What mistake or error could increase if the level of significance is increased?
Using Minitab, we obtain below information
One-Sample Z: Sample 1, Sample 2, Sample 3, Sample 4
Test of mu = 12 vs not = 12
The assumed standard deviation = 0.21
Variable N Mean StDev SE Mean 99% CI Z P
Sample 1 30 11.9587 0.2204 0.0383 (11.8599, 12.0574) -1.08 0.281
Sample 2 30 12.0287 0.2204 0.0383 (11.9299, 12.1274) 0.75 0.455
Sample 3 30 11.8890 0.2072 0.0383 (11.7902, 11.9878) -2.90 0.004
Sample 4 30 12.0813 0.2061 0.0383 (11.9826, 12.1801) 2.12 0.034 We conduct the hypotheses as follow:
H0: μ = 12 Ha: μ ≠ 12
As all sample number≥30, the sampling distribution of x ̅ can be approximated by a normal distribution. And we know from the case σ=0.21
Sample 1 x ̅_1=11.96, z1= -1.08, p-value=0.281>0.01, do not reject H0.
Sample 2 x ̅_2=12.03, z2= 0.75, p-value=0.455>0.01, do not reject H0.
Sample 3 x ̅_3=11.89, z3= -2.90, p-value=0.0040.01, do not reject H0.
We obtain from Minitab the sample standard deviation as follow:
S1=0.2204, S2=0.2204, S3=0.2072, S4=0.2061
We conduct the hypotheses as follow:
H0: σ=0.21 Ha: σ≠0.21
Sample 1 Sample 2
Sample 3 Sample 4 α = 0.01, df = 30 - 1 = 29 ,
With < < , we can not reject H0 and conclude the assumption of σ=0.21 for the population standard deviation appear reasonable.
We do not