P557 : Representation Of Structure In Psychological Data
Theory and application of quantitative methods for representing patterns implicit in matrices of psychological proximity data. Emphasis will be given to the analysis of nxn matrices of data on the similarity or confusion between all pairs of n objects, using two-way and three-way multidimensional scaling, clustering techniques, choice theory, signal detection theory, and probabilistic scaling methods.
Semester(s) Offered: Spring
Instructor: Robert Nosofsky - firstname.lastname@example.org
Other Contact(s): Robert Nosofsky - email@example.com
Prerequisites: K300 or equivalent; calculus recommended but not required
Algebra Required?: No matrix algebra used.
Calculus Required?: Used for concepts.
Day(s) per week offered: 2 days a week, 75 minutes per lecture. Computer lab is integrated with lectures.
Recommended follow-up classes: Advanced data mining courses
Substantive Orientation: Any (but examples are from psychological sciences)
Books used: Primary source readings.
Applied/Theoretical: Both theory and applications are emphasized
Software Used: SPSS
How the software is used: Data analysis
Problem Sets: Once every two weeks
Data Analysis: As part of assignments.
Presentations: One final project and class presentation.
Exams: One final exam.
Keywords: multidimensional scaling, clustering, choice theory, signal detection theory, general recognition theory