During the (agonizingly) long period leading up to the 2016 presidential election, much ado was made about whether Republican Donald J. Trump's success was (partially) chalked up to his political outsider status. While it's hard to know for sure, some of my own research suggests, at the very least, that presidential candidates were motivated to spend much less time in 2016 focusing on their prior experience as government officials, at least when it comes to primary and general election debates.
Political scientists who study presidential debates have shown that candidates both influence and are constrained by public opinion. Presidential hopefuls must cater their debate performances to shifts in public opinion in order appeal to voters, while at the same time candidates must try to influence the agenda of debates to suit their own purposes. It's a delicate balancing act.
The content of presidential debates, therefore, can tell us a lot about public opinion at the time of the debates, as well as about the goals of presidential candidates.
The following visualization displays the expected proportion of every primary and general election debate from 2000 to 2016 (excluding town halls and vice presidential debates) that consists of discussion about prior government experience as a qualification for the presidency.
Political scientists who study presidential debates have shown that candidates both influence and are constrained by public opinion. Presidential hopefuls must cater their debate performances to shifts in public opinion in order appeal to voters, while at the same time candidates must try to influence the agenda of debates to suit their own purposes. It's a delicate balancing act.
The content of presidential debates, therefore, can tell us a lot about public opinion at the time of the debates, as well as about the goals of presidential candidates.
The following visualization displays the expected proportion of every primary and general election debate from 2000 to 2016 (excluding town halls and vice presidential debates) that consists of discussion about prior government experience as a qualification for the presidency.
Expected proportions were estimated using a structural topic model (STM) where topical prevalence across Democratic and Republican primary and general election debates was estimated for 20 topics from 2000 to 2016 using an unsupervised machine learning algorithm. "The Importance of Prior Government Experience" was one of the 20 topics identified by the STM. I was able to infer the topic based on highly associated key words identified by the machine learning algorithm.
It's clear from the model that 2008 was clearly the year to discuss prior government experience (especially for Republicans during the primary season). However, in 2012 and in 2016, the prevalence of this topic took an abrupt nose dive.
These results could suggest several things, but given prior research that has shown a symbiotic relationship between candidates' rhetoric and public opinion, these results may suggest a marked shift in public opinion regarding the importance of previous experience as a government official. While in 2008 past experience in government appears to have been more important, more recently candidates appear less motivated to discuss prior government experience (they even may have been motivated to avoid discussing prior government experience as a qualification for the presidency).
Of course, while these results are not perfect (text analysis does not offer a 100 percent guarantee that the results will be flawless), they have arguable merit given shifts in the political atmosphere leading up to the 2016 election. 2016 very well may have been the "year of the 'outsider.'"
______________________________________
For more details about how I created the above graphic, see this page (Making Better Graphics for Structural Topic Model Objects with R) on my new personal site hosted by GitHub.
Comments
Post a Comment