Questions From Yesterday课件.ppt

Questions From Yesterday Equation 2: r-to-z transform Equation is correct Comparable to other p-value estimates (z = r sqrt[n]) ANOVA will not be able to detect a group effect that has alternating + and – ICC Effect defined in terms of between and within group variability rather than being represented individually SPSS Advanced Models can be ordered at the VU Bookstore for $51 精品文档 Hierarchical Linear Modeling (HLM) Theoretical introduction Introduction to HLM HLM equations HLM interpretation of your data sets Building an HLM model Demonstration of HLM software Personal experience with HLM tutorial 精品文档 General Information and Terminology HLM can be used on data with many levels but we will only consider 2-level models The lowest level of analysis is Level 1 (L1), the second lowest is Level 2 (L2), and so on In group research, Level 1 corresponds to the individual level and Level 2 corresponds to the group level Your DV has to be at the lowest level 精品文档 When Should You Use HLM? If you have mixed variables If you have different number of observations per group If you think a regression relationship varies by group Any time your data has multiple levels 精品文档 What Does HLM Do? Fits a regression equation at the individual level Lets parameters of the regression equation vary by group membership Uses group-level variables to explain variation in the individual-level parameters Allows you to test for main effects and interactions within and between levels 精品文档 The Level 1 Regression Equation Predicts the value of your DV from the values of your L1 IVs (example uses 2) Equation has the general form of Yij = B0j + B1j * X1ij + B2j * X2ij + rij “i” refers to the person number and “j” refers to the group number Since the coefficients B0, B1, and B2 change from group to group they have variability that we can try to explain 精品文档 Level 2 Equations Predict the value of the L1 parameters using values of your L2 IVs (example uses 1) Sample equations: B0j = G00 + G01 * W1j + u