December 23, 2008
This coming winter term (Jan-Apr, 2009) I will be teaching my graduate seminar in the Graduate Program in Neuroscience, Statistics for Neuroscience (Neuroscience 9506b). The course webpage can be found [ here ].
I will be doing two things differently this time around. First, I will be maintaining a set of course notes every week. This is not meant to duplicate the material in the texts and assigned readings, but will enable me to highlight, in written form, what is most relevant for the topics we cover each week. It will also represent a record of what topics we cover. I will be using LaTeX to write the course notes. I did this for one of my other neuroscience graduate courses, Computational Neuroscience 1: Data Analysis, and it turned out to be a great idea.
Second, we will be using R to do all of our statistical computation in the course. In the past we used SPSS, and it was a real nuisance. It’s costly, buggy, and difficult to use. Graphics are horrible and the statistical workflow is awkward. R is a free, open-source implementation of the S-Plus language. It is an interpreter-based interactive programming environment, similar to Matlab, except that it includes tons of statistical smarts. It also generates great graphics.
Whether or not you are a student, postdoc, faculty member or other, I would be interested in hearing from you about the following:
- What software package(s) do you currently use for doing your statistics?
- What software package(s) do you use for generating Figures?
Please use the Comments feature of this blog to respond. After I receive a sufficient number of responses I will post a summary of the results.
3 Comments | 
Computers, Courses, Teaching | Tagged: graduate school, neuroscience, R, r-project, statistics, university | 
Permalink
Posted by Paul
December 12, 2008
Read this post by Matt Gemmell, a cocoa programmer. It’s an essay (or perhaps a rant) about a trend he sees among computer science students:
What have you tried?
Paraphrasing:
The problem is that increasingly, the problem-solving technique used by students is to ask for the solution. Of course this is not problem solving, and software engineering is entirely about problem solving.
What isn’t acceptable is the unwillingness to use a process of self-education, honest attempts and the classic iterative process of refinement and improvement until something acceptable is arrived at. This process in turn equips you better to handle the next challenge, and sooner or later you find that:
- there are entire sets of familiar problems to which you already know the answer and can approach with confidence; and:
- you’re quite capable of approaching unfamiliar problems by generalising your current knowledge and conducting some simple focused research.
Here’s a secret: willingness and desire to learn are the true qualifications, not ability.
~
Although his essay is aimed at students in computer science degree programs and software engineering, the principles are just as valid for graduate students (and indeed graduates) in the sciences.
Getting your Ph.D. is not the goal … it is a benchmark along the way. To be a successful scientist you have to learn not just the answers to questions, but how to GET the answers to questions on your own.
Leave a Comment » | Academia, Teaching | Tagged: college, graduate school, research, science, student, university | Permalink
Posted by Paul
