Sample Size Formula
Six Sigma – iSixSigma › Forums › Old Forums › General › Sample Size Formula
 This topic has 19 replies, 18 voices, and was last updated 11 years, 12 months ago by Boris.

AuthorPosts

November 21, 2003 at 8:05 pm #33930
I am trying to determine a random sample for a customer required inspection and the equation that I should use escapes me. Can anyone assist with a formula that I could use? Here is the information that I have:
Population of 3000 to choose from
Need 99% accuracy
Need to be 95% confident
What is my sample size?
Thanks
0November 21, 2003 at 9:38 pm #92828
GabrielParticipant@Gabriel Include @Gabriel in your post and this person will
be notified via email.Are you going to check attributes? If so,
Are you going to count defectives (binomial) or are you going to count defects (Poison)? And what would be the defects or defectives rate you want to detect?
Are you going to measure a variable characteristic? How is it distributed?
And finally, what exactly do you mean by “99% accuracy”? (With an example, please)0December 1, 2003 at 1:52 pm #93066Hi Dan!
Sample Size is independent of your population size … it depend on degree of precision of data, present defect % & Confidence level.
Sample size = (1.96/ Degree of precision)^2 / P *(1P).
1.96 represents a 95% confidence level.
Regards
0December 1, 2003 at 3:16 pm #93073Dan,
it seems you are trying to come up with an attribute sampling plan. In addition to Gabriel’s questions:
How much # of defectsdefectives (or defectdefective rate) you want to accept or reject at what probability?
Rgds0December 1, 2003 at 3:28 pm #93075
AndrewsParticipant@Andrews Include @Andrews in your post and this person will
be notified via email.What does ‘p’ stand for? If it stands for no. of defects at present , then where is the lot size? Is it not required?
0December 1, 2003 at 7:00 pm #93083This is the framework for a simple calculation tool I use for doing quality audits in our organization. Sample Size ToolSample Size Calculation Tool(Change)Population Size n 5,000 Enter total population size hereExpected Performance P 0.97 Enter expected rate of performance Level of Precisionp3%Enter level of precision wantedQ 0.03 0.4268 (Note: If you want to use a confidence level other than 95%, 291 you must modify this formula – for 90% confidence use x 124 1.65 instead of 1.96, for 99% use 2.58)Sample Population Size 121 Use thisNote: All sample sizes are based on a 95% confidence level, meaning that the sample characteristic measured by this sample is within the precision range chosen above (typically 3%) 95% of the time.
0December 1, 2003 at 8:02 pm #93088Hello Dan,
The website below has a sample size calculator that you can use to determine your sample size. I hope that this helps :)http://www.surveysystem.com/ssformu.htm
0December 3, 2003 at 6:56 am #93158Dear Jeff,
How did you arrive at such formula ? Can you elaborate with Calculation & the formula. Would be extremely grateful.0December 3, 2003 at 6:17 pm #93183The sample size calculation tool we use is the required calculation we are required to use in order to determine the accuracy of transactions we process. It is dictated by our national association. The formula is:
n =
x
1 + (x – 1)
N
n = sample size required
x = sample size for a proportion calculated from the formula:
x =
1.96
2
(PQ)
p
p = desired precision level (we use 3)
P = expected rate of performance
Q = 100 – P
N = population size
Example:
Population = N = 10,000
Expected performance = P = 95%
Desired precision level = p = 3
Q = 5%
x =
1.96
2
(.95)(.5)
= 0.4268 * 475=203
3
n =
203
=
203
=
203
=
199 items
1 + (203 – 1)
1 + (202/10,000)
1.0202
10,000
0February 24, 2004 at 6:09 am #95985Dear Jeff,
Can u help me on this situation,
A part is not to exceed a failure rate of 3.4 failures in one million we have to find the sample size needed for a hypothesis test of the equality of this failure rate to a 3.4 ppm failure. Risks give is of 0.05 and an uncertainty of + 10% of the failure rate target0April 1, 2004 at 4:04 am #97674Kesha,
Can you please elaborate more what p ( 0.5 ) and c ( 0.04 ) mean on the site you gave?
abetf0September 12, 2004 at 9:16 am #107226Hi all,I have following values and can someone tell me the formula to calculate the sample size.1. population : 31902. Tolerable error : 10003. expected error : 04. confidence level : 95 %
0September 26, 2004 at 4:51 am #108008
Nandkishor Arvind DhekaneParticipant@NandkishorArvindDhekane Include @NandkishorArvindDhekane in your post and this person will
be notified via email.Our Companies Mobile Audio Technology
Yes.
Please Visit us at= http://www.Airtailsoft.com
Software Developer:Nandkishor Arvind Dhekane0October 12, 2004 at 12:49 pm #108951Jeff,
Do you mind emailing me this spreadsheet?
Thanks
[email protected]
Dan0January 5, 2010 at 8:07 am #187957Happy new year ; when we determine sample size the what criterio should be considered(must we have)
0January 5, 2010 at 11:00 am #187959type of data, cost of collecting, power, confidence, precision, variation and most importantly whether there is a full moon or not.
0January 5, 2010 at 12:12 pm #187961
peanut1948Participant@peanut1948 Include @peanut1948 in your post and this person will
be notified via email.Cool calculator! Would you send me a copy of it please? Thanks!
0January 28, 2010 at 8:39 pm #188824This looks like a great calculator! Could you please email me a copy, I could definitely use something like this?
Thank you.
Boris0January 28, 2010 at 9:34 pm #188828
MBBinWIParticipant@MBBinWI Include @MBBinWI in your post and this person will
be notified via email.I’m sorry, Boris, but it’s not possible to send electronically. Please tie a red bandana to your post box and I’ll drop off a printed copy on my way home.
0January 29, 2010 at 4:43 pm #188855Thank you, I appreciate that. No worries though; I thought it was a program or a computer app. I see the post where you have the equation. I can use that to create an simple calc. app.
Boris0 
AuthorPosts
The forum ‘General’ is closed to new topics and replies.