introduction to data analysis in hierarchical linear...
TRANSCRIPT
![Page 1: Introduction to Data Analysis in Hierarchical Linear Modelsstatlab.stat.yale.edu/help/workshops/HLMworkshop/statlab... · 2017-05-04 · Introduction to Data Analysis in Hierarchical](https://reader031.vdocument.in/reader031/viewer/2022011802/5b1da05d7f8b9a173f8b8764/html5/thumbnails/1.jpg)
Introduction to Data Analysisin Hierarchical Linear Models
April 20, 2007
Noah Shamosh & Frank FarachSocial Sciences StatLab
Yale University
![Page 2: Introduction to Data Analysis in Hierarchical Linear Modelsstatlab.stat.yale.edu/help/workshops/HLMworkshop/statlab... · 2017-05-04 · Introduction to Data Analysis in Hierarchical](https://reader031.vdocument.in/reader031/viewer/2022011802/5b1da05d7f8b9a173f8b8764/html5/thumbnails/2.jpg)
Scope & Prerequisites Strong applied emphasis Focus on HLM software
Has special functionality Other options: SPSS, SAS, MLWin, R
Familiarity with regression assumed
![Page 3: Introduction to Data Analysis in Hierarchical Linear Modelsstatlab.stat.yale.edu/help/workshops/HLMworkshop/statlab... · 2017-05-04 · Introduction to Data Analysis in Hierarchical](https://reader031.vdocument.in/reader031/viewer/2022011802/5b1da05d7f8b9a173f8b8764/html5/thumbnails/3.jpg)
Road to HLM Happiness Conceptualize model hierarchically Prepare data Import data into HLM Build statistical models Estimate and interpret models Graph models
![Page 4: Introduction to Data Analysis in Hierarchical Linear Modelsstatlab.stat.yale.edu/help/workshops/HLMworkshop/statlab... · 2017-05-04 · Introduction to Data Analysis in Hierarchical](https://reader031.vdocument.in/reader031/viewer/2022011802/5b1da05d7f8b9a173f8b8764/html5/thumbnails/4.jpg)
What is HLM? Hierarchical Linear Model
A multilevel statistical model Software program used for such models
Deconstructing the name (in reverse) Model: It’s a statistical model Linear: The model must be linear in the
parameters Hierarchical: Nested data structures are explicitly
modeled
![Page 5: Introduction to Data Analysis in Hierarchical Linear Modelsstatlab.stat.yale.edu/help/workshops/HLMworkshop/statlab... · 2017-05-04 · Introduction to Data Analysis in Hierarchical](https://reader031.vdocument.in/reader031/viewer/2022011802/5b1da05d7f8b9a173f8b8764/html5/thumbnails/5.jpg)
When are data hierarchical? When units are grouped at higher units of
analysis Such data may be nested within higher levels
(i.e., units) of analysis Nesting can occur between subjects…
Children nested within classrooms Classrooms nested within schools
…and/or within subjects Repeated observations on the same individuals
over time (observations nested within individuals)
![Page 6: Introduction to Data Analysis in Hierarchical Linear Modelsstatlab.stat.yale.edu/help/workshops/HLMworkshop/statlab... · 2017-05-04 · Introduction to Data Analysis in Hierarchical](https://reader031.vdocument.in/reader031/viewer/2022011802/5b1da05d7f8b9a173f8b8764/html5/thumbnails/6.jpg)
Why not use regularregression on nested data? Increased Type I error Model misspecification Miss opportunity to examine potentially
interesting contextual questions These problems increase as
observations become less independent
![Page 7: Introduction to Data Analysis in Hierarchical Linear Modelsstatlab.stat.yale.edu/help/workshops/HLMworkshop/statlab... · 2017-05-04 · Introduction to Data Analysis in Hierarchical](https://reader031.vdocument.in/reader031/viewer/2022011802/5b1da05d7f8b9a173f8b8764/html5/thumbnails/7.jpg)
Hierarchical ModelConceptualization What kind of hierarchical relations might
be present? What factors could I incorporate in my
model to reflect this organization?
