Qualitative and quantitative data, discrete and continuous data, primary and secondary data, and categorical and ranked data.

What this dot point is asking

AQA wants you to classify data correctly: qualitative or quantitative, discrete or continuous, primary or secondary, and categorical or ranked. The data type decides which averages, diagrams and techniques are valid later, so getting it right matters: you cannot draw a histogram of car colours, and the mean is meaningless for ranked finishing positions.

Qualitative and quantitative data

This is the first split you make, because it limits everything afterwards. Qualitative data can only use the mode as an average and bar charts or pie charts as diagrams. Quantitative data can use the mean, median and mode, and a wider range of diagrams.

Discrete and continuous data

A useful test: if data comes from counting it is usually discrete; if it comes from measuring it is usually continuous. Continuous data can always be measured more precisely (a time of $12.4$ s could be $12.43$ s with a better stopwatch), which is why it is grouped into classes and drawn as histograms rather than bar charts. The distinction matters for choosing diagrams later in the course: discrete data uses bar charts with gaps, continuous data uses histograms with no gaps.

Primary and secondary data

Primary data fits your question exactly and you control its quality, but it is slow and expensive to collect. Secondary data is fast, cheap and often very large scale, but it may not match your question precisely, and it could be out of date or collected with bias you cannot see. Exam questions often ask you to weigh these trade-offs in context.

Categorical and ranked data

Categorical data sorts items into named groups with no natural order, such as nationality or eye colour. Ranked (ordinal) data places items in order of position, such as finishing places in a race or a satisfaction scale from "poor" to "excellent", where the order matters but the gaps between positions may not be equal. Because the gaps are uneven, you can find the median of ranked data but not a meaningful mean, and ranked data is exactly what Spearman's rank correlation coefficient works on later in the course.