There was a article in Scientific American about diversity in STEM collecting together the best demographic data available about the science and engineering workforce. It’s a useful collection of references, and comes with some very pretty graphics, particularly this one, showing the demographic breakdown of the US population compared to the science and engineering fields:
This is a very professionally made graphic, but also misleading in the worst way. When I first looked at this, I thought “Huh. It’s actually much less bad than I thought…”
This is, of course, an illusion, caused by the fact that the colored bars aren’t actually proportional to the percentages they represent. The “White Male” bar in particular is supposed to represent 51% of the workforce stack, but is only about 35% of the top-to-bottom pixel count.
I understand why they did this– if you size everything in correct proportion, the smaller ethnic groups are basically invisible– but it has the effect of undermining the point they’re after. If you want to really see what the situation looks like, the correctly proportioned stacked bar graph is in the “featured image” above, and I’ll reproduce it here for the RSS reader crowd:
(General population is on the left, STEM workers on the right; converting from SigmaPlot to .jpg ate the labels, which I didn’t notice right away, and don’t want to take time to fix…)
This makes the usual point a little more dramatically: white males and Asians are dramatically overrepresented relative to the population of the US as a whole, at the expense of blacks and Hispanics.
Anyway, this is a minor quibble, but given my initial surprised reaction, I thought it was worth a quick post. On the whole, it’s a useful article, and not a shameless and stupid bit of clickbait, so good for Scientific American.