![Page 8: Introduction to Data Analysis in Hierarchical Linear Modelsstatlab.stat.yale.edu/help/workshops/HLMworkshop/statlab... · 2017-05-04 · Introduction to Data Analysis in Hierarchical](https://reader031.vdocument.in/reader031/viewer/2022011802/5b1da05d7f8b9a173f8b8764/html5/thumbnails/8.jpg)
HLM Caveats Adding levels of nesting increases the
complexity of the model exponentially HLM can handle up to three levels Must have several times more lower
level observations than upper levelobservations
Parameter estimation uses maximumlikelihood instead of least squares
![Page 9: Introduction to Data Analysis in Hierarchical Linear Modelsstatlab.stat.yale.edu/help/workshops/HLMworkshop/statlab... · 2017-05-04 · Introduction to Data Analysis in Hierarchical](https://reader031.vdocument.in/reader031/viewer/2022011802/5b1da05d7f8b9a173f8b8764/html5/thumbnails/9.jpg)
Road to HLM Happiness Conceptualize model hierarchically Prepare data Import data into HLM Build statistical models Estimate and interpret models Graph models
![Page 10: Introduction to Data Analysis in Hierarchical Linear Modelsstatlab.stat.yale.edu/help/workshops/HLMworkshop/statlab... · 2017-05-04 · Introduction to Data Analysis in Hierarchical](https://reader031.vdocument.in/reader031/viewer/2022011802/5b1da05d7f8b9a173f8b8764/html5/thumbnails/10.jpg)
Prep, prep, prep! This is the most labor intensive part of
workflow, and is the source of many problemsthat come to us at the StatLab
Two obstacles HLM doesn’t do data manipulation or basic data
description HLM requires a special data structure
Solutions Plan ahead. Do all data screening, variable
transformations, exploratory analyses, andassumption-checking beforehand
![Page 11: Introduction to Data Analysis in Hierarchical Linear Modelsstatlab.stat.yale.edu/help/workshops/HLMworkshop/statlab... · 2017-05-04 · Introduction to Data Analysis in Hierarchical](https://reader031.vdocument.in/reader031/viewer/2022011802/5b1da05d7f8b9a173f8b8764/html5/thumbnails/11.jpg)
Data prep: SPSS example1
Data set: IQv & language achievement Two files
Level 1: dependent variable (languageachievement) and other childcharacteristics (e.g. IQv)
Level 2: school characteristics (e.g. SES) Children are nested within schools
1 Extensively adapted from Bryk & Raudenbush (2002) and Bauer (2005)
![Page 12: Introduction to Data Analysis in Hierarchical Linear Modelsstatlab.stat.yale.edu/help/workshops/HLMworkshop/statlab... · 2017-05-04 · Introduction to Data Analysis in Hierarchical](https://reader031.vdocument.in/reader031/viewer/2022011802/5b1da05d7f8b9a173f8b8764/html5/thumbnails/12.jpg)
Road to HLM Happiness Conceptualize model hierarchically Prepare data Import data into HLM Build statistical models Estimate and interpret models Graph models
![Page 13: Introduction to Data Analysis in Hierarchical Linear Modelsstatlab.stat.yale.edu/help/workshops/HLMworkshop/statlab... · 2017-05-04 · Introduction to Data Analysis in Hierarchical](https://reader031.vdocument.in/reader031/viewer/2022011802/5b1da05d7f8b9a173f8b8764/html5/thumbnails/13.jpg)
Creating the Multivariate DataMatrix (MDM) Making an MDM file
A caveat… The procedure… Check your summary statistics before
building any models (cross-reference) Main window: are all of your variables
there?
