PS296 Lecture Notes - Lecture 18: Analysis Of Variance, Statistical Hypothesis Testing, Type I And Type Ii Errors
WEEK 10 – LETURE 18
PS295 / JULY 12, 2018
Choosing the “right” test:
One score:
z test
t test
One mean
one sample z test
one sample t test
Two means
related samples t test
independent samples t test
Two or more means:
analysis of variance (ANOVA)
Terminology:
Factor:
-an independent variable in an analysis of variance
Levels:
-the groups that make up the IVs, which are also called conditions
One way ANOVA (single factor design):
Document Summary
One mean one sample z test one sample t test. Two means related samples t test independent samples t test. Two or more means: analysis of variance (anova) An independent variable in an analysis of variance. The groups that make up the ivs, which are also called conditions. An analysis of variance where the groups are defined on only one factor (ex one iv) Analysis of variance where the groups are defined on two or more factors, with every level of one factor paired with every level of the other factor(s) With alpha = . 05, we have a 5% risk of committing a type i error (rejecting null incorrectly) If we ran 20 hypothesis tests, we would expect to make one type ! If we ran 100, we would expect to make 5 errors (100 x . 05 = 5) The more tests you do, the more risk you have of making a type i error.