L4 - Methods of Psychology

Question 1

Research Methods:

Answer 1

The lesson of Clever Hans (horse that could do “math”): To test specific hypotheses you need to constrain extraneous variables and focus on what you are manipulating (IV) = ↑ Control and analysis (at the cost of a more artificial experiment)

Ordered from Least → Most control:

• Introspection
• Naturalistic Observation
• Case History
• Survey
• Test
• Correlation
• Experiment

Question 2

Introspection:

Answer 2

• “Looking inward”; Systematic/structured way of observing our own consciousness + a verbal report of our observations
• Non-scientific, subjective (no separation between observer and observed), “mentalist” psychology
• Underlies contemporary self-report methods

Question 3

Naturalistic observation:

Answer 3

• Objectively studying events naturally, without intervention
• Good starting point but often difficult to remain objective without intervention (e.g. observe social interaction whilst not participating in it)

Question 4

Case history/study:

Answer 4

• Biography for a single individually obtained retrospectively; often via interviews
• Essential in clinical psych etc. (using case studies as a form of “storytelling” is persuasive) but observations cannot be generalised
• Paradigms: Biological, psychodynamic, humanistic

Question 5

Surveys:

Answer 5

• Quantitative measure of responses to questions; often via an interview or questionnaire of a large sample
• Easy to perform but easily distorted (e.g. “What grade do you think you deserve?” Everyone would put a 7)

Question 6

Tests:

Answer 6

• Quantitative measure of performance relative to a preset norm
• E.g. Mid-semester exam

Question 7

Correlation:

Answer 7

• Statistical calculation of direction and degree of relationship between any 2+ variables (“what goes with what”)
• Correlation coefficient (f) = Degree (strength) and direction of covariance (relationship) between 2+ variables
  
  • Strength: How far value is from 0 with a range of -1 → +1 (correlation of -1 is same strength as + 1); closer to 0 = weaker correlation
  • Direction: Positive = Direct relationship; Negative = Inverse relationship; 0 = No relationship
• Correlation fallacy: Even when strong, correlations CANNOT infer causality; there may be an additional variable (e.g. foot size of children is strongly correlated with intelligence but this is only because children with larger feet are usually older)

Question 8

Experiment:

Answer 8

• Method of manipulating one set of variables (IV) while observing & measuring the effect on another set (DV), with other factors being held equivalent (random & control variables)
• If a change of IV = Significant change in DV, we can infer that the IV has a causal relationship with the DV
• Paradigms: Behavioral, cognitive (e.g. The Stroop Effect), biological
• “Research lives in correlation and experimental”

Question 9

Common Sources of Bias (Distortion) in Research:

Answer 9

• Sampling Bias
• Subject Bias
• Experimenter Bias
• Operational Definitions

Question 10

Sampling Bias:

Answer 10

• Is the sample studied representative of the population of interest?
• Affects generalisability of conclusions
• Solution: careful design – don’t just test anyone/everyone

Question 11

Subject Bias:

Answer 11

• ‘Hawthorne’ or placebo effects
  
  • Early (1920s) research in the applied area of Industrial Psychology
  • What environmental factors affect worker productivity? (Hawthorne was the name of the production company)
  • The main experiment related to workplace lighting
• Were the subjects responding to their expectations, rather than to the experimental manipulations?
• Remember, the experimenters were with the subjects when changes were made
• Solution: single ‘blind’ research – subjects are less aware of what’s going on

Question 12

Experimenter Bias:

Answer 12

• Rosenthal’ effects (Robert Rosenthal)
  
  • 1960s
  • Initial studies were on classroom students
  • Student ‘experimenters’ observed the learning behaviour of rats in two conditions: “dull” versus “smart” rats
• Are the researchers influencing the behaviours they are observing?
• Favours one group over the other
• Solution: double ‘blind’ research – experimenter and subjects less aware of what’s going on

Question 13

Operational Definitions:

Answer 13

• Defining variables in terms of the operations (methods) used to observe/measure/manipulate them
• Needs to be very clear what you are measuring