Introduction to Multilevel Analysis Introduction to Multilevel - - PowerPoint PPT Presentation

introduction to multilevel analysis introduction to
SMART_READER_LITE
LIVE PREVIEW

Introduction to Multilevel Analysis Introduction to Multilevel - - PowerPoint PPT Presentation

Introduction to Multilevel Analysis Introduction to Multilevel Analysis PSYC 575 PSYC 575 Mark Lai Mark Lai University of Southern California University of Southern California 2020/07/30 (updated: 2020-08-10) 2020/07/30 (updated:


slide-1
SLIDE 1

Introduction to Multilevel Analysis Introduction to Multilevel Analysis

PSYC 575 PSYC 575

Mark Lai Mark Lai University of Southern California University of Southern California 2020/07/30 (updated: 2020-08-10) 2020/07/30 (updated: 2020-08-10)

1 / 15 1 / 15

slide-2
SLIDE 2

2 / 15

slide-3
SLIDE 3

Learning Objectives

Other names used for MLM Types of data structure MLM handles

3 / 15

slide-4
SLIDE 4

Naming and History of MLM Naming and History of MLM

4 / 15 4 / 15

slide-5
SLIDE 5

Alternative names for MLM:

Hierarchical linear model: Education/Sociology/Psychology 1 Variance component model: Statistics 2 Mixed model/mixed-effect model: Statistics, biomedical Random coefficient model: Econometrics

[1]: Raudenbush & Bryk, 1986 [2]: Aitkin et al., 1981

5 / 15

slide-6
SLIDE 6

History of MLM

MLM naturally handles data coming from different "levels" Robinson (1950)1 State level: Correlation between % illiterate and % foreign born: Individual level: Within-state correlation between illiteracy and foregin born status: [1]: p. 354, https://doi.org/10.2307/2087176

r = −.53 r = .12

6 / 15

slide-7
SLIDE 7

Usage of MLM Usage of MLM

7 / 15 7 / 15

slide-8
SLIDE 8

Multiple units at a lower level nested within a unit at a higher level Level 1 Level 2 Clients Therapist Classrooms School Employees Organization People Family Citizens Country Measurements Person

Network Graph

1 A B 2 C D E 3 F G H

Hierarchical Data Structure

8 / 15

slide-9
SLIDE 9

9 / 15

slide-10
SLIDE 10

Once you know that hierarchies exist, you see them everywhere.

Kreft & de Leeuw (1998, p. 1) 10 / 15

slide-11
SLIDE 11

When it comes to regression, multilevel regression deserves to be the default approach

McElreath (2020, p. 400) 11 / 15

slide-12
SLIDE 12

Sampling the Literature

PsycINFO, past 3 months, with keywords

"multilevel model*" OR "mixed model*" OR "mixed-effect model*" OR "hierarchical linear model*"

12 / 15

slide-13
SLIDE 13

Critical factors influencing interorganizational relationships between juvenile probation and behavioral health agencies Factors associated with preferred pre-exposure prophylaxis dosing regimen among men who have sex with men in real-world settings: A mixed-effect model analysis Longitudinal investigation of the role of cognitive reserve in the evolution of dementia in outpatients prescribed AChEI Validation of the early childhood teacher experiences scale in head start The cost of academic focus: Daily school problems and biopsychological adjustment in chinese american families A randomized pilot study of mindfulness‐based stress reduction in a young adult cancer sample: Feasibility, acceptability, and changes in patient reported outcomes Do social workers lead differently? Examining associations with leadership style and organizational factors 13 / 15

slide-14
SLIDE 14

General stat software

R (lme4, brms, nlme) SAS SPSS Stata

Specialized software

HLM MLwiN Mplus

Software

14 / 15

slide-15
SLIDE 15

Resources

Centre for Multilevel Modeling, University of Bristol http://www.bristol.ac.uk/cmm/learning/multilevel-models/ Textbook examples (with syntax for different software), Institute for Digital Research and Education, UCLA https://stats.idre.ucla.edu/other/examples/ GLMM FAQ by Ben Bolker https://bbolker.github.io/mixedmodels-misc/glmmFAQ.html Curran-Bauer Analytics, Multilevel Modeling Archives https://curranbauer.org/category/news-and- updates/multilevel-modeling/ Multilevel Modeling Discussion List (Listserv) https://www.jiscmail.ac.uk/cgi-bin/webadmin?A0=multilevel 15 / 15