![Page 14: Introduction to Data Analysis in Hierarchical Linear Modelsstatlab.stat.yale.edu/help/workshops/HLMworkshop/statlab... · 2017-05-04 · Introduction to Data Analysis in Hierarchical](https://reader031.vdocument.in/reader031/viewer/2022011802/5b1da05d7f8b9a173f8b8764/html5/thumbnails/14.jpg)
Road to HLM Happiness Conceptualize model hierarchically Prepare data Import data into HLM Build statistical models Estimate and interpret models Graph models
![Page 15: Introduction to Data Analysis in Hierarchical Linear Modelsstatlab.stat.yale.edu/help/workshops/HLMworkshop/statlab... · 2017-05-04 · Introduction to Data Analysis in Hierarchical](https://reader031.vdocument.in/reader031/viewer/2022011802/5b1da05d7f8b9a173f8b8764/html5/thumbnails/15.jpg)
Build statistical models Basic model: random-effects ANOVA Test for mean group differences in
population Between-group vs. total variance
Key assumption check of HLM
![Page 16: Introduction to Data Analysis in Hierarchical Linear Modelsstatlab.stat.yale.edu/help/workshops/HLMworkshop/statlab... · 2017-05-04 · Introduction to Data Analysis in Hierarchical](https://reader031.vdocument.in/reader031/viewer/2022011802/5b1da05d7f8b9a173f8b8764/html5/thumbnails/16.jpg)
Random-effects ANOVA Choose outcome variable Terms… Toggle Level 2 error term
Level 1 (r) vs. Level 2 (u) error terms The “Mixed” window
![Page 17: Introduction to Data Analysis in Hierarchical Linear Modelsstatlab.stat.yale.edu/help/workshops/HLMworkshop/statlab... · 2017-05-04 · Introduction to Data Analysis in Hierarchical](https://reader031.vdocument.in/reader031/viewer/2022011802/5b1da05d7f8b9a173f8b8764/html5/thumbnails/17.jpg)
Random effects ANOVALa
ngua
ge a
chie
vem
ent
M1 M2 M3GM
![Page 18: Introduction to Data Analysis in Hierarchical Linear Modelsstatlab.stat.yale.edu/help/workshops/HLMworkshop/statlab... · 2017-05-04 · Introduction to Data Analysis in Hierarchical](https://reader031.vdocument.in/reader031/viewer/2022011802/5b1da05d7f8b9a173f8b8764/html5/thumbnails/18.jpg)
Road to HLM Happiness Conceptualize model hierarchically Prepare data Import data into HLM Build statistical models Estimate and interpret models Graph models
![Page 19: Introduction to Data Analysis in Hierarchical Linear Modelsstatlab.stat.yale.edu/help/workshops/HLMworkshop/statlab... · 2017-05-04 · Introduction to Data Analysis in Hierarchical](https://reader031.vdocument.in/reader031/viewer/2022011802/5b1da05d7f8b9a173f8b8764/html5/thumbnails/19.jpg)
Random effects ANOVA Results
Fixed effects: the intercept Is the grand mean significantly different from
zero? Variance components (random effects)
Level 2 (U0): significant variability betweengroups?
Level 1 (R): significant variability within groups?
![Page 20: Introduction to Data Analysis in Hierarchical Linear Modelsstatlab.stat.yale.edu/help/workshops/HLMworkshop/statlab... · 2017-05-04 · Introduction to Data Analysis in Hierarchical](https://reader031.vdocument.in/reader031/viewer/2022011802/5b1da05d7f8b9a173f8b8764/html5/thumbnails/20.jpg)
Random effects ANOVA Intraclass correlation (ICC)
Proportion of total variance accounted forby between-group differences
Level 2 variance component divided bysum of Level 1 and Level 2 variancecomponents
Ours is .23; HLM is warranted
![Page 21: Introduction to Data Analysis in Hierarchical Linear Modelsstatlab.stat.yale.edu/help/workshops/HLMworkshop/statlab... · 2017-05-04 · Introduction to Data Analysis in Hierarchical](https://reader031.vdocument.in/reader031/viewer/2022011802/5b1da05d7f8b9a173f8b8764/html5/thumbnails/21.jpg)
Road to HLM Happiness Conceptualize model hierarchically Prepare data Import data into HLM Build statistical models Estimate and interpret models Graph models
![Page 22: Introduction to Data Analysis in Hierarchical Linear Modelsstatlab.stat.yale.edu/help/workshops/HLMworkshop/statlab... · 2017-05-04 · Introduction to Data Analysis in Hierarchical](https://reader031.vdocument.in/reader031/viewer/2022011802/5b1da05d7f8b9a173f8b8764/html5/thumbnails/22.jpg)
