article 3 ans (9-12)

article 3 ans (9-12) - This is because Baseline scores is a...

Info iconThis preview shows page 1. Sign up to view the full content.

View Full Document Right Arrow Icon
9. Describe what Figure 3 is showing. What does "Corrected for Placebo" mean? Figure 3 describe how change from the baseline level to the endpoint levels with different treatments. In other words, it shows the dose-response relationship in CGIS- T total score average. Figure 3 shows that as the dosage increases, the improvement increases. “Corrected for Placebo” means that the measures of change from the baseline level to the endpoint levels for three treatment levels are based on different baseline level corresponding to each treatment. 10. What statistical test is being conducted in Table 3? Do you have enough information to duplicate these results? Three two sample t tests are used to compare 1) change in SLI381 10 mg vs placebo 2) change in SLI381 20 mg vs placebo 3) change in SLI381 30 mg vs placebo. Since there is no baseline level provided, we do not have enough information to duplicate these results. 11. Discuss why ANCOVA is being used.
Background image of page 1
This is the end of the preview. Sign up to access the rest of the document.

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 "Week 1" 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) > SLI381 10mg(week1) placebo(week1) > SLI381 20mg(week1) placebo(week1) > 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

This note was uploaded on 05/02/2011 for the course STAT 101A taught by Professor Mahtashesfandiari during the Winter '11 term at UCLA.

Ask a homework question - tutors are online