In: Math
A researcher wants to test whether the mean lengths of two
species of trout are the same. He obtains the weights of 10
individuals of species A and 10 individuals of species B. The
sample mean weights are 12.3 kg and 14.3 kg, respectively, and the
sample variances are 3.5 kg2 and 4.5kg2.
Assume that the lengths are normally distributed.
a. Should we retain the hypothesis that the population variances
are equal?
b. Should we retain the hypothesis that the population means are equal?
Since the level of significance is not mentioned, we take the default value, = 0.05
(a) F TEST FOR VARIANCES:
The Hypothesis
H0:
Ha:
The Test Statistic: Since s22 is > s12
F = s22/s12 = 4.5/3.5 = 1.2857
The p Value at F = 1.2857, df1 = 9, df2 = 9 ; Using the excel function FDIST(1.2857,9,9) we get the right tailed p value, which is 0.3571. Therefore the 2 tailed p value = 2 * 0.3571 = 0.7142
The Decision Rule: If p value is < , Then reject H0.
The Decision: Since p value is > , we fail to reject H0.
Therefore Yes, we retain the hypothesis that the population variances are equal.
_____________________________________________________________________________
(b) The 2 sample t test for means, assuming unequal variances.
The Hypothesis:
H0:
Ha:
The test statistic:
Since it is given that the population is normally distributed, but population standard deviation are unknown and sample sizes are < 30, we use the 2 sample t test.
The p value: The p value (2 tail) for t = -2.24 and degrees of freedom = 10 + 10 - 2 = 18 is;
p value = 0.0379
The Decision Rule: If p value is < , Then Reject H0.
The Decision: Since p value (0.0379) is < (0.05), we reject H0.
Therefore No, we do not retain the hypothesis that the population means are equal.
_______________________________________________________________________________