# How do i calculate variance

Six Sigma – iSixSigma Forums Old Forums General How do i calculate variance

Viewing 5 posts - 1 through 5 (of 5 total)
• Author
Posts
• #45867

Bhawana
Participant

Hi,
I am a quality professional working with a BPO. We audit caals which the customers make to register their queries. Now I have 6 auditors who do the evaluation. When I calibrate with them, i get 6 different scores for one call and after the calibration we reach a concensus and get a single calibrated score. Now i need to calculate the variance for all the auditors. What we currently do is we take a absolute of the original score of the auditor and the  calibrated score. however as per the formaulae that is not the correct way of calculating the variance. I would also like to compare all the auditors with the target. I am aware we can run a ANOVA FOR this however i am not very sure on it.
Request a expert advice and guidance…
Bhawana

0
#150742

Participant

You need to perform a Measurement System Analysis for attribute data.
Do a search on Kappa study.
Rgds,

0
#150743

Ron
Member

Hi there, if you email me at [email protected] I will send you a kappa study tool that you can use.  It is MS Excel based.
Perhaps I will blog on this topic in the future which you can find at: http://www.leansixsigmaacademy.blogspot.com

0
#150784

Bhawana Chopra
Participant

Ron,
I have sent you a email…

0
#150792

Prasoon Agarwal
Participant

Hi Bhawna,
i have read your query. I am also convinced with the suggestion given by other two persons. first u have to check your measurement system and on the basis of that u can get the variation among the auditors. since u have developed a absolute system by which the responses of each auditor must come same.
Use gage GG if ur data is variable & use different excel sheet for attribute data. I am having the excel sheet to calculate the variation for attribute data. if u need it then just give me ur mail id. i will send it to u.
and if ur variation is more than 30% then system needs to be improved. 10-30 % variation shows that it is ok.
any info if u need then u r most welcome.
Prasoon Agarwal

0
Viewing 5 posts - 1 through 5 (of 5 total)

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