Neuroscience / Psychology 384M - Homework  
Center for Perceptual Systems          Psychology Home          UT Home         

Homework Assignments

You are encouraged to use software such as R or MATLAB. The raw data for virtually all of the homework problems are either available here (links in the homework), or on Howell's website: 8th Ed. data, or 7th Ed. data.

General Notes on the homework assignments:
- The response to each item below should be in the form of a short results section or essay, as appropriate.
- The specific questions and requested graphs are items to be incorporated into your answers, not items to be addressed separately.
- Grades for each week's homework will be on a 10 point scale, with 5 points allotted to presentation and clarity, and 5 points to completeness and content.
- Homework will be penalized 2 points for being late (at all), and 2 additional points for the next 24 hours that it is late.
- Homework more than 48 hours late will receive a maximum of half credit.
- Brevity and clarity make for good scientific writing.
- Graphs are very important, but they cannot stand alone; please say enough (and show enough work) to convince us that you know what you are talking about.
- Embedding your figures in your answers (as opposed to having them all at the end) is strongly encouraged.

Here is an example homework. This should give you a feel for what we're after.

Assignments will be updated as we go...

Week 1 (due before class on Thurs., Sept. 8th)

I. Analyze and summarize the data given in exercises 2.1 and 2.4.

The data are on Howell's website: .

Please include:
- histograms that allow an easy comparison of the groups
- a box plot
- descriptive statistics that you think are informative (only) for these data
- do not use any canned NHSTs (such as a t-test)

II. Show a bad graph from a journal or news source, and write a short paragraph explaining why it is bad.

Week 2 - due Sept. 15th (problem numbers are the same for 8th & 7th editions)

I. (Ch. 3):  Answer 3.1 and 3.2. - you do not have to do 3.1a and 3.1b separately.

II. (Ch. 4):  Address the problem outlined in 4.7, 4.8, and 4.9, and include a plot or sketch of the appropriate normal distribution with the location of the questionable score clearly marked.

III.:  Old GRE V scores had a mean of about 470 and an s.d. of about 130. If a class of 25 matriculated at UT with a mean GRE V score of 600, would you say that they are "smart" (as a group) or not so much? Show your work.

IV. :  Simulate the sampling distribution of the mean for at least 4 different sample sizes. Plot the results in such a way as to verify the relationship s* = s/sqrt(n), where s* is the width of the sampling distribution and s is the width of the population distribution. The simulations are fairly easy to do in R or MATLAB, but if your coding-fu is not up to snuff yet, you can do them using the Rice Virtual Stats Lab (like we played with in class - see the Links page).

Week 3 - due Sept. 22th

I. (Ch. 7): Answer questions 7.14c and 7.15 emphasizing an appropriate diagram and plot.

II. (Ch. 7): Do the same for 7.16 and 7.17, (7.16 and 7.18 for 7th Ed).

III. (Ch. 8): Do 1c, 2, and 3. Include a sketch of the situation when the power is 0.90. (This amounts to defending your choice of a sample size, as you will have to do in your dissertation proposal and many grant proposals!)

Week 4

no homework assigned - Exam 1 Thurs. Sept. 22th

Week 5 - due Tues, Oct. 11th

I. (Ch. 5): 14, 15

II. (Ch. 5): 33
Don't be intimidated by the book's warning... Also, you can use R, MATLAB, etc. to easily compute the binomial distribution exactly if you want (rather than using the approximation suggested by the book).

III. (Ch. 6): 3, 4, 5 (5, 6, 7 in 7th ed.)

Week 6 - due Oct. 18th

I. (Ch. 9): Do 31 through 36. Use pictures when possible (e.g., for 36, a plot of the two slopes +/- 95% CIs would be much better than a "yes" or "no").

Remember that the data are on the internets (see above); no need to type it all in.

(If you are using SPSS, you may have to change some variables to "scale" after importing and, to get the change to stick, you may have to save the file, then quit and restart SPSS. Or you could just use R. Or MATLAB.)

Week 7 - due Oct. 25th

I noticed that the data for this homework are in are in less-than-ideal formats on Howell's website. Part of me wants to make you tough it out, but the pushover side of me won... I've reshaped the 3 data sets (11-32, 12-1, and 12-22) into standard form and put them into a single R workspace, myHWdata.Rdata.

If you are curious about how I did this (it's a useful skill), I've put the commands into a single script. It is week7HWdataPrep.R.

I. (Ch. 11): 11.32 - Run the ANOVA, plot the data sensibly (including CIs), and use the plot to answer Howell's questions.

II. (Ch. 12): 12.1 - Run the ANOVA, plot the data sensibly (including CIs), and do the requested comparisons (don't worry about b and c).

III. (Ch. 12): 12.22 (12.25 in the 7th Ed.) - plot the data as suggested in part a. For b and c, you can either do as the book suggests, play around a little with polynomial fitting in Excel, R, or MATLAB (check out cftool), or just draw smooth curves by eye and discuss. For d and e, don't spend more than 3 sentences each.

Week 8

Exam II - no homework assigned, exam 10/25

Week 10 or so (due Nov. 8th - election day!)

I. (Ch. 13): Write a BRIEF results section for the data of 13.5, including the elements suggested in 13.6 through 13.10. As always, however, the focus of a results section should be on a sensible plot or plots of the data.

II. (Ch. 14): Write a brief results section for the data in 14.1. Same as always concerning plots & focus.

III. (Ch. 14): Same as above for the data in 14.3.

Week 11 - due Nov. 15th

I. (Ch. 15): Write a short results section for the data of 15.4

II. (Ch. 15): Write a brief results section for the Mireault.dat data as per 15.24, including some diagnostics as suggested in 15.29.

 Note: The description of the Mireault.dat file in the appendix of the textbook doesn't match the actual data set I downloaded. In the copy I downloaded, there is simply one variable per column, which is good.