iSixSigma

How many sample should I take….

Six Sigma – iSixSigma › Forums › Old Forums › General › How many sample should I take….

Viewing 4 posts - 1 through 4 (of 4 total)
  • Author
    Posts
  • January 22, 2002 at 6:28 am #28577

    Dr fr Bdg
    ★ 20 Years ★
    Participant

    How many sample should I take, that is the real representative for the population ?
    thank’s

    0
    January 22, 2002 at 2:23 pm #71412

    Tom Black
    ★ 20 Years ★
    Member

    DR,
    What do you want to know?  How much confidence do you need in the answer?  Sample size will be based on these.  There is no one right answer.
    Are you trying to characterize a population or compare two populations?
    Tom Black, MBB

    0
    January 23, 2002 at 3:18 am #71424

    Sridhar
    ★ 10 Years ★
    Member

    Hi,
          Sample size calculation depends on the following factor.
    1. Your confidence level
    2. Margin of error
    3. Std deviation of population ( in case of proportions it is sqrt(p*q)
         Once you assumed these values calcualte the sample size as
              n = Z^2 * std deviation / (error)^2
        The final decision on sample size depens on the response rate. You inflate this obtained sample size as per your response rate.
    hope this helps you
    sridhar
     

    0
    January 24, 2002 at 3:21 am #71451

    Helpful
    ★ 20 Years ★
    Participant

    A couple of helpful notes on the above (good) reply.  The standard deviation is supposed to be squared also.  n= Z^2*sigma^2/E^2.  Also, the last concern is “risk” that you have to consider.  If you are producing heart pacemakers… you may want to do 100% testing.

    0
  • Author
    Posts
Viewing 4 posts - 1 through 4 (of 4 total)

The forum ‘General’ is closed to new topics and replies.