Lecture 9 Flashcards
Who invented ANOVA?
Sir Ronald Fisher (hence, F statistic)
ANOVA can be used in a very wide range of experiment designs which are?
- independent groups
- treated measures
- matched samples
- designs involving mixtures of independent groups and repeated measures
- more than one independent variable can be evaluated at the same time
What is the statistics used in ANOVA?
f statistic (the F ratio)
The F statistic is a ratio formed as?
Between-group variability divides by within-group variability
AND
Treatment + individual differences + experimental error divided by individual differences + experimental error (error term)
The _____ the treatment effect, the bigger the value of the __ ____.
Stronger; F ratio
What are the most important details in ANOVA?
F ratio and the significance level (labelled p in SPSS)
What does it mean if p value is less than .05?
You can conclude that there is a significant difference among your groups.
What is the F test?
It tells you that there is a significant effect somewhere among your groups, but doesn’t tell you the source of the difference.
What do you need to follow up in order to locate the source of the significant difference?
f test
When do you carry out a post-hoc test?
This test is carried out after youbhave ovtained the significant overall ANOVA in order to locate the source of the significant F. exploratory, fishing expeditions.
Overall F must be ______ in order to justify the use of ____ tests.
Significant; post-hoc
There are different post- hoc tests. The one you choose depends on your research situation. Some common post-hoc tests?
- Scheffé test
- Newman- Keuls test
- Tukey’s Honestly Sifnificant Different test
- Fisher’s Least Significant Different Test
What does a post- hoc test do?
Test every group with every other group
What does pairwise comparisons ask?
Is there a significant difference between this pair of means?
What is degrees of freedom?
Figures that the F test use to calculate its exact value.
What are the two degrees of freedom for each F test?
- Number of groups -1
2. (Number of subjects in each group-1) x number of groups
What is established hypotheses?
Having some very clear, set ideas about what is going to happen.
What is planned comparisons?
It’s a specific hypotheses test involving specific sub-groups of your experiment, and are carried out instead of overall ANOVA followed by post-hoc tests.
Define planned contrasts?
Planned contrasts can test complex comparisons that involve combinations of groups.
What are the 2 advantages of planned comparisons over post-hoc tests?
- For the same comparison, you are more likely to find a significant difference with a planned comparison that with an equivalent post-hoc test (I.e., they are statistically more powerful)
- Can design complex comparisons
What is an advantage of post-hoc tests over planned comparisons?
- Let you compare everything with everything else (planned comparisons permit a smaller number of focussed comparisons).
What are the rules for planned comparisons?
- you do them instead of an overall ANOVA (not as well as).
- the maximum number of planned comparisons you can have is the number of groups -1.
- your contrast coefficients must sum to zero.