The National Endowment for the Arts [pdf] plotted a curve based on percentage of 18 to 24 year old’s who read “literature.” Their idea of “literature” must be quantitative, because I’m pretty sure the release of Harry Potter accounts for the 2002 uptick.
When I was in undergrad fulfilling the requirements for a Biology and English Literature degree, there were, to put it nicely, a few bureaucratic brambles along the way. For instance, I had two advisers from different planets. My biology adviser only knew, and was only interested in knowing, the requirements for a biology degree. He thought the “Poetics of Dynamic Text” was a chapter on blood formation. The same gaping knowledge hole afflicted my English Lit adviser as well. This Confederacy of Dunces said I needed a statistics course, one especially geared toward my degree program. That meant two semesters of redundancy on a rather dry subject. As you can imagine, the Humanities stat. requirement was orders of magnitude more basic than the one for biology. Anyway, I had to take both, and so I resented both. I didn’t do so hot, but I learned enough to spot when someone is blatantly manipulating data in a graph. Dear reader(s), can you flush out the egregious, almost laughable, attempt to skew the data in the following graph from the National Endowment for the Arts?
Poor old National Endowment. They’re always the first, followed by education, to have their funding slashed when we feel the need drop bombs or bail out some captain of industry. Can you really blame them for propagandizing such benevolent government initiatives as increasing literacy? You betcha!
I made a new graph that righted two fibs. First, the spacing on the x-axis from 1982-85 is almost the same distance as it is from 2002-08. One is a three year gap and the other a six year gap. The NEA can judge what constitutes literature but they can’t count. By shortening the 2002-08 gap, they skew the percent increase in their favor. They even admit as much: “this dramatic turnaround shows that the many programs now focused on reading, including our own Big Read, are working [!]” I fixed that.
Second, unless you have a good reason it’s usually a good idea to have percentages start at zero. Not doing so may make the graph prettier, but also less accurate. I fixed that, too.
Graphs act as visual aids to a list of numbers, teasing out otherwise hard to detect trends. They are supposed to represent the data, not your interpretation of the data. If the NEA is lying by graph, what’s to stop them from fudging the data their fudged graph is based on?
I plotted their graph correctly, but am having problems importing it. (Was that a C or B I got in CS 101?) C’est la vie.