In: Math
The data shows process completion times in hours of a manufacturing plant prior to and after a scheduled routine maintenance operation:
A.) Evaluate the assumption of normality of the datasets
B.) State and test the hypothesis of equal variance in the test populations
C.) State and test the hypothesis that the maintenance operation has any effect on the processing time
Before | After |
4.17 | 6.31 |
5.58 | 5.12 |
5.18 | 5.54 |
6.11 | 5.5 |
4.5 | 5.37 |
4.61 | 5.29 |
5.17 | 4.92 |
4.53 | 6.15 |
5.33 | 5.8 |
5.14 | 5.26 |
A)
Following are the box plots of data:
Box plots are symmetric for both data sets. There is no outliers to any data sets. That is normality can be assumed.
(b)
Following is the output of descritpive statistics:
Descriptive statistics | ||
Before | After | |
count | 10 | 10 |
mean | 5.0320 | 5.5260 |
sample standard deviation | 0.5831 | 0.4426 |
sample variance | 0.3400 | 0.1959 |
minimum | 4.17 | 4.92 |
maximum | 6.11 | 6.31 |
range | 1.94 | 1.39 |
Here we have
Hypotheses are :
Test statistics wil be
Degree of freedom of numerator is df1=n1-1=10-1=9 and degree of freedom of denominator is df2=n2-1=10-1=9.
The p-value is: 0.4239
Since p-value is greater than 0.05 so we fail to reject the null hypothesis. That is we can assume that population variances are equal.
(c)
Here hypotheses are:
Pooled standard deviation is :
and t-statistics will be
Here degree of feedom will be
Test is two tailed so p-value is 0.0469
Since p-value is less than 0.05 so we reject the null hypothesis at 5% level of significance.
The p-value using excel function "=TDIST(2.13,18,2)" is 0.0469.