rescaling a variable from -45 - 43 to 0 - 10

[Johanna Bjarnadottir]

Hello,

I'm struggling to rescale a variable with the min -45.946 and max of 43.21 to 0 - 10 scale.
I've seen few threads about this here on the forum but usually from 0-10 to 0-100 which sounds a bit easier.
Does anybody have a simple solution for how I could deal with this problem?

Best,
Johanna

[Maarten Buis]

// open some example data
sysuse auto, clear

// get the min and max of price
sum price, meanonly

// create the rescaled price
gen price_rs = ( price - r(min) ) / ( r(max)-r(min) ) * 10

// admire the result
sum price price_rs

[Bruce Weaver]

Hi Johanna. This is probably obvious, but I'll say it anyway. If the lowest possible score < the observed minimum or the highest possible score > the observed maximum, use the lowest and highest possible scores in place of r(min) and r(max)--they give the observed min and max in your dataset. HTH.

[Johanna Bjarnadottir]

Thank you so much for the quick response! The code achieved the exact results I was hoping for!

[Alberto Burgos-Rivera]

I am trying to rescale a variable scaled from1-10 to -4.5 to 4.5. How do I do it.

[Rich Goldstein]

gen newvar=oldvar - 5.5

[Alberto Burgos-Rivera]

Thank you, but this only gave me tha minimum value of -4.5 and not a fixed value of 4.5. How should I do to obtain a fixed rescaling of -4.5 to 4.5

[Maarten Buis]

Apperently, you have no one in your data that reached the theoretical maximum of 10. If someone in your data had a 10 on the original variable then (s)he would have gotten a 10-5.5=4.5 on the new variable. So the new variable is correct, it is just that your data does not cover the entire theoretical range.

[Marcos Almeida]

I'm wondering whether you have tried the code in #6. If so, it will give a range from min(-4.5) and max(4.5). I gather this is what you asked. To end, I fail to understand what is meant by "fixed".

[Alberto Burgos-Rivera]

Yeah, at first I didn't understand but I ran my code and got what I asked for. Thanks you everyone for your help.