RSS
Home
Are Americans moving to a public option?
Print
Email
RSS
Subscribe
Some Democrats in Congress appear to be hanging all their hopes of success for a public option on polls which they interpret as indicating that Americans support a public option and this support is growing. Most of us have never been approached to participate in such a poll and have never even seen one. For those who are interested in seeing the questions and the tabulated results of the polling go to:
http://www.washingtonpost.com/wp-srv/politics/polls/postpoll_101909.html?sid=ST2009101902502
and
http://www.surveyusa.com/client/PollReport.aspx?g=5ba17aa2-f1b9-4445-a6b8-62b9d1ba8693
The Washington Post-ABC News poll was conducted by telephone between Oct. 15-18, 2009, among a random national sample of 1,004 adults including users of both conventional and cellular phones. The sampling, data collection and tabulation was performed by TNS of Horsham, Pa. Although 1004 adults were in the sample, only 97, 9.6% responded. Whether this poll is valid or not based on the sampling method, hidden biases in the format or any other internal or external threat to validity could always be argued; however, when only 97 randomly selected individual represent the attitudes of 300 million Americans, the results suffer from one thing – the lack of statistical power.
The power of a statistical test is the probability that the test will not make an error which accepts a public options, call the alternative hyothesis, in favor of the notion that the American people do not want a public option, called the null hypothesis. When there is enough power, then polling which accepted the hypothesis about wanting a public option can be relied upon. As power increases, the chances of making this error decrease. Power analysis is used to calculate the minimum sample size required to accept the outcome of a statistical test with a particular level of confidence. It is doubtful that any power analysis was done regarding an actual sample size of 97 individuals because if it was it would be stated in the analysis.
In a similar poll by SurveyUSA, four question from NBC News and the Wall Street Journal were asked of 1,200 adults sampled. This poll was commissioned by MoveOn .org. Again the question of statistical power can be raised but, even more importantly, is the question of whether or not the bandwidth and amplitude of information contained in just four questions are sufficient enough to be answered by 1200 individuals in a way that validly reflects the true attitudes of 300 million . Bandwidth and amplitude have to do with the specificity of questions. Broad questions containing too much information represents a threat to the validity of the results. Such is the case in this SurveyUSA poll.
Whether the majority of Americans support a public option and this support is growing cannot be determined by these polls and it would either be delusional or a dangerous miscalculation to accept them as valid. It may be convenient to rationalize them in this way so that the public option could be supported. However, a better poll will be held next Tuesday in Virginia, New Jersey and upper New York State. If Democrat candidates are defeated in these States, which went solidly for Obama just one year ago, a valid interpretation would be that those citizens, who now fully understand the direction Democrats are taking regarding many issue, including health care, have repudiated the Democrat Party – a lesson that will have even greater consequences if a public option is forced down the throats of most Americans. The American people are not irrelevant.
Related Articles:
Comments
