Skip to main content

About this free course

Become an OU student

Share this free course

Teaching and learning statistics
Teaching and learning statistics

Start this free course now. Just create an account and sign in. Enrol on the course to track your learning.

1.6 Sampling method advantages and disadvantages

Different sampling methods suit different contexts. Now take a moment in the following acivity to consider some of the pros and cons of the methods covered so far.

Activity 4 Sampling pros and cons

Timing: Allow 10 minutes

For the three sampling methods covered so far, convenience, random and systematic sampling, can you think of any advantages or disadvantages for each of these as methods of selecting a sample?

Some key points to think about include representativeness, bias and cost effectiveness.

Write down your thoughts about possible pros and cons of each sampling method.

Comment

Here are some thoughts for each method, you may also have come up with others.

Convenience sample:

The main advantage of this method is that it is quick and easy to do, requiring little planning. However, this leads us to the main disadvantage, the sample gained can be very unrepresentative of the population. For instance, taking the first five pupils who arrive to a lesson would be very unrepresentative if the survey for which they were selected was about punctuality!

Random sample:

Using a random number generator to select members of a sampling frame removes the potential bias of selecting an item, e.g. if selecting something by placing a pin in a page, most right-handed people will be more drawn to one particular part of the page.

However, it may not be possible to give each item a unique identifier prior to selection for instance imagine you wanted a random sample of all the fish in a lake, are you going to catch every fish and tag it first?!

Systematic sample:

This method generally gives a sample that covers the full spread of the population and selection is not biased by how something looks or feels if it is being chosen ‘by eye’.

Do be careful though, if the underlying population is structured in some way, systematic sampling could produce very unrepresentative samples. For instance, if there were a mechanical fault that meant every 50th item produced contained a defect and your sample happened to be systematically selecting this item, you would get a very skewed opinion of the quality of the product produced.