Reader Comments
Post a new comment on this article
Post Your Discussion Comment
Please follow our guidelines for comments and review our competing interests policy. Comments that do not conform to our guidelines will be promptly removed and the user account disabled. The following must be avoided:
- Remarks that could be interpreted as allegations of misconduct
- Unsupported assertions or statements
- Inflammatory or insulting language
Thank You!
Thank you for taking the time to flag this posting; we review flagged postings on a regular basis.
closeFalsifibility of data and methods?
Posted by reddesilets on 07 May 2015 at 14:36 GMT
Even one person using that single term in an academic search would falsify the entire study. Also, how is it possible to know the race of the people searching that term? I don't understand how looking at people searching for a word online is any evidence of racism. It doesn't prove any actions, or even thoughts, of those people searching it. Please understand I think the idea of the study wanting to investigate the alleged correlation is worthy research; however, I don't feel confident enough in the methods used to make the claim that such and such areas are this much racist. Therefore, I cannot come to the same conclusions about the correlation. Any further explanations or points are appreciated. Thank you.
RE: Falsifibility of data and methods?
nalkoff replied to reddesilets on 09 May 2015 at 08:38 GMT
Even one person using that single term in an academic search would falsify the entire study.
Categorically false. That would be true if the variable were a binary "yes, there was somebody who searched for the word in that geographic area" or "no, not a single person searched for that word." The variable studied was the proportion of all searches that included the word. Yes, it's a proxy measure, but it's not obvious—indeed, probably unlikely—that such academic searches are a large enough amount of the total to render the proxy non-useful.
The comment title mentioned falsifiability, but the comment itself didn't really raise the issue explicitly.1 The authors' method would have yielded a null result if large volumes of searches for "n****r" (sans bowdlerizing) in the areas studied had not been correlated with the mortality data. Ergo, their methods were, by definition, falsifiable. (Of course, falsifiability isn't the sole criterion of a study's worth, but I'm just noting that they've met that threshold.)
1Other than using the word in the sentence quoted above. But that usage leaves the impression—I'll assume just the impression—that the commentator doesn't actually know what the term usually means when discussing study methods.