21 comments

[ 2.8 ms ] story [ 34.7 ms ] thread
The worst graph dissection in a decade. Does he intentionally omit reading that the graph is titled "Relative consumer spending on media, by category (2008) per capita"? Does he intentionally skip reading the article while he tries to decipher what a graph says?

What a tool.

The purpose of a graph (and any other infographics) is to illustrate data in an adequate and user-friendly manner. If you have to think over and over what the graph represents, it means it's poorly designed.
I don't disagree with your point. I'm simply saying it's impossible to understand what a graph illustrates without reading few, simple sentences. Like its title.
The graph looks like a timeline while it isn't. I don't know what's the number in the peak point. What does some point in the middle of the line between f.ex. tv and newspapers represent?

It's silly and confusing.

So, rendering the graph as a bar set with proper numbering and perhaps a detailed legend would convey the information just as well and avoid confusion as to what points between "Movies" and "Books" mean (although that very question has a very elementary one-word answer followed by a look of ever-so-slight amusement).

Still far from worst graph of the decade. For an example of a worse and far more hilarious case, over the top of my head: http://www.mydailyroast.com/?p=311

Have a nice day.

(comment deleted)
Is 'Nothing' the one word answer?
I think so, but the graph is so bad (allegedly) that I can't be entirely certain (apparently). /sarcasm
Figures including their title and label should mostly speak for themselves.

The title implicitly promises the viewer a comparison of different countries. This graph is a prime example of how not to present that data. The linked article explains that in every detail and provides feedback on how it should be done.

They still shouldn't add per capita figures up, which the article mentioned.
Surely the worst part of this graph is it should never have been rendered as a line graph at all? Either way you put the axis, there's no progression over time and a bar graph or some other representation makes a lot more sense.
Precisely. The first thing my brain told me when I saw the graph was "It's gone way down." That's how we're used to interpreting line charts.
Holy cow, read the IEEE spectrum site[1] it's hosted on. The info on the page is loaded with info that's not even remotely represented by the graph. And see the "US $240" part? It's for Norway...

In most other categories Norway leads. Partly, that’s due to its high cost of living. But they also seem to love media—especially books, at $240 per year per person. Must be those long winter nights!

WTF? I think I agree, that's absolutely, fascinatingly bad.

[1] : http://spectrum.ieee.org/consumer-electronics/audiovideo/dig...

edit: does anyone else HATE those mouse-over previews? Know a way to get rid of them across as many sites as possible?

Throw this in your /etc/hosts file:

    127.0.0.1 spa.snap.com
and then just make sure you don't serve obnoxious mouse-over previews to yourself :)

Here's my full ad-block. I just add a hostname to it whenever I see an obnoxious bit of stuff on a webpage:

    127.0.0.1 spa.snap.com ad-yt-bfp.doubleclick.net doubleclick.net m1.2mdn.net ad.doubleclick.net media.fastclick.net cdn.fastclick.net ds.serving-sys.com spe.atdmt.com content.dimestore.com speed.pointroll.com img-cdn.mediaplex.com clk.pointroll.com img-cdn.mediaplex.com pagead2.googlesyndication.com pagead2.googleadservices.com pubads.g.doubleclick.net partner.googleadservices.com tenbux.somethingawful.com somethingawful.crwdcntrl.net content.yieldmanager.edgesuite.net kona.kontera.com core.videoegg.com servedby.advertising.com uac.advertising.com at.atwola.com bannerfarm.ace.advertising.com view.atdmt.com ads.forbes.com bid.openx.net d1.openx.org edge.quantserve.com altfarm.mediaplex.com ad.uk.doubleclick.net shots.snap.com snap.com resources.infolinks.com fe.brandreachsys.com www.ttzmedia.com lapi.ebay.com cache.blogads.com mm.chitika.net techeblog.pgpartner.com cdn5.tribalfusion.com www3.adservenow.com ad.z5x.net clicktorrent.info this.content.served.by.adshuffle.com ads.adsonar.com js.adsonar.com web.adblade.com optimized-by.rubiconproject.com content.pulse360.com cdn.optmd.com as.casalemedia.com ad.ca.doubleclick.net ilinks.industrybrains.com context3.kanoodle.com adsyndication.msn.com technology.msnbc.us.intellitxt.com an.tacoda.net a.rad.msn.com b.rad.msn.com c.rad.msn.com d.rad.msn.com mjxads.internet.com ads.adbrite.com tag.contextweb.com contextweb.com media.contextweb.com cdn.eyewonder.com cdn1.eyewonder.com amch.questionmarket.com  iacas.adbureau.net portal.mxlogic.com ads.revsci.net js.revsci.net images.intellitxt.com ads.cnn.com imagec12.247realmedia.com im.afy11.net ad.afy11.net afy11.net c5.zedo.com clickbangpop.com ads.clicksor.com clicksor.com pages.etology.com ads.okcimg.com okcupid.tags.crwdcntrl.net coop.crwdcntrl.net cdn.crwdcntrl.net d7.zedo.com ads.yimg.com us.a2.yimg.com
This is such an obvious anti-spam device I feel like an idiot for not using it.

It just never occurred to me.

Thanks.

hah, very nice. Particularly because I just tried to enter that into Adblock, and it's busy ignoring me.

Thanks!

I recently was tutoring a high school student who was instructed in their data-management course to create several graphs from the same set of data, comparing rainfall in different Canadian cities.

Unfortunately this data didn't make any sense to be plotted as a line-graph, a bar-graph, a histogram and a pie chart, as the comparisons didn't make sense for any but the bar-graph. It was if the students were actively being taught how to use graphs incorrectly. Creating pie charts when the entire pie doesn't mean the whole of anything I think was the worst culprit.

I have always disliked this type of graph. But some people think they are the coolest thing. Like this baby name website is really popular and it is based on a graph like that: http://www.babynamewizard.com/voyager

(A couple years ago I made a competing baby names website, so that's why I bring it up. This type a graph is a little pet peeve of mine.)

That baby name graph is showing how popular names are over time. The NYT Box Office graphic does the same - http://www.nytimes.com/interactive/2008/02/23/movies/2008022... .

This kind of graphic is bad for showing exact numbers, good for showing relative share during time and change over time.

But the graph in the parent article wasn't making that comparison, so yes, it was a stupid, pet peevey way to do a graph.