Obama Riding the Twitter Hashtag Waves, or the Half-Life of Hashtags

With the announcement of President Obama’s support for same-sex marriage, the hashtag #MarriageEquality has rocketed on Twitter.  That’s not surprising.

But what is surprising is how short lived these hashtags are.  They’re not wars, meant to last to November.  They’re not battles, defining a key component of the President’s strategy.  They’re just skirmishes — a quick engagement and then a fade out.

Here’s a chart of the four most recent Obama-related hashtags – #DontDoubleMyRate, #Forward, #Julia (that one was created by opponents), and now #MarriageEquality:

Click to enlarge image

As you can see, the pattern is for a big initial spike and then a quick decay until, about a week later, the tag is just noise.

This makes me wonder two things:

  1. Will the President and his team work to keep some hashtags (perhaps #Forward) alive? It will take near daily effort if they do.  They have to work to make the tag relevant over and over again.
  2. Or will we continue to see a sort of hashtag of the week between now and the election? If so, that will take a different kind of effort.  It’s like an endless cycle of week long marketing campaigns.

The one thing the chart shows is that, after a hashtag dies down, if the President’s team doesn’t (re-)launch a new one, someone else will.  Twitter is not a place for long term conversations, and the lesson for everyone is that it takes constant, sustained effort to try to dominate the conversation.

Did the Republicans hijack the Hashtag #DontDoubleMyRate ?

You see a lot of comments on Twitter and elsewhere that the republicans managed to hijack the hash tag #DontDoubleMyRate to use to express opposition to Obama.  Usually, these comments come with a lot of glee that the conservatives have out-smarted the liberals.

So, did the hashtag get hijacked?

In a word, No.

There were 48578 tweets between 2012-04-18 11:04:19 and 2012-04-28 12:49:03 that used the hashtag.  Of those, only 14% or approximately 7,000 contained anti-Obama or just conservative messages.  There were probably as many spam messages!

The conservatives can say, all they want, that they managed to take over #DontDoubleMyRate, but the facts in this case, as Steven Colbert says, have a liberal bias.


A random sample of 376 tweets out of the total 48578 were selected programmatically and then read, by me, to be assessed as being either “conservative” or not. I was generous in my assessment, as some tweets are not terribly articulate.  54 of them were plausibly conservative.  54/376 = 14.36%    Thus, the estimated percentage of “conservative” tweets is 14.4% +/- 3.5% with a 95% confidence level.  At a 99% confidence level, the +/- is 4.6%.