No, you didn’t catch us opinion-ating.
We’re getting ready to debut a new daily feature that will try to use data to validate assertions, or to uncover insights that would be impossible without the data collection at Factba.se
It’s not ready to go just yet, and it’s not as pretty as we want it. But the data’s more important than the prettiness.
To that end, please see our first try at this, and let us know what you think. This was based on an article in May in Stat that asserted Trump was potentially going through cognitive decline. The term “senile” was latched on to by the press (and thus this infographic), but it was not used in the article, just the comments. Much of the data cited was due to speaking style and vocabulary.
Well, said we. We have a definitive record spanning 37 years. Let’s take a look.
We focused on two areas: the Flesch-Kincaid Reading Level, that basically scores word complexity, and rate of speech. It’s worth noting that while the average American is between 135-160 words per minute speaking, the average New Yorker is close to 200 words per minute.
We are not offering opinions as to why, but what we can say definitively:
- Overall, his rate of speech has dropped consistently;
- This has coincided with a decrease in his unscripted public statements (e.g. interviews) and an increase in his scripted public appearances (speeches, remarks);
- There is a statistically significant difference in his rate of speech when looking at the type of appearance. Interviews and debates, he speaks much faster. Remarks and speeches, much slower (almost half speed)
- The complexity of words used in speeches is almost double the grade level of those used in testimony and debates, which are less likely to be scripted.
Statistically, the press conference is a bit of an outlier, though Jennifer (yes, Jennifer) pointed out that he often begins these with a prepared statement, which is included in the vocabulary and the rate of speech, which may skew the results.
That said, let us know what you think.
[Correction 7/1/17: Note the original infographic incorrectly compressed the X axis in terms of years. This does not change or alter the data, but does affect the two timeline charts in the appearance of the data. It has been updated.]