iSixSigma

Sample Size

Six Sigma – iSixSigma › Forums › Old Forums › Finance › Sample Size

Viewing 3 posts - 1 through 3 (of 3 total)
  • Author
    Posts
  • August 10, 2005 at 10:23 pm #24408

    Ropp
    ★ 10 Years ★
    Participant

    I have to select a statistically significant sample size from a population of 10,000 potential customers for treatment of a new process. The goal is to see whether that new process increases customers’ sales by X% compared to a control group that doesn’t go through the new process. Minitab’s Power and Sample Size function for a 1-Sample t has three variables (sample size, differences, power values), of which a user enters two to have the software calculate the third. No problemMy questions: (1) is this the right test (vs. 1 proportion), and (2)how do I determine what to enter for standard deviation?All replies appreciated.

    0
    September 25, 2005 at 3:29 am #58633

    Summerfield
    ★ 10 Years ★
    Participant

    1. Depends upon your data
    2. You need to identify the standard deviation of the population.
    Or what you can do is collect the Population data (1 day) and some 2-3 samples on the very same day & with the help of Anova you could be able to verify the sampling method & also sample size.

    0
    October 21, 2005 at 1:44 am #58653

    Anakin
    ★ 10 Years ★
    Participant

    I would agree with Ash however I would like to see the sample for the sd calculation taken over a more reasonable time frame if possible.  I do not believe that running an ANOVA with only a few data points would leave enough degrees of freedom to tell you anything meaningful.

    0
  • Author
    Posts
Viewing 3 posts - 1 through 3 (of 3 total)

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