Time-specific Errors in Growth Curve Modeling: Type-1 Error Inflation and a Possible Solution with Mixed-Effects...

Area of Science:

• Statistics
• Longitudinal Data Analysis
• Biostatistics

Background:

• Growth curve modeling (GCM) is widely used for analyzing longitudinal data.
• Existing GCM approaches often overlook time-specific errors, affecting all participants.
• This oversight can lead to inaccurate statistical inferences.

Purpose of the Study:

• To investigate the impact of unaddressed time-specific errors on GCM.
• To propose an improved GCM that incorporates time-specific errors.
• To demonstrate how accounting for these errors affects conclusions about growth trajectories.

Main Methods:

• Utilized mixed-effects models to develop a GCM accounting for time-specific errors.
• Compared results from the proposed GCM with a standard GCM.
• Applied the models to real-world longitudinal data for illustration.

Main Results:

• Failure to account for time-specific errors significantly inflates Type-1 error rates in fixed-effects tests.
• The proposed GCM with time-specific errors yielded different substantive conclusions compared to standard GCM.
• The choice of GCM significantly impacts the interpretation of growth patterns.

Conclusions:

• Time-specific errors are a critical consideration in growth curve modeling.
• Mixed-effects models offer a robust framework for incorporating time-specific errors.
• Accurate growth trajectory analysis requires models that properly account for all sources of error.