4.2.3.2 - Quantitative and qualitative data

Quantitative data

Psychologists often turn behaviour into numbers so that it can be measured, compared and analysed. If a researcher records how many words a participant recalls, how long they take to respond, or what score they give on a stress scale, the resulting data are quantitative.

Quantitative data are common in experiments, correlations, questionnaires and structured observations. For example:

The key feature is that the number represents a measured amount, score, frequency or category count. A participant number, such as "Participant 12", is not really quantitative data because it is only a label. AQA questions often reward students who explain what the number measures rather than just saying "it has numbers in it".

Quantitative data are useful because they can be compared precisely. A researcher can calculate averages, look for relationships, test for differences and present patterns clearly. This supports objectivity because another researcher using the same numerical dataset should usually reach the same basic calculations.

A rating also need not have equal intervals: the difference between ratings 2 and 3 may not represent the same change in feeling as the difference between 8 and 9. An 8 is not automatically twice as much anxiety as a 4. Numbers make comparison possible, but the meaning and validity of the measure still need checking.

However, quantitative data can be reductionist. A score of 8 out of 10 for anxiety records how the participant rated their anxiety on that scale, but it does not tell us what the anxiety feels like, what caused it, or how it affects daily life. The behaviour becomes easier to measure, but some meaning may be lost.

Qualitative data

Not all useful psychological data come as numbers. Sometimes the researcher wants to understand how a person thinks, feels or explains an experience in their own words.

Qualitative data are common in unstructured interviews, open questions, case studies and unstructured observations. For example, a clinical psychologist might ask a client, "Can you describe what your panic attacks feel like?" The answer might include details about fear, embarrassment, physical sensations and situations that trigger panic. That kind of data is richer than a rating scale because it preserves the participant's meaning.

Qualitative data can have high validity for subjective experience because it captures depth, context and meaning. It is especially useful when the topic is complex, sensitive or not yet well understood. For instance, interviews with people experiencing depression may reveal patterns that a fixed questionnaire did not anticipate.

The weakness is that qualitative data are harder to analyse objectively. Researchers often have to interpret long written or spoken accounts, identify themes and decide what each response means. Different researchers may emphasise different parts of the same transcript, so qualitative analysis can be affected by researcher bias and may be less reliable unless clear procedures are used.

Qualitative data can later be turned into numbers, for example by counting themes in interview answers. That does not change the original raw data: the raw data were still words. The later counts are quantitative summaries of qualitative material.

Worked example: one account, two outputs

A student writes, "The countdown helped me plan, but seeing the timer made me panic." The complete statement is qualitative data: it preserves both a benefit and a difficulty. A researcher could code whether each student mentions planning and whether each mentions panic. Across ten accounts, six mention planning and four mention panic. The counts 6 and 4 are quantitative outputs, while the original accounts remain qualitative. The categories may overlap, so the researcher must not assume these are two separate groups of students. Counting mentions does not measure how strongly each person felt.

Data collection techniques

The specification asks for the distinction between qualitative and quantitative data collection techniques. This is slightly different from the distinction between qualitative and quantitative data. The data are the product; the technique is the way the researcher collects them.

The best rule is:

• fixed numerical questions, rating scales and systematic frequency counts produce quantitative data;
• unstructured or open techniques usually produce qualitative data;
• some methods can produce both, depending on how they are designed.