Ordinal data, different scales, comparison between variables

I would appreciate any help you could give me I have ordinal data which I want to standardise in order to compare averages and data spread between variables. I need to standardise since the variables have different scales (one is out of 7, the other out of 15). I have transformed the original values into percentages and then found the mean and standard deviations for each variable in order to allow for comparison. Is this ok? Somehow converting into percentage values and then taking the mean and SD of this doesn't seem right... Any help would me much appreciated :)

• One possibility would be to re-express all items scores on the same scale, using a simple linear transformation like $x = (x - min) / (max - min)$, where $min$ and $max$ are lower and upper bounds of your common scale of measurement. – chl Mar 28 '13 at 12:10

1 Answer

Partially answered in comments:

One possibility would be to re-express all items scores on the same scale, using a simple linear transformation like $x=(x−\min)/(\max−\min)$, where $\min$ and $\max$ are lower and upper bounds of your scale of measurement. – chl