Maths

Common mistakes in sampling and data collection

4 mistakes learners make with sampling and data collection, each one named and explained.

All common mistakes

A sample tells you about the population only if the way it was chosen gave everyone a fair chance of being in it.

A question that tells the reader the answer

What it looks like

You ask: do you agree that too much homework is unfair?

Why it happens

The question reads naturally, because it is phrased the way you already think.

A worked example

Question. Is this a fair survey question: do you agree that too much homework is unfair?

A common answer. yes, it offers agree or disagree

The answer. no. It suggests the answer before asking

Why. Too much and unfair are the writer's opinion, and asking do you agree invites a yes.

Put the two side by side

A fair question leaves the answer open.
A leading question has the answer inside it.

Read it back as if you disagreed. If that feels awkward, rewrite it.

A sample trusted because it came from the population

What it looks like

You say a survey of the school football team represents the whole school, because they are all pupils at the school.

Why it happens

Everyone asked really is part of the population, so the sample looks fair by definition.

A worked example

Question. A survey about sport asks the football team. Is that a fair sample of the school?

A common answer. yes, they are all pupils at the school

The answer. no. People who chose football are not like everyone else on this question

Why. Coming from the population is not enough. Everyone has to have had a fair chance of being asked.

Put the two side by side

Drawn from the population.
Drawn fairly from the population.

Ask who could never have been picked. They are the ones the sample will miss.

Primary data defined by who may have it

What it looks like

You say data is primary because nobody else has collected it.

Why it happens

Primary sounds like first and therefore only, and secondary sounds like second hand.

A worked example

Question. What makes data primary?

A common answer. that nobody else has collected it

The answer. that you collected it yourself for your own question

Why. Two people can collect the same measurements. It is primary for whoever gathered it.

Put the two side by side

Primary: you gathered it.
Secondary: someone else gathered it and you are using it.

Ask who did the collecting, not how rare the data is.

Response boxes that overlap or leave a gap

What it looks like

You offer 0 to 10, 10 to 20, 20 to 30, so someone who is exactly 10 can tick two boxes.

Why it happens

The numbers look tidy and continuous, and the join is the only place it goes wrong.

A worked example

Question. Boxes read 0 to 10, 10 to 20, 20 to 30. What is wrong?

A common answer. nothing, they cover every value

The answer. someone with exactly 10 fits two boxes

Why. Every value must fit exactly one box, so write 0 to 9, 10 to 19, 20 to 29, or use a clear inequality.

Put the two side by side

Every value needs a box.
Every value needs exactly one box.

Test the joins. Put each boundary number through and count the boxes it fits.

WAJD spots these patterns in your child's answers and names the one behind their wrong answers, instead of just marking them wrong.

More maths topics