Research Methods Year 2 (not finished)

Question 1

Define ‘case study’

Answer 1

The detailed and in-depth study of a case of an individual, group or organisation.

Question 2

What is triangulation?

Answer 2

When several data sets/pieces of research are analysed to reach a conclusion
- Increases validity and reliability

Question 3

Give two strengths of using case studies

Answer 3

• Longitudinal studies allow a greater depth of information to be collected and it is possible to see change over time for a particular variable
• Allows a researcher to investigate rare or unusual behaviours
• Lack of representation is not an issue as it is designed for an individualistic approach
• Several research methods can be used within one case study, meaning data can be both qualitative and quantitative - increased validity and reliability (linked to triangulation)
• High ecological validity as most are carried out in ‘real’ environments

Question 4

Give two weaknesses of using case studies

Answer 4

• Time consuming and expensive
• Not representative or generalisable - contradicts the nomothetic approach of science
• Longitudinal studies usually lead to high drop-out rates
• Researcher Bias - ppts could pick up on the aims of the study (if they were hidden) and display demand characteristics or social desirability
• Volunteer Bias - ppts may try to manipulate the researcher, especially if they had built up a rapport over time
• Could be predeterminist?

Question 5

Give one example of a case study

Answer 5

• Freud’s Case Study of Little Hans - Believed that Little Hans’ fear of horses was actually the displaced fear of his father created by castration anxiety.
• Thigpen and Cleckley’s Case Study of Christine Sizemore - Study of Sizemore’s DID, with the analysis of Eve White, Eve Black and Jane using various methods including interviews, psychometric testing, EEGs and observations

Question 6

Who postulated the principle of falsification?

Answer 6

Karl Popper

Question 7

What is the principle of falsification?

Answer 7

• A hypothesis/theory is only scientific if it can be proven false
• Scientists should actively seek data to ‘test a theory to destruction’ rather than simply data that proves it as it is easier to falsify than verify
• If something cannot be falsified, it is a pseudo-science e.g. Freud’ Psychodynamics
• Popper uses the example of swans: the statement “all swans are white” would be almost impossible to verify, but very easy to falsify by finding a single black swan

Question 8

Which scholar developed the concept of scientific paradigms?

Answer 8

Thomas Kuhn

Question 9

According to Kuhn, what is the difference between physical sciences and social sciences?

Answer 9

• Physical sciences have a single unifying paradigm that is able to shift based on new evidence
• Social sciences have several contradictory paradigms, so there is not a single theory to test new evidence against - this suggests that psychology is unscientific

Question 10

Is a nomothetic approach or an idiographic approach considered more scientific? Why?

Answer 10

A nomothetic approach is more scientific
- It seeks to create generalisable laws, meaning it has predictive power (i.e. it can make behaviour predicable)
- Idiographic approaches focus on individual experience, meaning they are not generalisable and have limited predictive power

Question 11

According to Kuhn, what are three things a discipline must have to be classed as a science?
Does psychology have them all?

Answer 11

1. Shared set of assumptions - No: Psychology has several approaches, each with their own assumptions about behaviour
2. Shared set of scientific methods - Yes: Psychology uses the scientific method to systematically construct and test hypotheses (e.g. experiments)
3. Shared set of specialist terminology - Yes: For example, independent and dependent variables are used in all sciences

Question 12

Why can ethical considerations impede objectivity in research?

Answer 12

• To analyse variables in a completely objective way, they must be isolated from as many extraneous variables as possible.
• As psychology is the study of the HUMAN mind, this is almost impossible to do in an ethical way. For example, you cannot separate MZ twins purely to research the nature/nurture debate.
• Animals have been used to mitigate some of the human ethical issues, but animal ethics still apply, and you can’t necessarily generalise animal findings to humans.

Question 13

What does operationalisation mean, and why is it important?

Answer 13

• Operationalisation means making variables measurable e.g. no. of words recalled, seconds, metres
• It is important because operationalising variables makes them testable, which is a cornerstone of the scientific method

Question 14

What is the significance level typically used in psychology and why?

Answer 14

The 5% level
- There must only be a 5% chance that the results of a test are down to coincidence
- Seen as a balance between Type I and Type II Errors

Question 15

What is a Type I Error and how may they occur?

