The Personal Website of Mark W. Dawson

Public Polling

Political polling has become ubiquitous and nefarious in today’s society.

Ubiquitous because no matter how inane, vacuous, unimportant, or insignificant the topic of the poll, there are people and organizations that will poll the topic. And there will be a hubbub over the results. This is not much of a concern of mine, as people and groups are free to do whatever they choose with their time and monies, and I am free to ignore these polls.

Nefarious, however, is a big concern of mine. It is well established that the wording of the questions, the order in which they are asked, and the number and form of alternative answers offered can influence the results of polls, and therefore influence public policy.

Statistical Methods of Polling

Statistics, in of themselves, are subject to many problems as I have outlined in my article, "Oh What A Tangled Web We Weave". The statistical methods utilized to create a public poll also has many inherent problems. Coverage bias is a source of error involving the use of samples that are not representative of the population due to the polling methodology. This has become even more a problem as technology has made it more difficult for the pollsters to reach a representative sample of the public. At the beginning of polling, most public polls were by telephoning a person at their home. Many people participated in these polls if they had the time and inclination to do so. However, today, home telephones are dwindling as people switch to using their cell phones. While using your cell phone, you may be at work, out in about walking or shopping, or engaged in other activities that you do not want to be interrupted. With the increased use of text messaging and social apps for communication, this problem is getting worse. Therefore, not as many people are participating in a poll, and those that do are not a true representative sample.

Self-selection bias arises in any situation in which individuals select or unselect themselves into/from a group, causing a biased sample with non-probability sampling. Polls also have the potential for inaccuracy in non-response bias or response bias. This occurs when someone does not answer or improperly answers a question. Many people simply refuse to involve themselves in a poll, and they do this for many reasons. They may not believe in polling, or they believe the poll or pollster is biased, and they will not get involved in polling. If they do become involved in the polling, they may answer the poll question on not what they think or feel, but on what they believe the pollster thinks or feels (sympathetic response). They may also answer the question inaccurately because they feel that they may be disparaged for their answer. This is especially true in a politically correct environment as many people become reluctant to provide a non-politically correct opinion.

Pollsters attempt to correct for these situations by making statistical corrections using various statistical techniques. However, the question then becomes, are these statistical corrections valid, or are they a reflection of pollsters’ biases or prejudices.

The perfect example of these inherent problems with polling was in the Presidential election of 2016. Practically all the pollsters predicted that there was no possibility of Donald Trump defeating Hillary Clinton. Yet Donald Trump was the victor. Prior to the election, political pollsters informed us that there was no way Donald Trump could win the election. In one case, a pollster informed us that Hillary Clinton had a 98 percent chance of winning the general election. Most (if not all) pollsters said that there was no chance that Donald Trump could win the necessary 270 electoral votes needed to win the election. But an election is the only accurate poll worth considering. Despite these pollsters, Donald Trump is the 45^th President of the United States (by a wide margin in the electoral votes). And despite these inaccurate polls, the pollsters are continuing to report on the opinions of the American people regarding Donald Trump’s popularity and policy positions.

It should also be noted that Hillary Clinton relied extensively on polls and their analytics to determine her campaign strategy. The problem is that if the polls were incorrect, then the analytics would be incorrect, and the campaign strategy would, therefore, be faulty. And so it was, and Hillary Clinton lost the election, in part, due to an over-reliance on public polls.

Given the failure of pollsters in the 2016 Presidential Election, it is proper to ask if the pollsters are correct or incorrect in their political polls after the election of Donald Trump. Pollsters would answer that they have changed their statistical corrections based on what they learned from the election. However, I am dubious, as their statistical correction is but guesstimates, and they may be wrong guesstimates. In addition, society has become more polarized, and pollsters may not be able to make accurate guesstimates because of this polarization.

A good overview of political polling is the Brookings Institute articles on “Polling & Public Opinion: The good, the bad, and the ugly” and “Can We Trust the Polls?: It all depends”.

Polls also have a host of other problems. Entire books have been written on the problems of political polling. For these reasons, a poll can be conducted with nefarious intentions and inaccurate results. I will not go into any of these problems, as it is beyond the scope of this article. I will instead focus on two major nefarious issues that are my pet peeves on the utilization of political polls, polling the ignorant, and the proper utilization of polls.

