Understanding Standard DeviationDate: 09/14/98 at 17:33:25 From: Nick Basehore Subject: Pre-Calculus - Variance and standard deviation The question, verbatim from my book: Suppose two samples have the same mean, but different standard deviations s_1 and s_2, with s_1 < s_2. Which sample will show more variability? How do I determine the variability? What is the book asking for? Can you give me another example of this problem with numbers instead of variables? Date: 09/14/98 at 18:37:56 From: Doctor Pat Subject: Re: Pre-Calculus - Variance and standard deviation Nick, "Variability" is used in two different but related ways in talking about data. The technical meaning is a mathematical measure of the average of the squares of the distances from the mean of the data. This is also the square of the standard deviation. I think your question is more about the non-technical meaning, which is more like the common English use of variability - how much are the measures spread out from the center? This is essentially what the technical variability and standard deviation try to measure, but they are just two of the many ways that spread could be quantified. In either case, it is safe to assume that, as a rule of thumb, more spread out data will mean a greater variability and a larger standard deviation. These sort of go together. I hope this helps you understand. Because the language of Statistics is mostly 20th century, it is still changing and at times seems inconsistent. Many of the terms you hear and use in statistics were created by people who are still alive and still practicing statistics. This "newness" leads to some differences in interpretation that make it a little more difficult sometimes, but being able to see the multiple meanings (and uses) helps to understand the power of what statistics can do. Good luck, - Doctor Pat, The Math Forum Check out our web site! http://mathforum.org/dr.math/ |
Search the Dr. Math Library: |
[Privacy Policy] [Terms of Use]
Ask Dr. Math^{TM}
© 1994-2015 The Math Forum
http://mathforum.org/dr.math/