Twitter

Visualizing Verified Twitter’s Reaction to Robert Mueller’s Investigation

By Matt Stiles | Matt Stiles | Topics: Policy & Politics

Special Counsel Robert Mueller’s now-concluded investigation into the Trump campaign and Russian influence over the 2016 presidential election was obviously a hot topic on Twitter.

More than 400,000 tweets — an average about 600 per day — mentioned the word* “Mueller” since the former FBI chief was appointed to lead the investigation in May 2017, according to a dump (190MB csv) of verified user data pulled from the social network using Python and Twint.

That interest spiked during key news events before skyrocketing on Friday and Sunday, when Mueller concluded the investigation and then the Department of Justice released a letter summarizing the findings:

You can download the data here (190MB csv).

* This data dump, due to my general laziness and day job duties, includes a bit of noise from misspellings of the the German footballer Thomas Müller’s name and some other unrelated “Mueller” tweets — about 60 a day, I would say, on average. You get the idea, though.

[Photo courtesy The (Obama) White House, via Wikimedia Commons]

Visualizing a Year of @realDonaldTrump

By Matt Stiles | Matt Stiles | Topics: Policy & Politics, Social Media

President Trump thumbed his way through another year in the White House on Twitter, compiling a good (great) collection of 2,930 touts, complaints, defenses and rants.

He left 2018 with this perplexing New Year’s Eve missive extolling the old-fashioned endurance of “Walls” and “Wheels” as one of his last.

As the message shows, the president’s twitter presence lately is crowded by an increasingly evergreen list of grievances (Democrats, Russia, fake news, etc). Still, plenty of his messages actually correspond quite neatly with news events.

Notice how the #maga hashtag, a political rallying cry, disappears after the midterm elections. He talks about The Wall and shutdowns in and around the shutdowns, of course. And he decries Special Counsel Robert Mueller most often around the times his former aides have appeared (and been convicted or pleaded guilty) in federal court.

These examples are obvious when plotted on a timeline with annotation:

Through it all, the president’s audience of followers grew steadily by 10 million users. He now has 56.7 million followers (me included). He’s No. 15 on that measure, according to friendorfollow.com, sandwiched between heavy hitters like @selenagomez and @britneyspears!

During 2018, @realDonaldTrump spread his tweets throughout the days of the week, with the president even finding time on the weekends to sound off:

This large collection of messages, scraped using twint, drew more than 300 million of engagements, with “likes” being most common by far. This one about North Korean leader Kim Jong Un a year ago received a whopping 475,000 likes, topping the list.

Here’s how those engagements split proportionally:

Speaking of retweets, there were 57 million in 2018. They came at the rate of 200,000 per day in some months. This popular “they-just-don’t-get-it” mashup of video clips, for example, received more than 110,000 retweets alone in July:

And, finally, as in years past, those messages were a mix of endorsements, promotions, defenses and complaints. Among the more popular keywords (sorry, no word clouds here):

You can download the data as a CSV here. Happy New Year!

Tracking Historical Twitter Followers: @elisewho vs. @stiles

By Matt Stiles | Matt Stiles | Topics: Social Media

My wife (@elisewho) and I (@stiles) had a silly social media moment yesterday when I replied to one of her tweets — despite the fact that she was sitting in an adjacent room of our Seoul apartment.

USC professor Robert Hernandez (a.k.a. @webjournalist) captured it:  

The exchange, which we both “liked”, got me thinking (resenting?) about why she is killing me in Twitter followers — even though we’ve been using the service for nearly a decade.

It’s not even close, according to snapshots captured by The Wayback Machine (which didn’t start tracking me until I’d been on Twitter for years):

You can track your own follower counts over time with this tool.

Charting Tweets At #UNITY12

By Matt Stiles | Matt Stiles | Topics: Social Media

Last week I attended the UNITY journalism conference in Las Vegas, and during my stay I ingested more than 9,000 tweets that had the #UNITY12 hashtag. This line chart shows how the traffic ebbed and flowed each day:

As of dinnertime Saturday, when I stopped collecting the tweets, these attendees were the most prolific users of the hashtag: @NABJDigital (164 tweets), @sandhyadirks (136), @emmacarew (118), @joshstoffregen (102), @VictoriaLim (98), @DorisTruong (88), @webjournalist (77), @L2theS2theV (75), @barbaradozetos (70), @cecepmarshall (69).

Download the data, and let me know what you find.

Ranking TechRaking Tweets

By Matt Stiles | Matt Stiles | Topics: Social Media

Dozens of technologists and journalists today descended on Google’s beautiful Mountain View, Calif., campus for a discussion about technology and journalism. The conference, organized by the Center for Investigative Reporting, led to some prolific tweeting, as one might expect.

I used a simple script to ingest the 1,500-plus tweets with the search API into a sqlite database. This chart, made with Google Docs’ chart tools (when in Rome…), shows the top 25 most prolific tweeters (as of 4:30 p.m. pacific) who used the #techraking and #techrakingcir hash tags.

Congrats, Ian Hill, you top the list (which includes, I think, some spammers):

This is just a quick chart made in a rush. Feel free to download and check out the pipe-delimited data for yourself: #techraking | #techrakingcir. Send me your visualizations or thoughts, and I’ll post ‘em here. See the full list of Twitter user counts here.

Use Calendar Heat Maps to Visualize Your Tweets Over Time

By Matt Stiles | Matt Stiles | Topics: Social Media, Tutorials

Following Nathan Yau’s excellent tutorial for creating heat maps with time series data (he used vehicle accidents by day for a year), I visualized 3,559 of my tweets back to March 2009.

These maps, created with a modified R script from the tutorial, show how often I sent tweets (both personal and RT), with darker shades representing more activity. It’s fun to go back to the dark days and recall what sparked flurries of tweets:

The Sentiment on Bill Keller (Updated)

By Matt Stiles | Matt Stiles | Topics: Uncategorized

Reaction to New York Times editor Bill Keller’s recent piece about Twitter appears to be mixed — at least according to a quick sentiment analysis of related tweets. 

See…

Excellent piece

And…

Must you always nay say, @nytkeller?

Here’s how this application charted recent tweets mention Keller: 

UPDATE (10 a.m. Monday): Since the publication in the print edition of the magazine, it seems reaction is much more positive: