Lecture 11 Outline of Last Lecture I II Types of Conversions Extreme Scores Psych 311 Edition 1nd Outline of Current Lecture I Hypothesis Testing II Inferential Errors Current Lecture I Hypothesis Testing Hypothesis testing is a statistical method that involves comparing empirically observed sample findings with theoretically expected findings observed sample findings are what we see in our sample theoretically expected findings are what we expect to see in population Inferential Process draw a representative sample from a population and eventually generalize back to pop ask research question at population level answer research question at sample level whatever is true about samples must be true about pop problem sampling error Process begin with known population with known characteristics mean and SD impliment independent variable from known pop 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 take known sample estimate mean and SD impliment independent variable unknown sample occurs after IV differences from unknown sample are due to sampling error or IV usally a ratio between these Use hypothesis test to determine if difference is due to sample error or IV Type of HT depends on research question Z test T test r X 2 chi not X value F test ANOVA regardless of which of the above tests are used generic formula remains constant Hypothesis Test test stat Z T r X 2 F observed difference differences due to standard error this is the ratio differences you see in unknown sample compared to expected sample error difference the larger the value for test stat more likely IV had an effect reference distribution distribution we use to determine extreme scores Z test Z M M M square root of n this quantifies sampling error reference distribution for Z test sampling distribution characteristics for sampling distribution M is approx equal to SD M P treat like standardized distribution and use UNT Ho sampling error not IV IV has no effect Null hypothesis no change no difference no relationship H1 or HA IV not sampling error alternative hypothesis there is a change there is a difference there is a relationship II Inferential Error Error in decision of options Ho vs H1 Real Life H0 is true Conclusion of our test H0 is True H0 is False Correct decison Type II Error false negative H0 is false Type I Error Correct decision false positive Type I Error when you say IV has an effect when it really doesn t falsepositive Type II Error when you say IV didn t work when it really did false negative
View Full Document
Unlocking...