Poorly constructed graphs.

A collection of troubling visualizations.

March 31, 2012 at 1:00am
3,781 notes
Reblogged from barackobama
(Via)  Subtle partisanship aside, I was very close to really liking this visualization.  Really close.  It taught me something, it’s well designed, has a good layout.

But then they had to go and stick dark grey in the middle of a light red to dark red gradient:

no data - light grey
0-25% - grey
26-50% - red
51-75% - dark grey
76-100% - dark red
Really?

(Via) Subtle partisanship aside, I was very close to really liking this visualization. Really close. It taught me something, it’s well designed, has a good layout.

But then they had to go and stick dark grey in the middle of a light red to dark red gradient:

  • no data - light grey
  • 0-25% - grey
  • 26-50% - red
  • 51-75% - dark grey
  • 76-100% - dark red

Really?

March 28, 2012 at 10:37pm
5 notes
This was constructed to illustrate the Tautochrone curve: a basic physics concept wherein an object will take the same amount of time to reach the bottom of the curve no matter where you drop it (assuming uniform gravity.)  Cool, right?  Now that I’ve explained it, you get it, right?  It’s even easier to understand if you look at the little graph in the upper right hand corner which, obviously, tells you the relationship between (s)peed and (t)ime.

Wrong.

The Y axis actually measures arc length.  Visualization is about communication, and since this targets those unfamiliar with physics it does a poor job by using potentially misleading labeling simply because it is agreed upon by the professional community.

Jargon seepage is dangerous.

This was constructed to illustrate the Tautochrone curve: a basic physics concept wherein an object will take the same amount of time to reach the bottom of the curve no matter where you drop it (assuming uniform gravity.) Cool, right? Now that I’ve explained it, you get it, right? It’s even easier to understand if you look at the little graph in the upper right hand corner which, obviously, tells you the relationship between (s)peed and (t)ime.

Wrong.

The Y axis actually measures arc length. Visualization is about communication, and since this targets those unfamiliar with physics it does a poor job by using potentially misleading labeling simply because it is agreed upon by the professional community.

Jargon seepage is dangerous.

March 27, 2012 at 11:48pm
1 note
(Via) The reason we organize data in tabular form is so that we can find a point in the table and then quickly determine which metadata (the row and column) are associated with it.  Understanding this, the obvious next question is, “what is the quickest way to convolute and obfuscate this view?”

Curve the columns.

(Via) The reason we organize data in tabular form is so that we can find a point in the table and then quickly determine which metadata (the row and column) are associated with it. Understanding this, the obvious next question is, “what is the quickest way to convolute and obfuscate this view?”

Curve the columns.

March 26, 2012 at 10:02pm
1 note
(Via) An interactive cloropleth map about interactive and social media!  Aren’t you just salivating over the keywords?  While it’s tough to screw up something this simple, right on the heels of a previous win with a new demographic, Forbes buries the scant specifics of a study in the middle of a page-long article while describing the pictured map in a headline as visualizing “America’s Most Influential News Outlets”.

Given the description of the data source, however, this should probably be called The Interactive Media Map: Trends in bit.ly clickthrough rates for links to articles at name-brand news media company websites with very large budgets and old-world readerships.

The data show the results of a geolocational study done by a link shortening website through their own links.  As a result, the map does not display the interest in a site by location, but how likely users in that location are to click through a shortened link to get to an article at that website.  If the users tested clicked through to these sites in the same situations the rest of us do (twitter, forums, etc.) then whether or not they even know to which news outlet the link would take them is questionable, let alone their interest in that outlet.

(Via) An interactive cloropleth map about interactive and social media! Aren’t you just salivating over the keywords? While it’s tough to screw up something this simple, right on the heels of a previous win with a new demographic, Forbes buries the scant specifics of a study in the middle of a page-long article while describing the pictured map in a headline as visualizing “America’s Most Influential News Outlets”.

Given the description of the data source, however, this should probably be called The Interactive Media Map: Trends in bit.ly clickthrough rates for links to articles at name-brand news media company websites with very large budgets and old-world readerships.

The data show the results of a geolocational study done by a link shortening website through their own links. As a result, the map does not display the interest in a site by location, but how likely users in that location are to click through a shortened link to get to an article at that website. If the users tested clicked through to these sites in the same situations the rest of us do (twitter, forums, etc.) then whether or not they even know to which news outlet the link would take them is questionable, let alone their interest in that outlet.

March 25, 2012 at 5:38pm
2 notes
(Via) What the hell is that little circle where the edge bends and what do the node colors mean?

(Via) What the hell is that little circle where the edge bends and what do the node colors mean?