Quantitative and qualitative data - Research methods

AQA A-level Psychology: Revision Made Easy - Jean-Marc Lawton 2017

Quantitative and qualitative data
Research methods

(See also presentation and display of quantitative data, page 114.)


Quantitative data are numerical (occur as numbers), such as counting the number of times something happens, while qualitative data are non-numerical (occur in forms other than numbers), such as someone describing their feelings. Quantitative data tend to be objective, reliable and simple, while qualitative data tend to be subjective, less reliable and more detailed. The two forms of data can be combined to give deeper understanding.

Qualitative data give insight into feelings and thoughts, but analysis can be affected by researcher bias (the researcher’s interpretation of the data). However, qualitative data can be converted into quantitative data through content and thematic analysis.

Quantitative data are generally produced from experiments, observations, correlational studies, structured interviews and closed questions in questionnaires. Qualitative data are generally produced from case studies and from unstructured interviews and open questions in questionnaires.

Primary and secondary data

Primary data refers to original data collected specifically for a research aim that has not been published before.

Secondary data refers to data originally collected for another research aim that has been published before.

Primary data are more reliable and valid than secondary data, as the data have not been manipulated in any way. Secondary data drawn from several sources can help give a clearer insight into a research area than primary data can.


Meta-analysis is a process by which a large number of studies, involving the same research aim and research methods, are reviewed together, with the combined data statistically tested to assess the overall effect. For instance, Smith & Bond (1993) did a meta-analysis of 133 conformity studies using the Asch paradigm to assess conformity levels in different cultures. As meta-analyses use data combined from many studies, they allow identification of trends and relationships not possible with individual studies. Meta-analyses are helpful when individual studies find contradictory or weak results, as they give a clearer overall picture.

Content analysis

Content analysis (CA) is a method of turning qualitative data, such as written, verbal and visual information, into quantitative data. Coding units are used to categorise the material being analysed, such as the number of times positive words occur in a written description. Coding units can involve words, themes, characters and time and space.

Strengths of content analysis

Image Ease of application: CA is easy to do, inexpensive and does not require interaction with participants.

Image Complements other methods: CA can be used to verify results using other research methods and is especially useful as a longitudinal tool to detect trends (changes over time).

Image Reliable: it is easy to replicate, meaning checking reliability is simple.

Weaknesses of content analysis

Image Descriptive: being purely descriptive, CA does not reveal underlying reasons for behaviour, attitudes, etc.

Image Flawed results — as CA is limited to available material, observed trends may not reflect reality.

Image Lack of causality — as CA is not performed under controlled conditions, it does not show causality.


Thematic analysis (TA) is a means of identifying, analysing and reporting themes (patterns) within qualitative data, with themes being identified through data coding. TA goes beyond merely counting words and phrases and instead seeks to identify ideas within data. TA can involve comparing themes, identifying co-occurrences of themes and using graphs to show differences between themes.

Identified themes become categories for analysis, with TA performed by a process of coding involving six stages:

1 Familiarisation with data — involves intensely inspecting the data to become immersed in their content.

2 Coding — codes (labels) are generated that identify features of the data important to answering the research question.

3 Searching for themes — codes and data are inspected to identify patterns of meaning (potential themes).

4 Reviewing themes — potential themes are checked against the data to see whether they explain the data and fit the research aim. Themes are refined, which can involve splitting, combining or discarding them.

5 Defining and naming themes — each theme is intensively analysed so that it can be given an informative name.

6 Writing up — information gained from the TA is combined.