Biological practical: a correlational study

Design decisions

• Co-variables, operationalised: e.g. 'age in years' and 'self-rated aggressive tendencies on a 1 to 10 scale', or 'minutes on social media after 9pm' and 'hours of sleep last night'.
• Hypothesis: directional ('there will be a negative correlation between...') or non-directional ('there will be a correlation between...'); plus a null hypothesis ('there will be no correlation...').
• Sampling: e.g. opportunity or volunteer sampling of students.
• Controls: standardised instructions, same questionnaire, same time of day, private answers.
• Ethics (BPS Code of Ethics and Conduct, 2009): informed consent, right to withdraw, confidentiality, protection from harm, debriefing. Be careful with aggression questions: participants should not feel labelled.

Analysis

• Descriptive statistics: mean, median, mode, range and SD for each co-variable.
• Scatter diagram: describe the direction and strength of the pattern.
• Spearman's rank: rank each co-variable, find Σd², calculate rs, compare with the critical value (N, one- or two-tailed, p ≤ .05), and state whether to reject the null hypothesis.
• Explain the correlation coefficient: its sign (direction) and size (strength).

Discussion and write-up

Write up procedure, results and discussion:

• Relate results to theory (e.g. melatonin and screen light; hormones and aggression).
• Strengths: easy to replicate with standardised questions; ethical way to study variables that cannot be manipulated.
• Weaknesses: self-report may be inaccurate or socially desirable; no cause and effect; small opportunity sample limits generalisation.
• Improvements: say how and why, e.g. 'use a sleep diary over 7 nights instead of one estimate, to increase validity'. Examiners say a bigger sample alone is not enough unless it is more varied.

Worked example

A class practical with N = 12 predicts a positive correlation between daily minutes on social media and self-rated aggressive tendencies, and finds Σd² = 112. Calculate rs and decide significance for a one-tailed test at p ≤ .05.

1. rs = 1 - (6 x 112) / (12 x 143) = 1 - 672 / 1716 = 0.608.
2. Critical value (N = 12, one-tailed, p ≤ .05) = 0.503.
3. 0.608 ≥ 0.503, so the result is significant.

Answer: rs = 0.608; significant at p ≤ .05 (one-tailed). The sign is positive, as predicted.

Check yourself

1. Which is the best operationalised co-variable?
   • Number of hours slept last night, self-reported to the nearest half hour
   • How tired someone is
   • Sleep
   • Whether someone is a good sleeper
   Show answer

   Number of hours slept last night, self-reported to the nearest half hour. It states exactly how the variable is measured.

2. Which is a null hypothesis for this practical?
   • There will be a negative correlation between screen time and sleep
   • Screen time causes poor sleep
   • There will be no correlation between evening screen time and hours of sleep
   • Students sleep less than adults
   Show answer

   There will be no correlation between evening screen time and hours of sleep. The null predicts no relationship.

3. N = 12, Σd² = 112. Calculate rs to 3 decimal places.
   Show answer

   0.608. 1 - 672 / 1716 = 0.608.

4. Which ethical step protects participants answering questions about their own aggression?
   • Not telling them the aim at all
   • Publishing their names with their scores
   • Making participation compulsory
   • Keeping answers anonymous and confidential
   Show answer

   Keeping answers anonymous and confidential. Confidentiality protects them from embarrassment or labelling.

5. Which improvement is best justified?
   • Use more participants
   • Use a sleep diary over 7 nights to give a more valid measure of typical sleep
   • Do it again
   • Make the questionnaire longer
   Show answer

   Use a sleep diary over 7 nights to give a more valid measure of typical sleep. It says what to change and why it improves validity.

6. Explain one weakness of using a correlational design for a study of screen time and sleep.
   Show answer

   A correlation cannot show cause and effect, so we cannot tell whether screen time reduces sleep or whether people who cannot sleep use screens more, or whether another variable such as stress affects both.. Cause and effect.

Practise with instant feedback Print this lesson

Sources

• Pearson Edexcel International Advanced Subsidiary/Advanced Level in Psychology specification, Issue 5 (February 2021)
• Pearson IAL Psychology component guide: WPS02 (Issue 2, November 2024)
• British Psychological Society, Code of Ethics and Conduct (August 2009)
• Pearson WPS03/01 question paper, January 2022 (formulae and critical values tables)
• Pearson WPS02/01 examiners' report, June 2025

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