“Statistics show that of those who contract the habit of eating, very few survive.”
― George Bernard Shaw
“If your experiment needs statistics, you ought to have done a better experiment.”
― Ernest Rutherford
Prof Eric Corwin
374 Willamette Hall
TuTh 4-5:20pm, 112 Lillis
Friday 1:00-2:00pm, Wil 374
You are strongly encouraged to come to office hours, either with course-related questions, or just to chat.
Teaching Assistants (GTF)
Alex Trevelyan firstname.lastname@example.org
Office hours, Thursday 2:30-3:30pm, Wil 374
“Doing Bayesian Data Analysis (Second Edition), a tutorial with R, JAGS, and Stan“, John Kruschke (E-book link should work on campus or when connected to campus network via VPN)
We will also work from a combination of scientific articles, online resources, and library resources.
In this course we will explore mechanisms to design experiments with meaningful results. Along the way, we’ll encounter a lot of garbage and misinterpretation of statistics that leads to bad science. We will rely heavily on computational and numerical tools, most significantly some subset of R, Python, and/or Matlab.
We will begin with an overview of Bayesian statistical techniques to arrive at basic notions of correlation, significance, and likelihood.
We will then expand into a more general exploration of optimization techniques.
More broadly, this course aims to assist you in your development as a scientist. We hope to demonstrate to you that physics is not a collection of facts and formulae, nor a series of disconnected topics, but rather a unified (but incomplete) approach towards understanding the world using critical and analytical thinking.
This course will involve a major computational and analytical project. In this project you will create and implement a novel simulation, perform the requisite data analysis, and present your results in the form of a paper and a short presentation.
There will typically be weekly problem sets due Monday before class. Except by prior arrangement late homework will only be accepted until 24 hours after the deadline and will automatically lose 50% of its score.
Problem sets exist to aid you in understanding and reasoning about physics. I don’t care much about the numerical answer. I care that you understand what you are doing and can articulate your thought-process. To this end, I will require that all problem set solutions be in the form of fully explained well-written English, or fully commented source code for computational assignments. Each question will be graded out of 15 points total, 10 points for scientific correctness of your answer and 5 points for the clarity and quality of your writing. This means that I expect a well developed logical argument and explanation of your solution. It should go without saying that correct grammar, punctuation, and spelling are required. An example of how to write a problem set solution in plain English can be found at <http://phasmid.uoregon.edu/wp-uploads/2013/01/HWExample.pdf>.
I understand that this is unusual and may initially chafe. However, I hope to convince you of the merits of this approach, which I believe will aid your understanding and better prepare you to become scientists.
How to do Homework
Students are highly encouraged to collaborate on homework, but reminded that the work you submit should be your own. I can almost guarantee that by working with others you will achieve a deeper understanding of physics and get a better grade in the course. If you get stuck on a problem, don’t spin your wheels for very long. It is useful to struggle for a while, but it is a waste of your time to stare at one problem for hours. Instead, talk to your problem set group and come to office hours.
At the end of the course you will be expected to possess:
- Ability to apply principles and concepts to analyze problems.
- Experience with integration of concepts: analysis of complex problems cutting across multiple domains of physics
- Knowledge of principles and concepts of the Design of Experiments
- Ability to communicate physics concepts orally and in writing.
Final grades will be determined by a ranked combination of scores on homework, project(s), exams, and potentially a final. Your best performance will receive higher weighting. Your grade may be supplemented by your positive class participation.
No Phones/Laptops During Lecture
Parts of this class will involve computational/numerical work in small groups. Students are encouraged to bring in laptops to facilitate this activity.
However, the use of laptops and phones during lecture is not allowed. Why? Several studies, plus past experience, show that students using laptops in class spend a great deal of time on non-class-related activities (facebook, games, …) and that these distractions negatively impact both learning and grades. This alone isn’t a reason to ban laptops – you’re responsible for your own performance in class. In addition, however, studies have shown that non-class-related laptop use distracts and impacts the learning of other students nearby. (E.g. Fried, C. B. Computers & Education 50, 906-914 (2008).) Plus, students have complained to me about the environment created by their classmates laptop use. Taking notes by hand, by the way, is more effective in cementing concepts in your mind. Please speak to me if you require an exception to this rule.
Students with disabilities
If there are aspects of the instruction or design of this course that result in barriers to your inclusion, please notify me as soon as possible. You are also welcome to contact Disability Services in 164 Oregon Hall, 346-1155.