stats Flashcards
what are parametric tests and what do they typically assume
statistical tests that make certain assumptions about the underlying population distribution of the data
- follows a normal distribution
- equal variance across groups
- measured on an interval or ratio scale
define dependent variable
data your participants provide, and variable you measure
-e.g reaction times, score on a test etc
whats the independent variable
the variable you manipulate (change), the conditions or groups that are compared, either by comparing diff groups or stimuli
when is an independent samples test used
when comparing the means of two groups that are independent of each other
- meaning observations in one group do not influence those in the other group
when is a repeated measures design used
when participants are exposed to more than one condition, and their responses are measured multiple times
- used to observe how participants behaviour or performance chnages over time or under different conditions
what do inferential tests (t-test and ANOVA) check
how likely it is that the results from the different conditions in your experiment came from the same population
what do you say/do when the inferential test(s) are highly unlikely that they came from the same population
- usually due to difference between conditions in experimental manipulation
- say the difference is statistically significant
- if likely results from condition came from same population then we CANNOT conclude the independent variable had an effect
- say the difference between conditions is non-significant
what does null hypothesis mean
there is no different between your conditions, they were sampled from the same population
whats an experimental hypothesis
there is a difference between conditions, the conditions were sampled from different populations
whats the difference between a one tailed and two tailed hypothesis
- 1: tests for effect in one specific direction
- 2: tests for an effect in both direction (greater or smaller)
when can H0 be rejected and not rejected
can be rejected if p< .05 and we accept the H1
- cannot be rejected if p> .05
what does it mean if p= .05 and what do you conclude and what is this known as
there is a 5% chance its a false positive, and conclude that samples came from different populations even if they did not
- knows as a Type I error
what is possible if p> .05
may be a false negative, and do not reject H0, even though the samples came from different populations
- type II error
whats a type I error
a false positive
- when we incoreectly reject the null hypothesis
- rejecting it when its actually true
whats a type II error
false negative
- we fail to reject the null hypothesis
- its actually false (missing a real effect or difference)