Task 7/Lecture 6

Question 1

Causation

Answer 1

one variable directly or indirectly influences another
- possible to trace an unbroken chain of influences from initial event to final result.

Unidirectional: A –> B, not vice versa
Bidirectional: A –> B, B –> A

Question 2

Correlation

Answer 2

changes in one variable accompany changes in another, but they do not cause change.

covariation: changes in one variable tend to be accompanied by specific changes in another. This does not necessarily mean that either variable has a causal influence.

Question 3

Correlational research

Answer 3

determine whether two (or more) variables covary and if so, establish directions, magnitudes and forms of the observed relationship.
- involves developing measures of the variables of interest and collecting your data.
- nonexperimental - no manipulation

also referred to as passive observation study/naturalistic study

Question 4

issues with drawing causal inferences from correlational data

Answer 4

Third-variable problem: possibility that correlational relationships may result from the action of an unobserved “third variable” → may influence both of the observed variables.

Directionality problem: even when there is a causal relationship, one cannot determine which variable causes what.

Question 5

Demonstration

Answer 5

looks like an experimental design but lacks the use of an independent variable.
- Expose a group of subjects to one (and only one) treatment condition.

Question 6

Experimental research

Answer 6

high degree of control over the variables of your study, which gives you the opportunity to establish causal relationships among your variables.

1. Manipulation of the independent variable
2. Control of extraneous variables.

+ one can identify a causal relationship
- can’t be used when the hypothesised variables cannot be manipulated.
- more control of extraneous factors –> less external validity.

Question 7

Extraneous variables

Answer 7

variables that may affect the behaviour you wish to investigate but are not of interest for the present experiment.

control:
- Holding them constant so they cannot cause uncontrolled variation
- Randomise their effects across treatments

Random assigment: assign participants to treatments randomly.

Question 8

Confounding variables

Answer 8

occurs when the effects of two variables on a response variable cannot be distinguished from each other, the confounded variables may be either explanatory variables or lurking variables.

prevention:
- random assignment
- blind or double blind technique (prevents experimenter bias)

Question 9

Quasi-independent variables

Answer 9

correlational variable that resembles an independent variable, but lacks the aspect of manipulation. (age, gender, IQ)

+ may allow testing generality of the results across different levels
+ reduce error variance
- may cause misinterpretation
- more effort/costs or added complexity.

Question 10

Developmental design: Cross-sectional design

Answer 10

select several participants from each of a number of age groups. You create groups based on the chronological ages of your participants at the time of the study.

+ useful developmental data in a short period of time.
- generation effect: influence of generational difference in experience, which becomes confounded with the effects of age → threatens the internal validity.

Question 11

Developmental design: Longitudinal design

Answer 11

following a single group of participants over an extended period of time.

+ no issues of generation effect
- cross-generational effects: results from one generation may not apply to another –> threatens external validity
- subject mortality: loss of subjects, can lead to a biased sample
- Multiple-observation effect: there may be improved performance over time –>carryover effect
- History effects.

Question 12

no-treatment control group

Answer 12

participants do not receive the treatment - provides a baseline for the experiment

Question 13

placebo control group

Answer 13

use an inert or innocuous medication or fake medical treatment that has absolutely no medicinal effect.

Outcome research
- Investigates the effectiveness of a treatment.
- Goal: does the treatment produce a substantial or clinically significant effect?
- More concerned with the general outcome than the factors causing the effect.
Process research
- Wants to identify the active components of treatment.
- Focused on components of the treatment, which part makes it effective? → less focused on the outcome

Question 14

Mediator variable

Answer 14

via
a –> b –> c

Question 15

Moderator variable

Answer 15

subgroup
a – (b) –> c

Question 16

Simpson’s paradox

Answer 16

association or comparison that holds for all of several groups can reverse direction when the data are combined to form a single group.

• Extreme form of the fact that observed associations can be misleading when there are lurking variables
• Conclusions that seem obvious when we look only at aggregated data can become quite different when we examine the data in more detail