IB Mathematics: Analysis and Approaches · Topic 4 · Statistics and probability
4.1 Sampling concepts
Describe common sampling methods, explain why unbiased samples matter, and identify sources of bias in data collection.
1/5A sample is used to learn about a population
Part 1 of 5: A sample is used to learn about a population
A school wants to know how its 1200 students travel to school. Asking everyone takes time, so it asks a smaller group instead. The whole group of interest is the population.
The smaller group that is actually asked is the sample. Results from the sample are used to estimate facts about the population. A good estimate needs the sample to be representative, meaning that it mirrors the population.
A sample that is poorly chosen can mislead, even when it is large. For that reason the method of choosing matters as much as the size.
Key terms · tap one for its definition
population
A population is the whole group of individuals or items that a statistical study aims to describe.
sample
A sample is a subset of the population from which data are actually collected.
Part 2 of 5: A simple random sample gives every member an equal chance
Now consider how the school should choose its sample. First it needs a list of all the students. This list is the sampling frame.
In a simple random sample, every member of the population has the same chance of being chosen. Chance alone decides who is selected. A random number generator can do the job. Number the students from 1 to 1200, then generate 60 different numbers.
If a number repeats, ignore it and generate another. The students with the chosen numbers form the sample.
Key terms · tap one for its definition
sampling frame
A sampling frame is a list of all the members of the population from which a sample can be drawn.
simple random sample
A simple random sample is a sample chosen so that every member of the population has an equal chance of selection.

Quick check
One question on the part you just read
Easy
Show answer and explanation
Answer: B · Worked explanation
Part 3 of 5: Other methods suit different populations and aims
Sometimes a simple random sample is slow to produce or misses small groups. Other methods then help. In systematic sampling, you pick a random start and then take every th member of the frame. For a population of and a sample of , use .
In stratified sampling, the population is split into groups that share a feature, such as year group. A random sample is taken from each group, in proportion to its size.
Four sampling methods
- Method
- Simple random
- How members are chosen
- Chance picks members from the whole frame
- Random choice?
- Yes
- Method
- Systematic
- How members are chosen
- Random start, then every th member
- Random choice?
- Start only
- Method
- Stratified
- How members are chosen
- Random sample from each group
- Random choice?
- Within groups
- Method
- Quota
- How members are chosen
- Interviewer fills set numbers from each group
- Random choice?
- No
Quota sampling looks similar to stratified sampling. The difference is that the interviewer chooses who to ask, so chance plays no part.
Worked example
Stratified sample
- Find the total: .
- Year 10: .
- Year 11: .
- Year 12: .
- Check: .
Answer: The sample has 36 students from Year 10, 28 from Year 11 and 16 from Year 12.
Key terms · tap one for its definition
systematic sampling
Systematic sampling is choosing every th member of the sampling frame after a random start.
stratified sampling
Stratified sampling is taking a random sample from each group of the population, in proportion to the size of that group.
quota sampling
Quota sampling is filling set numbers from each group of the population, with the interviewer choosing the members.
Part 4 of 5: Bias makes a sample unrepresentative of its population
Even a carefully planned sample can go wrong. A sample shows bias when its method of selection favours some members over others. Its results then shift in one direction.
Take a survey of weekly exercise held outside a gym. Gym users exercise more than most people, so the results will be too high. This is convenience sampling: choosing whoever is easiest to reach.
Bias matters because decisions based on the results may be wrong. A biased sample gives estimates that are wrong in a consistent way.
Key terms · tap one for its definition
bias
Bias is a systematic error that makes sample results differ from the population values in one direction.
convenience sampling
Convenience sampling is selecting the members of the population that are easiest to reach.

