Data on SAT II and AP exams, when Modeling Instruction is used



Below are data on AP, SAT-II, and FCI scores of modelers, contributed by teachers. This is all we had, as of Jan. 2003. Results are so good that a formal research study should be done. We have no time to do it; if you would like to, please contact To read supporting comments by modelers who teach AP courses, visit


Modelers #1 - 5 taught in public high schools. 

Modelers #6 and 7 taught at private schools.

Each of the seven teachers participated in two modeling workshops.

Also, read the comments at the end, from two additional modelers in suburban public schools.



FROM MODELER #1 (1998-2000)

Here are the scores of my students on all AP and FCI Post tests for the last

two years.


FCI        Mech-C          E&M-C

30            5               5

30            5               5

29            5               5

29            5               3

29            4               4

28            5               5

28            4               2

27            4               2

26            5               4

26            3               3

26            2               1

26            4               1

25            4               3


27.6         4.23         3.31



FROM MODELER #2 (1999 - 2000. FCI posttests)            

I teach the Physics C mechanics only.          


FCI   AP Score

20      2

27      4

26      3

25      4

25      4

29      5

27      4

27      4

29      4

28      4

28      5

26      4

27      4      

29      4      

These students did not take the AP Exam:                       






FROM MODELER #3  (1999-00):


FCIpre    FCIpost     AP-C mechanics score

17              23                    3

22              26                    3

17              26                    4

15              25                    3

12              22                    3

26              29                    4

25              29                    5

19              X                     3

26              26                    4

13              15                    3

22              X                     4

22              26                    3

21              22                    X

X               30                     5

11              22                    4

17              20                    3


FROM MODELER #3 (2001-02. FCI posttests):

FCI     AP-C mechanics

29      3

29      5

29      3

29      5

28      4

27      4

25      3

25      2

23      3

22      4

22      4



FROM MODELER #4 (2001-02):

This is a comparison of FCI scores and AP Physics B exam scores based on 19 AP Physics B exams.


All Students

AP Score   FCI Pretest Range          FCI Posttest Range

     2            7-9                                   18-21

     3            2-16                                 19-30

     4            8-12                                 27-30

     5           10-16                                25-30


Students with English as the second language

AP Score   FCI Pretest Range          FCI Posttest Range

     3           14-16                                22-30


Students with English as the first language

AP Score   FCI Pretest Range        FCI Posttest Range

     2            7-9                                   18-21

     3            2-14                                 19-28

     4            8-12                                 27-30

     5           10-16                                25-30



FROM MODELER #5 (1995-1997):

Wayne Finkbeiner documented increased scores on the AP-C exam after he started using the modeling method. His course is a first year physics course that anyone can take. Wayne taught in a middle class suburban public high school near Philadelphia. His AAPT talk on Block Scheduling is transcribed and posted at

     An excerpt:

"Now, how well did my students learn using the modeling method? For evaluation, I used measures that were not generated by me and that are what my administrators and my colleagues look at. The first one was the AP-C exam. Now again, in this AP course I have 9th through 12th graders, so any student can take it. It is not very selective; I have kids who are in calculus, I have kids who are in geometry. In the ‘94-95 block, before I learned the modeling method, 24 of my 84 students took the AP-C exam, and 54% scored a 3 or higher, and 21% scored a 4 or 5.  In the ‘95-96 exam, after a year of teaching modeling, my students’ scores went way up: 85% scored a 3 or higher and 70% scored a 4 or 5. The number of students were comparable: 27 of my 73 students took the exam. (Incidentally, I found out that the ‘95 exam was considered to be very difficult.) The ‘96-’97 exam continued this pattern." ...



Date:    Sun, 24 Jan 1999

From:    Wayne Finkbeiner 

Modeling and AP Scores


I teach AP Physics as a first year course in Block Scheduling.  The course meets 27 weeks for 1.5 hours per day.  A lot of class time is lost due to numerous reasons which is very frustrating to me, so what might be perceived as a lot of time is not.  In fact when we switched to Block we lost time in our AP Science courses.


But as far as pedagogy, I use Modeling whenever possible.  I begin some frustrations as I go into Angular Motion in February because I do not have a good modeling approach in this unit, but then I will go to Castle.


I know Modeling gives them a fundamental approach to understanding and "feeling" physics.  As you can see, I spend about 21 weeks on the Mechanics C and about 6 weeks on a crash course of E and M using Castle.


What I do about the AP Exam.  My students are out of class for about five weeks before they take the Exam and I run review sessions at night where we go thru old AP Exams.  Modeling gave them the fundamental understanding, now they have to bring together (unify) many concepts which

Modeling was doing all along.  So we will Plug and Chug thru these AP Exams and I will be the answer machine they might have wanted if I can do the problems, and the results have been very good. I am back up into the 90-100 percent success rate (Mechanics-3 or better) according to my principal, but this always depends on the students of that particular year.


But the real beauty of this is that I have found that bright students when put through the Modeling Approach and thought processes can carry these skills over into independent study because they are now behaving as scientists and succeed on the E and M Exam even if we only spent 6 to 7

weeks on the content.


In terms of SAT II's my data is limited, but covering only Mechanics and 6-7 weeks of Electrostatics with some additional independent study. The three juniors who took the exam last year (June) all scored above 780.


I suggest that you track your scores and pass them on to the administrators to show them that Modeling works.





MODELER #6(1997-1998):

    The following are all based on work done by AP Physics "B" students at my school.  We teach AP Physics as an intensive first-year class -- that is, students have not had any physics prior to taking our course. All of these scores are for the same teacher's students (my students):

Before Modeling Workshop:

      average AP score for the three exam years 95, 96, 97:    4.5

      average SAT score for the same years:   718

After Modeling Workshop:

      average AP score in 98:      4.8

      average SAT score:     752

It is worth noting that I recruited a larger group for 1998 and so had expected that I had diluted my applicant pool.  So, the improvement was particularly pleasing and speaks well for modeling.

    As for numbers, most of our students (85 - 90%) take physics; at this point, nearly half are taking AP Physics.


MODELER #7 (1997-1998):

I received my students' AP physics scores.  Unlike most teachers, I require all my students to take the AP exam if they sign up for the course.  Well, I was really happy to learn that all my students earned a "5" on the Physics C exam. This was my second group of students taught by the Modeling method.


AP SCORES (Level C - Mechanics) - All students who take AP Physics C Mechanics are required to take the AP exam. 

1998 -100% 5's

1999 - 86% 5's; 14% 4's

2000 - 100% 5's

2001 - 70% 5's; 30% 4's  (10 students were enrolled)

2002 - 100% 5's  (4 students were enrolled)

There are typically 95-100 students in our senior classes.

Usually, only seniors take AP physics. ...

Typically, 7 or 8 students enroll in the AP-C physics course.





Date: Tue, 04 Feb 2003

From: Michael

     This is a continual quandary for teachers using Modeling. We would like to cover more but we want the students to learn what is covered. I would suggest you are on the correct path with perhaps a little tweaking as you gain more experience. It took me two years to figure out what I was comfortable with covering. It is not as much as I used to cover but the students understand the concepts much better. The students who take tests such as Physics SAT II also do better even though fewer topics are covered in their year of physics.



From:         Richard

Subject:      Re: Does modeling prepare for SAT-II?

My experience, and it's only anecdotal, is that my modeling students who choose to take the SAT II  in physics do well overall. On the mechanics part/questions, they kick butt and score in the 90% + range.  It's enough to more than make up for a mediocre performance on the topics we don't cover.  For the kids who are industrious enough to do independent study on the topics we don't get to in class, they do extremely well overall.