# Lesson video

In progress...

Hello.

My name's Miss Parnham? In this lesson, we're going to learn about different sampling methods.

When we carry out surveys, the population is everybody or everything that can be surveyed.

If this is a really large number, then it would be impractical, costly, and time consuming to process all that data, so we choose a sample to represent it.

And this can be done in a number of ways.

The most obvious way is simple random sampling.

So, every person or item stands exactly the same chance of being chosen at random.

An adaptation of this is stratified sampling where people or items are divided into groups first before we do a random sample of each group.

Quota sampling is when we need data from a specific set of people, and we collect that data from them until we've got the quantity that we require.

And systematic sampling means there's a system towards picking people or things.

So, things are listed and numbered, and then starting from a random point in the list, every so often people are selected or items are selected from that list at regular intervals.

Here is a question for you to try.

Pause the video to complete the task, and restart the video when you're finished.

So, Ron's tutor group is probably about 30 people, and this tends to be a minimum for when we take a sample size.

So, asking the whole population is not going to be too difficult to do, so that's the best option here.

And the same goes with Eva's party.

The other two populations, we'd be talking millions or even tens of millions of people.

So, doing a sample would definitely be the best option here.

Here's another question for you to try.

Pause the video to complete the task, and restart the video when you're finished.

Probably the most unusual word here is stratified.

It comes from the word strata, which means layers, but in this context, it's sort of evolved to mean sections or groups.

So, it's a more refined version of random sampling where people are put into their respective groups first, and then a random sample taken of each.

Let's look at the suitability of certain sampling methods for particular situations.

It's important to use the most appropriate sampling method each time.

Starting with simple random sampling.

This is great for things like raffles or prize draws where everyone stands the same chance of being selected.

A refinement of this is stratified sampling.

If we can divide people into subgroups and then take the same proportion from each subgroup to form our sample, then that will be representative of the entire population as a whole.

Quota sampling is a bit different because we need data from a specific section of society.

So, we would gather it from those people, and when we reached the quantity that we desire, we would stop.

Systematic sampling where we list things or put things in some sort of order and pick every so often is particularly useful for quality control in factories.

Here are some questions for you to try.

Pause the video to complete the task, and restart the video when you're finished.

When you need data on a very particular topic, such as in question three, then you need to ask the people who it affects.

So, quota sampling is the best option here.

And then in question four, systematic sampling is the best for production lines because if you're checking, say, every 200 items, you could identify a fault that's occurred partway through that run, stop that, and do something about it.

Whereas if you take a random sample at the end of the production run, then you've probably potentially got some spoiled goods there.

Here's another question for you to try.

Pause the video to complete the task, and restart the video when you're finished.