Library Dusted Presidential Poll

Michael “Mr. Dust” McGrorty has posted the results of his very informal poll of librarian presidential voting intentions. The bonus, of course, is his always-eloquent commentary, sure to rankle some and set others’ heads to nodding. You have three guesses at who won, and the first two don’t count. Have a gander here.

October 28, 2004 by rochelle Uncategorized 7

madcow says:

October 28, 2004 at 11:29 pm

Flamewar in 3…2…
Look out! It’s gonna blow!!!

Reply
Grumpy Librarian says:

October 28, 2004 at 11:51 pm

Hmph.
It would be easy enough to claim that these figures are somehow skewed because they reflect the viewpoint of librarians who are connected to the listserv community, but the great disparity in the numbers would likely not be shifted significantly if another manner of query were employed.

On what basis does he suggest that the results “would likely not be shifted significantly” if he used a different sampling method?

It would be easy enough to dismiss out of hand potential research problems by saying they are unlikely to influence the results, but it is unlikely that those who see the methodology as problematic would by convinced by such a dismissal.

Reply
tomeboy says:

October 29, 2004 at 12:45 am

Giddy
From 223:1 to 7:1. Perhaps there is some truth about a “right-wing cabal” here at LISNews?

Note to flamethrowers: Joke

Reply
Fang-Face says:

October 29, 2004 at 3:12 am

Re:Hmph.

On what basis does he suggest that the results “would likely not be shifted significantly” if he used a different sampling method?

Perhaps he used the word “significant” from the viewpoint of a statistician. In laymen’s terms, the word means important; in statistical analysis, the word means “not likely to have happened by chance”. In that context, his statement reads: results from a different sampling method would not be enough to say the difference couldn’t have happened by chance.

The question is: What is McGortry’s background in statistical analysis?

Reply
Grumpy Librarian says:

October 29, 2004 at 10:42 am

Re:Hmph.
Ok, let me rephrase: On what basis does he suggest that the results would likely not be shifted to an extent that is not likely to have happened by chance?

My point is that you can’t dismiss a serious problem with the research (potential sampling error) by merely suggesting it is not likely to have a significant (more than expected by chance results).

What he probablyu means is that this results confirmed his own experience of the profession, but that is subject to the exact same sampling error.

Ok, this is not a scientific study and not intended as such, but I am grumpy, after all. (and research standards are low enough in our field already)

Reply
Fang-Face says:

October 29, 2004 at 7:09 pm

Re:Hmph.

On what basis does he suggest that the results would likely not be shifted

Almost certainly: none. If McGortry has no background in statistical analysis then he was voicing an opinion from the top of his head.

Reply
kgs says:

October 29, 2004 at 10:23 pm

Re:Hmph.
I’m not rankled but I’m not nodding. I agree it’s statistically weak.

Reply