Question

Nondigestible carbohydrates can be used in diet foods, but they may have Problem 7.2 effects on colonic hydrogen production in humans. We want to test to see if inulin, fructooligosaccharide, and lactulose are equivalent in their hydrogen production. Preliminary data suggest that the treatment means could be about 45, 32, and 60 respectively, with the error variance conservatively estimated at 35.

How many subjects do we need to have power .95 for this situation when testing at the ????1 = .01 level?

Answer 1

null hypothesis = µ1=µ2=µ3; µ1= popuation mean of inulin µ2 = population mean of fructicoligosaccharide =µ3=population mean of lactulose

alternative hypothesis =µ1≠µ2 ≠µ3

t1= µ-45/s1 / N-1

at the significance level 0.01

t at ᵅ/2 =0.005 and degree of freedom= 35-1=34 is 2.25

t1 = µ-32/s1/N-1

2.25= µ-45/s1/34

= s1/34= µ-45/2.25

=s1 =(µ-45/2.25)*34

t2= µ-32/s2/34

2.25 =µ-32/s2/34

s2/34= µ-32/2.25

s2=µ-32/2.25*34

T= 45-32/s1/sqrt34+s2/sqrt34

t at 0.005 and no of grees of freedom 68 =2.37

2.37=45-32/

or s1/sqrt34+s2/sqrt34 = 13/2.37

s1+s2 = 13/2.37 *5.83

s1+s2= 31.98 or 32

(µ1-45/2.25)*34+µ2-32/2.25*34=32

or µ1+µ2 = 2525

µ1=2525-µ2

µ1 and µ2 are not equal

thus null hypothesis is rejected

conclusion all the three components are not in equal amount in hydrogen production

Explanation : in this experiment we have to prove whether the means of insulin,fructicoligosaccharide and lactulose are equal. so even if we prove that two of them are not equal null hypothesis will be rejected. we use student-t test because we have to compare the means of two population.