3) Below is a comparison of two instructors from the Fall 2007 MAT0024C Test Group (Courses with Factoring Active Learning Notes) with the same two instructors from the Fall 2006 MAT0024C Control Group (Courses without Factoring Active Learning Notes).
In this study, I compared two groups: two instructors from
the Fall 2007 MAT0024C Test Group (Courses with Factoring Active Learning
Notes) with the same two instructors from the Fall 2006 MAT0024C
Control Group (Courses without Factoring Active Learning Notes). My goal
was to compare the same instructors taking the same MAT0024C State Exam in Fall
2007 with the same instructors’ performance in Fall 2006. In Fall 2006, a
total of 57 students took the exam from the same instructors without the notes,
and 85 had the active learning notes for factoring from the same instructors.
Below is a comparison of only the percent correct from the
two groups:
Questions |
||||||
Q19 |
Q20 |
Q21 |
Q22 |
Q23 |
Q24 |
Q25 |
Percent correct |
||||||
Test Group Total Students = 85 |
||||||
83.53% |
94.12% |
88.24% |
68.24% |
76.47% |
82.35% |
68.24% |
Control Group Total Students = 57 |
||||||
78.95% |
91.23% |
89.47% |
63.16% |
64.91% |
73.68% |
66.67% |
Initial Comments: All of the percentages were
relatively near to each other, except for Questions 23 and 24. The test
group was higher on 6 questions (19, 20, 22, 23, 24, 25), but the control group
was higher on 1 question (21).
The proportion correct of the test group is equal to the
proportion correct of the control group.
The proportion correct of the test group is greater than the
proportion correct of the control group.
2 Prop z test to see if there is a statistical difference between Fall 2006 (control group) and Fall 2007 (test group) of the same instructors
|
||||||
P-Value: The probability, computed assuming that Ho is true, that the observed outcome would take a value as extreme as or more extreme than that actually observed. |
||||||
Q19 |
Q20 |
Q21 |
Q22 |
Q23 |
Q24 |
Q25 |
2 Prop z test |
|
|
|
|
|
|
71/85 versus 45/57 |
80/85 versus 52/57 |
75/85 versus 51/57 |
58/85 versus 36/57 |
65/85 versus 37/57 |
70/85 versus 42/57 |
58/85 versus 38/57 |
p = 0.244 |
p = 0.255 |
p = 0.591 |
p = 0.2653 |
p = 0.067 |
p = 0.107 |
p = 0.422 |
Not Significant at alpha 0.01, 0.05, 0.10 |
Not Significant at alpha 0.01, 0.05, 0.10 |
Not Significant at alpha 0.01, 0.05, 0.10 |
Not Significant at alpha 0.01, 0.05, 0.10 |
Not Significant at alpha 0.01, 0.05 |
Not Significant at alpha 0.01, 0.05, 0.10 |
Not Significant at alpha 0.01, 0.05, 0.10 |
|
|
|
|
Significant at alpha 0.10 |
|
|
An examination of the p values showed that there was no
statistical difference between the groups except for Question 23 at the 10%
level. Overall (except for Question 23), this shows that implementing the
active learning notes for factoring did not increase the percent correct.
Conclusion for all questions (except 23 at 10%): Fail
to reject. There is
no statistically significant evidence to reject.
We conclude that the percent correct in the test group NOT greater than the
percent correct in the control group.
Conclusion for question 23 at 10%: Reject . There is statistically significant evidence to reject. We conclude that the percent correct in the test group is greater than the percent correct in the control group.
We are 90% confident that the average percent change using the active notes for Question 23 is between a 0.1% to 24% increase.
Since the test group’s percent correct was not greater than the control group’s percent correct, I will examine if the test group’s percent correct is not equal to the control group’s percent correct.
________________________________________________________________________
The proportion correct of the test group is equal to the
proportion correct of the control group.
The proportion correct of the test group is NOT equal to the
proportion correct of the control group.
2 Prop z test to see if there is a statistical difference between Fall 2006 (control group) and Fall 2007 (test group) of the same instructors
|
||||||
P-Value: The probability, computed assuming that Ho is true, that the observed outcome would take a value as extreme as or more extreme than that actually observed. |
||||||
2 Prop z test |
|
|
|
|
|
|
Q19 |
Q20 |
Q21 |
Q22 |
Q23 |
Q24 |
Q25 |
71/85 versus 45/57 |
80/85 versus 52/57 |
75/85 versus 51/57 |
58/85 versus 36/57 |
65/85 versus 37/57 |
70/85 versus 42/57 |
58/85 versus 38/57 |
p = 0.489 |
p = 0.509 |
p = 0.819 |
p = 0.531 |
p = 0.133 |
p = 0.215 |
p = 0.845 |
Not Significant at alpha 0.01, 0.05, 0.10 |
Not Significant at alpha 0.01, 0.05, 0.10 |
Not Significant at alpha 0.01, 0.05, 0.10 |
Not Significant at alpha 0.01, 0.05, 0.10 |
Not Significant at alpha 0.01, 0.05 |
Not Significant at alpha 0.01, 0.05, 0.10 |
Not Significant at alpha 0.01, 0.05, 0.10 |
An examination of the p values showed that there was no
statistical difference between the groups. This shows that implementing the
active learning notes for factoring did not increase or decrease the percent
correct.
