Lecture 2 Flashcards
Revision 2
What is the most accurate type of test design?
Experimental Design
What is a correlational design?
A description of the relationship between two variables, looking at how two things move together.
Not making an attempt to explain the relationship.
What is a non-experimental or quasi-experimental design?
To try and create a cause an affect explanation that falls short for some reason.
A design when you can’t truly randomise the control in the population.
What is an experimental design?
A design with two test conditions where you change one variable and test another variable and see the effect that the first variable has on the other. Must be randomly selected.
What is Semantic Memory?
a type of long-term memory involving the capacity to recall words, concepts, or numbers, which is essential for the use and understanding of language.
What is Episodic Memory?
a type of long-term memory that involves conscious recollection of previous experiences together with their context in terms of time, place, associated emotions, etc.
What is Working Memory?
Working memory is a cognitive system with a limited capacity that is responsible for temporarily holding information available for processing.
(Also called short-term memory)
What does ‘WMC’ stand for?
What is it?
Working Memory Capacity
Working Memory has a limited capacity, typically seen to be about 7 digits, 6 letters, 5 words (chunks)
Your study design looks like this:
Participants are divided into two groups based on their behaviour
Coffee-drinkers and non-coffee-drinkers
You use a standard measure of WMC
Your Predictor is which group are you in.
Your Outcome is measured WMC.
What sort of design is this?
A non-experimental design.
What type of design is this:
Two groups: group 1 gets caffeine, group 2 does not
People are assigned randomly to groups
Use a standard measure of WMC
Variables of interest:
Predictor is which group are you in.
Outcome is measure WMC
An experimental design.
What are some strengths of the experimental design?
Casual inference is possible in experimental designs.
Experimental intervention prevents the “outcome” causing the “predictor”
Therefore relationships arise from causal paths from predictor to outcome
What are some weaknesses of experimental designs?
Cutting off predictor from a normal environment can lead to “artifacts“
I.E. Coffee-drinkers in a non-caffeine group may get headaches, which may lead to artificial results
The experimental context can introduce new variables.
I.E. If you know you are in an experiment and being watched it may increase anxiety and that might influence results and bias response.
What does validity mean?
Validity refers to the accuracy of the study.
Does the answer inform us about the right theoretical construct? Are we measuring the wrong things somehow?
What does reliability mean?
Reliability refers to the precision of the study.
The stability or consistency of the measurements produced by a specific measurement procedure.
If the study is reproduced, will we get similar results?
What are the two types of validity?
Internal and External Validity
What is internal validity?
Refers to whether the study allows you to infer the correct causal relations and can the study tell us which direction causation goes.
E.g. does to study go predictor -> outcome or outcome -> predictor or predictor -> mediator -> outcome
What is external validity?
Refers to whether the study says anything meaningful outside of the experimental design.
Is it limited to the narrow range of people/situations studied?
What is face validity?
Whether a measure superficially appears to measure what it claims to measure (face value).
What is construct validity?
Do the scores obtained from a measurement behave exactly the same as the variable itself (outside of the lab)
What is ecological validity?
Does a score obtained from a measurement represent real-life behaviour?
(Stanford prison experiment tried to maximise ecological validity)
What are the three types of reliability?
Test-retest reliability (successive measurements)
Inter-rater reliability (simultaneous measurements)
Internal consistency
What is test-retest reliability?
Compares scores of two successive measurements of the same individuals and correlates the scores.
If the same person is tested twice, will they get similar results?
What is inter-rater reliability?
Agreement between two observers who simultaneously record measurements of the behaviours.
Does the person conducting the test influence the results?
What is internal consistency in regards to reliability?
Agreement between different parts of the same test.
Do all items in the test measure the same thing?
What is the relationship between reliability and validity?
Reliability is a prerequisite for validity as a measurement procedure cannot be valid unless it is reliable.
However, it is not necessary for a measurement to be valid for it to be reliable.
Name some sources of error that may occur in experiments.
History effects
Maturational effects
Repeated testing effects
Selection bias
differential attrition
Non-response bias
Experimenter bias
Demand effects (reactivity)
Placebo effects
Fraud, deception and self-deception
Regression to the mean
