The field experiment by Savage and Armstrong (1990) investigated the effect of consulting style on patients' satisfaction using a random sample of 359 patients from one general practice in London as participants.

Mark scheme

Marking points from the official mark scheme (Feb/Mar 2018). The mark or mark type is shown in brackets.

1. Marks: 1 mark for basic answer e.g. identification. +1 mark for elaboration/example.
2. Most likely answer (other appropriate responses to be credited): participants would be unaware that they were involved in a study which would not be the case if a different style were to be used (i.e. repeated measures);
3. participants may not visit the surgery often so testing them twice may have been difficult;
4. records would not need to be kept about what patient did what condition (i.e. which they did first so the other style used second).
5. Marks: Question always requires two ‘things'
6. 1 mark for each correct ethical guideline that was maintained, twice.
7. 1 mark for each associated reason, twice.
8. Note: no marks for identifying a guideline.
9. Indicative content: Informed consent – patients should not be studied without their agreement. 'We obtained patients' consent for the study by giving all patients attending each surgery a written request to allow audio tape recording of the consultation as part of a research project. If the patient refused to take part or was ineligible for the above reasons the next eligible patient to consult was selected.'
10. Psychological harm – patients should leave a study in the same state as they were before starting - 'Patients were also excluded if it was thought that they would be upset by taking part in the study or would be incapable of completing a part of it.'
11. Marks: Question requires knowledge of alternative sampling techniques and how they can be applied to this study.
12. 1 mark for each correct technique identified, twice
13. 1 mark for applying each suggested technique to 'this study', twice
14. Most likely answer (other appropriate responses to be credited): Self-selecting sample: advertisement placed in newspaper or in doctors' surgery asking for participants;
15. Opportunity sample: asking people sitting in waiting room; asking people on the street to participate.
16. Snowball sample: asking people who have participated to ask their friends/people they know to be a participant.
17. Note: 0 marks for 'random sample' as that was used in the Savage and Armstrong study, given in the question stem.
18. Question always plural of each argument. Question always requires conclusion.
19. 1 mark for each for/against argument up to 4 max.
20. 1 mark for conclusion.
21. 0 marks for candidates who write about random allocation.
22. Note: If three (or more) arguments for one side, best two credited. If one argument only, max 2 marks.
23. Note: answers must be related to 'this study' as the question states.
24. Most likely answer (other appropriate responses to be credited):
25. Strengths participants cannot chose to be in the study (they do not volunteer);
26. participants automatically drawn from a wide age range (range was 16–75); and varying illnesses.
27. anomalies are likely to be random rather than consistent;
28. there is no researcher (doctor) bias in selecting participants.
29. Weaknesses participants may not give consent (unlike volunteers);
30. participants may have to be excluded (if present with life threatening problem or is for administrative/repeat prescription);
31. patients may drop out (34 failed to complete initial assessment; 110 failed to complete the second assessment).
32. Conclusion: (a decision reached by reasoning) any appropriate conclusion drawn from the discussion that has been presented.
33. Marks: 1 mark if appropriate. 0 marks if the conclusion repeats what has already been stated.