Appendix 4: Formulae and statistical tables - Critical values for Spearman’s rank — Edexcel A-Level Psychology

In Pearson Edexcel A-Level Psychology, the critical values table for Spearman's rank correlation coefficient (Appendix 4) is used to determine whether an observed correlation between two co-variables is statistically significant. To achieve statistical significance, the calculated value of rs must be equal to or greater than the critical value at the specified alpha level (typically p ≤ 0.05) for sample size N.

Appendix 4 of the Pearson Edexcel A-Level Psychology specification provides the formulae and statistical tables required for psychological research methods. The critical values table for Spearman's rank correlation coefficient enables students to evaluate whether an observed monotonic relationship between two co-variables is meaningful or merely a product of chance variation.

Mastering the use of this statistical table is essential for Paper 1, Paper 2, and Paper 3 questions that evaluate empirical data from practical investigations. Students must confidently determine the number of participants or pairs (N), establish the level of significance (alpha), correctly identify directional or non-directional criteria, and apply the correct decision rule to accept or reject hypotheses.

• →Level of significance (p-value): The threshold probability (commonly p ≤ 0.05 in psychology) that observed results occurred by chance alone.
• →Sample size (N): In Spearman's rank, N represents the total number of paired observations or participants, not degrees of freedom.
• →One-tailed vs. two-tailed tests: Directional hypotheses necessitate a one-tailed critical value column, whereas non-directional hypotheses require a two-tailed column.
• →Decision rule for Spearman's: The calculated value of rs must be greater than or equal to the critical value to reject the null hypothesis and declare statistical significance.

• Name the test and justify it: a test of correlation, ordinal data or data that can be ranked, and two scores from each participant rather than two separate groups of people.
• Show that the coefficient is compared as a magnitude, so the minus sign is set aside for the table check but kept in the conclusion, where it means high scores on one measure go with low scores on the other.
• Read the table with the number of pairs of participants and the significance level, and say whether a one tailed or two tailed column is being used and why the hypothesis makes it so.
• Finish with the decision: because the coefficient reaches or passes the tabled figure at five per cent, the null hypothesis is rejected and the relationship is significant, although a correlation cannot establish cause.

• 💡Calculation questions supply the summed squared rank differences and the number of pairs, so the marks sit in substituting correctly and in remembering to subtract the fraction from one.
• 💡Explain questions often print an extract from the table with several rows and both tail columns, and the first mark is for choosing the right cell before any comparison is made.
• 💡Scattergram questions pair naturally with this, so be ready to describe the direction and strength shown in the plot and then tie that to the sign and size of the coefficient.
• 💡A significant correlation invites the third variable point and the reminder that causation is not shown, which is the evaluation examiners expect alongside the arithmetic.
• 💡Always state all four comparative elements in your written answer: the calculated value, the critical value, the sample size (N), and the significance level (e.g., p ≤ 0.05) alongside whether it is one- or two-tailed.
• 💡Check the wording of the stem carefully to determine if the hypothesis is directional or non-directional before looking at the columns in Appendix 4.
• 💡When explaining conclusions, always contextualise your answer by explicitly naming the co-variables from the scenario rather than just stating 'the result is significant'.

• Ranking both variables together in one long list instead of ranking each variable separately, which produces rank differences that carry no meaning.
• Comparing a negative coefficient as a negative number and concluding it falls short, when a coefficient of minus 0.79 is stronger than a tabled figure of 0.564.
• Using degrees of freedom or the number of individual scores instead of the number of pairs when reading the table, so the wrong row is used throughout.
• Borrowing the small is significant rule from Wilcoxon, so a coefficient that clears the table is reported as non-significant.
• Misidentifying N as degrees of freedom (N - 2): Unlike Pearson's r or a t-test which use degrees of freedom (df), Spearman's rank critical value tables in Pearson Edexcel directly index sample size by N (pairs of scores).
• Ignoring the negative sign in negative correlations: When comparing a negative calculated rs against the critical table value, students must compare the absolute value |rs|, because critical value tables display unsigned magnitudes.
• Applying the 'calculated must be less than critical' rule: Students confuse tests where significance requires calculated ≤ critical (e.g., Mann-Whitney U, Wilcoxon T, Sign Test) with Spearman's rank where calculated ≥ critical.

1. 1Day 1-2: Review correlation fundamentals, including ranking raw data, calculating rs, and matching directional/non-directional hypotheses to tails.
2. 2Day 3-4: Practice locating critical values in Appendix 4 tables across various sample sizes (N) and alpha thresholds (p = 0.05 and p = 0.01).
3. 3Day 5-6: Complete past paper questions focusing on structured justification paragraphs that compare calculated vs. critical values.
4. 4Day 7: Timed mock exam questions assessing statistical test choice, hypothesis writing, and interpretation of Appendix 4 tables.

• 📋Data analysis and interpretation questions: Short 2-4 mark questions providing a calculated rs value and requiring you to look up the critical value in an extract and state whether the null hypothesis should be rejected.
• 📋Research design evaluation questions: Longer 6-8 mark questions asking you to justify the choice of Spearman's rank and interpret calculated data in relation to an apprentice or contemporary study scenario.
• 📋Methodological calculation steps: Questions assessing knowledge of ranking paired scores and checking statistical significance using critical values.