Did women and men react the same way to Obama’s State of the Union Address? In many ways, No. The Temnos Lab looked at over a half-million data points to find that out.
During January 2016 at Temnos, we used our own semantic analysis engines to get a picture of how people reacted, in public commenting, to President Obama’s State of the Union Address (SOTUA). Utilizing the DISQUS data stream, we took comments made by users on 3,782 news articles and blog posts within US media, from the time the President’s address began (6:00 pm PST on January 12th, 2016) until midnight PST the following day, i.e. the first 30 hours following the start of the speech. This yielded 402,035 posted comments and 211,839 up-or-down votes on those comments from other people. We analyzed these according to geographical region and gender, and compared the results to Obama’s speech itself, and to the media coverage during the same time period. Numerous interesting patterns emerged. In this post, I’ll highlight some of the differences we noticed between male and female commenters.
Mexico and immigration
US immigration policy and the relationship of the US to Mexico are apparently of more concern to women than to men, among people who commented on Obama’s SOTUA. Of the top ten most common topics related to Mexico and immigration issues (see Figure 1, below), women far exceeded men both in total comments made, and in the rank of those topics compared to other topics of discussion. Overall, this indicates a much higher interest in these issues on the part of female commenters — they made 55% more comments in these topics than the men did, despite that female commenters were outnumbered 2½-to-1 by their male counterparts (there were 151,462 identified males compared to 60,407 females in our SOTUA commenter population). This level of gender difference was the highest of all major topics that were present in the data.
Issues surrounding gun violence
Another major topic area where women and men differed strongly in their interest level, was on gun violence, where in three of the top five related topics women exceeded men in commenting volumes, while matching the men in the remaining two topics –- all while being outnumbered by men in the commenting population by 2½-to-1.
Mention of other political leaders besides Obama
Generally, males and females both made frequent mention of other political leaders while they were commenting on Obama’s address. This is not unlike the patterns we see all year long –- pick any politician, and in comments on an article dedicated to him or her, you will see names of dozens of others political leaders being mentioned. It was no different in comments responding to Obama’s SOTUA; we saw frequent mention of Hillary Clinton, Donald Trump, Joe Biden, etc. Generally, men and women did not differ greatly in the frequency with which they mentioned other political leaders beside Obama –- mentions of Hilary Clinton, Donald Trump, etc. were at similar levels between male and female commenters.
However, there were a couple of big exceptions: female commenters were five times more likely than males to reference Nikki Haley (the female Governor who made the Republican response to Obama’s SOTUA) and were seven times more likely than men to mention former Governor Sarah Palin. (This pattern did not extend to Carly Fiorina, who was mentioned scarcely at all by male and female commenters alike). Conversely, men mentioned Ben Carson considerably more than women did (see Figure 3, below), and they also mentioned Bill Clinton more than women did.
Emotive language in comments
It was striking how little males and females differed overall in their use of emotive language (“ridiculous”, “I hate”, “I admire”, “glad to hear”, etc.). Both males and females used at least some emotive language in approximately 95% of comments, with positive language well exceeding negative:
When we started to zero in on specific issues, however, we could see men and women begin to differ in their sentiments. One such difference was in responding to Obama’s stance on education and schools: men’s sentiments were rather balanced, with a positive-to-negative ratio of about 3-to-2; women were far more positive about the topic, with a ratio of 7-to-1 (positive-to-negative).
Men were more positive about the President on taxes, with about a 5-to-3 ratio of positive-to-negative sentiment, while women were evenly split on the issue, with almost a 1-to-1 ratio of positive/negative sentiment.
It was mentioned previously that women commented far more than men did on topics related to immigration and border control. We found a difference in sentiment here as well.
Women expressed both positive and negative sentiment about Obama’s stance on immigration, but their positivity far outweighed their negativity. Although the same was true of men, it was by a slimmer margin of difference.
We did not find large, significant differences in sentiment between men and women on other issues.
Forthcoming reports from Temnos will outline differences in political commenting across different topics and across various regions of the US (Pacific states vs. Midwest, South, Northeast, etc.).