2Research methods
Research methods and statistics for DClinPsy selection
The 150 questions cover statistics, research design, qualitative methods and ethics across 10 topics. Every answer is explained.
A sample question
Scores on a symptom questionnaire are moderately positively skewed in the population. A researcher draws many random samples of 100 people and records each sample’s mean. What does the central limit theorem predict?
Answer and explanation
The sample means will be roughly normally distributed
The central limit theorem concerns the distribution of sample means: as sample size grows, it gets closer to normal whatever the population’s shape, provided the variance is finite, as it is for a bounded questionnaire score. With 100 per sample and moderate skew, the means will be close to normal and far less skewed than the scores. The scores within each sample still mirror the skewed population, however many are collected. The means vary much less than the scores: their spread is the standard error, SD/√n, here a tenth of the SD. They centre on the population mean, which in skewed data usually differs from the median. How large n must be depends on the skew; 'about 30' is a rule of thumb, not part of the theorem.
Source: Field (2024), Discovering Statistics Using IBM SPSS Statistics, sixth edition. SAGE, checked .
Revise by topic
- Foundation
- what the other topics build on
- Core
- the main methods, from testing to ethics
- Advanced
- reading results and judging study designs
- Describing data
- Distributions and sampling
- Hypothesis testing and p-values
- Choosing the right test
- Effect size and confidence intervals
- Reliability and validity
- Qualitative methods
- Research ethics
- Interpreting results
- Research design and critical appraisal
Try a sample question
Scores on a symptom questionnaire are moderately positively skewed in the population. A researcher draws many random samples of 100 people and records each sample’s mean. What does the central limit theorem predict?
Answer and explanation
The sample means will be roughly normally distributed
The central limit theorem concerns the distribution of sample means: as sample size grows, it gets closer to normal whatever the population’s shape, provided the variance is finite, as it is for a bounded questionnaire score. With 100 per sample and moderate skew, the means will be close to normal and far less skewed than the scores. The scores within each sample still mirror the skewed population, however many are collected. The means vary much less than the scores: their spread is the standard error, SD/√n, here a tenth of the SD. They centre on the population mean, which in skewed data usually differs from the median. How large n must be depends on the skew; 'about 30' is a rule of thumb, not part of the theorem.
Source: Field (2024), Discovering Statistics Using IBM SPSS Statistics, sixth edition. SAGE, checked .
Critique and the research test
What a research test can cover
Salomons, which arranges its tests with UEL, describes a 2026 research multiple-choice test that “may include a range of questions relating to statistical issues, qualitative and quantitative research methods, and ethics”. It says its 2027 process is under review. The questions here are written for this site.