When Jon Stewart announced that he was leaving The Daily Show, media reports of how Americans were relying on the satirical news show as their news source had become commonplace. As the show had become increasingly political over the years, its popularity and influence had also soared. The political division of the country was also intensifying quickly which led to Jon Stewart leaving the show in 2015, announcing Trevor Noah as his replacement as host of The Daily Show. Shortly after Trevor Noah was announced as the new host, he faced backlash over a series of tweets that were largely deemed sexist and anti-Semitic. Noah’s response at the time dismissed the resurfaced tweets as “jokes that didn’t land.” He did not address any of the accusations directly, or explain how they might be unfounded or how he may have grown past the prejudice. This became a moment that I revisit whenever I consider how misogyny works on “the left”, especially misogyny in the perceived “left leaning” media. It was this moment that inspired my final project, when I came across the data that was published on FiveThirtyEight when the Jon Stewart’s run at The Daily Show ended.

The questions the visualizations in my final project address are about Jon Stewart’s run between 1999-2015. These visualizations were created to provide a view of the show’s guests as a case study of representation of women on a liberal leaning, or widely perceived as liberal, show. More specifically these visualizations focus on if/how the demographic makeup of guests shifted as The Daily Show became more political. Were women represented more or less in the show’s guest roster? If any patterns emerge, do they indicate a gender difference among professions of the guests or their age? Are there any identifiable trends within these categories? While representation alone cannot speak to the content and the nature of the conversation that took place, it speaks to which demographic and whose voices the show provided a platform for. At the very least an examination of the show’s guests can reveal attempts to be more exclusive by including more guests. The primary audience for this project are gender studies, media studies, and political science researchers interested in media representation of women and queer people. More specifically any researcher investigating spaces of more insidious and unexamined forms of misogyny, especially on the left. 

The data for this project originated from the collection of a dataset from FiveThirtyEight and it was originally collected from Google Knowledge Graph, The Daily Show clip library, and Wikipedia. One of the major obstacles in working with this data was the lack of detail about the guests, and the constant conflation of the categories that denoted industry and occupation. The original FiveThirtyEight data contained two categories that required cleaning; they were titled “group” and “google knowledge Occupation.” Group was meant to indicate the industry where a particular guest worked, and google knowledge occupation was intended to indicate the occupation or profession that the show identified them as. In reality, for many of the guests these two categories overlapped, or as was more common, they were confused one for the other. In order to rectify this, I retitled “group” as “industry”, and replaced “google knowledge occupation” with “occupation.”

Apart from cleaning these categories, in order to explore further the background of the guests, I added data about gender, age at time of appearance, political orientation of guests (when applicable). The age data was collected from google, official sites, IMDB, and press releases put out by publishing houses. Additional information on gender and political affiliation was collected from publications, interviews, official sites, and news items. Because these visualizations seek to show gender representation, I marked all guests as the gender they presented as guests on the show. One of the earliest decisions about the data was about trans and gender fluid people who were featured on the show. Guests whose gender identity does not match what they presented at the time of their appearance on the show would not be identified by their dead names on the visualizations. At the end, the cleaning process revealed that there were three guests who had affirmed their gender to be different from what was represented at the time of their appearance on the show. Cleaning over 2,600 lines of data while researching and entering birthdays and nationalities took an inordinate amount of time, and required adjusting the original scope of this project. At their final presented stage, the visualizations investigate trends that appear over time in regards to occupation, industry, and gender of guests.

The first question that the visualizations explore is the overall representation of female guests over time and in total for the entire run of the show. To best visualize breakdown of guests by gender over time I used a stacked graph, and a line chart to best visualize the major patterns and differences between genders. The total percentage of guests is also represented in a simple pie chart. These visualizations together give a striking first look at the major patterns of the gender of The Daily Show guests, especially the outstanding disparity of guests; only about 23% of the total guests in the entire run of the show were women. The design of this page and the project as a whole was inspired by Edward Tufte: it seeks to bring absolute attention to the data, with visual clarity. The visualizations are ordered sequentially as an investigation looking for patterns, and then further exploring them into detail. 

The next step of the project investigates patterns within each gender, looking closely at the industry and the occupation of each guest as it relat. A tree map graph is the ideal visualization to show the breakdown of a particular group. For this enquiry, I produced two tree maps side my side for each gender represented in the show, separating the entertainment and politically related projects by color. The graph also shows a breakdown of professions within each group clearly; more female guests come from the world of entertainment and of this group the percentage of actors is greater than that of the male guests. 

After visualizing the gender breakdown of guests over time and the professional make up of male and female guests, the next step was to look closer at the patterns that the visualizations revealed. The 2006-2009 period shows the greatest gender disparity in number of female vs male guests. This dashboard shows the trend in entertainment guests vs guests from politically related professions, to show any particular pattern about this time period. It again lists both genders side by side to highlight the differences or similarities in patterns. What the line graph of occupations reveals is that in the 2006-2009 period most of the guests came from politically related fields. Women guests dropped considerably in number and do not show the same increase in politically related professions. This spike is most significant during the 2008 election year showing a sharp increase in male journalists. A similar increase at a modest scale is shown in the female guests, who made up an even smaller percentage of the guests that year. The tooltip in the line graphs show the total number of guests for each gender by field, highlighting the disparity in gender representation that year. Overall these visualizations show a pronounced disparity of gender in The Daily Show guests, which female guests making up less than a quarter of the total guests. They also show that the more political the show became, the fewer female guests were featured on the show. In terms of the professional breakdown of each gender, the visualizations also show that the show’s political bent was driven by male guests.

The first stage of visualizations for this project that was presented in the original pinup included fewer visualizations, each shown separately. The visualizations reflected the earlier stage in the investigation, before a clearer story emerged. The decision to present the visualizations as a story was based on the feedback received after the first pinup, which showed that viewers responded to the premise of the investigation but were looking for a more cohesive story. This also affected the number of visualizations and allowed for a minimal amount of text needed for each page of the story. 

Building the data to include age, ethnicity, and political orientation can add new steps in the visualizations. But it is also important to bring this data in conversation with Trevor Noah’s run at The Daily Show and with Stewart’s new show, The Problem with Jon Stewart. Gender issues are much more central to our public conversation since the end of Stewart’s run at The Daily Show. These visualizations might reveal a much more balanced gender representation and specific patterns in response to more recent political moments. The time that bringing together this data demands in order to produce these next steps is outside the scope of a single course, but the data can be made public to allow others to expand on these questions.