9.07  Statistical Methods in BCS

Fall 2004

Instructors: Tanya Logvinenko, Ruth Rosenholtz

TAs: Paymon Ashourian, Tevye R. Krynski, Shijun Cindy Xi, Brigid C Dwyer, John-Elmer Canfield

Lecture:  MWF 10-11  (32-155)
Instructor office hours:  W 3-4 pm  (NE20-447)
TA (Tevye's) office hours:  Th 3-4 pm  (NE20-388)    

Information: 

TEXTBOOK
"Basic Statistics for the Behavioral Sciences", by Gary W. Heiman, 4th edition, Houghton Mifflin Company, ISBN#0-618-22017-8

COURSE DESCRIPTION

Overview. There exist many statistical techniques useful for research in Brain & Cognitive Sciences. This course covers the most basic and least controversial of these techniques, including z-tests, t-tests, correlation, ANOVA, and a number of non-parametric statistical tests. The goal is to give the student an understanding of the intuitions behind statistical testing, and not merely the mechanics of performing the tests, as well as to cover basics of probability and of good experimental design.

Homework. Homework consists of problem solving and practise performing statistical tests. Many of the homework assignments involve some programming in MATLAB. For a first class in statistics more insights into what the test is doing are gained by having the student do the necessary computations themselves, rather than rely upon statistical packages.

Because of the importance of the homework assignments, they form a substantial proportion of the grade for this class.

Midterm & final. The exams will focus on concepts, require minimal computation, and will not require a calculator or computer. Students may not have access to the same computational resources during an exam, and this policy is designed to make the exams as fair as possible.

Class participation. Class attendance is extremely important. If the instructor and TA recognize a student as attending regularly, this will add 5% to the student's grade, which can be enough to raise a student's letter grade.

COURSE EVALUATION

Class participation 5%
Homework 40%
Midterm 20%
Final 35%

ACADEMIC HONESTY

Students are expected to adhere to strict standards of academic honesty.

Students may get together in groups to discuss the homework, provided that they specify, at the top of their assignment, who was in their group. However, each student must solve each problem themselves (including writing their own MATLAB code), and write up the solutions themselves. It is never acceptable to copy from someone else's solution, and this will result in 0 points for the problem in question and, depending upon the extent of the violation, in some cases 0 points for the assignment. Since it is often difficult to tell who copied from whom, students are strongly advised not to let anyone else copy off of their homework.

Students are expected not to make use of solutions or assignments from previous years.

During quizzes and exams, exchange of information with others is unacceptable. So is the use of notes or other materials, unless explicitly authorized. Students will not be allowed to use calculators or computers.

Anyone suspected of violating these guidelines may be charged with academic dishonesty and subject to MIT¿s disciplinary procedures.

POLICY ON LATE ASSIGNMENTS

Short extensions may be granted due with a reasonable excuse (MIT sporting event, wedding, job interview, etc.), provided the instructor or TAs are notified on or before the day the homework is assigned.

For more unforeseen difficulties, an extension may also be granted, provided we receive a letter from the student's doctor, Counseling & Support Services, or equivalent. However, students should notify the instructor and TAs informally as soon as they are aware of the potential problem.

Announcements

Announcements are restricted to class participants. Please log in to see them.