https://josh-jackson.github.io/longitudinal/" />
ALDA
1
Syllabus
2
LDA basics
2.1
Motivation, terms, concepts
2.1.1
Why longtiduinal?
2.1.2
Types of change (most common)
2.1.3
Between person versus within person
2.1.4
Trajectories, curves, change, growth… oh my
2.2
Data analysis and data structures
2.2.1
Modeling frameworks: MLM & SEM
2.2.2
Wide and Long form
2.2.3
Graphing
2.3
Putting it together example
2.3.1
a person level
2.4
What does this look like?
2.5
Design considerations
2.5.1
number of assessment waves
2.5.2
measuremnt
2.6
Threats to validity
2.6.1
Missing data
2.6.2
Attrition/Mortality
2.6.3
History/cohort
2.6.4
Maturation
2.6.5
Testing
2.6.6
Selection
2.7
Why not RM ANOVA?
2.8
Now you try
3
Growth curves
3.1
Between person models and cross sectional data
3.2
Within person models e.g., 2-level models
3.2.1
thinking about random effects
3.2.2
random effects
3.2.3
Empty model equation
3.2.4
Putting it together
3.2.5
Empty model
3.2.6
Putting it together
3.2.7
ICC
3.3
Adding time
3.3.1
What does this look like graphically?
3.3.2
Adding a random slope?
3.4
Individaul level random effects
3.4.1
Calculation of individaul level random effects
3.4.2
How are these random effects calculated?
3.4.3
random effects and residual (standard) assumptions
3.4.4
Random effect decomposition
3.5
working with models in R
3.5.1
basic lmer code
3.5.2
Example
3.5.3
How to calculate ICC?
3.6
Exploring beyond the summary
3.6.1
what do the random effects look like?
3.7
Adding time to the MLM
3.7.1
fixed slope
3.7.2
Random slope
3.8
Random effects
3.8.1
Calcualtion of random effect confidence interval
3.8.2
Individaul level random effects
3.8.3
adding time
3.8.4
caterpillar plots
3.8.5
Density of individaul random effects
3.9
comparing to a standard linear model
3.10
Other types of models
3.11
Matrix notation (as a way to help understand what is going on)
3.12
Estimation
3.13
Testing significance (adapted from Ben Bolker)
3.13.1
P values are not included
3.13.2
Likelhiood ratio test
3.13.3
Likelihood tests for random effects
3.13.4
AIC and BIC
3.13.5
MCMC
3.13.6
Bootstraps
3.14
Predictions and prediction intervals
3.14.1
Predictions and prediction intervals
3.15
Coefficient of determination equivalents
3.15.1
batch analyses
3.16
Now you try:
4
Conditional Predictors in growth models
4.1
Intercept effects
4.1.1
Seperatinng these into intercept and slope
4.2
Slope and Intercept effects
4.2.1
Equations necessary for plotting
4.3
Need for thinking about scaling your predictors
4.4
Time-varying covariates (TVCs)
4.5
Now you try
5
Polynomial and Splines
5.1
Polynomaials
5.2
polynomial example
5.2.1
importance of centering
5.2.2
random terms
5.3
Splines aka piecewise
5.3.1
seperate curves
5.3.2
incremental curves
5.3.3
splines example
5.4
splines + polynomail = polynomial piecewise
6
Intensive Longitudinal Designs
6.1
within versus between person processes
6.2
Centering
6.2.1
within person
6.2.2
Grand mean
7
SEM
7.1
Structural Equation Modeling
7.2
Latent variables
7.2.1
More pretty pictures
7.2.2
Classical test theory interpretation
7.2.3
Generizability interpretation of latent variables
7.2.4
measurement error
7.2.5
regarding means
7.3
goal of SEM
7.3.1
What questions can be asked?
7.4
Setting the scale and defining variables
7.4.1
identification
7.4.2
types of identification
7.5
lavaan
7.5.1
lavaan language
7.5.2
How to run lavaan
7.5.3
lavaan defaults
7.6
additional SEM details
7.6.1
coding revisited
7.6.2
plotting
7.6.3
Fit Indices
7.6.4
Comparing models
7.6.5
Parcels
7.6.6
Estimators
7.7
Types of longitudinal models other than growth models (brief intro)
7.7.1
Longitudinal CFA
7.7.2
Longitudinal Path Model
7.7.3
Longitudinal Cross lagged model
7.7.4
Longitudinal mediation model
7.7.5
Summary of panel SEM models.
7.8
SEM Growth models
7.8.1
Coding time
7.8.2
latent basis model
7.8.3
constraining slope to be fixed only
7.8.4
introducing covariates/predictors
7.8.5
introducing time varying covariates
7.8.6
multivariate growth curves
7.9
Measurement Invariance (MI)
7.9.1
types of MI
7.9.2
Testing MI
7.9.3
Comparing the models
7.10
Second order growth model
7.11
Multple groups
7.11.1
measurement invariance revisited
7.11.2
When to use
7.12
Missing data
7.12.1
Planned missing data
7.13
Power
7.14
Now you try
8
Advanced Multivariate Longitudinal SEM designs
8.0.1
Cross-lagged panel model
8.1
Difference scores
8.2
Analyzing change with two time points
8.3
multi wave latent change model
8.4
ALT and ALT-SR
9
lcmm
9.1
latent class / mixture models
9.2
Longitudinal Mixture models and Latent Class growth models– what is the difference?
Applied Longitudinal Data Analysis
Applied Longitudinal Data Analysis
Applied Longitudinal Data Analysis
Josh Jackson
Fall 2017
Applied Longitudinal Data Analysis