Multilevel Modeling with R Schedule

Multilevel Modeling with R:

A one-week (4.5 day) workshop on Multilevel Modeling Using R will be held at the University of Connecticut from Monday, June 11, through Friday, June 15, 2018. This workshop covers the basics and applications of multilevel modeling with extensions to more complex designs. Participants will learn how to analyze both organizational and longitudinal (mostly growth curve) data using multilevel modeling and to interpret the results from their analyses. Although the workshop does not require any prior knowledge or experience with multilevel modeling, participants are expected to have a working knowledge of multiple regression as well as some experience using statistical software (such as SPSS, SAS, R, Stata). Instruction consists of lecture, computer workshops, and individualized consultations. The emphasis will be practical with minimal emphasis on statistical theory, but those seeking more statistical information can arrange an individualized session with the instructors.  A teaching assistant will also be available for consultation and assistance.

All analyses will be demonstrated using R. We can also provide set-ups in HLM for some of the analyses that we present. Instruction will consist of lectures, computer demonstrations of data analyses, and hands-on opportunities to analyze practice data sets using R. The workshop emphasizes practical applications and places minimal emphasis on statistical theory.  The workshop takes place in a computer lab, so you do not need to bring a laptop. No prior familiarity with R is required, but if you have never used R and want to gain a general proficiency working with data in R, we encourage you to take the two-day DATIC Intro to R and RStudio workshop held on Thursday, June 7, through Friday, June 8, 2018.   In this Multilevel Modeling workshop we will focus solely on running multilevel organizational and longitudinal models in R. Several R setups will be given for data organization, analysis, and write-ups. Although prior experience with R is not required, it is certainly helpful. Also in meetings with individual participants, alternative programs and approaches (e.g., SPSS, HLM, MLwiN, SAS, Stata, and Mplus) can be discussed.

Logistical details: The Multilevel Modeling workshop runs Monday-Thursday from 9-5 and Friday from 9-12. A light continental breakfast is provided every morning. There is also a closing group lunch on Friday from 12:00-1:00 pm.

Benefits:

The following materials and events are included in the cost of the workshop:

  • Tape bound workbook with workshop outline, computer setups and outputs
  • R programs to assist in the write-ups of the results of multilevel analyses
  • Meals: Continental breakfast each morning, complimentary luncheon Friday

 

Day 1: Monday, June 11

8:30 – 9:00 Continental Breakfast, 319 Gentry
9:00 – 9:15 Introduction to Multilevel Modeling – overview of the week
9:15 – 10:45 Basics of Multilevel Modeling

Regression Example – High School and Beyond

10:45-11:00 Break
11:00-12:15 Continuation: Regression Example

·         High School and Beyond (HSB) Example: Two-level model

o    Within school regressions

o    Comparing regressions across schools

o    Conditional model example and interpretation

o    The logic of 2 level models- the multilevel and the combined form equations

12:15 – 1:30 Lunch
1:30- 3:00 Using R – Analysis walkthrough

·         Brief intro to R

·         Data Structures for MLM

·         Manipulating data using R

o    Aggregating variables

o    Creating new variables (e.g., centering)

·         Getting started in R for Multilevel Modeling

3:15-4:45 Building two-level models

·         Model 1 One-way random effects ANOVA (totally unconditional model)

o    ICC

o    Interpretation of fixed and random effects

·         Model 2 Random coefficients model (unconditional model)

o    Change in level-1 variance

o    Variance and covariance components

o    Interpretation of fixed and random effects

·         Model 3 Intercepts and slopes as outcomes (conditional model)

o    Reduction in level-2 variance

o    Residual variance and covariance components

o    Interpretation of fixed and random effects

4:45-5:00- Recap and Q&A

Day 2:  Tuesday, June 12

8:30 – 9:00 Continental Breakfast, 319 Gentry
9:00 – 9:15 Q & A from Day 1
9:15 – 10:45 Model Assessment Issues

  • Deviance
  • Model fit, model comparison procedures, AIC and BIC
  • Estimation and Maximum likelihood
  • Testing Variance Components
10:45 – 11:00 Break
11:00-12:15 Centering

 

12:15 – 1:30 Lunch (on your own)

 

1:30 – 3:30 Longitudinal Models in R

  • Re-conceptualizing Longitudinal Data for Multilevel Models
  • Person period data sets
  • Centering Time
  • Growth trajectories and individual growth curves
  • Fitting individual growth trajectories in R
3:30- 3:45 Break
3:45 – 4:45 Structured Lab I – Organizational Data

 

4:45 – 5:00 Questions and Answers – Lab review

 

Day 3:  Wednesday, June 13

8:30 – 9:00 Continental Breakfast
9:00 – 9:15 Q & A from Day 2
9:15 – 10:30 Longitudinal Growth Models in R: PIAT reading examples

Time structured versus Time unstructured data

Running and interpreting a two level model with level 2 covariates

10:30 – 10:45 Break
10:45 – 12:00 §  Intro to time varying covariates

NYS Example

12:00 – 1:15 Lunch (on your own)

 

1:15 – 2:15 ·         Interaction between time and time varying covariates

o    Unemployment example

2:15 – 3:30 ·         Modeling discontinuity- change in slope; change in intercept

o    Wages example

3:30 – 3:45 Break
3:45 – 4:45 Structured Lab II – Longitudinal data

 

4:45 – 5:00 Recap and Q&A

Day 4: Thursday, June 14

8:30 – 9:00 Continental Breakfast
9:00 – 9:15 Q & A from Day 3
9:15 – 10:30 Piecewise and Polynomial Models for Non-Linear Growth

·         Fit and interpretation

o    NYS as a non-linear model (Piecewise)

10:30 – 10:45 Break
10:45 – 12:00 Polynomial Models for Non-Linear Growth

o    NYS as a non-linear model (Cubic)

Comparison to piecewise model

12:00 – 1:15 Lunch (on your own)

 

1:15 – 2:45 Checking assumptions: Consideration of more complex error variance-covariance structures; Heterogeneous Variances
2:45-3:00 Break
3:00-4:00 Graphing
4:00-5:30 Open lab time and Individual Consultations

Day 5: Friday, June 15

8:30 – 9:00 Continental Breakfast
9:00 – 10:15 Graphing your results (and presenting your results)
10:15-10:30 Break and Evaluations
10:45-12:00 Power Analysis and Sample Size
12:00 Closing Luncheon –  Gentry 144

Register for the Multilevel Modeling workshop here.