Category Archives: Romney

Mitt Romney’s Twitter Gender Gap

Posted on by

There’s been much discussion lately about which presidential candidate is better for women, favored by women, and likely to win the women’s vote.  And while Twitter is a very unscientific way of estimating that, it is interesting to look at who’s actively supporting the candidates on Twitter.

One easy way to identify supporters is to look at people who retweet the candidate’s message — there’s no purer form of support on Twitter than that.  And when we look at who is retweeting each of the candidates, an interesting statistic emerges:

Mitt Romney's Retweet gender gap -- only 30% of retweeters are women

Barack Obama’s retweeters are evenly split between men and women, while Mitt Romney’s retweeters are 70% male and 30% female.

These numbers are even worse than the look, if this assessment of the overall Twitter community is correct: it says overall population of Twitter is 55% female, which means women are over-represented.  If we compensate for this so that gender mix on Twitter  matches that of the USA,  Romney drops to only getting about 26% of his retweets from women.

It is clear that, when it comes to Twitter, there’s a gender gap for Romney.  It will be interesting to see how this progresses through the campaign season.

Methodology:

I looked at all retweets for each candidate for the period 5/14 through 5/20, and estimated the breakdown of the retweet population by a statistically valid sampling.

There are two factors which come into play in this kind of analysis.  First is the sample size.  I looked at a large enough random sampling of retweets to make an observation with +/- 5% accuracy at a 95% confidence interval.  Second is determination on gender — for a discussion on that topic, see this page.

Who Gets Retweeted More Frequently: Obama or Romney?

Posted on by

I’ve seen a bunch of stories  lately that say that Romney gets more retweets than Obama does.  Actually, that’s not quite true — they say that, on average, each of Romney’s tweets gets retweeted more often than each of Obama’s does.

This is usually said as a means of explaining that, while Romney may have far fewer followers than Obama does (498,759 to Obama’s 15,409,788) Romney’s are far more enthusiastic than Obama’s.

But I wondered if that was true.  So I looked at all tweets each candidate’s official account sent (@MittRomney and @BarackObama) from 4/15 to 5/12 (four weeks), and determined the average retweet count for each:

Click to Enlarge

Obama does get more retweets, on average, for each of his tweets.  But given Obama’s 30-to-1 advantage in followers over Romney, it doesn’t seem like that much of an advantage!

But there is a reason why.  Let’s start with the number of actual tweets each candidate’s account sent during that time period:

Obama sent slightly more than 15 times as many tweets as Romney did.  So each Obama supporter had 15 times as many tweets to choose from to retweet.

When you take that all into consideration, the chart of total number of retweets is very telling:

As you can see, in total, the Obama followers are much more active at retweeting than Romney’s — by a factor of 20 to 1.

As a consolation, though, the average Romney follower is 50% more likely to retweet Romney than the average Obama follower is to retweet Obama:

But that’s pretty slim consolation in given Obama’s commanding lead in followers.

So if Team Romney wants to say their followers are more enthusiastic than Obama’s, that would be true. But it would be reasonable to suppose that if Romney grew his follower count to be in the same ballpark as Obama’s his average follower would be considerably less enthusiastic.  (Why? Because the more enthusiastic followers are likely to be the ones who followed him through the primaries — at least that’s my guess).

But Team Obama can take heart in knowing that their presence on Twitter still dominates Romney’s.

Presidential Campaign: Twitter Week in Review

Posted on by

It’s been an interesting week on Twitter for the presidential candidates.  Mitt Romney is finally the last man standing in the GOP.  Barack Obama has ramped up his campaigning and attacks against Romney. Let’s look at the campaigns’ week on Twitter in review:

Click on Chart to See Full Size

What does it all mean? Let’s walk down the stats…

Barack Obama still has a commanding lead in followers and tweets over Romney — Mitt is going to need to renew his engagement with Social Media if he wants to compete.  He’s been relying on the masses (urged on by the GOP and others) who like to stomp over Obama’s hash tags (see my recent post on how effective that has or has not been).  But that’s all negative — it doesn’t make the case for Romney very effectively.

We can see that last week Romney’s efforts on twitter have been lackadaisical: he barely managed to get out more than a tweet a day.  That’s not engagement, that’s disengagement.  Did he give the social media team the week off?

Obama, on the other hand, is on fire.  He’s averaging over 20 tweets a day, and set off Twitter fireworks on Saturday with 80 tweets.  I wonder if he’ll keep it up.

And the response from the tweeple matches what the candidates have been doing.  Obama gets mentioned nearly 5 times as often as Romney does.  Twitter is an idea amplification device: no signal in equals no signal out.

You might wonder: maybe most of the tweets about Obama are negative? After all, the republicans have made a concerted effort to hijack Obama’s hash tags.  Sadly for the GOP, that’s not happening.  Only 20% of the tweets mentioning the President were critical of him.  On the other hand, Romney negatives are are almost double at 36%.  And those didn’t seem to be driven by an organized democratic campaign.

