Variables, designs and hypotheses Flashcards
Define an experiment
Changing the IV whilst measuring the DV
How can we infer causality?
Changes in DV must be caused by changes in the IV
What is a Quasi-experiment?
IV can’t be manipulated
What is one issue with a Quasi-experiment?
Confounding variables
What is a Correlational design?
There’s no manipulation, and you measure 2 variables and determine whether they’re related
What is one issue with Correlational design?
Can’t infer causality
What is an IV?
Variable that is changed
How many levels can an IV have?
2 or more
What is a DV?
Variable that is measured
What are the 4 measurement scales for a DV?
nominal, ordinal, interval and ratio
What is a nominal scale?
Categorical- numbers refer to different class
What is an ordinal scale?
Ranking- numbers indicate a rank in a list
What is an interval scale?
Equal steps are meaningful
What is a ratio scale?
Equal steps are meaningful and theres a meaningful zero point
What is a confounding variable?
Variable that confuses the interpretation of the results
When would a confounding variable occur?
When some aspect of the experiment varies systematically with the IV
What is the aim of an experimental design?
To eliminate any potentially confounding variables from the experiment
What is a between-subjects design?
Each condition is applied to a different group of participants
How can you balance individual differences?
By assigning participants randomly
What is a within-subjects design?
Same participant performs at all levels of the IV
What is within-subjects design also known as?
Repeated measures design
What is one disadvantage of within-subjects design?
Order effects
How can you combat order effects?
Randomly or use counterbalancing
What can you do when you can’t run a within-subject design?
Use a matched design
What can we do when we can’t actively manipulate the variables we want to test?
Consider pre-existing variables and measure the extent to which they are co-related
What is an experimental hypothesis?
Questions we wish to address in experiments, based on our theories
What is a statistical hypothesis?
Precise statements about collected data
What is a null hypothesis?
states the different sample we look at come from the same population
For parametric stats, what does the null hypothesis state?
all the means are equal
For non-parametric stats, what does the null hypothesis state?
All the distributions are the same
What is an alternative hypothesis?
The logical opposite of the null hypothesis
When do we reject the null hypothesis?
When the probability of null hypothesis being true is less than the criterion
What do we set the criterion as?
a=0.05
What does p<0.05 mean?
There a less than 1 in 20 probability of it happening by chance
How do we determine p?
Calculate a test statistic
What is p?
The probability of collecting this data assuming the Null hypothesis to be true
(the probability the effect we measured is simply due to chance)
When can we think the events we measured are unusual?
When p is less than a