Random effects regression Test for relationship between a Level 1
IV and the DV Test whether an IV explains any
between groups variance Terms… We are assuming a fixed slope
![Page 23: Introduction to Data Analysis in Hierarchical Linear Modelsstatlab.stat.yale.edu/help/workshops/HLMworkshop/statlab... · 2017-05-04 · Introduction to Data Analysis in Hierarchical](https://reader031.vdocument.in/reader031/viewer/2022011802/5b1da05d7f8b9a173f8b8764/html5/thumbnails/23.jpg)
Random effects regression
IQ
Lang
uage
ach
ieve
men
t
![Page 24: Introduction to Data Analysis in Hierarchical Linear Modelsstatlab.stat.yale.edu/help/workshops/HLMworkshop/statlab... · 2017-05-04 · Introduction to Data Analysis in Hierarchical](https://reader031.vdocument.in/reader031/viewer/2022011802/5b1da05d7f8b9a173f8b8764/html5/thumbnails/24.jpg)
Road to HLM Happiness Conceptualize model hierarchically Prepare data Import data into HLM Build statistical models Estimate and interpret models Graph models
![Page 25: Introduction to Data Analysis in Hierarchical Linear Modelsstatlab.stat.yale.edu/help/workshops/HLMworkshop/statlab... · 2017-05-04 · Introduction to Data Analysis in Hierarchical](https://reader031.vdocument.in/reader031/viewer/2022011802/5b1da05d7f8b9a173f8b8764/html5/thumbnails/25.jpg)
Random effects regression Results
Fixed effects Level 1 intercept: Mean of DV where IV is zero Level 1 slope: Change in DV with one unit of
change in IV (just like OLS regression) Random effects
Intercept: Between-group variance that is notexplained by IV
Residual variance: Within-group variance thatis not explained by DV
![Page 26: Introduction to Data Analysis in Hierarchical Linear Modelsstatlab.stat.yale.edu/help/workshops/HLMworkshop/statlab... · 2017-05-04 · Introduction to Data Analysis in Hierarchical](https://reader031.vdocument.in/reader031/viewer/2022011802/5b1da05d7f8b9a173f8b8764/html5/thumbnails/26.jpg)
Random effects regression Variance accounted for by IV
Level 1: Compare residual variancecomponent to random effects ANOVAmodel
(8.0 - 6.5) / 8.0 = .19 Level 2: Do the same for the random
intercept variance component (19.6 - 9.6) / 19.6 = .51
![Page 27: Introduction to Data Analysis in Hierarchical Linear Modelsstatlab.stat.yale.edu/help/workshops/HLMworkshop/statlab... · 2017-05-04 · Introduction to Data Analysis in Hierarchical](https://reader031.vdocument.in/reader031/viewer/2022011802/5b1da05d7f8b9a173f8b8764/html5/thumbnails/27.jpg)
Fixed slopes
IQ
Lang
uage
ach
ieve
men
t
![Page 28: Introduction to Data Analysis in Hierarchical Linear Modelsstatlab.stat.yale.edu/help/workshops/HLMworkshop/statlab... · 2017-05-04 · Introduction to Data Analysis in Hierarchical](https://reader031.vdocument.in/reader031/viewer/2022011802/5b1da05d7f8b9a173f8b8764/html5/thumbnails/28.jpg)
Random slopes
IQ
Lang
uage
ach
ieve
men
t
![Page 29: Introduction to Data Analysis in Hierarchical Linear Modelsstatlab.stat.yale.edu/help/workshops/HLMworkshop/statlab... · 2017-05-04 · Introduction to Data Analysis in Hierarchical](https://reader031.vdocument.in/reader031/viewer/2022011802/5b1da05d7f8b9a173f8b8764/html5/thumbnails/29.jpg)
Random slopes Goal: test whether the IV - DV
relationship varies between groups Add only if supported by theory Toggle Level 2b error term In output, look at slope variance
component
![Page 30: Introduction to Data Analysis in Hierarchical Linear Modelsstatlab.stat.yale.edu/help/workshops/HLMworkshop/statlab... · 2017-05-04 · Introduction to Data Analysis in Hierarchical](https://reader031.vdocument.in/reader031/viewer/2022011802/5b1da05d7f8b9a173f8b8764/html5/thumbnails/30.jpg)
Slopes as outcomes Goal: test cross level interactions
Does the between-group variability in theIV - DV relation vary by a systematicfactor?
