Hacker News Reader

Comment by metabrew | Hacker News Reader

Those graphs are absolutely vile.

I kept wondering how anyone who chooses hard-to-read shonky 3d bar charts like that could possibly perform a decent statistical analysis.

View on HN

Agreed, not the prettiest graphics, but you shouldn't dismiss the data & findings because of it.

Unless of course you judge the validity of a study based on the appearance of its graphs?

Well, the title of the blog post is "First impressions count".

Also, the irony of that title, and the way this is a second go at this blog post...

To me, authors decision to use terrible 3D graphs and non zero Y axis makes makes me suspicious of his other decisions and comments.

Agree that 3D graphs are terrible, but it's a canard that y-axes always need to start at 0 and that doing anything else is somehow misleading.

On this point, see http://krugman.blogs.nytimes.com/2011/09/14/axes-of-evil/

Y-axis need to start at zero when the data being presented is a bar graph and the bar is filled in. If it is a scatter plot with points being plotted, then it's generally not misleading to shift the axis.

I don't disagree with that - a bar chart that doesn't start at zero can be misleading. But for what he was trying to do---compare means---it would have been a perfectly fine choice to do a lineplot/very thin bar plot and start the y-axes somewhere other than zero if he included error bars and clearly labeled things.

http://www.edwardtufte.com/bboard/q-and-a-fetch-msg?msg_id=0...

Annoyed me so much I quit reading half way through.

Not just that it is ugly, but also hard to read. ugh.

Graphs updated, as noted above.

Reply on news.ycombinator.com