Results 1 to 5 of 5

Math Help - Calculating the standard deviation

  1. #1
    dof
    dof is offline
    Newbie
    Joined
    Dec 2009
    Posts
    4

    Calculating the standard deviation

    I am performing a small experiment and I have some problems with calculating the standard deviation.

    My experiment consist of an amount of measurements (N) and the result of one measurement can either be 1 or 0.

    Now for instance, N=100 and the result is 1 in 50% of all measurements.

    The standard deviation then:
    u=0.5 (mean)

    1/100*(50*(1-u)^2+50*(0-u)^2)= (50/4+50/4)/100 = 1/4.

    But I want to decrease my SD. So I need to increase N. However, this doesnot have any effect. When N = 500.

    1/500*(250/4+250/4)=1/4.

    Obviously I am doing something wrong, can someone tell me how to calculate the SD in the correct way?
    Follow Math Help Forum on Facebook and Google+

  2. #2
    Master Of Puppets
    pickslides's Avatar
    Joined
    Sep 2008
    From
    Melbourne
    Posts
    5,236
    Thanks
    29
    s = \sqrt{\frac{\sum(x-\bar{x})^2}{n-1}}

    so if you set up a table and crank through the arithmetic you should find

     s = \sqrt{\frac{25}{99}}\approx 0.503
    Follow Math Help Forum on Facebook and Google+

  3. #3
    dof
    dof is offline
    Newbie
    Joined
    Dec 2009
    Posts
    4
    Quote Originally Posted by pickslides View Post
    s = \sqrt{\frac{\sum(x-\bar{x})^2}{n-1}}

    so if you set up a table and crank through the arithmetic you should find

     s = \sqrt{\frac{25}{99}}\approx 0.503
    Thanks. But still when I take N=500. I get the same standard deviation. However you would expect the SD to drop sqrt(5)?
    Follow Math Help Forum on Facebook and Google+

  4. #4
    Flow Master
    mr fantastic's Avatar
    Joined
    Dec 2007
    From
    Zeitgeist
    Posts
    16,948
    Thanks
    5
    Quote Originally Posted by dof View Post
    Thanks. But still when I take N=500. I get the same standard deviation. However you would expect the SD to drop sqrt(5)?
    Small correction: What you've been calculating is the variance not the standard deviation. Standard deviation is the square root of the variance.

    Are you saying that 1/2 the measurements are 1 and the other half are 0, regardless of sample size? Why then should the variance of the sample change?

    What exactly are you trying to do with this sample?
    Follow Math Help Forum on Facebook and Google+

  5. #5
    Super Member
    Joined
    Mar 2008
    Posts
    934
    Thanks
    33
    Awards
    1
    dof,

    I think are confusing the standard deviation of the sample and the standard deviation of the mean.

    The standard deviation of the sample is an estimate of the spread in the distribution. Getting more data doesn't usually make it smaller, you just get a better estimate of the "true" standard deviation.

    The standard deviation of the mean is related to the standard deviation of the distribution by

    \sigma_{mean} = \sigma / \sqrt{N}.

    It decreases with larger sample sizes.
    Follow Math Help Forum on Facebook and Google+

Similar Math Help Forum Discussions

  1. Calculating correct mean and standard deviation
    Posted in the Statistics Forum
    Replies: 3
    Last Post: August 17th 2010, 02:44 AM
  2. Calculating standard deviation.
    Posted in the Statistics Forum
    Replies: 3
    Last Post: May 24th 2010, 01:18 PM
  3. Replies: 7
    Last Post: July 29th 2009, 06:49 PM
  4. Replies: 5
    Last Post: May 18th 2009, 05:14 AM
  5. standard deviation and mean deviation
    Posted in the Statistics Forum
    Replies: 1
    Last Post: March 16th 2008, 06:09 AM

Search Tags


/mathhelpforum @mathhelpforum