I have two data sets, one that ranges from 0-200, and another that ranges from ~400-~2500.

I would like to compare the two according to a score from 0-10. I remember about normalizing from a statistics class that I took that in order to normalize you need to find the z-score which depends on the population mean and standard deviation (which I have). But I don't remember what to do with that z-score, or how to normalize both of these data sets down to a 0-10 scale so that they can be scaled down and compared against each other.

Anyone remember how to do this?

  • $\begingroup$ If you have z-scores do you mean that you have both in standard normal form? You could then simply design some scale from 0-10 that depends on z scores $\endgroup$ – 123 Jan 15 '15 at 18:58
  • $\begingroup$ I'd like to get both distributions in standard normal form. But by z score, I took (random variable - mean)/stddev from the original data set. I'd like to know how to scale it down to standard normal form. $\endgroup$ – johncorser Jan 15 '15 at 19:00
  • 1
    $\begingroup$ The right approach depends on why you want the score to run from $0$ to $10$. If you intend this score to be proportional to a quantile, you'll need $10\Phi(z)$, with $\Phi$ the $N(0,\,1)$ cdf. $\endgroup$ – J.G. Sep 1 '19 at 17:20

The z-score is the standardisation that you should plot. Full-stop. (And you have the correct formula for the z-score.) The z-score might usually range from -3 to +3 and you can then plot both z-score distributions on the same graph. The z-score distributions plot with their centres at z=0. You mention you want to plot on a 0-10 scale. What do you mean by this ?


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.