New fact-checking aesthetics on Twitter

Recent image-tweets from political fact-checking projects

Chris Blow France
Meedan Updates
3 min readNov 17, 2015

--

The U.S. election season is a perfect stimulant for a classical canon of fact-checking specific political claims. It’s a very satisfying format. Readers love it. Careful user interface design in this area can make a meaningful dent in the public consciousness, as discussed in a previous post, Antiviral social media: focus on visual design.

New examples are being released by the day; some are excerpted below. Each is an attempt to inject accuracy into a writhing beast of misinformation. Designers and journalists can take such examples quite seriously and, in the spirit of healthy design iteration, try to push them further. There is very little room for error here. We need to think about every pixel. Misinformation in political debate is an urgent issue — journalists and designers have a responsibility to advance the cause of clarity.

In this spirit, here are a some examples of political claim-checking that surfaced during the latest Republican debate, each with a bit of design commentary.

Good high-contrast white on blue. Nice integration of photo, branding, fact-checking status. The typography is a bit Tahoma, like overalls that get the job done but don’t fit perfectly. It looks like it was made just a little too fast — could use typographic refinements like curly quotes and an apostrophe instead of that straight single quote. I think an em-dash would be appropriate for the attribution line. The aspect ratio is not ideal for Twitter.
The quote is cut off when the tweet is quoted — but at least it still shows the “false” indicator. (This small square is one of the more difficult views on Twitter because it’s not well-documented exactly which part of the image will be cropped out.) Good use of the tweet text to add context to the quote.
Politifact has improved the rendering of these assets recently (much clearer now!), but when seen in this context it seems to beg for a slight redesign: move the “Half True” label down just a little, to fit it into the 2:1 ratio that Twitter uses. The arrow in the meter covers the lower text a bit. The beveled edges and heavy drop shadow feel a bit old-fashioned, but perhaps that’s the charm of these classic Politifact assets, which evoke dependability since they’ve been around through so many elections.
CNN’s new scale has simple text and follows a predictable pattern: a truth meter. Unfortunately the text of the tweet asks if Huckabee could fire the chairperson, but the Fact Check image text asks if he would. That’s a pretty serious error for this type of high-profile graphic, which is so narrowly focused on exactly this one fact! But the colors are nice.
A desaturated scheme connotes sophistication, level-headedness.
This visual vocabulary from the NYT seems extremely effective and clear because it suggests a range of consistent fact-checking outcomes that are color-coordinated. A good short quote and a verdict. This is the widest aspect ratio of the batch, which makes it feel lighter when it displays in the primary feed.
Very effective! So compact.
The photoshop filter used here feels like it’s turned up at least 20% too high. Arguably, a red filter makes sense for a false rating — but it’s a bit too scarlet letter-ish perhaps? Maybe their goal is to show a color of shame. But then will “TRUE” quotes make people sickly green? The high-contrast white on black is good, but the all-caps treatment hinders readability. Good emphasis on the “FALSE” conclusion, but the quote’s right-alignment feels unnatural.
Interesting color scheme and custom typography for their project branding at top left. Classic 2:1 layout. The text of the quote is paraphrased in the tweet, then the correction — and only the correction! — lives in the composite text. Nice. Overall a very effective technique. And a clever attribution — they tagged Ben Carson into the image, so his name appears above the retweet button at bottom.
Disqualified from serious evaluation because it does not have a link to a real fact-checking effort! :) But still, it’s actually a decent use of question format and hashtag. It’s goofy but don’t dismiss it totally — the font, Impact, especially with that heavy 3px black border could survive on top of any image.

Several elements are common to almost all of these — the clearly-stated claim or quote, attribution and a photo of the speaker, a clearly-labeled verdict about the truthiness of the claim, and a bit of branding around the newsrooms fact-checking campaign. These are the patterns that are being explored by fact-checking projects globally.

For the most part these stories are good for the public interest and popular with readers. These new designs each have their strengths and lessons — and they can immediately inform projects that work on broader verification efforts.

If other examples or observations come to mind, let us know. We are actively working on new verification design strategies.

For more examples see Antiviral social media: focus on visual design.

--

--