Lesson video

Hello, my name's Miss Parnham in this lesson, we're going to learn how to design questionnaires avoiding bias.

Let's look at some questions that Eva's designed.

The first question says, how often do you go to a cafe? Now she's not specified whether she means how often in a week you go, in a month, in a year? So she's given no timeframe in her question.

And then her response boxes are worded.

And we all understand what never means, but sometimes and frequently mean different things to different people.

So it's quite ambiguous.

So a better question would be to ask something like how many times a month do you go to a cafe and then give some response boxes which are numerical.

And here we always have a zero option.

Let's look at the next question.

How much do you spend in a cafe? Again, she's not specified a timeframe.

She's not said per visit, per person, per week.

And in addition to that, a response boxes, if someone's spent exactly five pounds, which box would they tick? So you've got overlapping categories there.

What if they spent 12 pounds? They wouldn't know which box to tick because we've got a gap in between categories.

And then beyond that, if they spent over 20 pounds, then they wouldn't have a box to tick at all.

So a better option is to be a bit more specific in the question.

What do you expect to pay for a main course in a cafe? Less than four pounds, four pounds to five 99, and then six pounds to seven 99 leaves no gap between that category and then eight pounds plus.

Here's a question for you to try pause the video, to complete the task and restart the video when you're finished.

Here are the answers.

The first question will be improved by asking how many times a week, a month, a year someone's downloading films, and we need an option for zero.

Believe it or not some people don't download films. The second question is even more vague, because again, it doesn't reference whether it wants visits to the cinema to be per week per month, per year.

And then the labels on the response boxes are worded.

And whereas most people would agree what never means sometimes and always, they're open to interpretation because they mean different things to different people.

Numerical labels would be best here.

The third question is well written because it does include a timeframe but it's response boxes let it down.

There's no option for zero and time is a continuous measurement.

So we can't have any gaps between the options.

So say someone had watched three and a half hours worth of films in a week, which box would they tick? Here's another question for you to try pause the video, to complete the task and restart the video when you're finished.

Here are the answers.

Hopefully what we talked about in the previous question has helped you write the suitable question for Mo, it doesn't have to be worded in exactly the same way as the solution, but as long as a timeframe is included and in terms of the response boxes, as long as you've got an option for zero and as long as there are no gaps between the boxes, then go ahead and Mark that right.

If something's biassed, it means it's unfair.

So when we design questionnaires, we need to avoid bias because we want the most accurate information possible.

So avoiding bias means asking a broad spectrum of people the question.

So you get a broad view and also asking questions that don't influence the outcome or answer.

So Eva is asked her family the following question, do you agree that toasties are better than Jacket potatoes? The two reasons why she's not avoided bias, are who's she's asked, her family, her family will be a small number of people.

So not enough quantity, and they will probably have similar views and she needs to get a broader spectrum of information.

Second thing she's asked is a leading question.

She started it with, do you agree most leading questions start with do you agree because they're trying to influence the responder to answer in a certain way.

Another way of having a leading question is maybe to make a negative or positive statement about something first, before that asking about it.

So this leading question does not avoid bias.

Here's a question for you to try pause the video, to complete the task and restart the video, when you're finished.

Here are the answers.

I did include a leading question that didn't start with, do you agree? So don't want you to fall into that trap of thinking it always starts with, do you agree? So one of these questions made a negative statement about fast food first before asking a question.

So it is trying to influence the response questions need to be neutral, so that give an honest response.

Here's another question for you to try pause the video, to complete the task and restart the video when you're finished.

Here are the answers.

Jack's friends are probably a similar age and have got similar life experiences.

So he needs to make sure that he's asking a broader spectrum of society, as well as, as a sample size that's big enough that will not give him some biassed results.

That's all for this lesson.

Thank you for watching.