Study your flashcards anywhere!

Download the official Cram app for free >

  • Shuffle
    Toggle On
    Toggle Off
  • Alphabetize
    Toggle On
    Toggle Off
  • Front First
    Toggle On
    Toggle Off
  • Both Sides
    Toggle On
    Toggle Off
  • Read
    Toggle On
    Toggle Off

How to study your flashcards.

Right/Left arrow keys: Navigate between flashcards.right arrow keyleft arrow key

Up/Down arrow keys: Flip the card between the front and back.down keyup key

H key: Show hint (3rd side).h key

A key: Read text to speech.a key


Play button


Play button




Click to flip

16 Cards in this Set

  • Front
  • Back
What does the test for homogeneity of variance test?
It tests for equality of group variance.
How do you interpret the SPSS output for the homogeneity of variance test?
If the sig value is below .05, you do not have homogeneity. Under these circumstances, you will have to worry about heterogeneity.
What does low power mean (in terms of effect size and variability)?
Low power" (too little data; meaningful effect sizes are difficult to detect).

Low power also means high variability
are based on theory and/or earlier literature. They are planned before the data are analyzed. Usually, only a small set of planned contrasts is used.
Planned Contrasts
Use when you have more than dfA planned contrasts or when your contrasts are not orthogonal.
Bonferroni Correction
Essentially the same as the Bonferroni, but is a little less conservative. Ensures that familywise α is exactly what you set (e.g., α = .05) whereas Bonferroni assures you that you are less than that. Usually used for planned contrasts, but can be used for small sets of post-hoc contrasts.
Šidák--Bonferroni Correction
means that there was no specific planned set of comparisons, usually each mean is compared to all other means.
Post-hoc Pair-Wise Comparisons
In Post-hoc Pair-Wise Comparisons,
We usually find the difference between the two means and compare the difference to a critical value ___ The observed difference must be greater than____
D bar, D bar
Use this only when all you want to do is compare each group with only one other group (usually a control group). There will be (a-1) comparisons
Dunnett Correction
What are some examples of Post-hoc Pair-Wise Comparisons?
Dunnet, Tukey, Fisher-Hayter Correction
What are some examples of Post-Hoc Contrasts?
Scheffé Correction,Pairwise Comparisons
Use this for all pairwise comparisons. Compute the critical mean difference using the q value in Table A-6. Find q for the total number of groups you have and the error df. Remember that your critical mean difference must be less than your observed difference.
Tukey Correction
Use this for all pairwise comparisons. It is less conservative than the Tukey. Use Table A-6, but use the q value for a-1, rather than for the total number of groups. Remember that your critical mean difference must be less than your observed difference.
Fisher-Hayter Correction
Usually, you use weights and make a fairly large number of contrasts that you select after seeing your data (e.g., {+2, -1, 0, -1, 0} for five groups). These were not planned (or you only planned a few of them and now you want to do more.) You will end up with a t (on SPSS) or an F (if you do it by hand
Post-Hoc Contrasts
- This is a VERY conservative correction, protecting you for all possible contrasts. Usually, you would calculate the contrasts using weights (using Contrasts in SPSS is fine) and then use a Scheffé F. (Square the t from SPSS contrasts.) You wouldn’t use a critical mean difference with Scheffé because many contrasts involve more than two groups. The Scheffé option in Post Hocs in SPSS does not correct contrasts that you enter through Contrasts. The Scheffé option puts the Scheffé correction on the pairwise comparisons and it is too conservative for that (use the Tukey or Fisher-Hayter instead).
Scheffé Correction
You can also compute each pairwise comparison (e.g., {0, +1, 0, 0, -1} for five groups) using weighted contrasts. If you do this, there are F versions for the pairwise corrections (square the ts from SPSS). For between-subjects designs, however, finding the critical mean difference and each actual difference is easier.
Pairwise Comparisons