The Village Idiot

Political polling is done with a representative sample of the population. As such, this sample only captures the sentiments and feelings of those sampled, as very rarely are the people that are being sampled have knowledge or experience of the subject that is being polled. Therefore, when a pollster asks what a person thinks, they are usually getting what a person feels.

Feelings are a poor method of crafting public policy decisions. Or as Warren Buffett has said:

"A public-opinion poll is no substitute for thought."
- Warren Buffett

When polling people, you cannot know their scope of knowledge or experience on the subject. You could be polling the village idiots or the village wise men. Therefore, the results of the political polling will often not be a good basis for setting public policy.

When reviewing political polls, you should keep in mind an ancient Chinese proverb:

“A wise man makes his own decisions, an ignorant man follows public opinion”

Utilization of Polls

Which brings us to the question of the proper utilization of the polls. Political polls are often utilized to mold public opinion and influence public policy. However, this results in two important issues when developing public policy.

Think vs. Feel - or - Emotions are Easy, Thinking is Hard

Many (if not most) of today's political debate is about feelings. But feelings do not make for good policy. Facts, intelligence, and reasoning should be utilized to create policy, with feelings being used as a supplement to thinking. The difference is that feelings are emotionally based, while thinking is reason-based. But emotions are easy, while thinking is hard. Always remember that the only good way to create public policy is by an open and honest discussion of the issues based on “Facts, Intelligence, and Reasoning” (reasoned based). All sides of an issue should be heard and debated to ensure that the best public policy is implemented. To do so otherwise often creates more problems than it solves. Doing so also reduces "The Law of Unintended Consequences", as discussed previously.

With Facts, Intelligence, and Reasoning

Facts should be utilized with intellectual reasoning to determine the truth. To do otherwise would abrogate the truth and lead you into misjudgments. To allow emotions into your facts and reasoning will also lead to falsehoods. You must, however, obtain all the facts before you apply intelligence and reasoning. This requires that you listen and read all sides of an issue, not just the side you agree with. You need to winnow out the facts to determine their applicability, their importance, and their usefulness to your reasoning. You must also weigh the facts in their importance and give more credence to the more important facts in your reasoning. Intelligent reasoning then requires that you apply formal and informal logic to the remaining facts. At the end of this process, and not before, you can claim your decision is based on facts, intelligence, and reasoning.

However, public sentiment is very important in the crafting of public policy. To craft public policy that goes against the sentiment of the public is not wise, as it will often not be supported by the public. As Abraham Lincoln noted:

“In this age, in this country, public sentiment is everything. With it, nothing can fail; against it, nothing can succeed. Whoever molds public sentiment goes deeper than he who enacts statutes, or pronounces judicial decisions.”
- Abraham Lincoln

Misuse of Public Opinion Polls

Many people are trying to utilize political polling to shape public policy decisions. But making good public policy decisions requires more than public opinion. Or as a famous Greek philosopher has stated:

"Just as it would be madness to settle on medical treatment for the body of a person by taking an opinion poll of the neighbors, so it is irrational to prescribe for the body politic by polling the opinions of the people at large."
- Plato

Politicians should utilize public polling to guide them in how to best explain their reasoning for the public policy and as a means to change public sentiment. As Ronald Reagan has been quoted as saying:

"I utilize polls to determine how much convincing I need to do".
- Ronald Reagan

Any politician who utilizes political polling to determine public policy is doing a disservice to their country, as stated in the following two quotes about the utilization of political polling:

"I did not enter the labor Party forty-seven years ago to have our manifesto written by Dr. Mori, Dr. Gallup and Mr. Harris."
- Tony Benn

"If you are guided by opinion polls, you are not practicing leadership -- you are practicing followership."
- Margaret Thatcher

Political Polling Conclusions

Whenever political polling is being utilized, you should always:

1. Beware the Poll Takers
2. Beware the Poll Results
3. Beware the Poll Utilizers

Or, you can do what I generally do:

1. Ignore the Public Polls.
2. Assume the Polls are Probably Inaccurate.
3. Remember - Things Change - So will the Public Polls.