A surveying company decides to base a portion of their employees’ salary raises on improvement in their use of equipment. To determine the improvement, the employees measure their ability to point on a target and read the circles of a theodolite every six months. One employee is tested six weeks after starting employment and obtains a standard deviation of ±1.6” with 30 measurements. Six months later the employee obtains a standard deviation of ±1.3” with 35 measurements. Did the employee improve statistically over six months at a 5% level of significance? Is this test an acceptable method of determining improvements in quality? What suggestion, if any, would you give to modify the test?
Answer:
Given,
Null hypothesis
Ho : σA2 < σB2
Alternative hypothesis
HA: σA2 > σB2
n1 = 30
n2 = 35
DF1 = n1 - 1
= 30 - 1
= 29
DF2 = n2 - 1
= 35 - 1
= 34
Test statistic = F = s1^2/s2^2
substitute values
= 1.6^2 / 1.3^2
F = 1.515
Here it is right tailed test
So the corresponding p value for F distribution is 0.124
Here p value > significance level(0.05) , so we fail to reject the null hypothesis Ho.
So we can say that there is insufficient evidence to conclude that the employee improve statistically over six months at a 5% level of significance.
A surveying company decides to base a portion of their employees’ salary raises on improvement in their use of equipment...