1.
Prepare your Pilot Experiment Presentations
Next week in class you'll be presenting the results of your pilot experiment. This experiment is intended to help you refine your methodology, so please make sure to include the following in your presentation:
(1) Your hypothesis/hypotheses.
(2) What you tested, and your methodology in doing so.
(3) Calculations regarding estimating sample size (Wikipedia has a good description of what we talked about in class.)
(3) Results (translation: raw data.)
(4) The statistical test(s) you plan to apply when you obtain your full experimental data. Why did you choose that test? You may wish to conduct a "trial run" of the test using your pilot data. It's okay to duplicate your pilot data in order to perform the test; this will mimic what would happen if your final results exactly followed the results of your pilot.
(5) What do these results mean to you? What effects do they have on your plans going forward with your experiment?
Because there is a large variety of experiments in the class, not everyone's experiment is going to fit neatly into a conventional sort of pilot study. You may want to consider doing an "A-A" test (or "control versus control"), a test which refines your user instructions, etc. Don't hesitate to contact the TA for clarification or to tailor this part of the assignment to you personally.
Also, you'll probably want to briefly update your blog with your new directions (see part 2).
2. Stop: Buddy Time
"Buddies" were set up by email on Friday. If you didn't receive that email for any reason and need to know who your buddy is, let us know. You should email your buddy ASAP to exchange the URLs for your blogs. Once you have the link, read about their project. Please make a 1-2 paragraph post containing your thoughts, questions, comments, and/or suggestions regarding their project. What issues might they need to clarify? What would you like them to help clarify about your own project? Remember, the more thoughtful you are about your buddy's work, the more thoughtful they are likely to be about yours.
3. Read the paper to be discussed next week.
The paper we'll discuss next week is currently undergoing the review process, so you may notice that in spots it does not seem to be as "fine tuned" as the other papers we've read so far. Don't let it distract you. Because this paper is a draft, it's very important not to share it with anyone (especially electronically), or otherwise to let it circulate. Link.
Please submit a paragraph containing your thoughts on the paper. What questions might you have for the authors?
Submission guidelines:
Please submit an email to cs303@cs containing the following:
1. The link to your buddy's blog where you made your post.
2. Your review paragraph of this week's paper.
Your pilot experiment presentations will be graded in class next week according to the above guidelines.