Practical investigation: a questionnaire study

Planning

Example aim (from Pearson): is there a difference in self-reported obedience between males and females?

• Hypothesis: 'There will be a difference in self-reported obedience scores between males and females' (non-directional).
• Sample: at least 10 per group; justify the technique (opportunity is acceptable only with a reason beyond convenience).
• Questionnaire: record each person's group; standardised written instructions (for example, complete alone); ethical statement covering consent, right to withdraw and confidentiality; closed Likert items (strongly agree to strongly disagree, plus don't know) scored to give a total; open questions for qualitative data.
• Ethics: no distressing scenarios; under-16s need parental consent; debrief afterwards.

Analysis

Quantitative: summary table with mean, median, mode, range and standard deviation for each group; frequency table; bar chart of the means. Example totals out of 40:

• Males: mean 30.5, median 30.5, range 9, SD 3.03
• Females: mean 34.5, median 34.5, range 9, SD 3.03

Qualitative: thematic analysis of the open answers, with quotes for each theme.

Example data invented for practice: total self-reported obedience scores out of 40, 10 per group.

Write-up and evaluation

Write the procedure (enough detail to replicate), results (tables, graphs, themes) and discussion (link results to theory such as agency theory and to research such as Milgram's Experiment 8 and Burger (2009), which found no gender difference).

Strengths of questionnaires: quick, many participants, standardised and replicable, anonymous (reduces social desirability).

Weaknesses: self-report may not match real behaviour (low validity); social desirability; leading questions; Likert items force choices.

Improvements: pilot the questionnaire, use a larger stratified sample, include reverse-scored items, use two coders for thematic analysis.

Worked example

Using the example data, describe the difference between the groups.

1. Females had a higher mean (34.5) than males (30.5).
2. The spreads were similar (SD 3.03 and 3.03).
3. This is descriptive only; an inferential test would be needed to say whether the difference is significant.

Answer: Females scored higher on average (34.5 against 30.5) with similar spread.

Check yourself

1. Calculate the mean of the female example scores: 36, 33, 38, 31, 35, 37, 34, 30, 39, 32.
   Show answer

   34.5. 345 / 10 = 34.5.

2. Calculate the standard deviation (n - 1) of the male example scores 32, 28, 35, 30, 27, 33, 29, 31, 34, 26. Give your answer to 2 decimal places.
   Show answer

   3.03. Mean 30.5; squared deviations sum to 82.5; divide by 9 and square root: 3.03.

3. Why should a 'don't know' option be included in Likert items?
   • So the data become qualitative
   • So participants are not forced into an answer that does not reflect their view
   • To make the questionnaire longer
   • To increase social desirability
   Show answer

   So participants are not forced into an answer that does not reflect their view. Forcing a choice reduces validity.

4. Which graph best compares the mean obedience scores of the two groups?
   • A histogram
   • A scatter diagram
   • A bar chart
   • A pie chart of the raw scores
   Show answer

   A bar chart. Two separate categories: bar chart of means.

5. Which statement should appear at the start of the questionnaire?
   • You may withdraw at any time and your answers will be kept confidential
   • Your name will be published with your answers
   • You must complete every question
   • The true aim will never be revealed
   Show answer

   You may withdraw at any time and your answers will be kept confidential. Right to withdraw and confidentiality are required.

6. Give one weakness of using a questionnaire to measure obedience.
   Show answer

   People may give socially desirable answers, saying they would disobey unjust orders, so what they report may not match how they would really behave.. Self-report may differ from real behaviour.

Practise with instant feedback Print this lesson

Sources

• Pearson Edexcel International Advanced Subsidiary/Advanced Level in Psychology specification (issue January 2026)
• Pearson International A Level Psychology component guide: WPS01
• IAL Psychology revision site: Practical (social)

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