Multilevel Modeling with HLM and SPSS (Amherst, MA)

Instructor(s):

  • Holly Laws, University of Massachusetts at Amherst
  • Aline Sayer, University of Massachusetts at Amherst

Multilevel models (MLM), also known as hierarchical linear models (HLM) and mixed effects models, are widely used across a range of disciplines including sociology, psychology, political science, education, economics, and public health. They provide a conceptual framework and a flexible set of analytic tools to study a variety of social, political, and developmental processes. In this workshop, we will teach in parallel the use of both the HLM and SPSS software packages to fit two- and three-level multilevel models, focusing on both linear and nonlinear outcomes. We will demonstrate the new features of the HLM 8 program, which include analysis of missing data through multiple imputation of both dependent and independent variables, and the use of fixed intercept models with random effects.

In MLM, one set of applications focuses on data in which persons are clustered within social contexts, such as couples, families, schools, neighborhoods, or organizations. A second set of applications concerns individual growth or change over time, where time series or longitudinal data are clustered within persons. A third set of applications involves a combination of the two: persons changing over time who are also nested within organizational contexts. Participants will be exposed to a wide variety of hands-on examples, with emphasis on computing practice and interpretation and reporting of results. Topics include an introduction to the basic two-level model for continuous outcomes, assessment of fit, checking model assumptions, single and multiparameter hypothesis testing, the extension to three-level models, and nonlinear models for binary and count outcomes. Although the workshop will focus on practice with SPSS and HLM, syntax for other statistical software packages (Mplus, R, Stata, and SAS) will be provided to participants as needed.

Prerequisites: Participants should have strong backgrounds in multiple regression analysis.

Registration Fee: Members = $1700; Non-members = $3200

Tags: hierarchical, HLM, multilevel models, MLM

Course Sections

Section 1

Location: University of Massachusetts -- Amherst, MA

Date(s): June 29 - July 2

Time: 9:00 AM - 5:00 PM

Instructor(s):

  • Holly Laws, University of Massachusetts at Amherst
  • Aline Sayer, University of Massachusetts at Amherst

Syllabus: