Hacker News new | comments | show | ask | jobs | submit login

Interesting article, but the first graph is one of the most blatantly misleading charts I have seen in a long time. Showing two graphs overlapping on completely different scales is a really nasty trick.



Interestingly, I think it would be more illustrative of their point if they presented the graphs in some normalized form (such as percentage increase). The private inmate population has risen ~48% (~89k to ~132k) over that time period, while the total inmate population has risen ~14% (~1.95m to ~2.22m).


Extremely. Also, it looks like they picked random years for the data points as well. They cross in 2010 like it has some significance.


Right, good point - when you choose arbitrary scales you can make them cross whenever you want.


> Interesting article, but the first graph is one of the most blatantly misleading charts I have seen in a long time. Showing two graphs overlapping on completely different scales is a really nasty trick.

They are the same in 2010 because that's where the graph starts. The rest of the graph should be interpreted as "cumulative growth as of $DATE, relative to 2010".

This is actually a common way of presenting two stocks. I believe Google Finance does it. Search for any ticker symbol (e.g. GOOG) and then type in another where it says "compare" (e.g. TWTR). No matter how far back you pull the graph, the leftmost points on each line will always overlap.

I agree that it's a bit weird if you're not used to it.


I don't think so - in Google Finance compare both graphs are actually on the same scale. Since the scales here are independent, it depends on the chosen scale where they will cross. The graphs do not start in 2010, they start in 2000.




Guidelines | FAQ | Support | API | Security | Lists | Bookmarklet | Legal | Apply to YC | Contact

Search: