Beside the point, but this data visualization is misleadingly bad.
Eyes first draw to the heading, which primes us to think temperature. Then we see the graph, where the unlabeled Y axis is assumed to be average night temperature. Finally, we read the subheading and it says that the Y axis is not temperature, but counts of days over a certain temperature.
I think that this metric is more useful than “avg. overnight temp.”, but please label axes.
Also, it would help to rephrase the subheading to use “80” since that’s obviously the cutoff. I spent a moment wondering what was special about 79F.
And now I see that this was made by the NYT. I guess they’re pumping out charts (maybe automatically) and thinking more about making them pretty than legible.
Are you looking at the same article as me? On both the NYT app and the website using this link, I see a heading that exactly matches the data displayed. It’s a dynamic page that adjusts the figure as you scroll and the heading clearly matches the data. It says “abnormally hot nights” in every bar chart, and temperature for all of the line graphs. NYT has some really nice visualizations, with the notable exception of the potato graphic the other week with your states electric production sources - that was hot dog shit. There’s a different baseline temp for the hot night graphs depending on the city - this clearly responds to a low level baseline pre-warming.
I showed this to my partner who isn’t an engineer and she thought it made perfect sense too. Not that my anecdotes are special, but I truly don’t understand the confusion.
Beside the point, but this data visualization is misleadingly bad.
Eyes first draw to the heading, which primes us to think temperature. Then we see the graph, where the unlabeled Y axis is assumed to be average night temperature. Finally, we read the subheading and it says that the Y axis is not temperature, but counts of days over a certain temperature.
I think that this metric is more useful than “avg. overnight temp.”, but please label axes.
Also, it would help to rephrase the subheading to use “80” since that’s obviously the cutoff. I spent a moment wondering what was special about 79F.
And now I see that this was made by the NYT. I guess they’re pumping out charts (maybe automatically) and thinking more about making them pretty than legible.
Are you looking at the same article as me? On both the NYT app and the website using this link, I see a heading that exactly matches the data displayed. It’s a dynamic page that adjusts the figure as you scroll and the heading clearly matches the data. It says “abnormally hot nights” in every bar chart, and temperature for all of the line graphs. NYT has some really nice visualizations, with the notable exception of the potato graphic the other week with your states electric production sources - that was hot dog shit. There’s a different baseline temp for the hot night graphs depending on the city - this clearly responds to a low level baseline pre-warming.
I showed this to my partner who isn’t an engineer and she thought it made perfect sense too. Not that my anecdotes are special, but I truly don’t understand the confusion.
They’re technically labeled the same way an “exit” sign is labeled on the Walmart doors. There but shitty.