Other's On-Line Resources (Chapter 13)

 

Two-Way ANOVA (A)
  • Description:
    In using this interactive on-line resource, you'll see a table of cell and marginal means, an ANOVA summary table, and a graph of the interaction. However, you can make changes in the cell means, n, and/or the error mean square, and then look to see the effect your change(s) on the ANOVA's 3 F-tests and the graph of the interaction.

  • What to Do:
    1. Click on the colored title of this on-line resource: "Two-Way ANOVA (A)."
    2. After clicking on "Begin," you'll get a new screen that you should carefully examine simply so you know what's where.
    3. First change one or more of the cell means to see the effect of your change(s) on (a) the marginal means, (b) the ANOVA summary table, and (c) the graph of the interaction.
    4. Click "Default Means" to return to the original set-up. Now change the "Group Size," observing the effect of this change on (a) the marginal means, (b) the ANOVA summary table, and (c) the graph of the interaction.
    5. Click "Default Means" to return to the original set-up. Now change the ""MSE" (the error MS) and see what happens to (a) the marginal means, (b) the ANOVA summary table, and (c) the graph of the interaction.

  • Sky Huck's Puzzle Question:
    First, click on "Default Means." Now, leave everything as it is but just imagine that you change each entry in the top row of cell means from 5 to 19.5. If this change were to be made, do you think the significant main effect for B would evaporate? After making your guess, change the cell means to find out whether or not your guess was right.
     
Two-Way ANOVA (B)
  • Description:
    Although this site is non-interactive, it contains helpful text-based information, an ANOVA summary table, and a graph of an interaction that will reinforce things you've read in Chapter 13. The emphasis is on the meaning of terms and proper interpretation, not the computations needed to obtain results.

  • What to Do:
    1. Click on the colored title of this on-line resource: "Two-Way ANOVA (B)."
    2. Read carefully what's said on each screen, and move to the next screen by clicking the "Next" button.
    3. You'll come across several words that are links to brief but helpful explanations of what a term means. Don't pass up the opportunity to gain a handle on these terms by using the available links.

  • Sky Huck's Puzzle Question:
    Assuming equal sample sizes, how many individuals were in each cell of the study involving the effects of drug dosage and task complexity on completion time?
     
Two-Way ANOVA (C)
  • Description:
    This interactive on-line resource allows you to specify the means, SDs, and sample sizes for each of the cells in a 2x2 design. Then, you'll be given the calculated F-values and accompanying p-levels for the 2 main effects and the interaction.

  • What to Do:
    1. Click on the colored title of this on-line resource: "Two-Way ANOVA (C)."
    2. In the menu on the left (with options presented in black type on a yellow background), click "2X2 Analysis of Variance."
    3. On the next screen that pops up, you'll see 4 gray boxes, each representing a cell in the 2x2 design. (These cells are labeled A1B1, A2B1, A1B2, and A2B2, with the "A" and "B" representing the columns and rows, respectively, while the "1" and "2" represent the 1st and 2nd level of each factor.)
    4. Enter any values you'd like for the mean, SD, and n of each cell.
    5. Click on the "Calculate" button and then scroll down on the box to the right of the 4 cells to see the F- and p-values.
    6. Repeat Steps 3 and 4 several times until you get a feel for the impact on the F-values of (a) making n smaller or larger, (b) making the SDs smaller or larger, and (c) having means that are closer together or further apart.

  • Sky Huck's Puzzle Question:
    Start with each M = 10, each SD = 10, and each n = 100. Next, verify that all 3 Fs = 0. Now, change 1 or more of the means by only 1 point so as to have p<.05 for the main effects but p=.999 for the interaction.
     
Post Hoc Tests (A)
  • Description:
    Although this site is non-interactive, it contains helpful text-based information and tables that will help you understand why and how researchers sometimes perform post hoc tests on the main effect means. A clear example is included.

  • What to Do:
    1. Click on the colored title of this on-line resource: "Post Hoc Tests (A)."
    2. Read carefully what's said on the 1st screen, and move to the next screen by clicking the "Next" button. The 3rd of the 3 screens will probably not be of much help.
    3. You'll come across several words that are links to brief but helpful explanations of what a term means. Don't pass up the opportunity to gain a handle on these terms by using the available links.

  • Sky Huck's Puzzle Question:
    How many comparisons would be made in the hypothetical study if all possible pairwise comparisons were to be made between Factor A's main effect means?
     
Post Hoc Tests (B)
  • Description:
    Although this site is non-interactive, it contains helpful text-based information and graphs that will help you understand why and how researchers sometimes perform tests of simple main effects. A clear example from a 2x2 ANOVA is included.

  • What to Do:
    1. Click on the colored title of this on-line resource: "Post Hoc Tests (B)."
    2. Read carefully what's said on each of the 3 screens, moving from one to another by clicking on the "Next" or "Previous" buttons.
    3. You'll come across several words that are links to brief but helpful explanations of what a term means. Don't pass up the opportunity to gain a handle on these terms by using the available links.

  • Sky Huck's Puzzle Question:
    If tests of simple main effects compared the two tasks, first at the "Control" condition and then a second time at the "Treatment" condition, both of these tests might yield a statistically significant result. Cite 2 reasons why the 1st of these tests might turn out to be significant even though the graph shows that the cell means are quite close together.
     

Copyright © 2012

Schuyler W. Huck
All rights reserved.

| Book Info | Author Info |

Site URL: www.readingstats.com

Top | Site Map
Site Design: John W. Taylor V