Collecting data - CCEA GCSE Statistics guide to sampling methods, types of data and questionnaire design

Collecting data well is the second stage of the statistical enquiry cycle and a heavily examined part of CCEA GCSE Statistics. This guide covers how to choose a fair sample, how to classify data, and how to design questionnaires and collection sheets that avoid bias.

Sampling frames and methods

A sampling frame is a complete list of the population, such as a register or an electoral roll; an incomplete frame causes bias because some members can never be chosen. Simple random sampling numbers every member and uses random numbers so each has an equal chance. Systematic sampling takes every nth member from a random start, where n is the population divided by the sample size. Stratified sampling divides the population into groups and samples each in proportion. Quota, cluster and convenience sampling are faster but more prone to bias, because the interviewer chooses, whole clusters may differ, or only the easiest people are reached.

The stratified sample

Stratified sampling is the calculation CCEA tests most. Work out the sampling fraction, which is the total sample size divided by the population size, then multiply each stratum size by that fraction to find how many to take from each group. A sample of 50 from 600 gives a fraction of one twelfth, so a group of 360 contributes 30 and a group of 240 contributes 20. Round carefully so the parts add to the required total, and select within each stratum by simple random sampling so the method is genuinely fair.

Types of data

Classifying data correctly decides which charts and averages are valid. Data is qualitative (categories such as colour) or quantitative (numbers). Quantitative data is discrete (separate values from counting, such as the number of pets) or continuous (any value in a range from measuring, such as height or time). A simple test is that counted data is discrete and measured data is continuous. Continuous data is grouped into class intervals, while discrete data can be listed value by value.

Primary and secondary data

Data is also classified by its source. Primary data is collected first-hand for the enquiry, so it fits the question exactly and you know how it was gathered, but it takes time. Secondary data comes from existing sources such as a census, a database or a report; it is quick and often large, but it may not match the question and its reliability may be unknown. The exam often asks candidates to state an advantage and a disadvantage of each.

Designing questionnaires

A good questionnaire uses clear, relevant, unbiased and non-leading questions, and closed questions need response options that are exhaustive and non-overlapping. Common faults to spot and fix are leading questions that push an answer, emotive wording, overlapping boxes (such as 0 to 10 and 10 to 20 sharing the value 10), a missing time frame, and sensitive questions that reduce honesty. A data-collection sheet organises data as it is gathered with clear classes, and a pilot survey trials the questionnaire on a small group to catch problems before full use.

How CCEA examines collecting data

CCEA rewards the stratified-sample calculation, naming a sampling method and explaining its bias, classifying data, and criticising and improving questionnaire questions. These are high-frequency, technique-driven marks. Use the dot points for specification-level detail and worked CCEA-style questions, then test yourself with the quiz.