Daniel Cubilla-Bonnetier

Explicit phonics instruction and decodable texts are well supported by evidence, yet little is known about how such programs perform when implemented at scale in low-resource, Spanish-speaking public school systems. This study evaluated Aprendiendo a Leer en Primero, a first-grade program combining the Kalulu phonics methodology with decodable books, in 22 public schools in the Dominican Republic, using a mixed methods intervention design. Schools were cluster-randomized, with stratification by district, to an experimental (12 schools) or control condition (10 schools) (N = 830 children assessed at pretest and posttest). Analyses of covariance adjusting for baseline letter knowledge, phonological awareness, and oral comprehension showed that the experimental group outperformed the control group in letter knowledge and in syllable, word, and pseudoword reading accuracy, with small effect sizes, and no differences in phonological awareness or reading comprehension. The advantage was concentrated among children who entered first grade with greater letter knowledge. Implementation monitoring revealed wide variation across schools in instructional time lost to strikes and reduced schedules and in the use of the decodable books, and school results did not map directly onto dosage. A teacher survey (n = 28) and two focus groups documented high acceptability, with close instructional coaching identified as the main perceived success factor, alongside tensions with competing national programs. The findings support continuing the program while adding differentiated support for children with the weakest initial skills, protecting coaching quality, and planning for teacher turnover in future scale-up phases.

Children of adolescent mothers face elevated risks of early childhood development (ECD) delay. Whether this reflects maternal age or the socioeconomic disadvantage that predisposes young women to early childbearing remains debated. We conducted a secondary analysis of Multiple Indicator Cluster Survey data (rounds 5 and 6, 2013-2022) from 15 Latin American and Caribbean countries, comprising 23,650 children aged 36-59 months. Using survey-weighted regression with cluster-robust standard errors, we examined associations between adolescent motherhood and ECD scores, discipline practices, and beliefs about physical punishment and partner violence. Children of adolescent mothers had lower weighted mean ECD scores in 11 countries. After adjusting for wealth, maternal education, and urban/rural residence, significant adverse associations persisted only in Paraguay and Suriname. Violent discipline was significantly higher among adolescent mothers only in Honduras after covariate adjustment. Adolescent mothers were significantly more likely to justify partner violence in Belize, Paraguay, and Suriname. Justification of partner violence was associated with lower ECD scores among non-adolescent mothers in the Dominican Republic and El Salvador. The ECD disadvantage associated with adolescent motherhood is largely accounted for by socioeconomic factors in most countries. Addressing poverty and gender inequity remains the most broadly applicable strategy for reducing ECD inequity in the region.

Low reading achievement is a persistent challenge in Latin American education, particularly in the Dominican Republic, where most primary school students perform below minimum proficiency levels. Standardized, culturally adapted assessment tools are scarce in Spanish-speaking low-resource contexts, limiting early identification of reading difficulties at scale. This paper presents the development and validation of SARA (Self-Applied Reading Assessment), an AI-assisted tablet-based tool for autonomous administration and automated scoring of reading assessments in Dominican primary school children (grades 2–6). SARA integrates ECLEC (Evaluación Centroamericana de la Lectura), a battery grounded in the Simple View of Reading and the Dual-Route Model, covering decoding (letter identification, syllable, word, and pseudoword reading) and oral and reading comprehension. Following convergent validity testing against the PROLEC-R and two pilot studies, SARA was administered to 1,860 students from randomly selected public and private schools in Santo Domingo. Oral responses were scored by two independent human rater teams, yielding an overall inter-rater agreement of 86.11% (73.07%–94.70% across subtests) and a Cohen's Kappa of.550, a conservative estimate reflecting the class imbalance inherent in reading assessment data rather than poor rater reliability. Machine learning models were trained using the expert rater team's labels and evaluated against a conservative human consensus, defined by cases in which both teams agreed. Results showed good to excellent internal consistency across decoding subtests and adequate reliability for reading comprehension. Structural equation models confirmed construct validity by replicating the predicted developmental shift from decoding ( β = .70 in 2nd grade) to oral comprehension ( β = .60 in 6th grade) as predictors of reading comprehension. On the held-out test set, AI scoring achieved 91.97% overall accuracy, comparable to the 86.11% agreement between human raters, with class-specific accuracy higher for correct (90.0%–96.2%) than incorrect responses (73.9%–80.1%). Grade- and school-system-differentiated norms were established, reflecting the documented performance gap between public and private schools. SARA shows promising initial evidence of validity and reliability as a reading assessment tool for Spanish-speaking low-resource contexts and may constitute a scalable solution pending further validation in other regions and contexts.

