”
I can prove anything by statistics except the truth.
”
This quote by George Manning is an interesting comment about the
issue of statistics and polls.
“I can prove anything by statistics except the truth.” This quote by George Manning is an interesting comment about the issue of statistics and polls.
So when I read Denise Apuzzo’s column dated April 22 titled “Important to understand the pulse of the community,” I came away with few chuckles.
Now while I certainly agree that it’s important to try and understand the pulse of our community, I disagree with Ms. Apuzzo’s conclusion that the results of The Dispatch online polls are a viable measurement for taking Gilroy’s pulse. Why even the online polls conducted by big news sites like Fox News or CNN expressively say words that the poll being taken is not a scientific poll.
But to the contrary, Ms. Apuzzo states that “The results of these polls provide an insight into the heart of Gilroy.” I’m surprised that someone of her intelligence would arrive at such a faulty conclusion. Here’s why.
In the first place, it can be argued that these online polls by their nature contain a built-in logical error known as the “inductive fallacy”. To refresh your memory, inductive reasoning consists of inferring from the properties of a sample to the properties of a population as a whole.
For example, suppose that we have a Gilroy adult population of 25,000 people out of a total population of around 45,000. Now further suppose that we take a random sample of 300 people from this population with a basic polling question in the format of “yes”, “no” or “undecided.” And finally suppose that our poll answer turns out to be 200 “yes,” 80 “no,” and 20 “undecided.” We could then infer inductively that two-thirds, or 66 percent of the total adult population of Gilroy would concur with a “yes” answer to the given poll question.
Now while statistics say that the more similar the same answer is to the population as a whole, the more reliable will be the inductive inference, on the other hand, if the sample (in this case the 300 people taking the online poll) is relevantly dissimilar to the population, then the inductive inference will be unreliable.
My point is that the sampling population taken with an online poll is not in fact a fair random representation of the total adult Gilroy population, and therefore cannot be a fair or accurate measurement of the community’s pulse on any issue.
As a further example, the very fact that the poll can only be taken online excludes the adult Gilroy population that does not have access to the Internet. And yes, there are adults in Gilroy who either don’t own a computer, or even if they own one, are either not familiar with the Internet and don’t care or don’t know about the e-version of The Dispatch. And it certainly seems fair to conclude that these people would probably have an opinion about the poll question if they were able to answer it.
Next, because the Internet is world-wide, it’s possible that someone living in Pixley, Iowa who has happened to find The Dispatch online could take the poll, thereby distorting the poll’s interpretation. Remember, under current online Dispatch polling, there is no way to exclude participants that are non-Gilroy residents. Why even someone who lives in Morgan Hill could be voting on Gilroy issues. How dare them! We Gilroyans don’t monkey around in their community affairs, do we?
Additionally, the poll results only show percentages. So while the first Dispatch online poll asked “Were the Gilroy High School skits too racy?” the results were that 58 percent said yes, 34 percent said no, and 8 percent were undecided. As I recall about 276 people voted on that issue. Ms. Apuzzo’s comment about this result was “Clearly, sexually explicit skits performed by and for Gilroy students are not supported in this community.”
Now that’s true for those people who took the poll, but again, Ms. Apuzzo is basing her conclusion on the inductive fallacy error. It would be interesting to conduct a real scientific poll on this same issue among the GHS students and one among the teachers. I’ll bet that the majority of students would vote “no” to this question. Also, does teenage opinion have validity for the measurement of a community “pulse?” You answer that question.
So, while these polls are entertaining, I don’t pretend to really think that they prove anything other than the views of a very small group of poll participants. The results simply don’t determine the real pulse of our community. Scientific polls withstanding, I’d like to know your thoughts on how to measure the pulse of this community. Send me a note with your comments.
