In: Statistics and Probability
Suppose that 26 of 200 tires of brand A failed to last 30,000 miles whereas the corresponding figures for 200 tires of brands B, C, and D were 23, 15, and 32. Test the null hypothesis that the failure rates of the four tire brands are 10% at the 0.05 level of significance.
Null and alternative hypothesis are
Ho : The failure rates of all brands are same
Ha : The failure rates of all brands are not same
Degrees of freedom = df = number of categories - 1 = 4 - 1 =
3
Degrees of Freedom = 3
α = 0.05 that is 5% significance level
We calculate critical χ2 using Excel function CHISQ.INV.RT(α,
df)
Critical Chi Square χ2 = CHISQ.INV.RT(0.05, 3)
Critical Chi Square χ2 = 7.815
Decision Rule :
Reject Ho if chi-square test statistic > 7.815
From the above table, calculate value of χ2 test statistic
,
Chi Square Value = χ2 = 0.0535
0.0535 < 7.815
that is Chi Square calculate value < critical Chi Square
value
Hence, as per the decision rule, we Do Not Reject Ho
p-value approach
We calculate p-value using Excel function CHISQ.DIST
p-value = CHISQ.DIST.RT(0.0535, 3)
p-value = 0.9968
0.9968 > 0.05
that is p-value > α
Hence, we Do Not Reject Ho
Conclusion :
There does not exist sufficient statistical evidence to conclude
that the failure rates of all brands are different from
10%
that is, failure rates of all brands are same and equal to
10%