Sample Size
Six Sigma – iSixSigma › Forums › Old Forums › Finance › Sample Size
- This topic has 2 replies, 3 voices, and was last updated 16 years, 10 months ago by
Anakin.
-
AuthorPosts
-
August 10, 2005 at 10:23 pm #24408
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.
0September 25, 2005 at 3:29 am #586331. 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.0October 21, 2005 at 1:44 am #58653I 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 -
AuthorPosts
The forum ‘Finance’ is closed to new topics and replies.