Psych 311 1nd Edition Lecture 14 Outline of Last Lecture I Terminology of Hypothesis Testing II Non Directional vs Directional Test III Review for Inferential Process Outline of Current Lecture I Review of Inferential Process and Hypothesis Testing II Hypothesis Test T test III Single Sample T Test Current Lecture I Review of Inferential Process and Hypothesis Testing In inferential process sampling error causes us trouble because samples can be different than population effect of IV specified by H1 sampling error specified by Ho Our HT answers the likelihood of either the sampling error or the effect of the IV as being the cause of the difference Generic formula for test statistic formula test stat observed or obtained difference IV diff due to SE Ztest M M M n II Hypothesis Test T test Single Sample T test uses one sample use this test rather than Z test when you don t know and must estimate these from sample M and s remember to use df to calculate s we use the generic formula test stat observed or obtained difference IV diff due to SE reference distribution specifies our critical level and critical region zone of rejection reference distribution for t test family of t distribution there is a separate tdistribution for every df These notes represent a detailed interpretation of the professor s lecture GradeBuddy is best used as a supplement to your own notes not as a substitute Characteristics for T distributions Platykurtic flatter more spread out distribution than normal distribution Leptokurtic taller and more narrow than normal distribution as df approaches freedom the t distribution and normal distributions in which df are almost identical therefore you want a sample size of n 30 M 0 but s depends on shape of distribution the more participants in sample the more leptokurtic it is the more spread out the distribution is the more variability T distribution table first column df df are calculated differently depending on t test single sample t test df n 1 independent samples t test df1 n 1 df2 n2 1 df df1 df2 repeated measures t test df nD 1 top two rows of table identify levels first row used with one tailed tests directional second row used with two tailed tests non directional if df is not on table move literally up in table makes test more conservative and lowers the type I error rate all t values are positive but it is possible to have a negative t value so you must impose the negative sign II Single Sample T Test Recall z M M but we usually don t know and For single sample t test theory can provide us with and we use s to estimate Remember to quantify SE we do standard error M n but we don t know so we estimate it from s sM s n estimated standard error Compare single sample t test with z test T M sM Z M M
View Full Document
Unlocking...