week 2 Flashcards
Hypothesis
is a specific, testable claim or prediction about what you expect observe
Research hypothesis
Statement you’re testing
What you expect to find
One-tailed: direction
Two-tailed: no direction
Null hypothesis
Provides a baseline against which to evaluate alternative hypothesis
Interpreting correlation
Correlation means that two variables vary together as one changes so does the other
Extraneous variable
Anything other than the independent variable the could affect the dependent variable
Confounding variable
A type of extraneous variables that affects dependent variable
It is uncontrolled
Between subjects
Allocate people to different conditions and to only one of these conditions
Av: no carry over effects
Dv: greater expense, less statistical sensitivity
Within subjects
each participant is tested in all conditions
Av: less expense, better statistical sensitivity
Dv: Carry over effect
Sample size
a lot to do with statistical power and effect size related to study
Effect size
Is a statistical measure of the magnitude of an observed effect in a population
Small effect size= larger sample
Type 2 error. Statistical power
Failing to detect an effect that exists
Statistical power: is the probability of detecting a true effect when it actually exists in your population
Internal validity
We can be sure that the changes we observe have actually been caused by our manipulation
Pre test
the observation or measure before the intervention
Experimental treatment
the different interventions or conditions
Post-test
the observation or measure after the intervention
Threats to internal validity
Confounding variables
In between subjects: Maturation effect: participants behaviour changes over time
History effect: something changes about the participants circumstance that influences the variables
Testing effect: merely having been testing before may have changed how they do on the post-test
Regression towards the mean: an extreme score is likely to become average
Control group
Help minimise any influence maturation
don’t remove them
Types: passive- participants do nothing. Active- they do something. Wait list- participants are waiting to take part
Eliminating confounding variables
Matched pairs design- participants are matched on a third variable and placed across the group
Random-participants design
Within-participants design
Differential attrition
When people leave one condition or treatment more than any other
Order effect
Practice effects-> participant perform a task better in later conditions
Fatigue effect-> participants perform a task worse in later conditions because they become tired or bored
Habituation -> participants may become less sensitive to stimulus through repetition
Reduce order effect Counterbalancing
testing different participants in different orders
A then B
B then A
When lots of conditions use Latin-square design
Doesn’t remove order effect just balances them across the study
Participants reactivity
Control groups awareness> active control
Evaluation apprehension> single blinding
Experimenter effect> double blinding
Blinding
Single blinding: participants don’t know
Double blinding: neither participants or researchers know
Stimuli
Should represent a good range of mor modalities
with manipulation check
Ceiling effects: the task is too easy or stimuli is extreme
Floor effects: task is too easy and stimuli is too weak
External validity
linking experiment to real world increases generalisability
