« Light entertainment: the chartjunk crisis | Main | Things they don't teach you at Harvard »

Comments

Francis

Nice, I agree that it could be cleare than the pie, but I can't understand your bar chart. The range can't surely be in "thousands of respondents"? And if it's in thousands of dollars, then isn't it a bit low? Maybe tens of thousands? I don't understand the order either: it is not by minimum nor maximum. Then what? Some non-marked average? Thanks!

Jorge Camoes

Yes, this is a very creative way to hide the data. If that's the author's intention I also respect his/her creativity...

Where is the pie chart? I see an "inspired by" chart but surely the poor pie can't be blamed for yet another crime.

Your chart misses the point. This has nothing to do with presenting the data effectively. This is a "if I wanted an Excel chart I wouldn't hire you" point of view.

This is graphic design at its worse. But it's cool, right?

1c1cle

Ah, the pie chart pulls the eye better than boring, objective bar chart. I even have issues with range, itself. It is only one descriptive statistic of a set. How many elements are in each set? Do the endpoints represent employees or theoretic ranges of a position.

I'm sure there is atleast one direct @ google who is real pissed.

Matthew F.

I wouldn't even call the original a "Pie chart"; it's basically a circular bar chart. I just didn't even realize what it was trying to convey until seeing yours in bar-chart form.

Kaiser

Francis: thanks for pointing out the scale error. It should be from 125 to 245, in thousands of dollars; will fix it when I get the chance. The "of respondents" is there to emphasize the fact that the salaries are self-reported numbers from users of a website, and one should be careful when using it to talk about "Google salaries" in general.

Paresh Shah

Kaiser - Not commenting specifically on either of the charts here but arising there from never the less.

There are natural semantics to images. When we talk parts of the whole, a pie, whether circular or square or some other shape, symbolises the whole. The bar chart can never never [ emphasis intended] fulfil that role.

Yes that tool is sometimes misused as has probably been done here. But we all make mistakes don't we ? I have written on this subject as a tangential reference to a graph I had observed in my last post on www.visualquest.in

There is no need to ban pie charts - use it carfully

Paresh Shah

Proof what happens when you use a tool carelessly. Carefully becomes carfully.

Talosman1

I totally agree that the chart is confusing if you want information from it. It's not a true pie. The slice angle shows the salary range for each category and the radial dimension is just arbitrary. I have no idea what the size of the stick figures means. Plus, since the salary range is listed, the chart is not really needed.

But, as graphic art, it's interesting and eye catching and is probably appreciated by other graphic artists who don't really care so much about the data.

Sunflower

The bar chart brings out information you can't get from the pie chart. It's easy to see the ranges and the variability among them in the bar chart.

Jim Lemon

Now what I would like to see is a point-with-dispersion bars where the point represents a useful measure of central tendency (e.g. median) and the dispersion bars show the range. More info, less ink. The horizontal layout is very good.

Dland

A bit late to the game, I realize, but the job of the shiny noisy crap at the top of this entry is not to communicate something about Google salaries, but to sell advertising to the likes of Toyota, IBM and SimplyHired.com (at least as I see the BusinessInsider.com page bearing this atrocity today.

I am in a protracted, if friendly, conflict with the product owner where I work over the pointless 3D chartjunk generated FusionCharts in the reporting component of one of our products. I insist on charts that communicate, simply and fairly, the data our clients need to do their business. The product owner insists that the 3D charts look "cool" and the developers love the way the bars in bar charts grow up from the bottom as the chart appears.

I have not done any end-user testing to determine whether "charts that communicate fairly and clearly" or "really cool charts that are fun to watch" is what the clients need.

Or, rather, what the clients want.

Or, to be completely honest, what the clients will buy.

ezra abrams

sigh
how does one decide if a chart is good or bad ?
well there is the GOP science free method , to assert ones opinion, which seems to be the main method on this blog
Then there is the science method, which is to devise an experiment to ask if the chart communicates well
I actually find this blog really offensive, as it gives the authors opinions as facts; no different whatsoever from H Cain

Verify your Comment

Previewing your Comment

This is only a preview. Your comment has not yet been posted.

Working...
Your comment could not be posted. Error type:
Your comment has been posted. Post another comment

The letters and numbers you entered did not match the image. Please try again.

As a final step before posting your comment, enter the letters and numbers you see in the image below. This prevents automated programs from posting comments.

Having trouble reading this image? View an alternate.

Working...

Post a comment

Marketing analytics and data visualization expert. Author and Speaker. Currently at Vimeo and NYU. See my full bio.

Book Blog



Link to junkcharts

Graphics design by Amanda Lee

The Read



Good Books

Keep in Touch

follow me on Twitter

Residues