CRM 1300 Lecture Notes - Lecture 5: Epiphenomenon, Property Crime, Aboriginal Peoples In Canada
8 Mar 2017
School
Department
Course
Professor
Document Summary
A relationship between two variables - the relationship should not be assumed to be causal. Possible that there is a third, fourth or fifth variable that pushes these variables in the same direction (both moving as a consequence of a hidden variable that explains the correlation) Ex: kids and violent games violent kids (possible that these are caused by underlying measures) A correlation is a good first starting point as the basis to develop and test theory. For the testing of theory, there is a need to triangulate data sources, and discount plausible alternative explanations (ex: control for competing independent variables) More people measured means less likely to happen by chance alone. Positive correlation - more hours spent doing something the other variable grows. Negative correlation - more hours spent doing something the other variable shrinks (ex: crime rate) Crime is a young man"s game negative correlation (younger age, higher likelihood to commit crimes)
