Can newspaper endorsements predict elections? [INFOGRAPHIC]

posted by Kerri Sorrell

Newspaper endorsements are a hot topic recently, as papers all around the nation are choosing their candidates in these final campaign days. The legitimacy of endorsements on their ability to sway readers has become increasingly questioned – the majority consensus now seems to write off the influence of endorsements.

The Des Moines Register made national headlines this weekend with its endorsement of Mitt Romney (the paper hasn’t endorsed a Republican candidate since Richard Nixon). At least 4 other Iowa newspapers, including the Cedar Rapids Gazette and Quad-City Times, have also endorsed Romney. Obama has seen his lead in Iowa slip in the past few weeks – a stronghold in the 2008 election, Iowa is now a tossup. This peaked my curiosity – what did endorsement spreads look like in other battleground states? Could these endorsements predict how a battleground state would vote, even if they don’t influence those votes?

I did some research and compiled the infographic below. I looked at 8 states historically categorized as battleground states (states in which presidential elections were decided by single digits in at least 6 elections in the last 44 years). For each state, I looked at the percentage of newspaper endorsements for each candidate, and compared that data to the winner of the popular vote. I did this for the last 4 elections to determine if there was an overarching pattern between who newspapers endorse in battleground states and how those states vote. Side note: I also included national data for comparison.

What I found was interesting. From this data, I calculated that endorsements and votes in battleground states matched up in 59% of the elections (including 2012 projected data – the percentage is 75% without it). The two data fields matched up the most in Colorado, Ohio and New Hampshire (both with 75% matches), and the least in Wisconsin, Iowa, Virginia and Nevada (25% matches).

Because only 59% of the endorsements matched how the state voted (again, 2012’s projected data skews this data), it’s hard to say if newspaper endorsements in battleground states can predict how a state will swing. Colorado, Ohio and Virginia are three of the most crucial states in this election – interestingly enough, all three states are projected to swing Obama, but have a higher percentage of Romney endorsements. As election day draws nearer, it will be interesting to see how these states lean, and if the 2012 endorsement records strengthen the pattern.

What patterns do you find interesting in this data? Anything that surprises you?

*data and figures collected from the NYT, Smart Politics, Editor & Publisher, George Washington UniversityThe American Presidency Project and Daily Kos

Advertisements

6 responses to “Can newspaper endorsements predict elections? [INFOGRAPHIC]

  1. One question I have would be what percentage of newspapers in Colorado, Ohio, and Virginia are located in liberal versus conservative areas. For example, does Colorado have more liberal people centered around Denver and one or two large newspapers, while the rest of the state being mostly conservative has many more newspapers to draw endorsements from?

    • Beau, thanks for your comment. This is something I thought about with all of the states – I considered just drawing endorsement data from the major newspapers in the state, but decided to draw them from all over the state. This by no means included ALL of the states’ newspapers, but did give equal weight to smaller newspapers.

      This is where I found a majority of my endorsement data: http://www.gwu.edu/~action/2008/media08/endorsements/endorse08newspgen.html
      George Washington University has this data for at least the last three elections, and seemed to make a valuable effort to seek out endorsements from a variety of publications. To the best of my knowledge, it is as exhaustive a list as possible.

  2. Your infographic proves that newspapers’ endorsements receive excessive hype. I agree that newspapers’ endorsements, like polls, can’t predict an election’s outcome.

    Plus, I feel newspapers’ endorsements detract from the issues at hand. When the Register endorsed Mitt Romney recently, the reasons behind its endorsement didn’t capture readers’ attention. Rather, the Register’s endorsement history caught readers’ attention, overshadowing the motives behind that endorsement (i.e. Mitt Romney’s economic plan).

    Also, thanks for providing an easy-to-read, useful infographic.

  3. I have to agree with Taylor. Newspaper endorsements receive too much hype and can’t predict elections. More people are concerned about the endorsement than the reasons for giving it.

    The most interesting paper is the how many newspapers have giving Romney their endorsements is much different than the projected results of the elections. Does this represent the shift in corporate ownership of newspapers versus the centrist nature of the electorate.

    I think the important thing to remember with any poll or study is this correlation doesn’t equal causality. You can predict what may happen from a poll, but that doesn’t mean it will.

  4. I’m sorry, I can’t really comment on anything because I am too distracted by how beautiful your infographic is. Wow, you have a lot of talent.

    But I agree with Jeff and Taylor, newspaper endorsements receive too much hype and I think they shouldn’t get as much attention. They take away from the true purpose of a newspaper–to deliver news and information factually.

  5. Pingback: Can newspaper endorsements predict elections? [INFOGRAPHIC] « Kerri Sorrell

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s