Lesson video

In progress...


Hello everyone, it's Mr. Miller here and welcome to the fourth lesson on statistics.

And today we're going to be looking at survey questions.

So first of all, I hope that you're doing well and welcome to today's lesson.

And we're going to start with this slide which you've seen before, so hopefully you're not getting too bored of it, but it's an important one because it tells us where we are in the overall context of statistics.

So we have already, we've looked at what a hypothesis was.

And in the last lesson we looked at sampling.

So that's all about who are we going to ask and where are we going to ask them? And this is the final lesson before we actually go and collect some data.

So once we've decided who we're going to ask, we need to decide what kind of questions are we going to ask them? And this lesson is all about survey questions and the responses that we offer to those questions.

So it should be pretty straightforward.

Let's go ahead and check out the Try This slide.

So let's imagine that you're running a survey about people's eating habits.

Here are four questions and these are all very bad questions.

So I want you to have a think about each of these questions.

First of all, fruit makes you really healthy.

Do you eat fruit? Why is that a bad question? Well, if you're thinking that, you know, if you say something like fruit makes you really healthy, then that person is going to be more likely to say, yes, I do eat fruit, because you know they might feel guilty if they don't eat fruit, that they should be eating it.

So, anyway, this is a leading question or a bias question because you're giving some kind of information that might change the answer before you actually ask the question.

Next one.

What's your hair colour? Why is that a bad question? Well, your survey is about people's eating habits.

Hair colour has nothing to do with it, so it's irrelevant.

Don't bother asking a question that has nothing to do with your survey.

Next one.

Are you overweight? Would you ask that one? Well, hopefully you wouldn't ask that question.

Like imagine if someone asks you that.

It's really, really embarrassing and yeah.

Any kind of question that someone might be uncomfortable answering, you should definitely avoid.

Next one, how many times have you eaten pizza? Well, it sounds like a reasonable question, but you haven't in the question you haven't, kind of mentioned what timeframe you're talking about.

So are you talking about the last week, you talking about the last month, that kind of thing.

So it's fine if you mentioned those timeframes in the answers, so your answers might be, you know, once a week, twice a week, whatever, whatever, but yeah, really important that the question is making the timeframe very clear.

Anyway, these are all examples of some bad questions.

Now we're going to have a look at some possible answer options that we can get.

So this time, imagine you want to run a survey about playing sports.

So what is wrong with these answer options? The first question, what is your favourite sport? That question is fine.

There's nothing wrong with that question, but there is something wrong with the different answer options that have been provided.

Can you think what's wrong here? Well, what's clearly wrong here is that if you liked football or hockey, then it's fine.

But what if you like a sport that isn't on here, like tennis or rowing or athletics or whatever it is.

So the problem is is that these answer options don't cover all possible answers.

So you might want to create another box.

Sometimes we see people doing this, another box with just says other.

And so if, if your favourite sport is not one of those four, you can just tick the other box and that survey question will be fine.

Next question.

How often do you play football? Fair enough question.

But what is wrong with the answer options? Well, that is really unclear what these answer options actually mean.

What does sometimes mean? Does that mean once a week? Does it mean once a month? Does it mean three times a week? It's really not clear.

So yeah.

Make sure that you avoid these kinds of answer options because people will interpret the meaning of often, the meaning of sometimes very, very differently.

So you're much better off giving them a quantitative answer option.

What I mean by quantitative, you know that word already, is a numerical value.

So you're much better giving answer options like never or zero times a week, once a week, twice a week, three times a week, four or more times a week.

That would be much, much better.


Let's have a look at some questions now.

Okay, for both of these questions, critique both the question and the answer option.

So two questions here for you to have a think about.

So pause the video now to have a think, have a read through these questions.

And you might think that there's something wrong with the question.

You might think that there's something wrong with the answer it's up to you.

But what I want you to do is pause the video and spend a couple of minutes writing down a couple of sentences for each of these questions.

Pause the video now.

Okay, great.

So let's go through this.

So the first question our customer service is amazing.

Isn't it? How satisfied were you with our customer service? If you're thinking that this bit here is a bit ridiculous to put at the start of a question, then I'd agree with you.

Because if you're saying our customer service is amazing, isn't it? You're kind of wanting people to agree with you.

And that might put people off.

It might make people more likely to say that they were satisfied.

So that is a leading question.

And you should definitely avoid that because you want your answer, your survey questions to be reliable.

In terms of the answer options, very satisfied, satisfied, et cetera.

I'm actually fine with this because when you see surveys in real life, you often get something very similar to this.

This is very common and it's absolutely fine to have these kind of answer options.

Next one.

How often do you exercise for 30 minutes or more in a week? What do you think? Well, I think the question is absolutely fine.

You are clearly talking about exercise here 30 minutes or more in a week.

That question is absolutely fine.

What's wrong is with the answer options.

And hopefully you picked out that these answer options overlap because if you exercise once a week for 30 minutes or more, which box do you tick? Do you tick the first one, zero to one, one or the second one one to two? You don't know? Same thing if you exercise two more times, two times a week.

You don't know whether to pick the second or the third box.

So you must make sure that your answer options don't overlap.

So you could have something which is zero to one, two, three, or four or more.

That would make perfect sense.


I hope that this was really clear.

Let's take a look at the explore task.

So here is the explore task.

And as I've been saying throughout this week, this is the question we have been leading up to.

So we are looking at the Cool Sports Inc.

, the designing clothes brand.

We want to hire people to do some data analysis.

So over the course of these past few lessons, you've been thinking about the different parts of this analysis.

So the first lesson you talked about, you were looking at different types of data you could use.

And so on.

Now, today you need to put all of this together to write a pitch to Cool Sports, explaining why they should hire you to run their study.

So they're interested in a study to do some market research, some data analysis on the clothing market.

In your pitch, you should include any hypotheses you might have.

Anything that they might be interested in.

The sample you want to take, the questions you want to ask, and finally why they should hire you.

So it's up to you.

What you write is up to you, be creative.

You can write as much as you want.

And on the next slide I'm going to show you the hashtag that you can use to show us your work.

We might even give some prizes for the best piece of work here.

So here we have it.

So have a go at this explore task.

Spend some time doing it.

I hope that you enjoy it.

Thanks so much for watching and I'll see you next time where we're going to be looking at analysing some data.

Thanks very much for watching.

See you next time.