Development and Validation of Writing Assessment Tools for Grade 7 Second Language Learners of the Division of Abra

Abstract

This study developed and validated writing assessment tools for Grade 7 second-language learners in the Division of Abra, addressing the gap in systematically validated teacher-made instruments for assessing writing proficiency. Anchored on Constructivist Learning Theory (Vygotsky, 1978) and Messick's Unified Theory of Validity (1989), the research employed a developmental research design involving 27 Grade 7 learners from Abra High School–Calot Extension and three expert validators in English language teaching, curriculum development, and language assessment. The researcher-developed instruments comprised five contextualized writing tasks aligned with the Most Essential Learning Competencies (MELCs), evaluated using a holistic rubric assessing content, organization, language use, grammar, and mechanics on a four-point rating scale. Data analysis utilized Weighted Mean for validity levels, Fleiss' Kappa for inter-rater agreement, and Cronbach's Alpha for reliability. Results demonstrated substantial improvement from initial to final validation, with weighted mean scores increasing from 4.50–4.70 to 4.75–4.96, indicating Very Highly Valid ratings. Fleiss' Kappa coefficients improved from −0.154–0.375 (weak agreement) to 0.391–1.000 (moderate to almost perfect agreement) following expert revisions. Reliability analysis revealed acceptable internal consistency for Tool 1 (α = 0.770), Tool 2 (α = 0.700), and Tool 5 (α = 0.710), while Tool 3 (α = 0.521) and Tool 4 (α = 0.631) require further refinement. The validated instruments provide English teachers with curriculum-aligned, evidence-based assessment tools that promote objective and consistent classroom assessment. The study recommends full implementation of the validated tools in Grade 7 classrooms, enhanced validation procedures with orientation sessions for experts, and further refinement of lower-reliability instruments through clearer rubric descriptions and rater training. Future research should involve larger and more diverse samples and additional expert validators to strengthen the generalizability and applicability of the developed writing assessment tools.