ANOVA module Flashcards
How is the Sum of Squares Total Calculated?
How is the Sum of Squares Within Calculated?
Sum of squares within = sum of data points - mean of group
How is the Sum of Squares Between Calculated?
Sum of squares between = sum of squares total - sum of squares within
When would be use an ANOVA?
Whilst an ANOVA can be used to determine whether there is a difference between two groups, it is generally used when trying to determine whether there is a difference between three the means of three or more groups at a population level.
What does the F statistic in an ANOVA tell us?
F statistic/ratio tells the ratio of the variation between groups/variation within groups and is calculated as mean square between/mean square within.
Is the Sum of Squares affected by the sample size?
Yes.
If Sum of Squares is affected by degrees of freedom, how is this accounted for when calculating the ratio of variation within and between groups, aka the F statistic?
The F statistic does not use just SS between and SS within.
F statistic uses mean square between = SS between/df between
And mean square within = SS within/df within.
How is the F statistic calculatied?
F = Mean Square Within/Mean Square Between, where Mean Square Within is calculated by taking the SS within and dividing by df within and Mean Square Between is calculated by taking Sum of Squares between and dividing by df between
An ANOVA can tell us whether there is significant difference between groups for a given variable, but it cannot tell us how big the difference is, i.e. the effect size. Correct?
Yes.
What is the definition of the Total Sum of Squared Deviations, also known as SS total?
The SS total is the sum of the squared difference of each of the data points and the grand mean.
In a one-way between-groups ANOVA, how many independent variables are there and are the participants of the groups the same or different?
We have one independent variable and the participants in the groups are different.
In an ANOVA, what does the p-value tell us?
The p-value tells us the probability of observing a given set of data if the null hypothesis of no difference between groups is true.
Why do we need to do post-hoc tests when we do an ANOVA test?
An ANOVA test can tell us whether there is a significant difference between groups, but does not tell us the effect size of the difference or where the statistically significant difference occurs, i.e. does it occur between all groups, between only certain groups etc?
What is the Bonferoni test used for, how is it calculated and what does it measure?
Bonferoni test is a post-hoc test to look at where the difference occurs (i.e. between which groups).
What is Tukey’s HSD?