Quick check
One question on the part you just read
Easy
Show answer and explanation
Answer: D · Worked explanation
Part 5 of 5: Bias can enter through the frame, replies and questions
Bias can enter at several stages of data collection. These are the common sources.
- A poor sampling frame leaves out part of the population, so those members cannot be chosen.
- Non-response bias arises when chosen members do not reply and differ from those who do.
- A leading question is worded so that it pushes people towards one answer.
Several steps reduce bias. Use a complete frame, select by chance, follow up people who do not reply, and word questions neutrally.
Key terms · tap one for its definition
non-response bias
Non-response bias is bias caused when chosen members do not reply and differ from those who do.
leading question
A leading question is a question worded so that it encourages one particular answer.
Check your recall[15 marks]
6 written questions. Answer each one in your head or on paper, then open the mark scheme.
- 1.Distinguish[2 marks]
Distinguish between a population and a sample.
Show mark schemeHide mark scheme
Model answer
A population is the whole group of interest. A sample is the smaller group from which data are actually collected.
- 1 mark for stating that the population is the whole group being studied
- 1 mark for stating that the sample is a smaller group taken from the population
- 2.Describe[3 marks]
Describe how a simple random sample of 60 students could be chosen from a school list of 1200 students.
Show mark schemeHide mark scheme
Model answer
Number the students from 1 to 1200 using the list. Generate 60 different random numbers in that range. Select the students with those numbers, ignoring any repeats.
- 1 mark for using the list of all students as the sampling frame, numbered 1 to 1200
- 1 mark for generating 60 different random numbers in that range
- 1 mark for selecting the students whose numbers appear, so each has an equal chance
- 3.Calculate[3 marks]
Calculate the number of students from each year needed in a stratified sample of 50, given 240 first-year, 160 second-year and 100 third-year students.
Show mark schemeHide mark scheme
Model answer
Total . First year: . Second year: . Third year: .
- 1 mark for finding the total of 500
- 1 mark for using each year's fraction of the total multiplied by 50
- 1 mark for obtaining 24, 16 and 10
- 4.Explain[2 marks]
Explain why increasing the size of a biased sample does not remove the bias.
Show mark schemeHide mark scheme
Model answer
Bias comes from the method of selection, which favours some members. A larger sample chosen the same way repeats the same error, so estimates stay shifted in the same direction.
- 1 mark for stating that bias is caused by the selection method
- 1 mark for explaining that a larger sample repeats the same shift in estimates
- 5.Suggest[2 marks]
Suggest why the results may be biased when a researcher surveys only people leaving a gym to estimate how often adults exercise.
Show mark schemeHide mark scheme
Model answer
This is a convenience sample. People at a gym are more likely to exercise often, so the estimated exercise frequency will be too high.
- 1 mark for identifying that the sample is chosen for convenience and is not representative
- 1 mark for stating that the results will overestimate how often adults exercise
- 6.Compare[3 marks]
Compare stratified sampling with quota sampling.
Show mark schemeHide mark scheme
Model answer
Both split the population into groups and take set numbers from each. In stratified sampling the members of each group are chosen at random. In quota sampling the interviewer chooses, so the sample may be biased.
- 1 mark for stating that both use groups with set numbers from each group
- 1 mark for stating that stratified sampling selects members at random within groups
- 1 mark for stating that quota sampling lets the interviewer choose, which can introduce bias

Practice this topic
10 quick-fire multiple-choice questions, easy to hard. Tap an answer to see if you're right and the worked explanation. (The written questions with mark schemes are under “Check your recall” above.)
0 right0 of 10 answered
Includes the 2 quick checks in the note above.
EasyIdentify[1 mark]
Show answer and explanation
Answer: A · Worked explanation
MediumCalculate[1 mark]
Show answer and explanation
Answer: C · Worked explanation
MediumCalculate[1 mark]
Show answer and explanation
Answer: A · Worked explanation
MediumIdentify[1 mark]
Show answer and explanation
Answer: B · Worked explanation
MediumIdentify[1 mark]
Show answer and explanation
Answer: D · Worked explanation
HardIdentify[2 marks]
Show answer and explanation
Answer: D · Worked explanation
HardCalculate[2 marks]
Show answer and explanation
Answer: B · Worked explanation
HardDetermine[2 marks]
Show answer and explanation
Answer: A · Worked explanation