Add Level 2 predictor Terms…
![Page 31: Introduction to Data Analysis in Hierarchical Linear Modelsstatlab.stat.yale.edu/help/workshops/HLMworkshop/statlab... · 2017-05-04 · Introduction to Data Analysis in Hierarchical](https://reader031.vdocument.in/reader031/viewer/2022011802/5b1da05d7f8b9a173f8b8764/html5/thumbnails/31.jpg)
Slopes as outcomes Fixed effects
For Level 1 intercept Intercept: predicted score on DV at mean value of L-1 IV Slope: Influence of Level 2 IV on DV
For Level 1 slope Intercept: Influence of Level 1 IV on DV Slope: Influence of L-2 IV on L-1 IV - DV relation
Random effects (same as before)
![Page 32: Introduction to Data Analysis in Hierarchical Linear Modelsstatlab.stat.yale.edu/help/workshops/HLMworkshop/statlab... · 2017-05-04 · Introduction to Data Analysis in Hierarchical](https://reader031.vdocument.in/reader031/viewer/2022011802/5b1da05d7f8b9a173f8b8764/html5/thumbnails/32.jpg)
Road to HLM Happiness Conceptualize model hierarchically Prepare data Import data into HLM Build statistical models Estimate and interpret models Graph models
![Page 33: Introduction to Data Analysis in Hierarchical Linear Modelsstatlab.stat.yale.edu/help/workshops/HLMworkshop/statlab... · 2017-05-04 · Introduction to Data Analysis in Hierarchical](https://reader031.vdocument.in/reader031/viewer/2022011802/5b1da05d7f8b9a173f8b8764/html5/thumbnails/33.jpg)
Graph: Simple slopes Useful for visualizing cross-level
interactions Just like simple slope plots in
regression Graph Equations > Model graphs Useful for categorical or continuous
data
![Page 34: Introduction to Data Analysis in Hierarchical Linear Modelsstatlab.stat.yale.edu/help/workshops/HLMworkshop/statlab... · 2017-05-04 · Introduction to Data Analysis in Hierarchical](https://reader031.vdocument.in/reader031/viewer/2022011802/5b1da05d7f8b9a173f8b8764/html5/thumbnails/34.jpg)
Graph: Level-1 equations Useful for:
Visualizing variability in intercepts andslopes
Identifying moderators Graph Equations > Level 1 equation
graphing
![Page 35: Introduction to Data Analysis in Hierarchical Linear Modelsstatlab.stat.yale.edu/help/workshops/HLMworkshop/statlab... · 2017-05-04 · Introduction to Data Analysis in Hierarchical](https://reader031.vdocument.in/reader031/viewer/2022011802/5b1da05d7f8b9a173f8b8764/html5/thumbnails/35.jpg)
Recommended Reading Bickel, R. (2007). Multilevel analysis for applied research: It's
just regression! New York: Guilford Press. Bryk, A. & Raudenbush, S. (2002). Hierarchical Linear Models:
Applications and data analysis methods (2nd ed.). ThousandOaks, CA: Sage.
Luke, D. (2004). Multilevel modeling. Thousand Oaks, CA:Sage.
Heck, R. H., & Thomas, S. L. (2000). An introduction tomultilevel modeling techniques. Lawrence Erlbaum Associates.
Kreft, I. & de Leeuw, J. (1998). Introducing multilevel modeling.Sage.
Singer, J. D., & Willett, J. B. (2003). Applied Longitudinal DataAnalysis: Modeling Change and Event Occurrence. Oxford Univ.Press. (Longitudinal focus)
![Page 36: Introduction to Data Analysis in Hierarchical Linear Modelsstatlab.stat.yale.edu/help/workshops/HLMworkshop/statlab... · 2017-05-04 · Introduction to Data Analysis in Hierarchical](https://reader031.vdocument.in/reader031/viewer/2022011802/5b1da05d7f8b9a173f8b8764/html5/thumbnails/36.jpg)
HLM Resources on the Web UCLA’s HLM portal
http://statcomp.ats.ucla.edu/mlm Excellent example of analysis
http://www.ats.ucla.edu/stat/hlm/seminars/hlm_mlm/mlm_hlm_seminar.htm