Tweets vs. Likes: An Analysis of Monkey Cage Data

A couple weeks ago I wrote a post about the question of why some blog posts are more likely to be tweeted and why others are more likely to be “liked”. I threw out some hypotheses, and got a lot of interesting suggestions in the comments.

Anton Strezhnev, a Georgetown University undergraduate and soon-to-be Ph.D. candidate in political science, took things one step farther, and actually tested two of my propositions using data from The Monkey Cage. Here’s how he got the data:

I wrote a quick screen scraping script and went through all posts from this February up until about May of last year. At some point after that, no likes or tweets appear to be recorded for most of the posts. In total, I scraped around 860 posts, 492 of which had both tweets and likes.

He has a long post on his blog that details his analysis (as well as providing some background on other similar types of analyses of data from other blogs), but in particular he finds that, despite some fairly rough data:

it does seem to suggest that Joshua’s hypotheses have some validity. Posts that have graphics or are funny are more “likable” while wonkier posts are more “tweetable.” I would add that the first relationship is a bit stronger than the second since its difficult to find a good measure of “wonkiness” (especially since almost all posts on the Monkey Cage are relatively wonky). That the more “tech” categories (Data and IT/Politics) had a positive effect on the Tweet Rating might lend support to Edwin Chen’s argument that the Twitter ecosystem is geared specifically towards technology nerds.

What does Strezhnev think this says about Twitter and Facebook?

I’m not sure that the results say much about the composition of Facebook vs. Twitter – a lot of people use both. I do, however, think that they may hint at a key difference in the content-sharing incentives behind both services. Facebook is much more graphically-oriented than Twitter. The new “Timeline” profile structure makes this absolutely clear. Moreover, pictures and videos receive much more visual prominence in a user’s Facebook feed than simple text. Therefore, there is a much greater chance that shared content will be noticed if it contains an eye-catching photo or graphic. Conversely, Twitter feeds are pure text, which means that graphics are not a means of distinguishing one’s tweets from those of others. The value of graphics is greater on Facebook, which gives users a strong incentive to share content that has some visual component in order to get noticed.

Certainly there are other possible explanations. Tweets tend to be more public than Facebook posts, which are aimed more at one’s circle of known friends and acquaintances. Even if the bases of users for both services is similar, there may be a difference in the types of people who prefer to use Twitter vs. those who prefer Facebook (nerds vs. “normal people”?)

Personally, I think the public vs. private dimension is an important one. I know, for example, that I do not tweet personal information about my family, but will put things about them in Facebook posts, precisely because I (hope!) I have some control over who sees the latter but not the former. Whether this translates to the types of material one is likely to “share” from third party sources is an interesting question. Again drawing on personal experience, I am much more likely to share something that is overtly partisan (thus revealing my own partisan preferences) on Facebook than I am in the public forum of Twitter, although that may be related to my own particularly idiosyncratic situation as someone who writes for a non-partisan politics blog and gets most of his Twitter followers (@j_a_tucker) through that blog. But as we move into the domain of trying to understand how social media may impact political behavior, it strikes me that this public vs. private dimension may be useful for distinguishing between the effects of different types of social media.

You can find Strezhnev’s full post here. As an added bonus, he’s made the data available in STATA format or in a tab-delimited text file. He also has made the screen scraper written in Python that he used to get the data available  here along with the conversion script to make the results usable in statistical software.  I am open to more guest posts on this topic if anyone is interested doing more data analysis!

5 Responses to Tweets vs. Likes: An Analysis of Monkey Cage Data

  1. John February 16, 2012 at 12:59 pm #

    My question would be on how the results may be different if there was a focus not just on likes (which are trackable) but also on the direct posting to Facebook without liking the page (which I would assume would be much harder, if not impossible, to track). I’m not sure it would change anything, but I think the two are definitely different actions. The like would seem to simply be a quick action, while the post being one that involves the need to occasionally introduce the topic. That might increase the desire to post wonkier things to Facebook if a good intro can bring non-academics into the conversation.

  2. Andrew Gelman February 16, 2012 at 8:35 pm #

    It’s all about incentives. John Sides pays us double for posts that get at least 10 tweets or 10 likes, and triple for posts that get at least 50 tweets+likes.

  3. Jose C Silva February 18, 2012 at 3:42 pm #

    I’ll probably write a deeper analysis on my blog later, but for now here’s a couple of observations, from running poisson models on tweets and likes (separately, which I find better than using a ratio):

    (Results: )

    Joshua Tucker doesn’t influence tweetability, but his authorship decreases likability; ditto for Andrew Gelman and John Sides. Sorry, guys. James Fearon writes tweetable but not likable content. 🙂

    Potpourri is the least tweetable tag and somewhat likable; International relations is the most tweetable but not likable; Frivolity, on the other hand is highly likable. That says something about Facebook, no? 🙂

    Newsletters are tweetable but not likable… again Nerds on Tweeter, Airheads on Facebook.

    (Not to be taken too seriously; but what else is there to do on a Saturday morning?)


  4. Jose C Silva February 18, 2012 at 5:24 pm #

    Posted the details of the analysis at:

    And concluded:

    Given my violation of the Prime Directive of model building (make sure you understand the data before you start building models), I wouldn’t start docking the — I’m sure — lavish pay and benefits afforded by the Monkey Cage to its bloggers based on the numbers above.