Hacker News Reader

Comment by milliams | Hacker News Reader

When I see a graph like [1] with the y-axis not zeroed I stop reading. Furthermore, the assertion that it "ticked back up" is not really supported by that graph any more than saying "it ticked back up and then it ticked down again"

[1] https://cdn.vox-cdn.com/uploads/chorus_asset/file/24341618/K...

View on HN

I agree it is totally unnecessary in this case, but sometimes truncating the y-axis serves a purpose. Many times it is for stylistic reasons, but also if the trend is more interesting then the scale, e.g. most stock charts truncate the y-axis because people don’t really care how much a stock has dropped relative to the value, but are more interested in the overall trend in the time period and the specific times when the trend reversed (I think, I don’t trade in stocks).

It is unnecessary to do that in this case because the scale isn’t that vast and you can easily see the trend on an absolute scale. However, this graph serves the same purpose as—I gather—stock charts do. They are interested in what happened while the trend was downwards, and the specific time at when the trend stopped (or at least halted).

There’s a lot of red flags in the article for me as well.

I’m genuinely amenable to the idea that mass antibiotics is bad because it makes intuitive sense. But the article starts out with “For decades…” and then the relevant plot has the X-axis start in 2012 for some reason. I strongly suspect, but don’t have time to confirm, that if you extended that axis back to, say, the 90s you’d see more nuanced trends that would threaten the doom-porn nature of the standard Vox “explainer”.

Yes, that's a deceptive graph, very likely intentionally deceptive in order to reinforce the clickbait headline.

The article is terrible.

At the very least break down the use by animal and mass.

Never mind cows are not raised like pigs which are not raised like chickens. To lump them all together is silly and insulting.

But they don’t lump them together, and they acknowledge your concern specifically.

What I would like to see a break down by animal mass. For meat production the US produces almost equivalent amounts of beef and pork. But the US also has about 10 million dairy cows. The FDA report is for food producing animals – which includes dairy cows.

Unfortunately the FDA report does not include mass either, but I reckon if you were to look at the antibiotics/animal mass you would see that swine is a huge outlier.

Reply on news.ycombinator.com