Wednesday, 2 September 2015

TANDQ 01: Bias Confirmed

In 2014-2015 I wrote an education column called "There Are No Dumb Questions" for the website "MuseHack". As that site has evolved, I have decided to republish those columns here (updating the index page as I go) every Wednesday. This first column originally appeared on Wednesday, March 26, 2014.

How can we ask good questions?

In brief, start by not knowing the answer, not knowing where to find the answer - and perhaps not even knowing if there IS an answer. For more depth, read below. For context, it may help you to know who I am. Which would be Gregory Taylor (@mathtans) a high school mathematics teacher from Ontario, Canada who knows a thing or two about education and statistical curiosities. As a geek, I’ve written a fair bit of fan fiction, as well as my own serial (ongoing), “Taylor’s Polynomials”, which personifies mathematical relationships in the same way the anime “Hetalia” personifies countries. I hope to bring a different perspective to things, along with a sense of who’s going to be heading out into the workforce in 4-5 years time.

Now, back to the problem of questions. Many questions that we ask ourselves today are not very good, simply because they end up being answered using a search engine. Yet in doing that, we often bias our own results! For instance, if you believe Asimov wrote the Foundation series of books, you will likely include the term “Asimov” as part of your search. Or if you believe health care is unaffordable, you will include the term “unaffordable” in your search. In both cases, there will be some results that confirm that you were right all along… and so you will not ask any more questions. (Questions like: Is it just me, or is Asimov’s “psychohistory” becoming reality?) This problem is known as Confirmation Bias.

Confirmation Bias is the tendency for an individual to seek out information that confirms what they believe already. We have all become guilty of it, in large part because (if you’re logged into an account) web searches remember the sort of pages you like to visit - ensuring that you will constantly cycle back to the same sorts of pages, reinforcing your beliefs. By the same token, we also don’t tend to follow people on Twitter who disagree with us, or initiate online chats with people who have dissenting viewpoints. In short, the internet is a terrible place for discovering new information - unless we are first able to reject our own preconceived ideas. (Could there be a connection here with Steven’s thoughts on Deep Geeks last week?)

Highlighting the Problem

Watch the following to see what I’m talking about. The Veritasium video “Can You Solve This?” elegantly demonstrates how our initial beliefs can lead to asking the wrong questions. In this other video, Numberphile summarizes the experiment behind a recent newspaper article: “Politics wrecks your ability to do math”. Basically, people stop questioning as soon as their liberal or conservative beliefs are confirmed - even if there is a different interpretation. But it gets worse: being presented with facts to the contrary merely has a tendency to strengthen our initial (flawed?) preconceptions! It’s something called the backfire effect.

So what are we to do? Now, that’s a good question. I know I try to include a few people in my Twitter feed with dissenting viewpoints. I also try to make a point of checking out page 6 in my Google searches, rather than always going with it’s top choices. And, of course, I consider the sort of questions I ask - a topic that is becoming big in education these days. (At least in MY circles - oops, bias is creeping in!) In particular, it’s more important that students learn to question WHY things are the way they are, rather than simply accepting “facts” like “a negative times a negative equals a positive” on blind faith.

To conclude, while I (obviously) believe that there are no stupid questions, this is largely because I don’t want you, my students, or the public in general to ever stop questioning things. There’s more problems out there than just confirmation bias… like the fear that asking a question will brand you as ignorant, or argumentative, or subversive. But who cares? Ask anyway. It sure beats the alternative. Feel free to ask me questions in the comments below too!

For further reading:

1. The 12 cognitive biases that prevent you from being rational

2. The importance of questions

3. Eli Pariser video: Beware online “filter bubbles”

1 comment:

  1. (Comments on the original March 2014 post)
    From SERDAR:
    "I've noticed that the folks in my circles who tend to ask the most questions also tend to ask the *best* questions -- the most informed, most genuinely curious ones. They're not simply looking to have their prejudices confirmed. They don't mind being a little uncertain about things. (See John Holt's "How Children Fail" for some more notes on this score: children with a dread fear of uncertainty make poor learners, because they want answers rather than submit to the process of discovery.)"

    From ME:
    "It's also probably true that the more you ask questions, the better you get at it. I think this is partly why education these days is pushing that angle, as opposed to just memorizing formulas that you can look up on your iPhone. Thanks for the reference! ...hope to remember to look into it once I'm done with the school musical."

    "Uncertainty really is necessary to truly learn. You need to have a certain level of a clean slate to really process information, not just slot it."

    From ME:
    "Not too clean though. I was at a Professional Development session where at one point the presenter said that learning is when I have an idea, and thanks to you, I change it. If I start with no ideas, and thanks to you, I gain some, that's merely acceptance. Unless you then question them on your own."

    "It's like exercising a muscle, really. We need to learn how to ask, answer, and KNOW we have answers."