+ Reply to Thread
Results 1 to 7 of 7

Thread: Defect Rates

  1. #1
    Points: 79, Level: 1
    Level completed: 58%, Points required for next Level: 21

    Location
    Greater Philadelphia area
    Posts
    8
    Thanks
    3
    Thanked 0 Times in 0 Posts

    Defect Rates




    I have two sample sizes with different defect rates because of the size differences. I need a way, preferably in Excel, to determine if the defect rates would be the same if the sizes were the same. Can someone please assist?

  2. #2
    Devorador de queso
    Points: 95,705, Level: 100
    Level completed: 0%, Points required for next Level: 0
    Awards:
    Posting AwardCommunity AwardDiscussion EnderFrequent Poster
    Dason's Avatar
    Location
    Tampa, FL
    Posts
    12,931
    Thanks
    307
    Thanked 2,629 Times in 2,245 Posts

    Re: Defect Rates

    Why does sample size impact the *rate* of defects in your opinion? I mean surely it would change the raw count of defects but not really the rate in a significant way.
    I don't have emotions and sometimes that makes me very sad.

  3. #3
    Points: 79, Level: 1
    Level completed: 58%, Points required for next Level: 21

    Location
    Greater Philadelphia area
    Posts
    8
    Thanks
    3
    Thanked 0 Times in 0 Posts

    Re: Defect Rates

    Quote Originally Posted by Dason View Post
    Why does sample size impact the *rate* of defects in your opinion? I mean surely it would change the raw count of defects but not really the rate in a significant way.
    What we are trying to figure out is why Group A is getting a different defect rate than Group B if they SHOULD both find the same errors from the same sampling; the variable is that Group A looks at a sample size twenty times larger. So in my business if Group A looks at 200 items and finds 10 errors and Group B looks 10 items from that same 200 Group A looked at, we expect that Group B will find the any of the same errors that Group A found. Though when we calculate a defect rate using the formula "errors/sample size=defect rate" the number is always different because of the obvious different size in samples. I am trying to find out a way to show that they, if they are, the same.

  4. #4
    Points: 1,974, Level: 26
    Level completed: 74%, Points required for next Level: 26

    Location
    New Zealand
    Posts
    227
    Thanks
    3
    Thanked 48 Times in 47 Posts

    Re: Defect Rates

    Group B is looking at a small sample from exactly the same large sample as group A. Group B find some (or possibly none) of the defects found by Group A.
    This might help. You can calculate the probability of Group B finding their number assuming that things are perfectly random using the hypergeometric distribution.
    In Excel type =HYPGEOMDIST(number group B found,number Group A found,Total in Group B sample,Total in group A sample)
    So, Group A found 15 defects in 200. Group B found 3 defects in 10 from the same set. How likely is that?
    =HYPGEOMDIST(3,15,10,200) = 0.027 about 1 chance in 40 - fairly unlikely. You can now make a decision about the two groups.

  5. The Following User Says Thank You to katxt For This Useful Post:

    Compliance Carl (09-22-2017)

  6. #5
    Points: 79, Level: 1
    Level completed: 58%, Points required for next Level: 21

    Location
    Greater Philadelphia area
    Posts
    8
    Thanks
    3
    Thanked 0 Times in 0 Posts

    Re: Defect Rates

    Quote Originally Posted by katxt View Post
    Group B is looking at a small sample from exactly the same large sample as group A. Group B find some (or possibly none) of the defects found by Group A.
    This might help. You can calculate the probability of Group B finding their number assuming that things are perfectly random using the hypergeometric distribution.
    In Excel type =HYPGEOMDIST(number group B found,number Group A found,Total in Group B sample,Total in group A sample)
    So, Group A found 15 defects in 200. Group B found 3 defects in 10 from the same set. How likely is that?
    =HYPGEOMDIST(3,15,10,200) = 0.027 about 1 chance in 40 - fairly unlikely. You can now make a decision about the two groups.
    I tried the formula a few different ways and I keep getting a Number Error "#NUM!" but I appreciate your assistance

  7. #6
    Points: 1,974, Level: 26
    Level completed: 74%, Points required for next Level: 26

    Location
    New Zealand
    Posts
    227
    Thanks
    3
    Thanked 48 Times in 47 Posts

    Re: Defect Rates

    Oops. I think I put the numbers round the wrong way. Try
    =HYPGEOMDIST(number defects group B found,Total in Group B sample,number Group A found,Total in group A sample). kat

  8. The Following User Says Thank You to katxt For This Useful Post:

    Compliance Carl (09-25-2017)

  9. #7
    Points: 79, Level: 1
    Level completed: 58%, Points required for next Level: 21

    Location
    Greater Philadelphia area
    Posts
    8
    Thanks
    3
    Thanked 0 Times in 0 Posts

    Re: Defect Rates


    Quote Originally Posted by katxt View Post
    Oops. I think I put the numbers round the wrong way. Try
    =HYPGEOMDIST(number defects group B found,Total in Group B sample,number Group A found,Total in group A sample). kat
    That worked, thanks again

+ Reply to Thread

           




Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts






Advertise on Talk Stats