EEB225H1 Lecture Notes - Lecture 9: Analysis Of Covariance, Null Hypothesis, Dependent And Independent Variables
28 Apr 2016
School
Department
Course
Professor
Document Summary
Ha: atleast one of the slopes is different from another. Control for (categorical or continuous) factors which cannot be randomized and may influence the dependent variable. Steps: compute each regression line, compare the slopes (null: slopes are same, compare using anova with interaction term like two-way anova b. If slopes are the same- the interaction term will not be significant. Test the null hypothesis that the y-intercepts of the regression lines with a common slope are the same. If null1 is rejected, do multiple comparisons to see which intercepts differ (if there are more than. 2 levels for xx (level 1 is lower than level 2)). If it is accepted, proceed to fit common regression model. Because the lines are parallel, saying that they are significantly different at one point (y- intercept) means that the lines are different at any point. If the y- intercepts are significantly different this implies that the treatment effect is significant.
