268627_en-US

liliconcom

Mean from several columns in ANOVA

Is there a way to perform an ANOVA analysis where the mean of each column is represented? For example, I have 4 different columns with answers from 1-7. I want to represent these answers using only one model. If i use "distribution" i get 4 different models, and I only want one. Anyone? 

Re: Mean from several columns in ANOVA

@txnelson you're a data munging genius. Nice solution and post too.

Re: Mean from several columns in ANOVAWow, just what I was looking for. Thank you so much!!Re: Mean from several columns in ANOVA

To expand on @Dan_Obermiller , if you have a data table

You can move the data into the form required for the Anova analysis by using

     Tables==>Split

Fill the dialog box out as shown below

Click OK, and you will get a data table like this

Now go to

     Analyze==>Fit Y by X

and fill out the dialog box as:

Click OK and the platform will generate

Now go to the red triangle and select "Means Anova" and you will get

Your Analysis of Variance analysis

Re: Mean from several columns in ANOVA

Not exactly sure what you are looking for. ANOVA is a method to compare several groups of data to each other. For example, you have four groups so have your data structured similar to this:

 

Group     Data

A             7

A             9

B             3

B             5

...

then you use Fit Y by X. The group column is the X. The Data column is the Y. 

You should start by going to the Help menu and choosing Tutorials > Many Means Tutorial.