Calculating SD/Variance

[MedHopeful234]

Hey Guys,

with regards to calculating Variance/SD do you use n or n-1?

I recall Chad mentioned to do n-1 and a problem in achiever also used n-1, but then in other sources (Destroyer I believe) it was done with n.

---

Members don't see this ad.

 

[StumpMT]

This may be a little late since your DAT is either today or tomorrow from what I remember but I'm pretty sure that you use n-1 for smaller populations ie smaller sets of data and just n for giant populations.

 

[MedHopeful234]

Hey Thanks StumpMT! It's actually tomorrow... thank you so much for answering this... do you know what's considered large pop? None of the practice questions (that use n-1 or n) were more than 10 numbers so I wasn't sure... I guess I'll just use both and see which one fits the answer choices.

This may be a little late since your DAT is either today or tomorrow from what I remember but I'm pretty sure that you use n-1 for smaller populations ie smaller sets of data and just n for giant populations.

 

[StumpMT]

I think <100 would be considered a small population. I Googled it earlier after I originally answered your question and you use n-1 to estimate the variance of some larger population. I'd say that if your sample size is only 10 then use n-1. To be safe you could probably do the calculation for both n and n-1 without spending too much extra time on it to see how different the answer becomes and what your answer choices are.

 

[datthirty]

Hey Guys,

with regards to calculating Variance/SD do you use n or n-1?

I recall Chad mentioned to do n-1 and a problem in achiever also used n-1, but then in other sources (Destroyer I believe) it was done with n.

My understanding is like this, I am not major in Math and Stat, so I may wrong.

n-1 is for SAMPLE population SD.
n is for COMPLETE population SD.

If the sample is complete population, use n.