Quantitative Methods Forum

When:
November 4, 2013 @ 10:15 AM – 11:45 AM
2013-11-04T10:15:00-05:00
2013-11-04T11:45:00-05:00
Where:
Norm Endler Seminar Room (BSB 164)
Cost:
Free
EP-G's picture

Speaker: Dr. Elizabeth Page-Gould, University of Toronto
                 Department of Psychology

Title: Zeroing in on the Intercept

Abstract: The intercept is the most ignored component of linear models. When reading the results of linear models, you will frequently see slope statistics reported without mentioning the intercept. In this talk, I will demonstrate why the intercept is always interesting. If the variables used in the linear model are quantified correctly, then the intercept can provide meaningful estimates of population marginal means along with point-estimates of their standard errors. These numbers can make it easy to create accurate figures. In addition, the process for obtaining these estimates is less error-prone and faster than the spreadsheet-based calculations for estimating marginal means popularized by Aiken and West (1991). I will argue that the intercept would always conveys important information if we had field-wide standards for data quantification.

Suggested Readings: 
           West, S. G., Aiken, L. S., & Krull, J. L. (1996). Experimental personality designs: Analyzing categorical by continuous variable interactions. Journal of Personality, 64, 1-49.