Item Parceling in Second-Order Latent Growth Model

• In Event: Research Methodologies in Education

Abstract

Item parceling involves summing or averaging together two or more items and using the resulting sum or average as the basic unit of analysis. Item parceling has been widely applied in structural equation modeling (SEM) but has been less discussed in the Latent Growth Modeling (LGM) context, especially for second-order LGM. The present study investigated whether item parceling could enhance the performance of second-order LGM in terms of psychometric properties and model fit using Monte Carlo simulation study. The results shows that under conditions with item parcel and large sample size, parameter bias was negligible and the model fit improved.

Authors

• Suyoung Kim, University of Texas at Austin
  Presenting Author

• Jiwon Kim, University of Texas at Austin
  Presenting Author