Here is an opportunity to create facts(?) where only fiction has existed before. Our press and politicians seem more impressed when some polster reports the results of a new survey that shows ...
But what do the results mean? Three things: (1) the pollster asked some questions, (2) the interviewee actually answered in a some coherent fashion and (3) the pollster dutifully reported the statistic showing the percentage response (and sometimes the margin of error. On this political decisions are made, funding for programs authorized, products are introduced to market and all sorts of other events are set in motion.
Consider the following by the Pew Research Center for the People and the Press. A survey was conducted In early May 1997, prior to the Senate voting on the bill to ban partial birth abortions. One must assume the results of the survey were considered by some of legislators on how they should vote so as not to enrage or offend their constituents. The poll was of 615 people; no mention of men or women or children, voting age or not, rich, poor or middle class, whether they voted in the last election or not, married or single, religion, race, age, handicapped, having had an abortion or know someone who has, owns property or rents, a felon, convicted but not sentenced, free on bond, incarcerated, &c, &c, &c. If one takes the 50 states and divides the total (615) you get about 12 people for each state, other divisions and categories and you begin to doubt if the size survey has significance.
Then, you qualify the respondents by their knowledge of people and events, to with, 82% knew the name Tiger Woods as the new black golfer, 80% identified Dennis Rodman with his wild hair color and basketball fame, 62% were aware of Ellen DeGeneres of television fame as an out-of-the-closet lesbian. Surprisingly, 40% identified Alan Greenspan of the Federal Reserve. No one else came close to being known to these citizens of the United States. John Huang (although we can't find him) was correctly identifed by 20% of those polled as having something to do with Democratic fund raising. He shared this percentage with Kenneth Starr (hopefully not confused with Bart Starr) who was identifed as the independent counsel investigating President Clinton. Then followed closely behind was Garry Kasparov, the loser in the match with IBM's computer at 18%. And bringing up the rear was Trent Lott, Webster Hubbell, and Tony Blair at 15%
It doesn't take much imagination to conclude that there is really no difference between the 15 and 20% recognition statistics, just the random lack of specific knowledge of an individual in a particular role. When you move to questions answered correctly by 40 to 60 percent of the respondents, the pollsters have obviously expanded the population of people. And at, 80% recognition statistic, they have included yet another group. Until finally, and surely, there must be some question that everyone could have got right, you would have the entire population making up the 615 interviewees. What a bunch of dunderheads (the pollsters, the population polled and the people who paid good money for this nonsense).
Now the above just gives you an idea of the nature of people responding the survey. But what did the survey's intended question, i.e., should partial abortions be outlawed? Why they split about 50/50.
With politicians depending so heavily on polls to decide which wheel should be greased, I guess this shows better than anything else, that it's time to revise the old maxim: You can fool some of the people part of the time and all the people part of the time, but you can't fool all the people all of the time. Now it should read: most of the people are fools, some more so than others, therefore why bother, just fool them as much as you like, they wont know the difference.
I hope that the results of the survey didn't sway any of the Senators in their voting. Eighty percent of the public (if you believe this survey) wont know or care which way you voted and probably can't identify you as their Senator, regardless.
But the more serious matter is the professionals, masquerading as scientist, professing that their surveys have meaning. Sham on them! Here's another fine example of learned ignorance, the art of selective use of data for the purpose at hand. Where are the Menckens, Branns, and Huggins of today who will so depreciate the report at hand that the pollsters will either clean up their act or go silently into the night.
Of course, you may say the survey has real meaning, it provides a characterization of our society, our educational system and our quest for data, not information.
--- N.B. The results of the survey were from an Associated Press, wire datelined Washington and reported in the Sioux City, Iowa and Yankton, South Dakota papers on Friday, May 23, 1997
ABOUT Joe Wortham