Facts, not Fantasy

Sunday, January 23, 2011

Stat Girl: Healthday anti-vax Propaganda

A blogger by the name of "StatGirl" posts a blog where she states:  "Ever notice how the media abuses statistics to write headline? Have you ever fallen for a poorly written "conclusion" from a scientific study? This blog attacks the poor reporting of statistics by the media to save people from making poor decisions based on misinterpretations of data. Also, it's just good clean fun for me."  I suppose one area where people are constantly getting it wrong is in the manufactured vaccine debate. Not only that, it seems that people aren't above intentionally misrepresenting things for their agenda.  But then we knew that about the anti-vax pro-disease charlatans.

Healthday anti-vax propaganda --Parents believe the MMR Autsim Link and then a third party PAC makes sure you still question rationality

Slightly More than Half of Americans Say Vaccines Don't Cause Autism 18% Don't Trust MMR Vaccine and 30% are Undecided

Seriously? Have 18% of American adults lost their minds? Probably.  I want to skip my tirade about how dangerous this belief system is (based on questioning result after result by claiming conflicts of interest on the part of researchers). Not to mention that it's poor logic.

Time to look at how badly the reporting screwed up the numbers. In all surveys, there is a margin of error. What was it here? No idea. Went to Harris interactive and they can't seem to figure one out.  Sooo... 18%? Who knows. If the margin of error is +/- 5 percentage points. So could be 13%? Could be 23%? Whatever. Poor reporting. Also, isn't there some belief that at any given point in time 20% of American are mentally ill? 18% anti-vaxx, 20% mentally ill... if only... but I try not to commit crimes against logic.

At the very bottom of their tables, they offered some commentary/disclaimers about the survey, (they pay their sample and have to "adjust for" propensity to be online, propensity to respond to a survey, age, race, education level and income to make sure that the sample looks like the American public. Gimme a break. That's A LOT of adjusting.

Here's their explanation,
"All sample surveys and polls, whether or not they use probability sampling, are subject to multiple sources of error which are most often not possible to quantify or estimate, including sampling error, coverage error, error associated with nonresponse, error associated with question wording and response options, and post-survey weighting and adjustments. Therefore, Harris Interactive avoids the words "margin of error" as they are misleading. All that can be calculated are different possible sampling errors with different probabilities for pure, unweighted, random samples with 100% response rates. These are only theoretical because no published polls come close to this ideal.
Respondents for this survey were selected from among those who have agreed to participate in Harris Interactive surveys. The data have been weighted to reflect the composition of the adult population. Because the sample is based on those who agreed to participate in the Harris Interactive panel, no estimates of theoretical sampling error can be calculated."

No comments:

Post a Comment

Please keep posts here respectful. Those that cross boundaries will be deleted, and then placed in a special place for future ridicule.