A systematic tendency to favour some outcomes or groups.
Discrete data
Counted values, usually separate whole numbers.
Key ideas
Begin with a question that needs data and expects variation, rather than one with a single known answer.
Define the population, decide how to sample fairly and choose what to measure.
Distinguish categorical data from numerical data; numerical data may be discrete or continuous.
Use neutral questions, consistent units and sensible recording methods. Do not collect unnecessary personal information.
Consider sampling bias, sample size, non-response and measurement error before generalising.
Describe conclusions in context and state limits. An association does not establish that one variable causes the other.
03 · Follow the working
Worked examples
Read the question first. Cover the working and decide what you would do next.
01Improve a sample
A student asks only their basketball team how much exercise first years do. Is the sample suitable?
Why this method? Decide what the question is trying to find out. Then check whether the data and method can support that conclusion.
Population: all first-year students.
Why?The question is about the whole year group, so identify that group first.
Sample used: the basketball team.
Why?This is only one part of the year group.
The sample may overestimate exercise.
Why?Team members are likely to do more organised exercise than many other students.
Choose students randomly across the first-year classes.
Why?This gives students outside the team a chance of selection. Use the same question and time period for everyone.
A sample still has uncertainty.
Why?Students may estimate their exercise inaccurately, and another random sample could give a different result.
02Classify variables
Classify favourite sport, number of siblings and journey time to school.
Why this method? Decide what the question is trying to find out. Then check whether the data and method can support that conclusion.
Categorical: values are categories.
Why?Favourite sport
Numerical and discrete: it is a count.
Why?Number of siblings
Numerical and continuous: in principle it can take any value in a range, even if recorded to whole minutes.
Why?Journey time
03A statistical question
Improve: “Do you agree our brilliant new lunch menu is healthier?”
Why this method? Decide what the question is trying to find out. Then check whether the data and method can support that conclusion.
Use neutral wording rather than brilliant or healthier.
Why?Remove leading language
For example: “How many days in the last school week did you buy a school lunch?”
Why?Define what is measured
Survey a suitable sample and report the distribution. This measures lunch purchases, not health itself.
Why?Collect and interpret
04Choose a simple random sample
A school has 240 students. Select 30 for a survey about journey time.
Why this method?
Number the students 1 to 240.
Why?Start with a complete list of the population.
Draw 30 distinct numbers randomly.
Why?Select without replacement so each student has the same chance.
State the limit of the conclusion.
Final stepA sample from one school does not represent every school.
Common mistake
Assuming a larger sample automatically removes bias. Asking many members of one sports club about favourite sports can still misrepresent the whole school. Use a sampling method suited to the population.
Exam technique
Explain how your sampling method reaches different groups, not just your friends.
Show your method, use units where appropriate and check whether the question asks for an exact or rounded answer.
04 · Try it yourself
Self-check questions
Work on paper first, then open the answer. These are practice questions, not submitted homework.
Question 1
A student surveys only the school choir about how much music all students listen to. Explain one problem and suggest an improvement.
Show answer & check
The sample may over-represent students interested in music. Select students from across the school using a suitable random sampling method.
A large sample from one club can still be biased. Match the sampling method to the population.
Question 2
Classify eye colour, number of pets and measured arm span as categorical, discrete numerical or continuous numerical.
Show answer & check
Eye colour: categorical. Number of pets: discrete numerical. Arm span: continuous numerical.
Counts take separate whole-number values. A measured length can take any value within a range.
Watch an explanation
Videos for this chapter
Choose one explanation, pause to work an example, then return to the notes. All Maths videos →
Corbettmaths · Start here
Sampling: random
The YouTube player loads only when you choose this button.