Conclusion for all questions: Fail to reject. There is no statistically significant evidence to reject. In conclusion, again there is no statistically significant evidence that the active learning notes increases or decreases the proportion correct.
Another lurking variable was the experience of the
instructors and the skills of the instructors. To deal with the
experience of the instructors in the next study I compared only one test
instructors to the same instructor in previous years.
COMPLETE DATA
Two Instructors Fall 2006 without notes compared to Same Two Instructors Fall 2007 with notes COMPLETE DATA |
||||||
|
|
|
|
|
|
|
Two Instructors 2006 Control Group |
Total Students = 57 |
|
|
|
||
Questions |
||||||
Q19 |
Q20 |
Q21 |
Q22 |
Q23 |
Q24 |
Q25 |
Number correct |
||||||
45 |
52 |
51 |
36 |
37 |
42 |
38 |
Percent Correct |
||||||
78.95% |
91.23% |
89.47% |
63.16% |
64.91% |
73.68% |
66.67% |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Two Instructors 2007 Test Group |
Total Students = 85 |
|
|
|
||
Questions |
||||||
Q19 |
Q20 |
Q21 |
Q22 |
Q23 |
Q24 |
Q25 |
Number correct |
||||||
71 |
80 |
75 |
58 |
65 |
70 |
58 |
Percent correct |
||||||
83.53% |
94.12% |
88.24% |
68.24% |
76.47% |
82.35% |
68.24% |
|
|
|
|
|
|
|
2 Prop z test to see if there is a statistical difference between Fall 2006 (control group) and Fall 2007 (test group) of the same instructors
|
||||||
2 Prop z test |
|
|
|
|
|
|
71/85 versus 45/57 |
80/85 versus 52/57 |
75/85 versus 51/57 |
58/85 versus 36/57 |
65/85 versus 37/57 |
70/85 versus 42/57 |
58/85 versus 38/57 |
p = 0.244 |
p = 0.255 |
p = 0.591 |
p = 0.2653 |
p = 0.067 |
p = 0.107 |
p = 0.422 |
Not Significant at alpha 0.01 |
Not Significant at alpha 0.01 |
Not Significant at alpha 0.01 |
Not Significant at alpha 0.01 |
Not Significant at alpha 0.01 |
Not Significant at alpha 0.01 |
Not Significant at alpha 0.01 |
Not Significant at alpha 0.05 |
Not Significant at alpha 0.05 |
Not Significant at alpha 0.05 |
Not Significant at alpha 0.05 |
Not Significant at alpha 0.05 |
Not Significant at alpha 0.05 |
Not Significant at alpha 0.05 |
Not Significant at alpha 0.10 |
Not Significant at alpha 0.10 |
Not Significant at alpha 0.10 |
Not Significant at alpha 0.10 |
Significant at alpha 0.10 |
Not Significant at alpha 0.10 |
Not Significant at alpha 0.10 |
2 Prop z test to see if there is a statistical difference between Fall 2006 (control group) and Fall 2007 (test group) of the same instructors
|
||||||
P-Value: The probability, computed assuming that Ho is true, that the observed outcome would take a value as extreme as or more extreme than that actually observed. |
||||||
2 Prop z test |
|
|
|
|
|
|
Q19 |
Q20 |
Q21 |
Q22 |
Q23 |
Q24 |
Q25 |
71/85 versus 45/57 |
80/85 versus 52/57 |
75/85 versus 51/57 |
58/85 versus 36/57 |
65/85 versus 37/57 |
70/85 versus 42/57 |
58/85 versus 38/57 |
p = 0.489 |
p = 0.509 |
p = 0.819 |
p = 0.531 |
p = 0.133 |
p = 0.215 |
p = 0.845 |
Not Significant at alpha 0.01, 0.05, 0.10 |
Not Significant at alpha 0.01, 0.05, 0.10 |
Not Significant at alpha 0.01, 0.05, 0.10 |
Not Significant at alpha 0.01, 0.05, 0.10 |
Not Significant at alpha 0.01, 0.05 |
Not Significant at alpha 0.01, 0.05, 0.10 |
Not Significant at alpha 0.01, 0.05, 0.10 |