All subjects · Psychology · WPS04 Clinical Psychology and Psychological Skills

Decision making and interpreting clinical data

Edexcel IAL Psychology · WPS04 Clinical Psychology and Psychological Skills

Spec 8.3.4 asks you to use List A (descriptive) and List B (inferential) statistics on clinical data, and 8.3.5 to evaluate clinical research. This lesson applies the three Pearson tests to clinical examples. Formulae and critical values tables are given in the exam.

Practise this in the app: instant marking and a fix for every mistake. Free, no email needed to start.

Practise this lesson

Choosing the test

Aim	Design / data	Test
Difference	Related scores (repeated measures or matched pairs), at least ordinal	Wilcoxon signed ranks (T must be ≤ critical value)
Relationship	Two scores from each person, at least ordinal	Spearman's rank (rs must be ≥ critical value)
Difference or association	Independent groups, nominal (frequencies in categories)	Chi-squared (χ² must be ≥ critical value; df = (rows - 1)(columns - 1))

Wilcoxon: depression before and after CBT

Practice data (invented). Depression scores for 10 clients before and after CBT.

Before: 32, 28, 35, 30, 26, 33, 29, 31, 27, 34

After: 24, 27, 30, 30, 20, 25, 26, 22, 28, 27

Differences (before - after): 8, 1, 5, 0, 6, 8, 3, 9, -1, 7. One zero is dropped, so N = 9.

Ranks of the absolute differences: 7.5, 1.5, 4, 5, 7.5, 3, 9, 1.5, 6.

Sum of positive ranks = 43.5, sum of negative ranks = 1.5, so T = 1.5.

Critical values for N = 9, one-tailed: p ≤ .05 is 8, p ≤ .01 is 3. T = 1.5 is ≤ both, so scores fell significantly (p ≤ .01).

Spearman: criticism at home and symptom severity

Practice data (invented). Critical comments from relatives (an expressed emotion measure) and symptom severity for 10 patients.

Comments: 2, 9, 4, 12, 7, 1, 10, 5, 8, 3

Severity: 10, 18, 14, 22, 13, 9, 20, 11, 19, 12

Σd² = 12; rs = 1 - 6Σd² / (n(n² - 1)) = 1 - 6 x 12 / (10 x 99) = 0.927.

Critical value (N = 10, one-tailed p ≤ .05) = 0.564. 0.927 ≥ 0.564, so there is a significant positive correlation: but it does not show that criticism causes severity.

Evaluating clinical research (8.3.5)

Judge clinical studies on reliability, validity, generalisability, credibility, objectivity and subjectivity, ethics, and practical application. Clinical research has special ethical issues: participants may be vulnerable, consent may be affected by illness, and withholding treatment from controls can cause harm.

Criterion	Question to ask
Validity	Does it measure what it claims? Is it true to real life?
Reliability	Would the same result be found again, or by another clinician?
Generalisability	Does the sample represent the people the claim is about?
Credibility	Is it supported by other evidence and trusted?
Objectivity / subjectivity	Is it free from the researcher's views, or does it rely on interpretation?
Ethics and application	Was it ethical, and how does it help real patients?

Worked example

Before: [18, 22, 15, 25, 20, 17, 23, 19]; after: [15, 22, 16, 18, 14, 17, 20, 12]. Calculate the Wilcoxon T and decide whether the reduction is significant at p ≤ .05 (one-tailed).

Differences: 3, 0, -1, 7, 6, 0, 3, 7. Drop zeros: N = 6.
Ranks of absolute differences: 2.5, 1, 5.5, 4, 2.5, 5.5.
Positive total = 20, negative total = 1, so T = 1.
Critical value (N = 6, one-tailed p ≤ .05) = 2. 1 ≤ 2, so significant.

Answer: T = 1, N = 6: significant at p ≤ .05 one-tailed.

Key idea: Difference + related + ordinal = Wilcoxon (T ≤ critical). Relationship = Spearman (rs ≥ critical). Independent nominal frequencies = chi-squared (χ² ≥ critical, df = (r - 1)(c - 1)). State observed and critical values and the significance level; correlation is not causation.

Check yourself

Using the before and after CBT data in step 2, what is the Wilcoxon T value? Give your answer to 1 decimal place.
Show answer
1.5. Positive ranks total 43.5, negative ranks total 1.5; T = the smaller = 1.5.
Using the critical comments and severity data, calculate Spearman's rs to 3 decimal places.
Show answer
0.927. Σd² = 12; rs = 1 - 6 x 12 / 990 = 0.927.
For the worked example data (before [18, 22, 15, 25, 20, 17, 23, 19], after [15, 22, 16, 18, 14, 17, 20, 12]), what is N?
Show answer
6. One zero difference is dropped, so N = 6.
A researcher compares the number of patients who relapse (yes/no) after family therapy or drug therapy alone, in two separate groups. Which test should be used?
- Chi-squared
- Wilcoxon
- Spearman's rank
- No test is possible
Show answer
Chi-squared. Independent groups and nominal (relapse yes/no) data: chi-squared.
Patients rate their anxiety before and after an fMRI scan on a 1 to 10 scale. Which test should be used?
- Chi-squared
- Spearman's rank
- Wilcoxon
- A thematic analysis
Show answer
Wilcoxon. Repeated measures, ordinal ratings, testing a difference.
What is the critical value of rs for N = 12 at p ≤ .05, one-tailed, from Pearson's table?
Show answer
0.503. N = 12, one-tailed .05: 0.503.
A study finds a significant positive correlation between critical comments from relatives and symptom severity. Explain why this does not show criticism causes worse symptoms.
Show answer
It is a correlation, so it cannot show cause and effect; more severe symptoms might lead relatives to criticise more, or a third variable such as stress could affect both.. Correlation lacks direction and may involve a third variable.

Practise with instant feedback

Sources

Last updated 8 November 2026. Spotted a mistake? Tell us.