Use LEFT and RIGHT arrow keys to navigate between flashcards;
Use UP and DOWN arrow keys to flip the card;
H to show hint;
A reads text to speech;
17 Cards in this Set
- Front
- Back
Comparison Contrasts |
Differences between a combination of treatment means, The differences to be tested are provided in a contrast matrix
|
|
Polynomial Contrast
|
The linear, quadratic, and higher-order terms from orthogonal polynomials fitted using the X values from the levels of the factor, or 1,2,3... by default if the factor only has labels or ordinal values defined
|
|
Regression Contrast
|
The treatment means regressed against the contrast values. The contrasts to be tested are provided in a contrast matrix, and are orthogonalized so that they are all statistically independent. A warning is given if the original set of contrasts was not orthogonal. (Two contrasts ai' and aj' are orthogonal if the sum of the products of their terms is zero, i.e. ai'aj=0) |
|
Contrasts
|
comparisons between the levels of a treatment factor that you are particularly keen to assess.
|
|
Simple Contrasts
|
an individual difference between two levels of a factor and these would be given values -1 and 1.
|
|
What is a contrast?
|
A set of weights (a vector) that defines a specific comparison over scores or means.
|
|
When are contrasts used?
|
To test more focused hypotheses than the overall omnibus test of the ANOVA
|
|
What does orthogonality mean?
|
The two contrasts are not correlated (the covariance between A and B is zero). When sample sizes are equal.
|
|
If the omnibus test is statistically significant, then
|
at least one contrast is significant.
|
|
If the omnibus test is not statistically significant, then
|
a contrast can still be statistically significant.
|
|
Three types of post hoc comparisons
|
Least Significan Difference, Tukey's HSD test, Scheffe's Test
|
|
Fisher Least Significant Difference Test
|
Explores all possible pair-wise comparisions of means comprising a factor using the equivalent of multiple t-tests.
|
|
Tukey's HSD Test
|
Procedure that accurately maintains alpha levels at their intended values as long as statistical model assumptions are met.
|
|
Scheffe's Test
|
Most flexible and most conservative of all post hoc procedures. Corrects alpha for all pair-wise or simple comparisons of means, but also for all complex comparisons as well. Least statistically powerfull procedure. Can lead to Type II errors unless complex comparisons are being made.
|
|
Bonferroni's Adjustment
|
Flexible (and simplest) post hoc method for making post hoc comparisons that ensure a family-wise type II error rate no greater than alpha after all comparisons are made.
|
|
Familywise Error
|
Alpha inflation or cumulative Type I error. Represents the probability that any one of a set of comparisons or significance tests is a Type I error.
|
|
Dunnet
|
used only if a set of comparisons are being made to one particular group. |