moldybluecheesecurds 2

Thursday, November 05, 2009

CDC says sick people should stay home, but in the U.S. that means giving up a paycheck

Wonk Room: "Actually, almost 50 percent of private-sector workers in the U.S. have no paid sick days. A survey last year by the National Opinion Research Center at the University of Chicago found that “68 percent of those not eligible for paid sick days said they had gone to work with a contagious illness like the flu.”" [emphasis mine]
And once again, when it comes to health care, "The U.S. is the only developed country without a policy mandating some form of paid sick leave."

This is just embarrassing.

2 comments:

rick said...

The merit of the proposal aside, I got to thinking "why those 22 countries?"

According to this source, they picked the top 22 countries on the UN Development Program's "Human Development Index"

http://preview.tinyurl.com/r5uf56

Seems like a reasonable choice. But still, I wonder, why 22? Surely the UNDP ranked more than 22 countries! Let's have a look.

http://hdr.undp.org/en/statistics/

Hmm, looks to me like a natural cutoff would have been the top 38 countries. Such a list would include all countries in the "Very high Human Development" category.

I'm guessing number 23, Singapore, would be the first country on the list other than the US that does not include benefits. (I don't know this to be true, I am just guessing.)

According to the report itself:
"Our sample consists of those countries that score at least 0.94 on the Human Development Index (HDI)..."

http://www.cepr.net/documents/publications/paid-sick-days-2009-05.pdf

Again, it is not clear why 0.94 is chosen. It could be that there are countries that are ranked .9399 and that also don't provide benefits. Wouldn't that also be interesting to know?

The study reports that among the top 22, the US is the outlier, but this smells like an "ESPN statistic" (i.e. look for a sexy way to slice and dice the data to report a "noteworthy" statistic). To the extent that this reporting is interesting, it would be more interesting with additional context.

The study authors should be more forthcoming regarding their methodology.

rick said...

More concisely, let's say I want to write a report showing that the US provides unusually low sick leave among developed countries.

Of course my report will be biased, but I want it to seem unbiased, so I will search for well-respected lists of developed countries. From all these lists, I will choose to use the one that allows me to maximize X in the following statement:

"The US is the only country in the top X countries that does not mandate sick leave."

I suspect this is what happened, but it shouldn't have.