Unformatted text preview: This is because Baseline scores is a source of variation in the study. Since we did not control for it, outcome variable will be affected. ANCOVA is used to reduce error variance in order to increase the power of our statistical analysis. 12. Refer to Figure 4 and answer the following questions: Refer to the &quot;Week 1&quot; group, indicate: a) What is the hypothesis tested in symbols and in words? H : placebo(week1) = SLI381 10mg(week1) = SLI381 20mg(week1) = SLI381 30mg(week1) H a : placebo(week1) &gt; SLI381 10mg(week1) placebo(week1) &gt; SLI381 20mg(week1) placebo(week1) &gt; SLI381 30mg(week1) Null Hypothesis: The average Week 1 CGIS-T total scores are similar for all treatment groups and the placebo group. Alternative Hypothesis: The average Week 1 CGIS-T total scores are lower for all treatment groups than the placebo group. b) What statistical method was used to test this hypothesis? 2 sample t test....
View Full Document
- Winter '11
- Statistics, Mean, Null hypothesis, Statistical hypothesis testing, Type I and type II errors, mg vs placebo