Data Types, Sources and Sampling

Two different questions about data

Psychologists collect data: information that helps them investigate behaviour and mental processes. When describing that information, two separate questions matter: what form does it take? and where did it come from?

Quantitative and qualitative describe its form. Primary and secondary describe its source. Sampling concerns whose information is included, which affects how widely the findings can be applied.

Quantitative data: measuring and counting

Quantitative data is information recorded in numerical form. Examples include the number of words recalled in a memory test, a rating given in response to a closed questionnaire question, or the number of times a particular behaviour occurs during a structured observation.

Numbers make comparisons easier. A psychologist can compare participants or groups, identify patterns and summarise results using descriptive statistics. Clear numerical scoring rules also make it easier for other researchers to repeat the measurement or check whether different observers agree. This can improve reliability, meaning the consistency of measurement.

In Piaget’s research (1952), the numbers of children in different age groups who correctly completed conservation-of-number tasks were quantitative data. In Heaven’s research (1996), numerical measures of personality and delinquency could be analysed to investigate relationships between them.

However, a number does not necessarily explain the experience behind it. A low memory score shows how many items someone recalled, but does not itself reveal why they struggled. Reducing complex experiences to scores can lose important detail. It may also weaken construct validity: how well the measurement captures the psychological concept being investigated.

Qualitative data: describing experiences and reasons

Qualitative data is non-numerical information describing behaviour, thoughts, feelings or experiences. It may take the form of interview transcripts, answers to open questionnaire questions, case-study descriptions or unstructured observation notes.

In Morrell et al.’s research (2011), interviews with young people and stakeholders provided descriptions of involvement in rioting or looting and the reasons for it. Wilson et al. (2008) recorded case-study descriptions of Clive Wearing’s neurological and psychological experiences following a viral infection that caused severe memory difficulties.

Such accounts can provide rich detail about why people behave as they do. Rather than reducing an experience to a single score, they preserve its complexity and context. This can help a researcher understand and measure the concept more meaningfully.

The difficulty is that detailed accounts take time to analyse. Researchers must interpret what participants mean, and different researchers may interpret the same description differently. This subjectivity can weaken the validity of the conclusions if the analysis reflects the researcher’s assumptions rather than the participant’s experience.

Qualitative information can also be grouped into categories and counted. For example, descriptions of reasons for taking part in an activity could be grouped by theme, then the number of responses in each theme recorded. The descriptions are qualitative; the resulting counts are quantitative. Counting makes comparisons easier, but some detail is lost.

Neither type is always better. Quantitative data is useful for comparing amounts and identifying patterns; qualitative data is useful for understanding meanings and reasons. The choice should fit the research aim.

Primary and secondary data: tracing the source

Primary data is information a researcher collects first-hand for their own investigation. A psychologist who runs a memory test collects primary data from the participants’ scores. A psychologist who interviews participants also collects primary data, even though the answers may be qualitative.

Collecting primary data lets the researcher choose questions and measurements suited to their aim and control how collection takes place. However, organising participants and collecting information requires time and resources. Researcher bias can still affect what is recorded or how it is interpreted.

Secondary data is existing information collected by someone else and used in a new investigation. For example, a psychologist might analyse another researcher’s existing memory-test results rather than recruit participants and run the test themselves. Existing interview transcripts could also be used as secondary data.

Secondary data can save collection time and provide access to information that would otherwise be difficult to obtain. However, it was collected for someone else’s purpose, so it may not match the new research aim. The new researcher has less control over the original procedures and must consider the quality and age of the information.

The two distinctions combine like this:

Sampling: whose behaviour does the data represent?

The target population is the whole group to whom a researcher intends the findings to apply. The sample is the smaller group who actually take part. Studying a sample is usually more practical than studying every member of a population.

A representative sample reflects the population’s relevant characteristics, such as age and background. This matters because researchers often want to generalise: apply findings from the sample to the wider population.

Suppose a psychologist wants to investigate study habits among all pupils at a school, but recruits only pupils attending an after-school revision club. These pupils may be especially motivated to study. Their answers might describe the club members accurately without representing all pupils. This is sampling bias, because the selection process systematically favours a particular group.

Applying sampling principles

How participants are selected affects the usefulness of the resulting data:

• Random sampling gives every member of the target population an equal chance of selection, for example by drawing names from a complete list at random. This reduces selection bias, although chance can still produce an unrepresentative sample.
• Opportunity sampling uses people who are available and willing at the time. It is quick and convenient, but people available in one place or at one time may share characteristics that the wider population does not.
• Self-selected sampling uses volunteers who respond to an invitation or advert. It can help recruit willing participants, but volunteers may differ from non-volunteers in motivation, interest or available time.

Sample size matters too. A larger sample generally reduces the influence of individual differences, but size alone cannot remove bias. Recruiting more pupils from the same revision club does not solve the problem of excluding other pupils.

When interpreting findings, connect the sample to the intended population. Consider who was eligible, how participants were recruited, whether important groups were left out, and whether the size and composition justify the proposed generalisation. Practical limits on time, cost and access may prevent an ideal sample, but those limits should shape the conclusions rather than be ignored. Whatever the selection method, participation should be voluntary and based on informed consent.

Data form

• Quantitative: numerical scores, ratings or counts. Easy to summarise and compare; may lose detail about meanings and reasons.
• Qualitative: non-numerical descriptions. Rich detail and insight into reasons; slower to analyse and open to subjective interpretation.
• Categorising qualitative responses and counting them produces quantitative data.

Data source

• Primary: collected first-hand for the researcher’s investigation. Control over collection, but requires time and resources.
• Secondary: existing information collected by others. Saves collection time, but may be outdated or poorly matched to the aim.
• Either source can provide either data type.

Sampling and generalisation

Target population → sample → findings → possible generalisation