Skip to main content
WalesMathsSyllabus dot point

How do you choose a sample, design data collection and classify data so that conclusions are reliable?

Understand populations and samples, use random and other sampling methods and recognise bias, design data collection including questionnaires, and classify data as qualitative or quantitative and discrete or continuous.

A focused answer to the WJEC GCSE Mathematics statistics content on sampling and data, covering populations and samples, random and other sampling methods, sources of bias, designing questionnaires and classifying data as qualitative or quantitative and discrete or continuous.

Generated by Claude Opus 4.811 min answer

Reviewed by: AI editorial process; not yet individually human-reviewed

Have a quick question? Jump to the Q&A page

Jump to a section
  1. What this dot point is asking
  2. Populations and samples
  3. Sampling methods and bias
  4. Designing data collection
  5. Classifying data
  6. Why this matters

What this dot point is asking

This is the data-handling foundation of WJEC statistics. You are asked to understand the difference between a population and a sample, to choose a sample using random and other methods while avoiding bias, to design data collection (especially questionnaires), and to classify data correctly. Statistics begins with how data is gathered, because a poorly chosen sample or a leading question makes every later calculation worthless. The exam rewards clear descriptions of sampling methods and sharp criticism of flawed survey questions, both worded-answer skills.

Populations and samples

Studying a whole population is often impractical, so we sample.

So to find the average height of students in a school of 900900, you might measure a sample of 6060 rather than all 900900, provided those 6060 fairly represent the whole school.

Sampling methods and bias

How the sample is chosen decides whether it is fair.

The exam often gives a flawed sampling plan (surveying shoppers on a weekday morning, say) and asks why it is biased. The answer names the group that is over- or under-represented.

Designing data collection

Questionnaires must collect data fairly and clearly.

A good survey question is clear (no ambiguity), neutral (not leading the respondent to an answer), and offers response options that do not overlap and cover all cases. For example, age groups "00 to 1010", "1010 to 2020" overlap at 1010; better are "00 to 99", "1010 to 1919". A data collection sheet or tally chart records responses efficiently. Recognising and fixing a leading or loaded question is a recurring two-mark task.

Classifying data

Data falls into types that decide which charts and averages suit it.

Why this matters

Sampling and data is the entry point to the whole statistics strand and is examined as worded reasoning rather than calculation, so it tests AO2 and AO3 directly. The skills (choosing a fair sample, spotting bias, criticising and improving survey questions, classifying data) appear in short, reliable mark-earning questions, and the data type determines whether you later use a bar chart or a histogram, a discrete or a grouped frequency table. Getting the foundation right makes the calculation topics that follow meaningful.

Exam-style practice questions

Practice questions written in the style of WJEC exam questions on this dot point, with worked answer explainers. The year tag is the paper they imitate, not the source.

WJEC 20192 marksA school has 900900 students. A teacher wants a sample of 6060 students that fairly represents the school. Describe one suitable method of choosing the sample. (Foundation and Higher, Unit 2, calculator.)
Show worked answer →

A suitable method is simple random sampling: give every student a number from 11 to 900900, then use a random number generator (or random number tables) to pick 6060 different numbers, and select those students.

The key idea is that every student has an equal chance of being chosen, which avoids bias.

Markers award a mark for describing a method that gives every student an equal chance, and a mark for the practical detail (numbering the students and selecting at random). Vague answers such as "pick people at random" without a method may not score the second mark.

WJEC 20212 marksA survey question reads: 'Do you agree that our excellent local park should get more funding?' Give one reason why this is a poor question and write an improved version. (Foundation and Higher, Unit 2, calculator.)
Show worked answer →

The question is leading (biased): the word "excellent" pushes the respondent towards agreeing, so the results would not be fair.

An improved version is neutral, for example: "Should the local park receive more funding?" with response options "Yes / No / Don't know".

Markers give a mark for identifying the bias (a leading or loaded question) and a mark for a neutral rewrite, ideally with clear, non-overlapping response options. Just saying "it is biased" without explaining why, or without an improvement, loses a mark.

Related dot points

Sources & how we know this