Comments on: Think you drink a lot?

By: James James

James James — Wed, 19 Jul 2017 12:37:58 +0000

If you hold the unadjusted top decile constant at 37.49, you have to multiply the others by 4.9 to keep the total the same. The ninth decile goes to 37.97, more than the top decile.

The unadjusted top decile must be underreporting somewhat, because the other deciles are probably not underreporting by 5 times.

By: Gaikokumaniakku

Gaikokumaniakku — Tue, 18 Jul 2017 22:41:24 +0000

“Wow, was this crap peer-reviewed? While drunk?”

No self-respecting academic would review the work of his peers while sober.

If reviewers were sober, they might start to question the paradigms of peer-reviewed research, and then they would be out of a job.

By: Kirk

Kirk — Tue, 18 Jul 2017 16:07:05 +0000

Why do these guys never “close the loop”, and go back to things like annual sales of alcohol and then work backwards to see if the results of their breathless mathamajiggering even begin to make sense?

You see this all the time in “journalism”, which based on the intellectual rigor displayed by its practitioners, barely qualifies as a discipline. Don’t even get me started on the crapola that comes up with regard to military and defense issues. The innumeracy burns, it does, it does…

By: Ross

Ross — Tue, 18 Jul 2017 13:18:32 +0000

Wow, was this crap peer-reviewed? While drunk?

“…we find that Cook corrected these data for under-reporting by multiplying the number of drinks each respondent claimed they had drunk by 1.97..”

Too hard to estimate under-age drinking? Just “multiply by two” and call it done?

Economists……SMH