspaper endorsements as the election results were a surprise to most Americans – to most American opinion pollsters anyway?

That may be determined by how the frames used by the newspapers are understood. American newspapers have been steadily getting out of the presidential endorsement business during recent elections, framing their exit in terms of questioning the influence of endorsements.

Yet, in an interesting twist to the trend, this year some newspapers, such as USA Today, that previously refrained as a matter of principle from endorsing candidates at presidential elections jumped into the fray. Also, some that never or almost never endorsed a Democrat did so this time.

Many endorsements framed Clinton as flawed but acceptable, although many also went out of their way to say she was the best prepared presidential candidate ever. She was framed as having the character and temperament to be president.

Trump was framed ‘dangerous’ and ‘unfit’ because of personal comments and behavior that stoked racism, stirred anti-immigration sentiment, and disrespected women. He was framed as not having the character and temperament to be president.

And so, is it that newspaper endorsements – despite framing the candidates so drastically differently – did not have any influence in the election?

According to preliminary results, Clinton actually won the popular vote total. Trump won the most Electoral College votes, which determines who wins the presidency. Each state has a number of electoral votes equal to the state’s number of members in the House of Representatives (which is based on the state’s population) and US Senate (each state has two Senators). The candidate who wins the most votes in a state wins that state’s electoral votes. 

An analysis of endorsements of top 100 circulation newspapers and voting results in swing states (whose election outcome typically cannot be predicted) suggests areas for further study.

Trump’s performance when endorsed shows he lost 46% to 48% in Nevada despite the endorsement of the Las Vegas Review Journal, a newspaper owned by one of his key wealthy supporters. He won Florida by the thin margin of 49% to 48%, after having been endorsed by the Florida Times-Union. In that state, four newspapers, including the largest, endorsed Clinton and one – the Palm Beach Post – did not endorse anyone.

Trump’s performance when not endorsed shows he won some swing states by comfortable margins: Iowa (52% to 48%) although its largest newspaper, The Des Moines Register, endorsed Clinton; North Carolina (51% to 47%) and Ohio (52% to 44%) despite being repudiated by multiple newspapers in both states which endorsed Clinton or did not make an endorsement.

Trump’s performance when not endorsed also shows some razor thin victories in swing states: Michigan (48% to 47%) and Wisconsin (48% to 47%). One might wonder whether the endorsement of Johnson on the Libertarian ticket by the Detroit News in Michigan and the recommendation of anyone but Trump by the Milwaukee-Wisconsin Journal Sentinel siphoned votes from Clinton. Many editorials framed a vote for Johnson as a vote for Trump.

Trump won Pennsylvania 49% to 48%. While Clinton won the big prize of the endorsement of the Philadelphia Inquirer, her reward might have been urban voters who would support her anyway. Neither she nor Trump earned the endorsement of the four other newspapers, two of which did not endorse and a further two did not endorse as a matter of principle.

Clinton won Virginia easily (50% to 45%) after its typically Republican leaning newspapers did not endorse Trump, although they did not endorse her either: one supported Johnson and the other did not endorse. Her vice presidential running mate was also from Virginia.

This analysis suggests that endorsements may still play an important role, and the frames used should be further explored. Perhaps the role of endorsements has changed and therefore the framing of endorsements should reflect that change. Maybe they already do so. This may be an invitation to other newspapers to come back to the presidential endorsement business. As some editorialists now say, endorsements no longer tell us how to vote, but rather they contribute to the conversation.