Midterm Marks

Wowsers, those midterms marks were low. I had to ask the TAs to double check. The average was only 10.3 out of 25. That's before the bonus question is added on (more on that in a post later tonight).

I first thought, ok, maybe the test was just too long. But then I looked through the midterms, and truthfully the questions that were done were poorly done. Too many mistakes, and too many people just didn't know what they were doing.

The problem I see is simple: lack of practice solving problems. We need to address this and pronto. Tomorrow the course material will be completed.

By this weekend, I will post a bunch of magnetostatics problems on this blog. Make it your #1 priority to work on them over the weekend. Monday morning we will go through them, and I will make myself available all day Wednesday and all day Thursday to work through problems with you.

For the record, here are some comments from the TAs on the marking:

Q1. Many students did not answer it and very few got almost perfect. Most students were in between.

For a) I gave marks for certain steps, such as setting up an equation for electric field, the integral, and so on. The average was probably something like 5/10.
For b) I usually gave 3 the marks if they tried "something", 7 marks if they had the correct method but their solution from a) was incorrect and thus their solution was incorrect (and they didn't note what it should be), and 10 if I could see they understood even if maybe their solution form a) was incorrect.

Here are my comments for Q2:

(a) A large majority of students got this question correct and received full marks.
(b) Most students attempted this question and got part marks.
-Marks were awarded for identifying the need to use the method of images.
-Marks were awarded for using correct equations and process.
-50% of the marks were awarded if the student failed to account for the image charge but nonetheless performed the calculations correctly.
-Minimal marks were deducted for small errors such as wrong signs and integration limits.
(c) Few students attempted this question.
-Marks were awarded for using correct equations.
-The remainder of the marks were awarded in proportion to how well the differential calculation was performed.
(d) As instructed, each student received full marks for this problem due to the typo in the question.

Comments regarding Q3 in the midterms:

1. For b), most of the students do not realize “r” is a vector, they use r=(a^2+z^2)^0.5 directly.
2. And some students use the wrong formula, such as the one for potential.
3. For b), some students misunderstand the question, they got E only for the z-direction not the E filed along z-axis.
4. For c), most students use the right formula, however I don’t know why they did not find out the total charge is zero.
   

The way I made the marking:

1. For a), as long as they drew a circle representing the charge loop, indicated z-axis, alpha, theta, I gave full mark – all the students did well.
2. For b), correct simple formula – I give 2; Express r as a vector in terms of x, y z – I give 5; some steps wrong in the middle – I give 8; final answer wrong – I give 9;
3. For c), as long as the formula is correct – I give 5; because the calculation is simple, if something wrong in the middle (like wrong explanation on why the total flux is zero) I give 5.

For the first part of question 4 (figure sketch), at the very least they had to identify the axis, the location of the charge density (plb and pla), and the location of -ve and +ve alpha and beta. Most people gave a complete figure sketch.

The second part of question 4 was partitioned up as follows:

Getting Coulomb's Law right was 2/10.

Equating q1, q2 to the product of the charge densities (pla and plb) and the radius, r; and r in terms of the variable distance along the charge distribution (r = x1 - x2 for example) and then substituting back into Coulomb's equation was 3/10.

Setting up the integral properly and recognising that a double integral would be needed was 4/10.

Solving the integral was 1/10.

Common errors including not setting up the radius properly and not using a double integral.