Trump’s Google polling remark shows we will believe anything

I saw a funny thing. This week we saw the first “presidential” (the quotation marks are there for a reason, do the math ;-)) debate between Hillary Clinton and Donald Trump. As usual, after the debate, a new debate arises: who won. This is usually done by polling.

When it comes to polling there are a lot of ‘buts and ifs’ surrounding the results. What I know from my time working at a market research company, is that what you are looking for as a poller is ‘significant’ results. You need proof based on a certain number of responses, within a specific group of people. If the numbers are too small or too biased, the results don’t mean anything.

Simply said: you can’t say “America thinks X, Y, or Z” if you only poll Republicans or only Democrats for example.

So polls should be looked at with the necessary criticism.

We don’t do that. And politicians do even less.

Today Donald Trump proved he is both a politician like any other (which he claims he’s not) as well as a completely ignorant person. Like most of us.

He proved he is a politician because he will use anything that might support him or his message, without even taking a second critical look.

He proved he is ignorant because he doesn’t understand the difference between a poll done BY Google and a poll where Google’s software is used. (Though chances are he might know, but just doesn’t care, making him even worse).

Here’s what happened.

At a rally in Wisconsin on Wednesday Trump claimed a “Google poll” showed his campaign is leading. He also claimed Google is trying to hide bad news about Clinton (not true, but a different discussion).

The problem is: this wasn’t Google poll. And on top of that, it wasn’t really a trustworthy poll either.

What is was, was a survey performed by the website “Independent Journal Review”. What they did is they used Google’s survey software to ask their readers what they thought about the debate and the campaign in general. You can see the survey and results here.

The survey showed Hillary winning the debate, but Trump leading on the question “As of today, which candidate do you support?”.

Case closed. Well, for Trump at least. He took that specific answer and then turned it into a lead in the polls.

However, any critical person who has to make decisions (something I think a President has to do quite a bit?) would look at the results a bit deeper. And what would they find?

Well for a start, the answer to the next question: “Which candidate was most honest during the debate?” When most people in the survey think Trump was (where many sources say he told most lies) telling the truth most, you should at least become suspicious.

And if you look even further it will become even clearer you should at least be careful with these results. The last sentence of the article on IJR says it all to me:

The people surveyed were ‘only’ 1,304 “likely voters” who watched all or some of the debate. So chances are it’s not even people who saw the debate, who still have an opinion, who might or might not vote. That’s a bit vague, isn’t it?

And then there is the issue of the software used. Where Trump claims it was a “Google poll”, he ‘forgot’ to leave out that it wasn’t Google who polled, but IJR using Google’s Consumer Survey tool.

The tool can be used in two ways: using a phone app and as a survey presented before you can enter a website. Which means the audience is a very specific one (readers of the website or those using the app). Knowing the website favours the Republicans, chances are most respondents would favour Trump anyway. Secondly, chances are people ‘just answered something’, to get to the website. Thirdly, the 1,304 respondents could just as well be 10 people, answering 13 times. We can’t check it.

Basically, this poll has so many things against it, it wouldn’t be seen as trustworthy amongst those in the surveying business. Yet Trump did a very political thing: he used it in his favour anyway. And I’m afraid these are marketing tactics as well…

Smart by Trump, but it also shows how most of us will believe almost anything we read or hear. Unfortunately. And that could be the biggest problem with this election and any other one that we will see in the next few years.