Interviews

Question 1

Structured Interview

Answer 1

Predetermined questions delivered by an interviewer who does not probe beyond the answers received.

Question 2

Semi-Structured

Answer 2

Some Qs are predetermined but new ones are developed during the interview.

Question 3

Unstructured

Answer 3

No questions decided in advance.

Question 4

+ and - of Structured

Answer 4

+ Easily repeated.

+ Easy to analyse as answers are predictable.

• Interviewers’ expectations may influence the interviewee.

Question 5

+ and - of Semi and Unstructured

Answer 5

+ More detail can be obtained from each respondent as they are specially shaped to them.

+ Can access info that may not be revealed by predetermined Qs.

• Requires well trained interviewers, which may be difficult to obtain and makes research more expensive.

Question 6

Open Question

Answer 6

Respondents choose their own answers.

+ Provides rich details on how people behave as they can freely express themselves.

• Difficult to draw conclusions as there are likely to be a wide range of answers, researcher may look for patterns instead.

Question 7

Closed Question

Answer 7

Fixed number of possible answers.

+ More objective as more likely to be interpreted in same way.

• Oversimplifies reality and human experience as it suggests there are simple answers.

Question 8

Ranked Scale

Answer 8

Scale such as 1-5.

+ Objective way to represent feelings and attitudes.

• PP may respond in the same way to all questions regardless of context.

Question 9

Vallentine Interview aim

Answer 9

Assess educational materials for groupwork with offenders in high-security psychiatric hospital.

Question 10

Vallentine procedure

Answer 10

42 detained male patients 80% had SZ or delusions.
4 x 20 session groups over 3 years.

2 outcome measures before and after:
CORE-OM assessed wellbeing, symptoms, social life and risk to others.
SCQ is 30 question questionnaire measuring self esteem. Has high test-retest reliability and good clinical validity.

Semi structured interview to evaluate experience, analysed using content analysis.

Question 11

Vallentine Findings

Answer 11

Inferential statistics showed that there was no significant difference between any pre and post tests between the groups.

Changes in CORE-OM but only one participant showed a reliable change. More reliability in SCQ but a negative shift.

Interview data showed what the participants valued and why, what was helpful about the group, clinical implications identified by pats, what was unhelpful.

Question 12

Vallentine Conclusion

Answer 12

Attention needed to absence of reliable changes and a further consideration of negative changes in some pats.

Qualitative analysis of interviews showed pats did value sense of hope and empowerment provided.