The analysis was odd for a number of reasons — not the least of which was that the results skewed far differently when the list of search terms excluded the word "bitch."
But a number of readers found more technical reasons to question the results — and their erudite responses speak positively about the quality of the education they received at the institutions in question.
Here are two examples.
Michelle Ess writes:
A lot of problems with this study that media (such as yourself) is not taking into account, & reporting as fact. A website conducted a study and posted it online without peer review and with a questionable operational definition, essentially implying tweets within 3 miles of a campus came from the campus. That is a huge area; the entirety of downtown is within 3 miles of Metro.
Nathan-Luke Anderson writes:
If you like this story, consider signing up for our email newsletters.
SHOW ME HOW
You have successfully signed up for your selected newsletter(s) - please keep an eye on your mailbox, we're movin' in!
I definitely have a problem with the data analysis here. First, there is no control for location, so the data can be skewed dramatically by the metro area surrounding it.
Second, if you actually analyze the rate at which these tweets are made, there are a couple of outliers, and an extremely long left tail distribution, which means that the difference between the mean of the dataset and the middle quartile range have a very low standard deviation, which in turn means there's likely no statistically significant correlation between the data when controlling for outliers.
Finally the outliers here, namely Clemson and Yale, are significantly above most of the rest of the data, which leads me to believe that one of the trigger keywords is affecting the data — for example, "bimbo" could be an endearing term for the school's mascot or something like that.
Overall, my DU education tells me this is a very, very poorly structured study.