Abstract: Political Humor and Anxiety

When a person is confronted with political humor that is dissonant with their political opinions, does this increase their levels of anxiety or does the presence of humor still reduce anxiety? Previous research has found that simple one-liners and jokes decrease anxiety, but political humor’s effect on anxiety remains untested. Satire targeted toward a favorable group might have an ego-depleting effect which would reduce the subject’s ability to manage anxiety. Alternatively, humor that makes the subject feel superior to an out-group can contribute to a boost in self-esteem that buffers anxiety. In our increasingly heated political environment, political humor is ubiquitous and unavoidable. Through this research project, I hope to determine the effect that political humor has on anxiety levels and whether this is significantly different from the effect of other kinds of humor.

While some research on this topic exists, it typically consists of experiments that use a single video stimuli of parody news programs. I specifically plan to create a stimulus formed from a series of multiple politically humorous images, similar to what one might find scrolling through a social media site, to replicate a more natural exposure to humor stimuli.

The primary focus of the summer phase of this project is to code a program in R or Python that is capable of webscraping multiple popular social media sites and pulling the most-liked politically humorous images, since picking stimuli that are consistently humorous, topical, and political is difficult to replicate for future researchers. By automating this process, I hope to reduce bias, variability, and time spent in running repeated trials or replicating the experiment. This code will be used to select the stimulus for a submission to the Omnibus Project (a survey given primarily to Government department students at William and Mary). While testing this code during the summer, I will also take notes on key aspects of the potential stimuli, including the word-to-image ratio, the presence of political figures (caricatured and non-caricatured), and the general topic of the humor.


  1. iawilliams says:

    Interesting topic! This has some ethical questions I’m wondering about though – will you pull from pages where people have a reasonable expectation that their words are completely public and available to any researcher as public information? How will you determine whether these posters expect privacy? It seems that public twitter accounts will have a reasonable expectation that their words are completely public, but what about Facebook? Will you access groups which must approve members and thus have some expectation of privacy, or just pull from pages rather than individuals?
    Also interested in seeing how you will quantify anxiety levels as a reaction to political humor. I’m sure there are some good sentiment analysis packages which aim to measure anxiety, but of course these only go so far. Additionally, how will you determine whether anxiety levels are a response to the humor, or the event itself?
    Very interested to see how this plays out! I think you have some come up with some interesting and creative variables to examine political humor. Good luck!

  2. Good points about the ethics. For that reason, I’m only pulling from public Facebook pages, which are visible to anyone, regardless of whether they have an account, so no issues with getting approved to get into groups (which would cause roadblocks if others tried to reproduce my method). Same for all other social media sites, I’ll pull from open/public accounts that any researcher could get access to.

    Anxiety analysis is going to be physiological (heart rate, skin pore dilation, etc), so I’m writing up protocols for BIOPAC sensors. Lots of work to be done there, since I’m not as familiar with phys equipment or baseline phys responses, but I’m excited to dive in!