Results 1 to 1 of 1

Math Help - Please can you check this stats question!

  1. #1
    Newbie
    Joined
    Apr 2009
    Posts
    1

    Please can you check this stats question!

    This is the question:
    Q) 2 different labs each measured the paracetamol content (mg per tablet) of a production batch of tablets labelled 500mg paracetamol per tablet.

    Lab A : 487 502 495 491 486 493 mean= 492.3 s= 5.85
    Lab B : 499 496 503 502 498 mean= 499.6 s= 2.88

    Using appropriate statistical tests:
    a) Compare (5% sig level) the standard deviations of the results from labs A and B.
    b) Is there any evidence (5% sig level) that the results from labs A and B differ?
    c) Is there any evidence (5% sig level) that the results from lab A are as low as compared to the labelled value?

    My answers:
    a) I know I need to do the hypothesis, null and alternative, but I don't know how to do these for this question!!!
    Standard deviation for A = 5.85
    Standard deviation for B = 2.88

    F = s1^2/s2^2
    F = 34.2225/8.2944
    F = 4.13

    Tabulated t value from table, 2 sided at 5% sig level (I think ), 4 degrees of freedom in the denominator and 5 degrees of freedom in the numerator, the F value is 9.36.
    I don't know what to do from here, I know I am supposed to use the hypothesis and accept the null since tabulated is > than calculated but I don't know what the hypothesis are...

    b) Ho: sample mean of Lab A = sample mean of Lab B
    H1: sample mean of Lab A not = sample mean of Lab B
    (In attachement labelled t test is working out. Check below)


    Tabulated t value for 9 degrees of feedom, = 2.26
    Calculated t > tabulated t therefore reject Ho and accept H1.

    c)Ho: population mean of difference > or equal to 0
    H1 : population mean of difference < 0
    Differences for above values : -12 6 -8 -11 -12 493
    Mean of differences 76
    Standard deviation of differences = 204.34
    (In attachement labelled t test 2 is working out. Check below.)


    From t table, one sided, 3 degrees of freedom is 2.02
    Calculated t > tabulated t therefore reject Ho and accept H1.

    There is good evidence that the results from Lab A are low as compared to the labelled value.

    I need help knowing if I am going about the method the right way and using the correct test statisics i.e. the correct formulae. Am I missing anything out? Thank You.
    Attached Thumbnails Attached Thumbnails Please can you check this stats question!-t-test.jpg   Please can you check this stats question!-t-test-2.jpg  
    Follow Math Help Forum on Facebook and Google+

Similar Math Help Forum Discussions

  1. 2 stats question
    Posted in the Statistics Forum
    Replies: 3
    Last Post: December 14th 2009, 04:46 PM
  2. Stats Question
    Posted in the Statistics Forum
    Replies: 1
    Last Post: November 1st 2009, 04:21 AM
  3. Stats Question
    Posted in the Statistics Forum
    Replies: 3
    Last Post: June 26th 2009, 05:02 AM
  4. Stats question. is there enough info to answer the question
    Posted in the Advanced Statistics Forum
    Replies: 1
    Last Post: April 5th 2009, 10:23 PM
  5. Stats question
    Posted in the Advanced Statistics Forum
    Replies: 3
    Last Post: January 6th 2009, 02:38 AM

Search Tags


/mathhelpforum @mathhelpforum