Answer 15

A false positive result - not being cautious enough
- Usually caused by using a significance level that is too lenient, meaning the null is wrongfully rejected

Question 16

What is a Type II Error and how may they occur?

Answer 16

A false negative result - being too cautious
- Usually caused by a significance level that is too strict, meaning the null is wrongfully accepted.

Question 17

What is nominal data?

Answer 17

Data is organised into separate named categories
- Usually applies to frequency

Question 18

What is ordinal data?

Answer 18

Data that can be ordered or ranked in some way
- Usually involves an arbitrary scale ( e.g. Rate X from 1-10)

Question 19

What is interval data, and how does it differ from ratio data?

Answer 19

Interval data looks at the dispersion (gaps) between data points
- The scale used has equal and meaningful intervals e.g. a percentage score

The difference between ratio and interval data is that the scales used for ratio data are not only equally spaced, but have a ‘true zero’, such as with seconds or kilograms.

Question 20

Does content analysis generate quantitative or qualitative data?

Answer 20

Quantitative data

Question 21

Does thematic analysis generate quantitative or qualitative data?

Answer 21

Qualitative data

Question 22

What does the process of content analysis entail?

Answer 22

• Carried out on secondary data such as newspapers, books, advertisements etc
• Certain relevant themes/words/images are coded and then counted by the researcher
• The data is then analysed and conclusions are drawn about what the frequency, relationship, meanings etc. of themes mean for that piece of media.

Question 23

Give two strengths of content analysis

Answer 23

• No ethical issues over ppt consent etc. as all content used is in the public domain
• Can be applied to a wide range of media types, both textural and audio-visual - flexible research method
• Depending on the media being analysed, this method is generally quick and easy to do.
• Quantitative data is objective and can be statistically analysed
  to understand patterns etc.

Question 24

Give two weaknesses of content analysis

Answer 24

• Vulnerable to researcher bias - researchers may interpret themes in different ways, or miss things entirely - subjective and unreliable method
• Cannot establish causality from the data as it only describes the themes

Question 25

What is the difference between top-down analysis and bottom-up analysis?

Answer 25

• Top-down analysis is when a researcher begins with an initial assumption about the content of the media and applies those ideas as they go through it.
• Bottom-up analysis is when a researcher begins with no initial assumptions about the content, and categorises themes as they appear.

Question 26

What is inter-rater reliability?

Answer 26

• When to or more researchers analyse data to see if they can come to the same conclusion.
• All the researchers must be trained in the same way
• The results can only be called reliable if the researchers have an 80% concordance rate

Question 27

Name the 5 types of validity

Answer 27

1. Face Validity - Do the results adhere to common sense?
2. Concurrent Validity - Do the results of two different tests compare with one another?
3. Predictive Validity - Can these results predict future results?
4. Ecological Validity - How far can the results be generalised outside of the context of a lab/study?
5. Temporal Validity - Are the results of a prior test still valid in today’s context?

Question 28

Reporting Psychological Investigations???

Question 29

Which statistical test would be appropriate for a test of difference with independent measures that gathered ordinal data?

Answer 29

Mann-Whitney U Test

Question 30

Which statistical test would be appropriate for a test of association using a correlation gathering nominal data?

Answer 30

Chi-Square

Question 31

Which statistical test would be appropriate for a test of difference with independent measures that gathered nominal data?

Answer 31

Chi-Square

Question 32

Which statistical test would be appropriate for a test of difference with repeated measures gathering ordinal data?

Answer 32

Wilcoxon (…matched pairs signed ranks test)

Question 33

Which statistical test would be appropriate for a test of difference using repeated measures gathering nominal data?

Answer 33

Sign Test

Question 34

Which statistical test would be appropriate for a test of association using a correlation gathering ordinal data?

Answer 34

Spearman’s Rank (…Order Correlation Coefficient)

Question 35

Which statistical test would be appropriate for a test of association using a correlation gathering interval data?

Answer 35

Peterson’s Test

Question 36

Which statistical test would be appropriate for a test of difference using repeated measures gathering interval data?

Answer 36

Related T-Test

Question 37

Which statistical test would be appropriate for a test of difference using independent measures gathering interval data?

Answer 37

Unrelated T-Test