1.17 Year 2 Research Methods - Reliability

Question 1

Reliability

Answer 1

How consistent finding from an investigation is

Question 2

Reliability of experiments

Answer 2

High due to high levels of control over CVs and EVs. Easier to replicate due to use of standardised procedures allowing cause and effect to be established

Question 3

Reliability of correlations

Answer 3

High due to quantitative data that is measured and statistically analysed

Question 4

Reliability of self-report

Answer 4

Variable, structured interviews are easily replicable so thus reliable. Unstructured are not easily replicated due to use of different questions so less reliable. Varies further depending on use of open or closed questionnaires.

Question 5

Reliability of Observations

Answer 5

Variable, reliable when coding is used but when observing can be open to interpretation so subjective making them less reliable. Controlled vs natural observation also has a role in this

Question 6

Reliability of Content Analysis

Answer 6

Quantitative CA is not open to interpretation so more reliable and easier to replicate in terms of coding. Qualitative CA is open to interpretation, can’t always be replicated so less reliable

Question 7

Reliability of case studies

Answer 7

Low as they are too subjective and open to interpretation. Affected researcher bias e.g. loss of insight and becoming invested in the study due to longitudinal nature so less reliable and very hard to replicate

Question 8

Test-retest method

Answer 8

Used to check research methods e.g. questionnaires level of reliability. Tools used in a test are the same but given at different times, if they are reliable then the results should be the same or similar each time someone else uses them. Timing has to be perfect to avoid participants remembering the answers of the test. In short, same tools, same people, different occasions – 80% similarity or more needed for reliability.

Question 9

Inter-rater/observer reliability

Answer 9

Used in observations/content analysis/interviews. For example with transcripts, these are open to interpretation and hold subjectivity bias. Often to increase the reliability, more than one will take part in the observation, observers will watch the same event multiple times via a recording and score independently before comparing

Question 10

Improving validity in questionnaires

Answer 10

• Test re-test method
• Re-Write questions
• Replace open with closed questions

Question 11

Improving validity in experimental methods

Answer 11

• Standardised procedures to allow replicability

- Ensure high control over the EV

Question 12

Improving validity in interviews

Answer 12

• Use same interviewer
• Ensure correct training given to interviewer, increases reliability
• Use structured interviews

Question 13

Improving validity in observations

Answer 13

• Pre-agree categories
• Controlled observations
• Standardised Procedures