Constructive critique of NYT’s “At the National Conventions, the Words They Used”: Part 1
Note: I’m taking the 6-week Introduction to Infographics and Data Visualization course offered by the Knight Center and taught by Alberto Cairo. I wrote this blog post as one of the second week’s assignments. Check out my related post, “Redo of NYT’s ‘At the Nation Conventions, the Words They Used’: Part 2.”
Overview
The New York Times’s graphics department continuously nails it when it comes to infographics and interactive data visualizations. Their recent “At the National Conventions, the Words They Used,” is no exception.
The visualization breaks down the word usage of speakers at the Democratic and GOP National Conventions into bubbles, based on transcripts from the Federal News Service. It compares the number of mentions of a word or phrase per every 25,000 utterance, with the bubbles increased in size based on higher usage by both parties.
Additionally, the graphics team divided each bubble by traditional party color with Dems on the left and Republicans on the right — oh the details! That doesn’t mean each bubble is divided in half by color. Instead below the word or title of the bubble they show the exact number of times it was used by both, for example, 69 - 26 for the word Women. Thus the blue portion of the Women’s bubble is much larger than the red.
If a user selects a bubble, then the lower portion of the data visualization updates to show example quotes where the word or phrase was used by speakers at both conventions. And the subhead of this part of the page repeats again the total times a word was used by both parties. For instance: “Democrats mentioned Women 69 times per 25,000 words,” or “Republicans mentioned Women 26 times per 25,000 words.”
One of the coolest features, though, is that the graphics team created a field so that users or readers can enter their own word or phrase to see how many times it was used.
As a test I entered “climate change.” A teeny tiny bubble appeared and dropped over to the far left side with 1 - 0, thus the bubble is completely blue. I created a “Dream Act” bubble, and it too fell to the far left with 2 - 0. “Generation Y” landed right in the middle with 0 - 0 and no bubble, as did “Millennials.” Finally I entered “Reagan” and this bubble dropped down on the right side with 1 - 11.
Critique
At first glance the “At the National Conventions, the Words They Used” data visualization appears functional. In reality it is only functional in broad terms. It allows a quick overview, and I believe that was the goal of the NYT’s graphics team. I think they wanted to target those readers that didn’t watch the conventions in full or at all.
The interactive graphic does present multiple variables: two parties and the various words they favor. It allows readers to make a quick comparison. Although I don’t think any of the outcomes are surprising, and readers cannot make an accurate comparison since the data is presented with bubbles. With the layout of the graphic as a whole, bubbles and speaker excerpts, the information is well organized and thought out — the reader doesn’t really have to think. And the piece does show a correlation between party platform and word usage.
One feature that really bothers me is that a user can move the bubbles around. I cannot figure out what’s the purpose of this feature. The bubbles don’t bounce back if you move them across party lines. As a test, I checked out the graphic on my smartphone, and the bubbles seem to jiggle more. So still, there is no real reason for their mobility.
The excerpts also bothered me slightly. There’s no real context — they’re just pull quotes — and the quote doesn’t even have to be a good quote. I suppose most politicians that have been quoted in the media before will say that they were misquoted or their quote was taken out of context. This part of the graphic made me think of such concerns. At the same time, I would have been heavily annoyed if this information was not provided. The data visualization would be a failure without the excerpts because they do, in the end, provide context for the bubble chart.
Overall it is an amusing way to take in the national conventions. I don’t think readers really learn anything from it. It is merely a fun way to explore the speeches. Yet I do wonder how the New York Times could have created this data visualization differently.
