Hello, This is the first time i am trying to figure out Ranova in PP and was wondering if i could get help/suggestions from forum members.
I have two sets of data (dataset 1, and dataset 0) each containing some selected properties (properties A-F) as shown below
-------------------------------------------------------
Dataset PropA PropB PropC PropD PropE PropF
0 192 14 0 6 176 90
0 132 11 0 0 125 56
0 33 4 2 6 28 19
0 48 1 2 6 42 24
0 71 2 0 18 55 33
1 45 2 2 6 40 23
1 49 1 2 6 44 26
1 80 7 0 0 81 33
1 90 8 0 0 83 41
1 85 12 0 0 86 39
--------------------------------------------------------
My aim is to do statistical analysis to find out which properties (PropA-F) are able to discriminate
most between dataset1 and dataset0. For this I want to do one-way analysis of variance (ANOVA) on
each of these properties and determine those that are best able to separate the two datasets (on the basis of F-ratios).
Now i am trying to plug this text file of data straight into RANOVA with "Dataset" defined in "ResponseProperty".
This is the output that i get:
-------------------
Term SumOfSquares MeanSquare Fvalue Pvalue
PropertyA 0.19376 0.19376 0.91079 0.41031
PropertyB 0.3177 0.3177 1.4934 0.30895
PropertyC 0.047495 0.047495 0.22326 0.66882
PropertyD 0.51014 0.51014 2.398 0.21925
PropertyE 0.11477 0.11477 0.5395 0.51584
PropertyF 0.67791 0.67791 3.1866 0.17223
Residuals 0.63822 0.21274 NaN NaN
------------------
Question is, is this the right way to do it?.
Also i did the same calculations in ANOVA with MATLAB and i got different results for F-values and others. Not sure where i am messing up.
Any suggestion/s would be very much appreciated.
Thank you
Nar