Evaluating the validity of the dual discrepancymodel in identifying students at-risk of reading disability in Philippine public schools

Date of Publication


Document Type


Degree Name

Doctor of Philosophy in Educational Psychology Major in Educational Measurement and Evaluation


Br. Andrew Gonzalez FSC College of Education


Counseling and Educational Psychology

Thesis Adviser

Bustos-Orosa, Alicia, Dr.

Defense Panel Chair

Garcia, John Addy S.

Defense Panel Member

Reyes, Jose Alberto S. Dr.
Magno, Carlo P., Dr.
Ouano, Jerome A., Dr.
Mingoa, Thelma A., Dr.


The study aimed to determine the validity of Dual Discrepancy model in identifying students at-risk of reading difficulty in public schools through comparison with other identification procedures based on Messick’s (1995) validity measures: construct validity, relevance and utility, social consequential validity, and values implications. It was conducted in two phases: Phase I involved the development of CBM-ORF probes while Phase II covered the evaluation of the validity of DD model to identify students-at risk of reading difficulty. It employed a multi-stage sampling technique involving Grade 2 students in one public school. Various tests and measures were administered to identify the following groups of at-risk students: Dual Discrepant (DD), IQ-Achievement Discrepancy (IQ-Ach), Low Achievement (LowAch), and Low Performing (LowPerf) groups. An informal reading intervention was also employed among the DD students. Progress monitoring students’ reading proficiency was conducted for eight weeks. The study was able to establish the construct validity of DD model. DD students registered significantly lower performance level and growth rate after 8 weeks than IQ-Ach and LowAch groups. The DD model also exhibited higher classification accuracy indices (i.e., sensitivity and hit rates) than the other at-risk groups when 33rd percentile rank in reading achievement was used as criterion. It also exhibited relevance and utility, as DD students registered lower performance level, growth rates, and reading test scores than the not-at-risk students, indicating that it can significantly differentiate between poor and proficient readers. Moreover, it was able to contribute to the enhancement of student achievement as shown by significant increases in DD students’ reading proficiency after 8 weeks of progress monitoring using CBM and in at-risk students’ reading achievement test scores after one grading period of exposure to the general education. It also did not exhibit potential bias against students

Abstract Format




Accession Number


Shelf Location

Archives, The Learning Commons, 12F Henry Sy Sr. Hall

This document is currently not available here.