Types of data and descriptive statistics

Quantitative and qualitative

Quantitative data are numbers: scores, counts, times. They allow statistical comparison and are easy to summarise, but they strip out meaning. Qualitative data are words and descriptions: they capture reasons and depth, but are harder to analyse objectively and open to researcher interpretation. Most good designs collect both.

Levels of measurement

Nominal: categories with no order (yes/no, colour of car). Ordinal: ranked but the gaps are not equal (rating scales, positions in a race). Interval: equal gaps on a real scale (time in seconds, number of words recalled). Getting this right matters because it decides which statistics and which test are legitimate.

Measures of central tendency

The mean uses every score, so it is the most sensitive — and for the same reason it is distorted by outliers. The median is the middle value once ordered; it ignores extreme scores, which makes it the right choice for skewed data and for ordinal data. The mode is the most frequent value, and it is the only option for nominal data.

Measures of spread

The range is quick but rests on the two most extreme scores, so one unusual participant changes it completely. The standard deviation uses all scores and tells you how tightly the data cluster around the mean: a small SD means the participants behaved consistently, and that is what you should say about it in an answer.

Choosing and justifying

Questions rarely ask 'what is the mean'. They ask which measure is appropriate and why — so always tie the choice to the level of measurement and to whether there are extreme scores.