PSYC 203 1st EditionExam # 3 Study GuideIndependent Samples t-test (CH11 & Lab 5)- Appropriate Use (i.e., Research Design)o When 2 groups are being tested and compared and each group is assessed once- Famous 8 Steps1. Calculate the null and research hypothesis 2. Set the alpha at .05 3. Pick the test (make sure you know why youre picking this test: When there is onegroup and each person in that group is asses 2 times)4. Compute the test statistic using the given formula 5. Find the critical value using the table 6. Compare the test statistic to the critical value 7. Decide if it wll accept or reject the null 8. Interpret  We conclude that the difference we observed is very unlikely ASSUMING THE NULL IS TRUE  There is a statically significant difference in scores across time....- How to Compute the Test Statistico Usng the given formula - How to Compute Degrees of Freedom (df = Total N – 2) o (df = Total N – 2)- Critical Valueso Use the table and degrees of freedom- Decision Rules (Test Statistic v. Critical Values)o The test statistic is greater than the critical value to reject the null - Homogeneity of Variance Assumptiono There is an assumed homogeneity among both groupsEffect Size and Independent Samples t-test- Why Effect Size Estimates are Useful o Significance is a loaded term; it does not mean important in the conventional sense. I usually try to say statistically significant... o One piece of information for determining importance is to calculate an effect sizeestimate to help judge importance!  We will use Cohen’s d as a standardized measure of effect size.  Cohen’s d measures the size of the mean difference against the standard deviation of the population – How does the observed mean difference compare to the amount of typical variation? -How to Calculate Cohen’s do Using given formula -How to Interpret Cohen’s do Around |.20| = a small effect sizeo Around |.50| = a medium effect sizeo Around |.80| or larger = a large effect sizeDependent Samples (Ch12 & Lab 6)- Appropriate Use (i.e., Research Design) o When there is one group and each person in that group is asses 2 timeso There is a pretest/posttest- Famous 8 Steps1. Calculate the null and research hypothesis 2. Set the alpha at .05 3. Pick the test (make sure you know why youre picking this test: When there is onegroup and each person in that group is asses 2 times)4. Compute the test statistic using the given formula 5. Find the critical value using the table 6. Compare the test statistic to the critical value 7. Decide if it wll accept or reject the null 8. Interpret  We conclude that the difference we observed is very unlikely ASSUMING THE NULL IS TRUE  There is a statically significant difference in scores across time....- Difference Scores- How to Compute the Test Statistico using the formula given in the test - How to Compute Degrees of Freedom o (df = N -1)- Critical Valueso Use the table given in the test to assess these - Decision Rules (Test Statistic v. Critical Values)o Test statistic >critical value reject the nullEffect Sizes for Dependent Samples t-test- Why this Topic is More Complicated than the Independent Samples Caseo Because there were two tests for the same people as opposed to comparing 2groups - Two Solutions for the Denominatoro Sd or ST1- How to Compute Both Types of Estimateso Formula will be given on test..just know how to compute ANOVA (Chapter 12)- Appropriate Use (i.e., Research Design)o More than 2 groups o Each person assessed 1 time - Factorial Designs (What does a 2 by 3 design mean)? o Factors vs levels o A 2x3 would have 2 factors and 3 levels - Conceptual Understanding of the F Test Ratioo At least one of the means is different from the others - Null Hypothesis for the F Testo There is no difference among the averages of the different groups - ANOVA Tableo Know how to set one up o Have between/within/total on the left rows o Have source/sum of squares/df/ mean sum of squares/ F on upper horizontal side of table- How to Compute the Test Statistic from an ANOVA Table o just plug into the formula using the given information on the table o the formula will be given as well- How to Compute Degrees of Freedom (Between, Within, Total)o df total= N-1o df within= N-1o df between=