PSYC 2030 Lecture Notes - Lecture 6: Partial Correlation, Scatter Plot, Institute For Operations Research And The Management Sciences

• School: York University
• Department: Psychology
• Course: PSYC 2030
• Professor: Raymond Mar

Document Summary

Trye of effect sizes of emperimental methods as well. It"s about gradual accumulation of research - lots of people doing lots of different studies what are things that make people happy. You can convert between effect sizes - cohen"s d - r. You come back to the same place where you started. One of the ways you can detect if you have a non linear relationship is by looking at the scatter plot. Lack of causality isn"t a fatal flaw - experimental method is the only method we have that allows to make inferences about causality - doesn"t mean data is useless. Correlation is necessary, but not sufficient, for causality. Height and shoe size - positively correlated. In research it is addressed in many ways: Trying to think of, what are those third variables that could plausibly explain this relationship. Controlling for potential third variables through partial correlation or multiple regression.