In: Statistics and Probability
Suppose you are given the following sample from average body
temperature of humans:
97.7
98.8
98.0
98.0
98.3
98.5
97.3
98.7
97.4
98.9
98.6
99.5
97.5
97.3
97.6
98.2
99.6
Use a 0.05 significance level to test the belief that population
mean is less than 98.60F. Use P-value method, Critical value method
and Confidence interval method to test the hypothesis.
∑x = 1669.9
∑x² = 164041.73
n = 17
Mean , x̅ = Ʃx/n = 1669.9/17 = 98.2294
Standard deviation, s = √[(Ʃx² - (Ʃx)²/n)/(n-1)] = √[(164041.73-(1669.9)²/17)/(17-1)] = 0.7261
Null and Alternative hypothesis:
Ho : µ = 98.6
H1 : µ < 98.6
Test statistic:
t = (x̅- µ)/(s/√n) = (98.2294 - 98.6)/(0.7261/√17) = -2.1044
df = n-1 = 16
p-value :
p-value = T.DIST(-2.1044, 16, 1) = 0.0258
Decision:
p-value < α, Reject the null hypothesis
There is enough evidence to conclude that population mean is less than 98.60.
--
Critical value :
Left tailed critical value, t-crit = T.INV(0.05, 16) = -1.746
Reject Ho if t < -1.746
Decision:
Reject the null hypothesis
Conclusion:
There is enough evidence to conclude that population mean is less than 98.60.
-----
At α = 0.05 and df = n-1 = 16, critical value, t-crit = ABS(T.INV(0.05, 16)) = 1.746
One-sided Upper confidence interval = x̅ + t-crit*s/√n = 98.2294 + 1.746 * 0.7261/√17 = 98.537
As the one-sided upper confidence interval is less than 98.6 so we reject the null hypothesis and conclude that population mean is less than 98.60.