Of course, Romney and Obama are most frequently mentioned by each other’s followers.  There’s a lot of anti as well as pro out there. Young Jeezy was an interesting blip in mentions, owing to Obama’s prediction of his future singing…

So if last week wasn’t so good for Mitt, will he step up his Twitter output this week? Will the GOP increase its hashtag hijacking to grow Obama’s negatives? We’ll see in a week!  Be sure to subscribe to updates from this website — we only send out notification of new postings, so there’s no spamming to worry about.

Methodology:

Want to know how I came up with this data?

  • Tweets by candidates look only at the @BarackObama and @MittRomeny accounts.
  • Mentions look for tweets containing one or more of @BarackObama, #Obama2012, @MittRomney, or #Romney2012
  • Sentiment was calculated by taking a random sample of all tweets during the week mentioning either candidate.  Each tweet was scored, after I read it, as being in opposition of the candidate (anti-), not in opposition of the candidate (pro-), or not really about the candidate (off-topic) The sample size was 383 for Romney and 384 for Obama, and the percentages reported are + or – 5% at a 95% confidence level.  There is a level of subjectivity in my scoring, but for most tweets it’s pretty darn clear.
  • The list of top mentions is fairly simple to calculate.

Tools used were custom Java analysis programs, MySQL database.  Charts were generated in Excel and the entire document was assembled in Photoshop Elements 10. All on a Mac.

Sorry you read this far 🙂 ?

The Limits of Automated Sentiment Analysis of Twitter

Posted on by

As part of my weekly analysis of the campaign race between Obama and Romney, I took a look at the sentiment of the tweets mentioning both of them.  That report will be up a bit later on today, but I wanted to share something interesting I’ve found while working on the report.

I did a manual sentiment analysis of the tweets to determine whether they are pro-candidate (or at least neutral) or anti-candidate.  The manual process consists of selecting a statistically valid sample at random and then reading each tweet to score it.  It’s slow, painstaking, but it produces the best results because I, as a human and one that is versed in current events, understand the text, the context, and can usually guess the writer’s intent.

But there are also automated tools for doing sentiment analysis, and I use these tools too.  These tools look at the text, do a simplistic parsing of the tweet, and assign a score based upon the words that are used: “Hate” gets a negative score while “Love” gets a positive score.  You might think that these tools are very crude, and you’d be right.  But that doesn’t mean they don’t deliver some useful insight.

Take, for example, last week’s (4/29-5/5/2012) tweets mentioning Obama.  Here’s how I scored them manually:

The margin of error is +/- 5% at a 95% confidence level (which is pretty much the gold standard for surveys).  How did the automated tool do:

Very close!  You’ll note there’s no “off topic” or “neutral” category in this chart.  That’s because a huge majority of tweets end up with a neutral score, and that’s really a failure of the scoring system rather than a real indication of indifference.  Still, both the 76% and 24% are within the margin of error of the manual survey, so we can say they produced the same results.  Fantastic! (Obama -Romney means I just looked at tweets that mentioned Obama but did not mention Romney — I didn’t want the score to be confused by negative-about-Romney tweets that mentioned Obama).

But does this always work? Let’s look at Romney’s sentiment as scored manually:

Mitt didn’t have that great of a Twitter week, although a lot of the negatives were other republicans not happy about losing Santorum and Newt, or people still supporting Ron Paul.  So a lot of the negatives are probably not supporters of Obama.

Let’s compare this to the automated sentiment analysis:

Quite different! Why? Mostly because although I tried to filter out posts that were really about Obama, there’s a lot of posts that snuck through without mentioning him.  #Julia and other anti-Obama hashtags were a common source.

The lesson here is that for automated tools you need to very carefully scrub the tweets being examined to make sure they are really on the topic you are interested in.  Or, put another way, because so many Romney supporters like to talk about how bad the President is instead of how great Romney is, it skews the automated analysis.

I’ll be trying to work in a better algorithm for analyzing sentiment, and I know others have made great strides in that direction as well.  But the key thing to remember is that automated sentiment analysis works best when the tweets are talking about the subject you’re interested in, and poorly when people mention your subject in passing while talking about something else.  When you see people give a score to a collection of tweets, you should generally assume that they are using automated tools rather than scoring them by hand.  And so you should keep in mind the limitations of those tools.

Obama Gets a Huge Boost from Jimmy Fallon – Week in Review

Posted on by

On Tuesday night, 4/24/2012. President Barack Obama went on the Jimmy Fallon show, and produced the best ratings in years for Fallon and a big spike in twitter mentions:

Click to Enlarge

We can see the next day Twitter was abuzz.  There’s been criticism of the President from the conservative side of things about the appearance, but it’s clear as a political move it was shrewd.  We’ll see in next week’s analysis whether Saturday’s appearance at the White House Correspondent’s dinner will have the sam effect.

Mitt Romney, on the other hand, got a bit of a boost from his primary victories on Tuesday, but it was no where near what the President saw.  Romney really needs to push harder on Social Media if he’s going to be competitive in the fall.  Fortunately, he has plenty of time to work on it…