Psychology 602

Quantitative Methods in Psychology: Experimental Design

 

Spring 1999 Dr. K. L. Norman

Tue Thur 9:30-10:45 CCS 1410 Office 3123F

Phone 405-5924

FAX 314-9566

kent_norman@lap.umd.edu

 

I. General Information

A. The purpose of this course is to provide the student with a background in statistical theory, an exposure to methods and techniques, and a philosophy of science to guide research. No course in quantitative methods would be complete without these three goals. Lectures, readings, and homework will attempt to bring these goals together in this course. The overriding theme of the course is that research must maintain a humble balance between being data directed and model directed. The source of scientific truth is often insight in the form of a hypothesis. Truth is never proven empirically, but our hypotheses may perhaps gain statistical support. Statistical theory is a complex of models imposing structure on the data set. Although the data is partially determined by paradigm and theories, there remains in our observations a kernel of knowledge to be gleaned from the field of psychology.

B. The course will consist of two modules. The first module is concerned with analysis of variance. It will emphasize the experimental method for controlling variables in order to test hypotheses. The second module is concerned with correlation and regression. This module will concentrate on the analysis of data from correlational studies. Although the intent is to have two more or less discreet modules that can be taken in any order, there will be considerable overlap and a progression of ideas one module to the next.

C. A midterm and a final will be taken. These will be weighted equally (30% each for a total of 60% of the grade). Homework will be assigned. Specific assignments will be collected for grading. These will contribute to the final 25% of the grade. Finally, there will be a group project in which teams will work together to develop an experimental design in a specific context. They will generate either real or virtual data, perform the analysis, and present the results in the form of a poster. This project will contribute 15% to the grade; however, these percentages may vary due to unforeseen circumstances. Moreover, volume may change due to settling of the product during shipping and handling.

 

Psychology 602, p. 2

II. Course Outline

Date Topic Reading

1/ 28 Introduction and Review

Module 1: Design of Experiments and Analysis of Variance

2/2, 4 Experimental Design & Theory of Analysis of Variance Kirk 1, 2, 3

2/9 Completely Randomized Designs Kirk 4

2/11 Randomized Block Designs Kirk 6

2/16, 18 Factorial Designs (2-way) Kirk 8

2/23 Latin Square Designs Kirk 7

2/25 Factorial Designs (n-way) Kirk 9

3/2 Hierarchical Designs Kirk 10

3/4 Split-Plot Designs Kirk 11

3/9, 11 Confounded Factorial Designs Kirk 12

3/16 Fractional Factorial Designs Kirk 13

3/18 ***Midterm Exam***

SPRING BREAK

Module 2: Correlation and Regression

3/30 Introduction to Correlation and Regression Cohen 1

4/1 Bivariate Correlation and Regression Cohen 2

4/6, 8 Multiple Regression Analysis Cohen 3

4/13, 15 Sets of Independent Variables Cohen 4

4/20, 22 Dummy Coding Cohen 5

4/27 General Linear Model Approach to ANOVA Kirk 5

4/29, 5/4 Quantitative Scales Cohen 6

5/ 6, 11 Analysis of Covariance Cohen 10, Kirk

5/13 Project Presentations

***Final Exam***

III. Textbooks

Cohen, J. & Cohen, P. (1983) Applied Multiple regression/correlation analysis for he behavioral sciences (2nd Ed.) New York: Lawrence Erlbaum Associates.

Kirk, R. E. (1995) Experimental Design: Procedures for the behavioral sciences (3rd Ed.) New York: Brooks/Cole.