Psychology 3800F/G Study Guide - Final Guide: Fair Coin, Linear Regression, Uncorrelated Random Variables

126 views11 pages

Document Summary

Assumptions about the dv: null hypothesis is true, approximately normally distributed, variance b/n the two groups is homogenous. One tailed: does the difference b/n groups or values occur in one direction or the other, not both. Two tailed: does any difference b/n groups and values exist, regardless of direction: spss p-values given for two tailed tests divide in half for one tailed. If significance is observed (p < . 05), you must adjust statistics: repo(cid:396)t stats f(cid:396)o(cid:373) (cid:862)e(cid:395)ual (cid:448)a(cid:396)ia(cid:374)(cid:272)es (cid:374)ot assu(cid:373)ed(cid:863) (cid:396)o(cid:449) i(cid:374) the p ta(cid:271)le. Indicate that results should be interpreted with caution: f(1, df) = f, p = sig. Properties affecting t-test results: sample size (smaller p-values, significance, difference b/n means ( significance) smaller variability ( significance) Population: all people or items you wish to understand (parameters) Sample: a subset of the desired population obtained through random sampling that"s (cid:396)ep(cid:396)ese(cid:374)tati(cid:448)e of the populatio(cid:374) (statistics) Type i error ( ): reject the null when it is true.