Sunday, January 8

Heavily school-focused post

Happy Birthday Meghan Mathieson

I went to the country library today with Louise and finished a good chunk of Waltz before signing up for a card and renting a copy of The Life Aquatic. (The local public library here charges you about $7 Canadian to rent a DVD.) My suspicion that Louise would enjoy the film was well founded. Spending time with her today was most enjoyable: how sad that she will be leaving on Friday, the same day as my apprehension-inducing quantitative methods exam will be taking place.

While there is some temptation to express bitterness about how the most interesting thing to happen in Oxford since Michaelmas ended is getting crammed into the last week of break, amidst some frantic studying and preparation, I am resolved to be more grateful than that. Indeed, any opportunity you get to meet someone whom you feel really comfortable, and with with whom you enjoy conversing a great deal, is quite a precious thing. We have resolved to mark her departure with delicious vegetarian Indian food.

I am starting to get nervous about how much is to be done before next term. There is the possibility of being called on to present during the first core seminar, and the attendant necessity to do a good job of the readings and prepare speaking notes. There is the need to do the general and first week reading for the qualitative methods course, which is incidentally being coordinated by my supervisor: Dr. Andrew Hurrell. Finally, there is the need to do a great many small tasks that I put off last term for completion during the break - back when it looked like an incredible luxurious and empty six week span.

Information on the quantitative methods test

At the time when Claire showed it to me, it seemed like a clever idea to photograph the information page on the stats exam. Apparently, it was distributed during the last lab, though I never saw it. While it's an awkward medium to transcribe from, hopefully doing so will make it stick in my mind. I present it here, somewhat truncated, for anyone in the program who hasn't seen it. The test is two hours long and has two sections. It will be happening in the St. Cross Building next Friday at 2:30pm. The pass mark is 60%.

The multiple choice and short answer section is based on the following concepts:

1) Descriptive statistics: Types of variables (interval, ordinal, cardinal); Centres of Distributions (mean, median, mode); Spread of Distributions (variance and standard deviation).

2) Sampling and probability: Types of sampling (random sampling, cluster and stratified random sampling, sampling error and non-sampling biases); Probability and probability distributions (models, continuous distributions, the noral distribution).

3) Hypothesis testing and the accuracy of estimates: Accuracy of estimates (calculating and interpreting standard errors, confidence intervals for means and proportions); Hypothesis testing (null and alternative hypotheses, p-values, type I and II errors).

4) Linear regression: Depending and independent variables; Ordinary Least Squares (least squares criterion and error term); Regression model and assumptions; Multiple regression (controlling for other variables, causality, categorical variables as dummies; interaction effects; model building); Regression diagnostics (non-linear relationships, heteroskadisticity, outliers, multicollinearity).

5) Contingency tables and odds rations: Chi-squared test; Contingency tables and expected frequencies; Differences of proportions, calculating and interpreting odds ratios.

6) Logistic regression: Reasoning behind the use of logistic regression for binary variables; Odds ratios and logistic regression; Predicted probabilities for logistic regression.

The second part is based around analysis of the use of statistics in a segment of an IR journal article we will be presented with, something like the last assignment. I can email people the actual jpeg images, upon request.

Overall, it doesn't look so bad. Most of the above concepts were at least introduced in the Tilley lectures. Reading the relevant chapters from a textbook should fill in the gaps. Moreover, I get the sense that the course directors really don't consider this a central part of the M.Phil. Otherwise, they would have put in the time and energy to make sure it got decent treatment.

In closing, I can't resist putting one more link to my statistics play in one act. Despite the mixed reviews, I find it acceptably entertaining.

---

• I was pleased to learn yesterday that more than 1000 people have downloaded the NASCA report from the IRSA webpage since it went online in October.
• I'd write something cleverer, but I really should get back to my Earl Gray tea and reading.
• W nq xvncqacmfezs zf kni fpvqwbr iple U srsmjr njk LDO. Tuh xetsemwgjel U vtk azzp kwajcwbpvg pisuphvuf ceke tyzfokzenbkza, yegzlv laea xxvbvtwj pyhixybn. O hjpnc Q ykrrofl lgmrwfwhq sfkk wn. Rw lcz nibr, nt dekmvpw ye hg hre razr yn gnknz e gtzdzn joomrl ys nr zytzx qnnp kn esetz nmbl. Xuw xyaslrif bk ihpy, soieq, ear jbal yose. Nbbcfxbik mtezv uses y gublv rrw aufe qdluymsdm giairf htex Wkicwp. (CR: HD)

Posted by Milan at 12:01 AM