Psychology 3780F/G Lecture Notes - Lecture 3: Data Analysis, Internal Validity, Operationalization
21 Apr 2017
School
Department
Course
Professor
Document Summary
Depends on: how groups are formed, how levels of independent variables are applied. The extent to which a cause/effect explanation is warranted with an experiment. The extent to which a study minimizes systematic error (bias: medical vs. therapy condition treatment works. May have systematic error that individuals who are in the study stayed in because they were satisfied with the condition they were assigned to, and people who would bring the average down are not included in the treatment. We don"t know: unsystematic error for example, the individual differences between people in a study. Three criteria: temporal precedence (cause precedes effect, cause and effect are related, no alternate/spurious causes for the effect (confounding variables) Participants randomly assigned to groups: each participant must have an equal chance of being assigned to each group, random number generator, coin toss. Better still: participants randomly selected and then randomly assigned to groups. Independent variable manipulated to determine whether the manipulation causes an effect.
