I have a simple monadic test where subjects are asked to rate their agreement/disagreement (5 scale) to liking a sample. There are 3 samples (3 columns of responses) and 100 total participants (rows). Thus, 33% of the subjects see sample A, another 33% see sample B and the last group see sample C.
The responses are numerical data where 0 = strongly disagree, and 100= strongly agree. The values are being treated as continuous (ok- 100 lashes of wet noodles on my back). My objective is to determine if there is a difference in consumer preference. My question: How to transform the data to be able to perform Tukey-Kramer Multiple Comparison.
I suppose I can Table--> Stack the 3 columns, delete all rows with no values, and do Fit Y by X.
Is there a shortcut to do this? or a more palatable way to get to my objective?
