| Volume 3 Index July 2009 |

Full articles
| PDF | SWF |

Title
Testing and Reducing L2 Vocabulary Learning Strategies
Inventory Using Rasch Model

 Authors
Johnny T. Amora
Statistician, Center for Learning and Performance Assessment
De La Salle-College of Saint Benilde, Manila
Alejandro S. Bernardo
Faculty of Arts and Letters
University of Santo Tomas, Manila

Bios.
Johnny Amora is presently the statistician of the Center for Learning and Performance Assessment of De La Salle-College of Saint Banilde. He has written several researches in assessment and evaluation for the said college. He has presented and published his works primarily using quantitative analysis in national conferences.

Alejandro S. Bernardo has been teaching English for seven years. At present, he is a member of the Faculty of Arts and Letters of the University of Santo Tomas. He is finishing his Ph.D. in Language Studies at the same university. His research interests include vocabulary acquisition and developmental reading.

Abstract
The Second Language Vocabulary Learning Strategies Inventory
(L2VLSI) instrument developed based on the framework of Schmitt (1997) has been intensively studied using the Classical Test Theory. This study was conducted to further test and refine the L2VLSI using the Rasch model. The refined instrument, called L2VLSI-R, was subjected to Rasch recalibration using the testing and validation samples. In addition, using the validation sample, the refined instrument was subjected to validation study through the latent structure methodology. Results of the Rasch analysis revealed that: the full form of L2VLSI fitted significantly the Rasch rating scale model; the 6-point scoring system of the instrument demonstrated logical order; the full form was successfully refined by removing the five misfit items; both the full form and L2VLSI-R have high reliability; the L2VLSI-R fitted significantly the hierarchical factor model; and the Rasch calibrations showed consistency between the two samples.