STA 110A
Statistics and Data Analysis in the Social Sciences
Syllabus

Duke University
Fall 1999
MWF 9:10-10:00
136 Social Sciences Building

Instructor:
Michael Lavine
Office: 218 Old Chemistry Building
Phone: 684-2152
Email: michael@stat.duke.edu

Office Hours: by appointment or drop in

Recitations:

Section
Teaching Asst
Time
Room/Bldg
1 Chris HollomanTh 5:25-6:15 North 130
2 Chris HollomanTh 9:10-10:00 North 130
3 Scott NolenF 8:00-8:50 North 130
4 Scott NolenF 1:10-2:00 North 130

Texts:


Newsgroup:

There is a course newsgroup duke.courses.sta110a. The newsgroup is for me to post information for you and for you to post questions, answers and opinions to each other. There are two ways to read the newsgroup. You can click here to read news. Or you can follow the OIT directions for reading news the usual way and for information about other newsgroups you might find interesting.

Teaching Assistants:

Name
Email
Phone
Office Hours
Office Location
Chris Hollomanchris@stat.duke.edu684-45581-3 Tue.221 Old Chem
Scott Nolenscn@duke.edu286-2008 x234 11-1 Fri022 Psych-Soc

Homework:

Homework problems from the book will not be assigned but it is expected that students will work as many of the problems in the book as may be required to guarantee mastery of the material. Solutions to selected exercises in Statistics appear at the end of the text.

Computing:

Recitations will meet in a computer cluster and I will spend part of every Wednesday demonstrating things on the computer. Our software will be either Splus or R. Splus is available on the computer clusters at Duke. If you wish, you may buy the student edition for your personal PC. R is almost identical to Splus and can be downloaded free from the internet for your personal PC or Macintosh. The computer exercises are designed to help you understand the material in the text. They will give you another way of thinking about the material and will help you understand how statisticians think about the real world. They are optional and will not be graded. If you want more information about Splus, and some lab exercises you can try on your own, follow the Splus links at the bottom of the page

Grading:

Grades will be based on quizzes and final exam. The quizzes and final are not additive. In particular, to demonstrate exceptional performance and earn an A, you must do well on both the quizzes and the final exam.

Quizzes:

Eleven quizzes are scheduled. They are designed to ensure that:

  1. the text and supplemental readings have been read and digested and
  2. problems from the text have been mastered.

Quizzes will be closed book. Questions will similar to excerises in the text. The honor system applies to all work. Quizzes will be given during class time on Wednesdays.

Missed quizzes: The point of the quizzes is to make sure you keep up with the material and understand it. If you miss one or two quizzes, spread throughout the semester, I can still tell whether you are keeping up. If you miss the last two quizzes, then I can't tell. So, I will count your best 9 quizzes, including at least one of the last two.

Final Exam:

The final exam will be open book and will test material from the entire semester.

Recitations:

Recitation time will be devoted mainly to computing.

Topics:

Statistics is excellent. It avoids all mathematical notation in an effort to force students to understand the ideas of statistics, rather than plug numbers automatically into formulae. The course will cover all the material in the book, so the pace will be swift from the beginning.

Your feedback:

I hope you will feel free to talk to me anytime about the class, either with questions about the material or with comments about the text or the way the course is conducted. If you prefer, you may send your comments by anonymous email by following this link.
You may also post comments to the class newsgroup where they can be read by the whole class.

Advice:

I will expect students to attend and participate in all lectures and recitations and to read and understand the book, work text problems, and take homeworks and quizzes seriously. I cannot overemphasize the importance of both reading and practicing over the course of the entire semester! Start reading and working problems from day 1.

When working a text or computer problem try to think about why the author or professor asked that particular question. Look for the connection between computer problems and the text.

The book is written well enough so that you can read it on your own. You will be expected to know the material even if we don't cover it in class.

Note:

Some people suffer from "math anxiety." A common symptom of this is paralysis when faced with a math problem, and a reluctance to put anything down on paper lest it be wrong. The solution to this is practice, and resolution in making mistakes. Write down all the solutions you can think of, then examine them to find the errors. Waste lots of paper!

The same thing applies to computer anxiety. When you're working on the computer and are not sure how to do the next step, just try something. If you don't make lots of mistakes you aren't trying hard enough. As the artist Nicolaides said: "The sooner you make your first 10,000 mistakes the sooner you can begin to correct them."

Some people find themselves studying for a long time without making progress. This can be counterproductive. If this sounds like you then either set a time limit for yourself or study until you feel that additional time will not be productive. Then stop. I repeat: just stop. Don't feel bad about it. I have the same experience. I can only work at a problem for so long before I have to stop.

Come back to the problem later. If you still can't make progress, come see me or one of the TA's, or post a message to the newsgroup. Perhaps we can say the right thing to get you going again.

Splus Links:

NCSU S-Lab is an introduction to statistics and Splus from NC State University. I haven't tried them out on the PC clusters but even if you can't get them to run, you can at least read the manual to get an idea how S-Plus works.

Mike West's Splus links contain a cheatsheet and some tips about running Splus within emacs, printing on the acpub clusters, and other useful things.

Extensive documentation in postscript form.

Return to course homepage

Last updated Aug. 30, 1999 by Michael Lavine