There was been a lot of discussion towards the end of last week about this particular chart that was mentioned in Ravi Parikh’s excellent article on how to lie with data visualisation. In this article he states that by ignoring standard chart conventions i.e. reversing the y-axis in this instance, we make it a lot harder for the viewer to understand what is actually going on.
Chances are, as a new viewer to the chart, upon seeing the timeline at the bottom and the thick dark line running across the graph, you'll automatically “read" the data in the white area, which appears to show a stark decline in gun deaths after the “Stand Your Ground” law was enacted. This would in fact be the wrong way to read this particular chart. After a few seconds you'll probably realise that the chart’s y-axis has been reversed, and that you’re actually supposed to be reading the red area. This is hugely significant, as the key point we’re supposed to take away from this chart, is that gun deaths have in fact increased after 2005, not decreased, which is what it appears to show at first glance.
Many have called out this chart as an example of deliberate deception*. That it was the designers prerogative to mask the real story from view. I however do not believe that this was the case. Unfortunately, I chalk this up to a case where the pull of a clever visual metaphor (i.e. dripping blood), was much stronger than actually telling the data’s story.
Here the designer had a chance to do something visually clever with this simple chart, but ironically, in order to do so, they had to forgo the key message they were in fact trying to show.
* For a little back story on the conversation / debate surrounding this chart, check out Andy Kirk's series of posts: