Case Studies, Correlations and Types of Data
Case Studies — Studying the Unique
A case study is a detailed investigation into one person, group or event, rather than a large sample. Psychologists use case studies when a case is too rare, complex or extreme to study through a typical experiment — for example, a person with an unusual brain injury, or a single family coping with a specific disorder. Because only one case is studied, researchers can build a rich, detailed picture rather than a broad but shallow one.
Case studies rely on qualitative data gathered through interviews, observation and documents, and they always use a small sample — often just one person or one group. This is the trade-off at the heart of the method: depth of understanding in exchange for a tiny, unrepresentative sample that may never occur again in quite the same way.
Case studies use small samples and produce qualitative data. This gives huge depth of insight into one case, but because the sample is so small — and often unusual by definition — findings cannot be generalised with confidence to the wider population.
Strength: case studies allow researchers to study situations that would be impossible or unethical to create experimentally, such as the effects of extreme early neglect. They can also generate new hypotheses for later research and capture rich, real-world detail that a controlled experiment would miss entirely.
Weakness: because the sample size is one, or a very small group, findings cannot be generalised with confidence to other people. The researcher may also grow close to the person being studied over time, which risks a subjective or biased interpretation of what they observe and record.
Correlations — Measuring Association
A correlation measures the relationship between two co-variables, using quantitative data. Rather than manipulating an independent variable as in an experiment, researchers measure two variables as they naturally occur and check whether they rise and fall together in a predictable pattern.
A correlation can take one of three forms. In a positive correlation, both variables increase together, such as hours of revision and exam score. In a negative correlation, one variable increases as the other decreases, such as hours of sleep and levels of daytime tiredness. In a zero correlation, there is no consistent pattern between the two variables at all.
Correlation shows association, not causation. Two variables can be strongly correlated because a third, unmeasured variable affects both of them — this is the third variable problem. Ice cream sales and drowning rates both rise in summer, but ice cream does not cause drowning; hot weather affects both.
Correlations are displayed on scatter diagrams, where each participant contributes one dot showing their score on both variables. A positive correlation produces dots trending from bottom-left to top-right; a negative correlation trends from top-left to bottom-right; a zero correlation produces a scattered cloud of dots with no clear pattern in either direction.
Strength: correlations let psychologists study the relationship between variables that could never be manipulated ethically or practically, such as stress and illness, and can point towards patterns worth testing later with a controlled experiment. Weakness: because neither variable is manipulated, a correlation can never establish that one variable causes the other — this is its most important limitation.
Types of Data
Research methods generate different kinds of data. Quantitative data is numerical — scores, frequencies, times — and is easy to analyse statistically and compare across groups. Qualitative data, introduced above through case studies, is expressed in words and captures detail and meaning that numbers alone cannot show.
Data is also classified by where it comes from. Primary data is collected first-hand by the researcher for their own study — for example, results from an experiment they ran themselves. Secondary data already existed before the study began, such as official crime statistics or hospital records, and is reused by a researcher for a new purpose.
Quantitative data is numerical and easy to analyse statistically. Qualitative data is descriptive and captures meaning. Primary data is collected first-hand by the researcher; secondary data already existed and is reused. Quantitative data allows comparison, qualitative data allows depth, primary data is tailored to the study, and secondary data saves time and resources.