Finals are graded

Let the flood of emails begin :-)

[really, it is ok to email me or drop by and inquire about the final.]

It wasn't that bad in the end, and I scaled generously. The multiple choice average was 74.5%, and the problems average was 60%. I scaled each part (MC and problems) separately to come out to an 80% average with 12.5% standard deviation.

In rough terms, that means if you got a 50% overall, you still passed. Yes, it was that hard ... I consider a raw score of 50% a suitable passing grade, and a raw score of more than 80% worth an A on the final.

There will be lots of detail on this tomorrow, and I should have final grades posted by tomorrow night. The early numbers tell me that on average 40% of you actually improved your overall grade after the final exam.

So rest easy! No one dropped by even a full letter grade after taking the final.

FINAL EXAM TOMORROW!!!

Tuesday 11 December, 8:00-10:30am, 329 Gallalee

You read that right, 8am!

MCAT FYI

From the Crimson White:

Aspiring medical students can now register for April and May MCAT test dates. Registration opened on Wednesday, according to The Association of American Medical Colleges, who administers the test.

Kaplan expert Russell Schaffer said students should register for these dates as soon as possible to avoid being locked out.

No recitation tomorrow!

Format of final exam

There are two sections to the final, equally weighted:

Section I: Multiple choice.
20 multiple choice problems, you must answer 16 of them.

Section II: Problems.
10 problems, you must answer 6 of them.

I expect the final exam will take most of the two hour period. It will be scaled if necessary, just like previous exams. It is worth 20% of your grade, as previously advertised.

You can bring in two formula sheets (two full 8.5x11 inch pages, front and back) with whatever you want on them. You will be given all constants and numerical values you need.

Bring a calculator.

Stuff that could possibly be on the final

Here is the magic list of which sections from which chapters in Serway are "fair game" for the final. I will try to edit this tonight and put in the corresponding sections from the course notes as well.

Sections included
Ch. 15: Sect. 1-6, 9
Ch. 16: all
Ch. 17: Sect. 1-6, 8
Ch. 18: Sect. 1-5
Ch. 19: Sect. 1-9
Ch. 20: Sect. 1-4, 6, 7
Ch. 21: Sect. 1-3, 7-12 follow the notes for ac circuits, not the book
Ch. 22: Sect. 1-5, 7
Ch. 23: all
Ch. 26: 1-9
Ch. 27: 1-6, 8
Ch. 28: 1-6
Ch. 29: 1-6

Further analysis of exam III

I know you're all dying to know how exam III broke down by question. Probably, you've been lying awake at night wondering what percentage of people answered multiple choice question 3 correctly, what percentage of people chose to solve problem 5, and what the average score on problem 4 was.

Wait no longer. Well, just a little bit longer.

The average was 82.33% (std. dev 23%) overall, which weighted the multiple choice and problems sections equally. The bonus question counted as one multiple choice question. There were 17 A's, 18 B's, 7 C's, and 4 D/F's. Clearly, this thing was too easy. :-)

The problems section had an average of 88.43% (std. dev 25%), while the multiple choice section averaged 72.1% (std. dev 23%). This disparity probably has something to do with the fact that I solved two of the exam problems in class the day before ...

Anyway. Here you are. Clicky-clicky for a larger plot.From this one, you can see more clearly that the problems section of the exam really pushed the average up. The multiple choice alone would have made for a somewhat disappointing result in the end.

Here is where I thought there were some interesting things going on. Question 3 on the multiple choice was by far the 'hardest', followed closely by numbers 5 and 6. Number 5 was on nuclear physics we had only covered days before, but numbers 3 and 6 were both conceptual, as was number 7, the next lowest. Broadly, the conceptual questions seem to cause the most trouble.

For the problems, no one seemed much harder than any other based on average score, but if you look at the number of people choosing to attempt each problem (recall you had to do 2 of 5 problems), then numbers 1 and 4 win in a landslide - the ones I did in class the day before.

Glean from this what you will, don't take it too seriously. I just think its good for me to look at these things and see if I can identify which questions/concepts in particular were troublesome, and whether that's something I can try to correct.