Minitab V17 - Analysis of Variance - Two-way
The Minitab Worksheet is available. It contains the following data:
Column N Comments IV1Code 10 Dummy codes for groups (1=0%EDC, 2=35%EDC). IV2Code 10 Dummy codes for groups (1=young, 2=older). TrialstoPA 10 Number trials to learn PA.
Version 17 of Minitab strengthened the ability of the program to handle factorial designs. There is probably more than one correct way to do this analysis using Minitab. I will present one straight forward method. Furthermore, with the factorial design, unlike previous analysis we have done using minitab, it best to ask for the descriptive statistics first and then do the analysis afterwards.
Thus, we need to get the means and standard deviations for each group in the design. Use the "Descriptive Statistics..." command off of the "Stat, Tables" menu. That is:
That will take you to the following dialog box:
For the "Catagorical variables:", type in (or double click on) the variables that code each IV. Type one for the rows and one for the columns. Everything else should be blank. Then choose the "Associated Variables ..." button which will take you to the following dialog box.
In the box for "Associated variables:", choose the variable that contains the DV scores. Then check "Means" and "Standard deviations". Click on "OK" twice and you will see the following output.
Tabulated Statistics: IV1Code, IV2Code
Rows: IV1Code Columns: IV2Code
1 2 All
1 3.600 6.000 4.800
1.140 1.581 1.814
5 5 10
2 15.600 6.400 11.000
2.881 2.608 5.497
5 5 10
All 9.600 6.200 7.900
6.653 2.044 5.098
10 10 20
Cell Contents: trialstoPA : Mean
trialstoPA : Standard Deviation
It would be helpful to plot these means, but Minitab is not especially friendly in this regard, thus I recommend doing that by hand or using a spreadsheet program. For the analysis, use the "General Linear Model...", "Fit General Linear Model..." command off of the "Stat, ANOVA" menu. That is:
It will lead to the following dialog box:
Type (or click) in the "Responses:" what the dependent variables are (in this case, we only have one) and then type (or click) on the two IVs for the "Factors:" box. Then click the "Model..." button and you will see the following dialog box:
This dialog box is necessary to tell Minitab that we want to know about the interaction. Thus, in the box labelled "Factors and covariates:", select both factors and then click the Add button next to the "Interactions through order: 2" and the interaction will then be added to the "Terms in the model:" box as shown above. Click OK.
Now click the "Results:" button and you will see the following dialog box:
Make sure that the only thing with a black check mark next to it is the "Analysis of variance". Clicking OK twice will then lead to the following output:
General Linear Model: trialstoPA versus IV1Code, IV2Code
Analysis of Variance
Source DF Adj SS Adj MS F-Value P-Value
IV1Code 1 192.2 192.200 40.68 0.000
IV2Code 1 57.8 57.800 12.23 0.003
IV1Code*IV2Code 1 168.2 168.200 35.60 0.000
Error 16 75.6 4.725
Total 19 493.8