The scale on the first plot is a bit disingenuous. Here's the same data replotted with a proper scale (i.e. ranging from 0 to 4 instead of 2.6 to 3.4).
The kind of false impression that at least I get when I look at this graph is that the grades were multiplied by a factor ~1.3 over the past 20 years when they were only multiplied by a factor 1.06. For anyone familiar with GPA, this is misleading.
The point is that a grade is an ordinal quantity (somewhat). You don't get anything useful from multiplying grades by any "factor". Anyone in danger of thinking otherwise should not blame the graph. Fast food articles such as "Top 10 secrets of misleading graphs your profs don't want you to know" are probably more culpable in that regard.
In principle I agree, but grade inflation preserves the order so why should we care about grade inflation if grades are only about order? I think the plot and the rest of the article goes beyond that.
Also an histogram is a poor representation choice for ordinal data.
It isn't really disingenuous. It's become fashionable to claim that "all graphs should start at zero", but that's not necessarily how you display data. Here is Edward Tufte:
"In general, in a time-series, use a baseline that shows the data not the zero point. If the zero point reasonably occurs in plotting the data, fine. But don't spend a lot of empty vertical space trying to reach down to the zero point at the cost of hiding what is going on in the data line itself. (The book, How to Lie With Statistics, is wrong on this point.)"
He's talking about time-series in this case, but the point remains: you need to understand the data. Sometimes (as in this case, where the trend is important, not the magnitude of change) showing the "zero" point is irrelevant.
Not all graphs no, but for histograms which are meant to compare quantities, maybe yes. See the other discussion in this thread.
The kind of false impression that at least I get when I look at this graph is that the grades were multiplied by a factor ~1.3 over the past 20 years when they were only multiplied by a factor 1.06. For anyone familiar with GPA, this is misleading.
Your graph still shows the trend. But in a way, that's the most dangerous lie, the sort where you "help truth along" a little bit, to be more compelling.