Overview of Research Methodology
Research methodology covers the decisions psychologists make before, during, and after a study: which method to use, who to include, how to analyse the data, and how to ensure the findings can be trusted. These decisions are not independent of each other. A sampling choice affects generalisability; a measurement choice affects what statistical test is appropriate; an ethical constraint can determine which designs are even possible.
This subtopic treats methodology as a connected set of choices, not a checklist. The sections below follow the logic of a research project: first, understanding what the data can tell you; then the methods used to collect it; then the people included; then the standards used to judge whether the whole thing was done well.
Quantitative Data Analysis
Data analysis is the process of examining collected data in order to draw meaningful conclusions about it.
Data analysis in psychology divides into two broad categories: descriptive statistics and inferential statistics. The distinction matters because they answer different questions.
Descriptive statistics summarise what the data actually look like. The main tools are:
- Measures of central tendency (mean, median, mode): these describe the typical value in a dataset.
- Measures of dispersion (range, standard deviation, semi-interquartile range): these describe how spread out the values are around that typical value.
The mean is sensitive to extreme scores; the median is more robust when outliers are present. Standard deviation tells you how far scores cluster around the mean, and the semi-interquartile range does the same job when using the median.
Inferential statistics go further: they allow researchers to draw conclusions beyond the immediate sample and to test whether an observed difference or relationship is likely to reflect a real effect in the wider population, or whether it could have occurred by chance. This is hypothesis testing. Examples of inferential tests include t-tests, chi-square tests, Mann-Whitney tests, and Wilcoxon tests. Each is suited to a particular research design and type of data.
A result that is statistically significant means it is unlikely to have occurred by chance alone, usually at a threshold of .
A p-value of 0.05 means that, if the null hypothesis were true, results at least as extreme as these would occur 5% of the time. If , the evidence is insufficient to reject the null hypothesis. If , the null hypothesis is rejected.
Effect size is a separate but important concept. Statistical significance tells you whether a result is likely to be real; effect size tells you how large that effect is in practical terms. A study with thousands of participants can produce a statistically significant result with a trivially small effect size. Both pieces of information are needed to evaluate a finding.