The Microsoft Excel function "STDEV" uses the n-1 formula: "estimates
standard deviation based on a sample."
"STDEVP"  uses the n formula: "calculates standard deviation based on
the entire population given as arguments."

A student suggested that this made sense.  If you are using the "n"
formula, which many of us would call the sample standard deviation, the
sample is your population.  You are describing variation withing the set
of data that you have in front of you.  So, the "n" formula gives you
the population standard deviation.  The "n-1" formula gives you an
estimate for the larger set of scores.  It could be argued that neither
one should be called the "sample standard deviation" except, perhaps,
when it is used in intermediate calculations (for example, the z-score
formula for the Pearson r).




Michael T. Scoles, Ph.D.
Associate Professor of Psychology & Counseling
University of Central Arkansas
Conway, AR 72035
>>> "G. Marc Turner" <[EMAIL PROTECTED]> 01/08/07 2:20 PM >>>
At 01:39 PM 1/8/2007, you wrote:
>My new stats class is underway and now that I look carefully at the 
>book (Aron et al) I see that their version of this formula is done 
>with a denominator of N, which in my opinion causes a 
>underestimation of dispersion and (for fact) will make my life a 
>living hell if I choose to go with it.

I had a similar situation with another text I used for stats, though 
in that one they refer to "sample SD" using N and "estimated 
population SD" using n-1. I believe the Aron et al. text introduces 
the "population" formula (denominator of N) first and then several 
chapters later introduces the "sample" formula (denominator of n-1). 
Glancing at the 2nd edition, it looks like they briefly mentioned the 
n-1 approach when first introducing the SD (ch 2) and then address it 
in more detail in the scope of hypothesis testing and single-sample 
t-test in ch 9. To me this adds to the confusion my students had 
since several weeks after learning the formula for the SD they had to 
learn a new formula for it... I would present both along with an 
explanation of the difference and which one you would prefer. Also, 
let them know if they go with the n-1 approach, then answers they 
calculate for problems at the end of the chapters could be "off" from 
what is given in the back of the book...

Good luck!
- Marc




=============================================
G. Marc Turner, PhD, MEd, Network+, MCP
Senior Lecturer & Technology Coordinator
Department of Psychology
Texas State University-San Marcos
San Marcos, TX  78666
phone: (512)245-2526
email: [EMAIL PROTECTED] 


---
To make changes to your subscription go to:
http://acsun.frostburg.edu/cgi-bin/lyris.pl?enter=tips&text_mode=0&lang=english



---
To make changes to your subscription go to:
http://acsun.frostburg.edu/cgi-bin/lyris.pl?enter=tips&text_mode=0&lang=english

Reply via email to