Answers are grounded in this paper and cite their sources.
This study examined the observed grammar performance and the assessment's exploratory latent structure among 150 first-year students at an autonomous university in Davao City, Philippines. Participants completed a 45-item assessment covering Subject–Verb Agreement, Punctuation, and Transitional Words. One zero-variance item was excluded from psychometric analyses. Exploratory factor analysis used a smoothed tetrachoric correlation matrix, minimum residual extraction, and geominQ oblique rotation. Binary parallel analysis recommended four factors, whereas the scree plot suggested three. A three-factor solution: Integrated Grammar Mechanics, Cohesion and Transitional Language, and provisionally labeled Sentence Grammar was retained as a parsimonious summary, with the four-factor solution acknowledged as a plausible alternative. Mean percentage scores were 71.33 ± 17.20 for Subject–Verb Agreement, 92.84 ± 7.96 for Punctuation, and 95.20 ± 9.43 for Transitional words. Overall raw alpha/KR-20 was 0.798; estimates for the EFA-derived item groups ranged from 0.698 to 0.758. The three-factor loading pattern was similar across two matrix-adjustment methods but changed when items with sparse response categories were excluded. Only Factor 3 differed across academic colleges after Benjamini–Hochberg adjustment ( p = 0.0139); its regression model explained 9.1% of score variance (adjusted R 2 = 0.066). However, opposing agreement and punctuation loadings precluded interpreting higher Factor 3 scores as stronger overall sentence grammar proficiency. The findings describe uneven observed performance and a provisional factor structure. Ceiling effects, sparse responses, matrix adjustments, and dimensional uncertainty limit interpretation and warrant evaluation in larger, independent samples.
No takes yet. Share an insight, caveat, or question.
Masangya et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: