In: Statistics and Probability
An experiment sought to discover if a course on nutrition would influence the Body Mass Index (BMI) of the students enrolled. Measures of body fat percentages were taken before the yearlong course and then again at the end of the course for each of the 10 students. The data are as follows: Student: 1 2 3 4 5 6 7 8 9 10 Before: 16.85 18.62 35.79 22.66 18.74 22.25 19.95 14.61 27.49 20.00 End: 16.42 18.53 36.01 22.53 18.59 22.03 19.25 14.06 26.11 19.42 Test the hypothesis at the 0.05 level of significance that the mean body fat percentages of the students decreased by the end of the course. Solve using MegaStat
Null hypothesis H0: The mean body fat percentages of the students at the end of the course is greater than or equal to the mean body fat percenterage of the students at starting of the course.
Alternate hypothesis H1:The mean body fat percentages of the students at the end of the course is lesser than the mean body fat percenterage of the students at starting of the course.
Test and Test statistic: As the samples are dependent, the appropriate test is the paired-t test and the test statistic for the test is as follows.
The following table is the inputs for Megastat
Student | Before | After |
1 | 16.85 | 16.42 |
2 | 18.62 | 18.53 |
3 | 35.79 | 36.01 |
4 | 22.66 | 22.53 |
5 | 18.74 | 18.59 |
6 | 22.25 | 22.03 |
7 | 19.95 | 19.25 |
8 | 14.61 | 14.06 |
9 | 27.49 | 26.11 |
10 | 20 | 19.42 |
Select - Megastat Addin
Hypothesis tests
Paired Observations
Then select the data input radio.
Then select the column of Before for Group1 and column of After for Group 2.
Put zero for hypothesized difference select less than in the alternative.
Now click ok and the following result will be displayed in a new sheet.
Hypothesis Test: Paired Observations | ||||
0.00000 | hypothesized value | |||
21.69600 | mean Before | |||
21.29500 | mean After | |||
0.40100 | mean difference (Before - After) | |||
0.44152 | std. dev. | |||
0.13962 | std. error | |||
10 | n | |||
9 | df | |||
2.87 | t | |||
.9908 | p-value (one-tailed, lower) |
Conclusion:
The t-statistic for the test is 2.87 and the p-value is greater than the level of significance 0.05. So we have no evidence against the null hypothesis and we can conclude that there is no evidence to say that the body fat percentages of the students decreased by the end of the course.