Latent class analysis of reading, decoding, and writing performance using the Academic Performance Test: concurrent...

Area of Science:

• Educational Psychology
• Child Development
• Psychometrics

Background:

• The Academic Performance Test (TDE) is used to assess reading and writing skills.
• Understanding distinct performance profiles is crucial for targeted interventions.
• Previous classifications may lack empirical validation.

Purpose of the Study:

• To identify and validate the optimal latent class solution for TDE reading and writing subtests.
• To establish empirically supported categories of academic performance.

Main Methods:

• Item Response Theory (IRT) was used to select high-discriminating items from TDE subtests.
• Latent Class Analysis (LCA) was performed on selected items.
• Solutions were validated using IQ, ADHD, and depression diagnoses.

Main Results:

• A three-class solution emerged as the best fit, identifying groups with good, moderate, and poor performance.
• This classification correlated significantly with estimated IQ and ADHD diagnoses.
• No significant association was found between the latent classes and major depression diagnoses.

Conclusions:

• The three-class solution demonstrates both concurrent and discriminant validity.
• This study provides initial evidence for a valid, empirically derived classification of reading and writing performance.
• The findings support further research into correlates of academic performance using the TDE.