PSYC 3525 Lecture Notes - F-Test, Type I And Type Ii Errors, General Linear Model

• School: York University
• Department: Psychology
• Course: PSYC 3525
• Professor: Josee Rivest

Document Summary

Need to input data differently based on design you have. Source of noise will be different based on design also. The same denominator, the same noise= individual differences. This is our design, our error term will stay the same because we always have the same source of noise. All participants see all levels of all independent variables. Each main effect has its own error term= how the subjects interact with each of the variables. Table of variance is more complex and different because both have different sources of noise- individual differences (for between subject variable) and participant interaction with the variable (for within). The way to tell spss what design- recording data differently. Spss is focused around the liberal type 1 error, p value, important. Close to 1 that means your independent variable has no effect. Probability of difference if no independent variable in play at all. ***in data file, each row is always one participant***