Psych 202 - Test 3 Flashcards
degrees of freedom
the number of scores that are free to vary when estimating a population parameter
why do we subtract by 0 in the mean
we assume that the population and sample mean are the same, so there will be no difference (0) when subtracted from each other
estimated variance
- we do not know the variance so we have to estimate it
- if the null is true and we pull the sample from the population, the means and standard deviations should be the same
- biased estimate with more likelihood of error
- sample variance will always be small than population variance
divide by N - 1
to correct for the biased nature of the estimated variance
t-distribution
t < 30 = platykurtic
t > 30 = in between platykurtic and normal
t-table
always go to lower number when looking at values
- rather be more conservative
- i.e. looking for 45 with choice between 43 and 50 –> go with 43 because we know we have 43 with our “group”
t-test
- used when population variance/standard error is unknown (our denominator in the final equation)
ANOVA
analyzes how much variance is due to randomness vs. how much is due to treatment/population differences
difference score
change (A) - base (B)
> always subtracting the base from the specific
between group variance
measures difference between the separate groups
- more difference expected than in within variance
within group variance
measures difference within the separate groups
- ideally 0
planned contrast
- within (group variance and df) stays the same
- between changes (group variance and df)
- recalculate the between variance and df with contrast scores instead of original data