ABCD Examples
Examples of longitudinal analysis methods using data from the ABCD Study® dataset.
Filters
Method Family
Statistical Engine
Covariates
All Tutorials
Difference Score: Paired Samples T-Test
Difference scores quantify change over time by subtracting initial measurements from follow-up measurements, isolating individual-level change within each participant. A paired samples t-test then eva...
Difference Score: Simple Regression
Difference scores quantify within-subject change by subtracting initial measurements from follow-up measurements. Simple regression models can then examine whether individual characteristics predict t...
GEE: Basic
Generalized Estimating Equations (GEE) analyze longitudinal or clustered data by extending generalized linear models to estimate population-averaged effects while accounting for within-subject correla...
GEE: Time-Varying Covariates
Generalized Estimating Equations with time-varying covariates extend standard GEE models to examine how predictors that change over time influence longitudinal outcomes, estimating population-averaged...
GLMM: Basic
Generalized Linear Mixed Models (GLMMs) extend linear mixed models to handle non-normally distributed outcomes such as counts or binary responses while modeling random effects to account for individua...
GLMM: Interaction
Generalized Linear Mixed Models with interaction terms test whether predictor effects on non-normal outcomes vary across levels of other variables, combining fixed and random effects to model moderati...
LGCM: Multiple Groups
Multigroup Latent Growth Curve Modeling (MG-LGCM) tests whether growth patterns differ systematically across groups by estimating separate intercept and slope parameters for each group while allowing ...
LGCM: Time-Invariant Covariates
Latent Growth Curve Modeling with time-invariant covariates extends basic growth modeling by explaining why individuals differ in initial levels and rates of change. By incorporating predictors like d...
LMM: Random Intercept
Linear mixed models with random intercepts extend ordinary linear regression by allowing each participant to have a unique baseline level in addition to the overall mean intercept, accounting for indi...
LMM: Time-Invariant Covariates
Linear mixed models with random intercepts and slopes extended with time-invariant covariates allow examination of how stable individual characteristics predict both baseline levels and rates of chang...
Residualized Change Score
Residualized change scores quantify within-subject change while controlling for baseline levels by regressing follow-up values on initial values and extracting residuals that represent deviations from...
LGCM: Basic
Latent Growth Curve Modeling (LGCM) analyzes longitudinal change by estimating growth trajectories as latent factors while distinguishing systematic development from measurement error. Using intercept...
LGCM: Nesting
Latent Growth Curve Modeling with clustering addresses dependencies where observations within the same family or study site are more similar than observations from different clusters. Ignoring this st...
LMM: Random Slopes
Linear Mixed Models with random slopes allow each individual to have a unique trajectory of change over time, recognizing heterogeneous developmental patterns across participants. By estimating both r...
LGCM: Multivariate
Multivariate Latent Growth Curve Modeling (MLGCM) simultaneously models trajectories of multiple outcomes, revealing how different developmental processes unfold together over time. By estimating inte...