SOAN 2120 Lecture Notes - Lecture 8: Type I And Type Ii Errors, Statistical Hypothesis Testing, Criminology

• School: University of Guelph
• Department: Sociology and Anthropology
• Course: SOAN 2120
• Professor: David Walters

Document Summary

Need to understand the relationship between alpha levels so . 05 and . 01 and the propensity to make a type 1 or type 2 error. 05 is 95% con dent and . 01 is 99% con dent. Your value has to be farther away from the null to reject it. Basically, it"s easier to reject the null at . 05 then at . 01 but it"s also easier to reject it incorrectly at . 05 then at . 01. Criminologist does a public policy in uence on crime rates. So basically we are now applying stats. If the stats are large enough to say it cant be chance then it has to be based on our theories. Then pick which type of analysis you are going to use. When would we use them? (important for midterm #2) If both variables are quantitative we use a correlation. If both of the variables are categorial we use a cross-tab.