Brian Beutler | June 23, 2011, 6:30AM
One of the nation's foremost experts on survey writing compared McKinsey & Company's controversial health care study to a push poll, calling into further question whether its results can be trusted as even a snapshot of employer sentiment.
"There is no doubt that the answers one would get after priming respondents the way they did would be expected to include more expressed interest in the possibility of not insuring employees than a question asked in a nonprimed context," said Floyd Fowler, a Senior Research Fellow at the Center for Survey Research at University of Massachusetts, Boston, and author of the book Survey Research Methods.
McKinsey's study concluded that employers are fairly likely to rescind insurance benefits after the health care law takes full effect in 2014. The firm claims to stand by the result, though it now admits the results of the survey aren't predictive. However, the results were based on some curious questions, pasted in full below, designed to lead survey-takers to conclusions at odds with the majority of expert analysis.
-snip
Republicans and foes of the health care law continue to cite the McKinsey study and stand by its results despite McKinsey's admonition that it's not intended to be predictive. So the political damage may already be done. But we now have authoritative reason to believe that McKinsey's result is misleading -- not simply an innocent outlier.
-snip
http://tpmdc.talkingpointsmemo.com/2011/06/top-expert-disputed-mckinsey-health-care-study-akin-to-push-poll.php?ref=fpblg