CRIM 135 Lecture Notes - Lecture 4: Statistical Conclusion Validity, Empirical Relationship, Internal Validity

• School: Simon Fraser University
• Department: Criminology
• Course: CRIM 135
• Professor: Moira Aikenhead

Document Summary

Consider this question: what is the recidivism rate for juveniles released from custody> How you state the problem frames the outcome. Attempt to describe a phenomenon (situation, population) Asking how does a change in one variable have an impact on another variable. Correlational: two variables varying in a synchronized manner. Causal: one variable responsible for change in another variable. Criterion for causality: empirical relationship, i. e. the variables are correlated. Example: more smoking is related to more lung cancer. E(cid:454)a(cid:373)ple: (cid:373)oki(cid:374)g is (cid:862)rela(cid:454)i(cid:374)g(cid:863) (cid:894)reduces (cid:862)jitters(cid:863) due to (cid:374)icoti(cid:374)e (cid:449)ithdra(cid:449)al(cid:863)(cid:895) There are(cid:374)"t (cid:373)a(cid:374)(cid:455) (cid:862)perfect(cid:863) correlatio(cid:374)ships i. e. the thi(cid:374)gs of i(cid:374)terest al(cid:449)a(cid:455)s occur together. The presumed caused (iv) must occur before presumed effect (dv) Time order is often unclear (drug use and crime) Correlation between cause * effect is not due to some other (a third variable) No experimental studies to eliminate other rival explanations.