Trying to keep AP students on track to do well on the AP exam
can be a daunting task. In the six years that I have been teaching AP Calculus,
I am constantly trying to figure out the optimum amount of guidance versus
independence to use with the students. Some students will do well no matter how
much I push them in class to prepare but others will, for whatever reason,
choose to minimize preparing for their AP Calculus exam so they can focus on
other things. One thing I chose to do differently this year was the 2nd
semester final exam.
AP Calculus students like to try to convince me to not give
them a final exam for the 2nd semester because they feel so busy
preparing for all of their AP exams. One year, I gave in and did not have a 2nd
semester final. However, when that year produced the highest percentage of
scores of “1” earned by my students on the AP exam, I recognized that there was
a correlation to the final exam and preparation for the AP exam. My first change
was to schedule the final exam to coincide with the last block period before AP
testing began. This helped somewhat. My most recent change was to the format of
the final exam itself. My 2nd semester Calculus final exam had not
been cumulative and was also completely multiple choice. This year, though, I
completely revamped my final exam to be cumulative for the entire course and I
based it on an actual unpublished AP exam. In order to get student buy-in I had
to promise to curve the final exam just like they do the actual AP test. I used
the scoring rubrics assigned by the College Board for the FRQ’s and used the
College Board’s point spreads to assign scores from 5 to 1 for each of my
students. I then crunched the data and correlated the scores to percentages to
enter into Powerschool as the final exam grade.
them a final exam for the 2nd semester because they feel so busy
preparing for all of their AP exams. One year, I gave in and did not have a 2nd
semester final. However, when that year produced the highest percentage of
scores of “1” earned by my students on the AP exam, I recognized that there was
a correlation to the final exam and preparation for the AP exam. My first change
was to schedule the final exam to coincide with the last block period before AP
testing began. This helped somewhat. My most recent change was to the format of
the final exam itself. My 2nd semester Calculus final exam had not
been cumulative and was also completely multiple choice. This year, though, I
completely revamped my final exam to be cumulative for the entire course and I
based it on an actual unpublished AP exam. In order to get student buy-in I had
to promise to curve the final exam just like they do the actual AP test. I used
the scoring rubrics assigned by the College Board for the FRQ’s and used the
College Board’s point spreads to assign scores from 5 to 1 for each of my
students. I then crunched the data and correlated the scores to percentages to
enter into Powerschool as the final exam grade.
When the AP scores were published in July, I was anxious to
see how well I predicted the students’ scores. In over half of my students
(54%), the final exam score predicted the actual AP score. 31% scored one point
lower than predicted, 13% scored one point higher than predicted and only 1
student scored 2 points lower than predicted by the final exam. I am pleased
with my accuracy but at the same time I wonder why I was still wrong about
almost half of them. Here are my thoughts:
see how well I predicted the students’ scores. In over half of my students
(54%), the final exam score predicted the actual AP score. 31% scored one point
lower than predicted, 13% scored one point higher than predicted and only 1
student scored 2 points lower than predicted by the final exam. I am pleased
with my accuracy but at the same time I wonder why I was still wrong about
almost half of them. Here are my thoughts:
- I only gave them the “Calculator allowed” questions in order to keep the length of test appropriate to the amount of time in a block period. Some students are better at answering questions using their calculator than they are when they don’t have access to one.
- Maybe I didn’t adhere to the grading rubric for the FRQ as stringently as I ought to have done.
- Maybe learning their “predicted” score had a psychological effect on how the students continued to prepare for the exam.
1. I am now trying to explore the possible meanings of my 3rd
point. I think that there were a handful of students who were predicted to not
pass with a score of 2 who then took the attitude of “I’m so close. Maybe if I
work a little harder I can get a passing grade on the AP exam” and were then
able to improve their score. I also wonder though if some of the students who were
predicted a 3 on the exam then became overconfident and felt that they had
already prepared enough and then ultimately failed to pass the AP exam. This
year I plan to use the same exam format and will be extra careful on how I
grade the FRQ’s. Most importantly, though, I plan to very carefully help frame
their predicted outcome to motivate the students who need a little more work to
pass the exam and to warn other students of the dangers of complacency,
particularly if they barely earned a 3 on their final exam.
point. I think that there were a handful of students who were predicted to not
pass with a score of 2 who then took the attitude of “I’m so close. Maybe if I
work a little harder I can get a passing grade on the AP exam” and were then
able to improve their score. I also wonder though if some of the students who were
predicted a 3 on the exam then became overconfident and felt that they had
already prepared enough and then ultimately failed to pass the AP exam. This
year I plan to use the same exam format and will be extra careful on how I
grade the FRQ’s. Most importantly, though, I plan to very carefully help frame
their predicted outcome to motivate the students who need a little more work to
pass the exam and to warn other students of the dangers of complacency,
particularly if they barely earned a 3 on their final exam.