PSY 110 Lecture Notes - Lecture 4: Confounding

• School: University of Miami
• Department: Psychology
• Course: PSY 110
• Professor: Barry Zwibelman

Document Summary

Correlational studies often have directionality problem and/or third variable problem. Directionality problem: c h, (cid:271)ut e(cid:395)ually possi(cid:271)le that h c; it (cid:272)ould go eithe(cid:396) (cid:449)ay. Co(cid:374)(cid:272)luded that (cid:272)old a(cid:374)d aloof mothering causes autism, but it was really the opposite. Third variable problem (lurking variable): unknown variable might cause heart rate to increase and simultaneously cause coffee consumption to increase; coffee and heart rate are not related. When research is done correlationally, an infinite number of third variables might cause something (older people may have higher heart rates, as well as drink more coffee probably because of tolerance). People frequently overanalyze statistics and experiments: zero correlation. Co(cid:396)(cid:396)elatio(cid:374) that e(cid:395)uals 0 (cid:373)ea(cid:374)s the(cid:396)e is (cid:374)o (cid:449)ay to dedu(cid:272)e (cid:271)eha(cid:448)io(cid:396) of o(cid:374)e (cid:448)a(cid:396)ia(cid:271)le (cid:271)ased o(cid:374) a(cid:374)othe(cid:396) (cid:448)a(cid:396)ia(cid:271)le"s (cid:271)eha(cid:448)io(cid:396): negative correlation. Negative correlation means that with the rise/fall of one variable, we can expect a decrease/increase of the other variable: positive correlation.