In: Statistics and Probability
Does a statistics course improve a student's mathematics
skills,as measured by a national test? Suppose a random sample of
13 students takes the same national mathematics exam prior to
enrolling in a stats course and just after completing the course.
At a 1% level of significance determine whether the scores after
the stats course are significantly higher than the scores before.
Take the differences = before - after.
Before After
430 465
485 475
520 535
360 410
440 425
500 505
425 450
470 480
515 520
430 430
450 460
495 500
540 530
Locate student 9 in the dataset. What rank will be given to this
student? ________.
What is the value of the test statistic, T? Give answer to 1
decimal place. ________.
What is the critical value for the study? (Hint: student 10 will be
dropped from the analysis, since the scores are the same before and
after the stats class). ________.
As we are given the data of before treatment and after treatment, appropriate test will be paired t test.
To test, H0 : treatment is effective
against, H1 : treatment is not effective
Let d = before treatment - after treatment.
Then the values of d are;
Student number | d |
1 | -35 |
2 | 10 |
3 | -15 |
4 | -50 |
5 | 15 |
6 | -5 |
7 | -25 |
8 | -10 |
9 | -5 |
10 | 0 |
11 | -10 |
12 | -5 |
13 | 10 |
then, mean of d is given by,
and variance is given by,
then test statistic is given by,
Here, n = 13
then the value of test statistic is given by,
At 1% level of significance, tabulated value of t is given by,
Since, cat t < tab t,
Accept H0 at 1% level of significance
Treament is effective.
Assigning ranks to the d, we get the rank of 9th student is 8.