(a) I think we're talking past each other. You said the following:
(b) That wasn't the purpose of the study. It was just to identify and analyze the content of tweeting that exhibited bot behavior. The headline is misleading and it's influencing the way you're reading the article. I made this diagram for another poster to better illustrate what was done. I think you're confusing the purpose of the analysis with the findings of the analysis. Where am I going wrong?
(a) Could very well be. Probably so.
(b) And this is where we're talking past each other. You seem to be satisfied with what the story relates. My concern goes past what the story relates to the methodology of the study. So it would be true I am confusing the purpose with the analysis except that I'm not. Meaning, I have no problem with what they wrote as pertains to the "opening America" aspect; my problem is that they chose to only look at one half of the issue.
I'm not trying to be obtuse or pedantic here. I'm also not trying to confuse you (or anyone). My issue (as a result of too many grad school classes drilling proper methodology deep into my cranial cavity) is that the thing CMU studied has at least two facets. It is only a single-sided issue if CMU had no problem with this study being used for political purposes (be it by CMU or anyone else, be it Left or Right). If that's the case, shame on those folks. However, if CMU wanted this to be seen as a fair, statistical take on bot behavior during this "lock-down/set me free" debate it should have looked at both sides or stated why it chose to only look at one side. Otherwise, the study is implying something that may give a false impression as to what is actually going on in Botland.
Maybe this will help illustrate what I'm trying to point out. If I did a study on Soviet collectivization in the Ukraine in the late 1920s/early 1930s I could legitimately say that thousands were saved from starvation by Stalin's efforts. I could legitimately say that because it would be true. However legitimate my
results were, this
study wouldn't be legitimate because it failed to look at the other side of collectivization: the millions who died of starvation (and other reasons such as execution) because of Stalin's war on the kulaks to force collectivization. In other words, doing one side produces good propaganda, doing both sides produces good history. Which angle results in better analysis? I would say the study that looked at both sides.
Bottom line, my issue with the press release is not the results of the study; rather, the methodology of the study. I take no issue with the former, I take huge issue with the latter.
I did find the press release interesting. I was even a bit surprised by what CMU found (though, I guess, in retrospect I shouldn't have been). But to me, the press release begged an additional question. So yup, perhaps I expected too much. I've been told that is a problem of mine. If so, my bad.
--- End of line (MCP)
