Definitions
This test is used to compare 2 or more group means.
one-way ANOVA
The main formula for one-way ANOVA is this ratio.
What is F = MSbg / MSe?
(must get all parts correct)
ANOVA compares ___, not just means directly.
varience
df for between groups =
k-1
F values can never be-->
Negitive
The independent variable in one-way ANOVA is also called...
What is a factor?
PS. One-way ANOVA = _ factor (how many)?
MS means-->
Mean square
(p.s. = variance is also called this)
A large F value suggests this.
That group means are (very) different
df for within groups =
N-k
(N not n)
The rejection region for ANOVA is located here.
Upper right tail (varience can't be negitive)
The groups within a factor (independent var) are called: ____.
Levels
EX:
You study how types of therapy affect anxiety.
- Levels:
Extra: whats the factor here?
Omega-squared corrects for this and this -->
(suprise ! this is worth 600 points, will only subtract 300 for getting it wrong though).
What are error (MSe) and number of groups (DFbg)?
The denominator (error) of the f-stat represents this type of variance.
within-groups variance (error)
(must say: within-groups varience)
Total df =
N − 1
The F distribution is shaped:
Positively skewed
Variation within groups is also called:
error
(or within-groups variance)
Eta-squared (η²) equals this:
SSbg / SSt?
If F ≈ 1, this is likely true.
We retain the null hypoth.
What's our df (between groups) if k = 3?
2
(k-1)
Compared to F-obt to decide significance
-->
F-crit
This value increases when differences between group means grow larger relative to variability within groups, and it determines whether results are statistically significant in ANOVA.
F-statistic (F-ratio)
In an ANOVA, this value is calculated by dividing the sum of squares for error (SSₑ) by its degrees of freedom (dfₑ) and serves as the denominator of the F-ratio.
What is mean square error (MSE)?
The numerator of F-stat represents this type of variance =
The denominator of F-stat represents this type of variance =
Between-groups variance
Within-groups varience
N = 15
k = 3
df (within-groups/error) =
12
This effect size measure tends to overestimate variance explained -->
What is eta-squared (η²)