MLM Cluster Level Moderator Exercise June 16, 2015 Data The Gambia Data (2009) Level 1 Variables (pupils) schoolID – (as ID) S2Q3_PP – Number of words correct in 60 seconds (in mdm) Level 2 Variables (head teacher) schoolID – (as ID) Q205 – One shift per day=0, Two shifts per day=1 (in mdm) WSD – Control=0, WSD =1 (in mdm) WSDxshift – Interaction of WSD and Q205* *Note that I created the interaction variable in SPSS prior to uploading the data to HLM. Load the data above into HLM. Use number of words correct in 60 seconds (S2Q3_PP) as the dependent variable for all analyses. 1. Fit the unconditional 2-level model a. What is the unconditional ICC. 2. Suppose a group of researchers are interested in whether the treatment effect is moderated by whether a school has 1 or 2 shifts per day. a. Fit the appropriate model. Is there evidence that number of shifts per day moderates the treatment effect? In your own datasets, choose an outcome of interest. 3. Run the unconditional model as a starting point for future analyses. What is the unconditional ICC? 4. Consider a cluster-level variable that you believe may moderate the treatment effect. Test the hypothesis.