WEIRD SCIENCE (ECONOMICS DIVISION): As I read yesterday's PD article about organized labor pushing a new convention center initiative, the following sentence caught my eye:

In the 1990s, employment gains in Northeast Ohio's hospitality sector - which includes, retail, museums and hotels - almost equaled lost manufacturing jobs, according to CSU analysis.

Read that sentence again. See anything strange? Here's a hint: Do you think of shopping malls as part of the "hospitality sector"? Probably not... but it seems that's the way CSU's study was counting in order to make its highly political point.

The study, by Levin College of Urban Affairs Dean Mark Rosentraub, is available here in PDF format. Published last May, it was waved around a lot during the Summer by Convention Center Tax advocates, but I never looked at it until this morning. The sentence from the PD quoted above is based on a passage that starts on page 14 and continues on page 15, referring to "changes in job levels throughout Northeast Ohio during the 1990s":

There was a loss of more than 22,000 manufacturing jobs... The number of new jobs in the retail, amusement services, museum, and lodging sectors -- which when grouped constitute the hospitality sector -- was almost equal to the loss of manufacturing jobs.

A chart follows which shows the following sectoral job gains (among others) for 1989-2000: 17,314 jobs in retail trade, 191 jobs in hotels and lodging, 4,115 jobs in amusement and recreation, and 478 jobs in museums. The total of these lines -- 22,098 jobs -- is apparently the "hospitality sector" gain that is supposed to almost offset the manufacturing loss. "Retail trade" is three-quarters of that total.

Now I don't know how these things work in the economics biz. Maybe there's some good reason for Dean Rosentraub and his colleagues to have a construct called "the hospitality sector" that includes Wal-Mart, K-Mart, Topps, Builders Square, the corner gas station, the video store, the Dollar Store and some apple stand in Hinckley. But it's hard to see how the performance of such a "sector" tells us anything useful about the economic impact of Cleveland's downtown convention and tourist business.

But who cares?. It's a statistic. It comes from the Dean of a college at CSU. And it appears to document the vital economic potential of conventions and hotels for a desperate Cleveland economy... at least until you give it a second glance. So put it in the paper and let the yokels try to figure out if it actually means anything.

Well, fellow yokels, it doesn't. It's junk science... a pre-judged conclusion in search of some data to support it.

Of course there's always a lot of this stuff floating around Cleveland -- remember the 25,000 good jobs Gateway was going to produce? But it sure would be nice if PD reporters had better smell detectors... or if CSU, which reportedly wants to be regarded as a real research institution, would start asking its faculty (and even its Deans) to act more like scientists and less like shills.