Combining value labels without changing actual numeric responses

#1
I have responses from 1-8 where respondents were asked to rank a set of indicators. I want to combine value labels such that 1,2=High 3,4,5=Medium 6,7,8=Low. However, I do not want to alter individual's responses in the data view. I saw a previous suggestion for using the Transform function to re-code into different variables, but this would alter my data. I don't want all 2's to become 1's and so on, but when I analyze the data I would like 2's to report as High in addition to 1's. Is there a way to play with the value labels so that 1;2= High...3;4;5= Medium similar to how excel does it?
 
#2
If you recode into different variables, your original variables will still be available, so you're not actually altering any information.

I don't believe you can produce output in SPSS that combines different values without merging those values together. You could just paste the output in Excel and then group the value frequencies in whichever way you see fit. Or, if you use the recode into different variables route, run frequencies for both the original and the collapsed variables.

In short, I don't think SPSS can do exactly what you want, but I don't understand why the alternative options are an issue either.
 
#3
Use Recode AND build a recoded NEW vairbale.

Use the command:
Recode .... into...

the values in the original viarbale will not be replaced.