There has been, and will be, a lot of comparing and contrasting among competing versions of health care reform in this election. The brouhaha surrounding Sen Obama’s not-quite-universal coverage consumed a good bit of the blogosphere earlier this year, and (if the Dems ever agree on a candidate) the election itself will see the two candidates holding forth on the likelihood that their proposals will reduce the number of uninsured, cut costs, make us all healthier, and brighten our teeth as white as they can be.
Before you condemn or condone, dig into their stats, the basis for their arguments. Do their claims make sense? Are they basing their position on shaky ground? How are they validating their assumptions?
A lot is going to come down to statistics. Unless you are a blind ideologue (which this blog seems to have an uncanny ability to attract), in which case you won’t be bothered by facts, data, or logic.
Which makes this article from Wharton particularly relevant. Here’s what to watch out for.
- Selection bias – do the data selected to assess the plan look to be slanted in a particular direction?
- Misrepresentation – did the advocate report all relevant results, and not just the ones that supported their position? (this is very common in the policy world)
- Unintentional errors – are there math mistakes?
Here’s an example. A survey from the NFIB (National Federation of Independent Business) purports to speak for America’s small businesses. The NFIB has long been an advocate of high deductible plans and an opponent of mandated health insurance coverage. Knowing their history, it is not surprising their survey questions appear to reflect selection bias. Here’s an example.
Question – What is the BEST (their caps) general approach to controlling health care costs (Mark ONE only.)
1. Individuals shopping for the best prices in health care and health insurance.
2. Government regulating health insurance and health care prices
3. Employers choosing/purchasing health insurance on employees’ behalf
Surprisingly, only 49% of respondents voted for number 1. Surprising, because the other options are anathema to the typical independent small business owner. Yet the survey report stated “Small business owners responding to the survey indicated they believe the price mechanism could work to reduce healthcare costs.”
The last quote is indeed accurate, but it does reflect a certain amount of ‘selection bias’, both in the way the question was worded, and the way the responses were described (note the quote did not say how many small business owners believed…)
I’m not picking on the NFIB, which is doing some good work on the national health reform front – merely pointing out that agendas tend to drive results.
As the campaigns continue, keep a skeptical eye out for candidates torturing numbers to get them to say what they want.