Background The Dominican Republic consistently ranks last among Latin American countries in international reading assessments, yet the relative contributions of socioeconomic conditions versus school system quality remain unclear. This study examined reading performance differences between public and private school students and explored the influence of socioeconomic factors on these outcomes. Methods A total of 1,168 students from second grade to the beginning of secondary school were assessed using standardized instruments evaluating psycholinguistic precursors of reading (phonological awareness, rapid automatized naming, short-term verbal memory, and alphabetic knowledge), oral language skills (vocabulary, grammatical comprehension, and oral comprehension), and reading performance (fluency and comprehension). Group differences were analyzed using t-tests and Welch’s t-tests; socioeconomic predictors were identified via linear regression; and ANCOVA was used to estimate system-level effects after controlling for socioeconomic status (SES). Results Significant performance differences favoring private school students were found across nearly all variables, with medium to large effect sizes, particularly in reading fluency. Family income was the most consistent predictor of reading fluency, vocabulary, and comprehension. Controlling for SES substantially reduced or eliminated most differences between systems, especially in comprehension and linguistic skills, suggesting that SES accounts for a substantial portion of the variance associated with the educational gap. However, medium effect sizes persisted in decoding fluency, indicating a possible residual impact of the educational system. At the transition between 2nd and 3rd grade, 40% of public-school students lacked functional reading ability. Conclusions Socioeconomic conditions account for much of the reading gap between public and private school students in the Dominican Republic, though residual system-level effects persist in decoding fluency. These findings highlight the urgent need to universalize pre-primary education, strengthen early reading instruction, and implement tiered support systems such as Response to Intervention in public schools.

La conciencia fonológica es un precursor esencial del aprendizaje lector inicial. Se evaluó la efectividad de una intervención en conciencia fonológica en estudiantes de primer grado de escuelas públicas de la región metropolitana de Panamá. Se empleó un diseño cuasiexperimental con 318 estudiantes, divididos en un grupo experimental (n=154) y un grupo control (n=164) de contextos socioeconómicos diferenciados. La intervención consistió en un programa estructurado de conciencia fonológica con actividades lúdicas multisensoriales diarias de 20 minutos durante un año escolar. Se tomaron medidas pre-post de conciencia fonológica y de conocimiento alfabético y post de rendimiento lector. La intervención tuvo un efecto significativo en la precisión de la lectura de pseudopalabras y un efecto diferencial por nivel socioeconómico, con mayores beneficios en estudiantes de zonas vulnerables, que no lograron eliminar la brecha entre contextos. No se encontraron diferencias significativas en la velocidad lectora ni en la comprensión lectora. Estos hallazgos subrayan la importancia de la estimulación de la conciencia fonológica en contextos de pobreza multidimensional, de la generalización de la educación inicial y sugieren la necesidad de complementar estas intervenciones con estrategias de fortalecimiento en velocidad de denominación y acceso equitativo a materiales de lectura.

In a low-reading-achievement context, this study examines the applicability of the Simple View of Reading in the Dominican Republic, a Spanish-speaking country, assessing elementary students across the first seven years of formal schooling, and it investigates the contribution of key psycholinguistic precursors to decoding fluency and reading comprehension across public and private schools. Structural equation models were estimated for a sample of 1168 Dominican students across three grade-level clusters and for each educational subsystem (public vs. private). Model fit indices support the adequacy of the proposed latent structure. Our findings reveal that decoding emerges as the primary predictor of reading comprehension in the early grades, whereas language comprehension becomes increasingly influential in later grades—although this shift appears to occur later than reported in other contexts, particularly within public schools. Moreover, phonological awareness contributes persistently to decoding (even in 6th–7th), supporting the hypothesis of delayed decoding automatization, which could account for the reading-comprehension difficulties identified in international assessments. Socioeconomic position exhibits a decreasing effect on reading comprehension as students’ progress through school, although it remains significant at all grades. These findings highlight the need for educational policy approaches that accelerate decoding automatization to free cognitive resources for comprehension processes, emphasizing effective phonological-awareness training and explicit, systematic literacy instruction.

This study examined differences in reading performance between public and private school students in the Dominican Republic—a country with historically low reading achievement in international assessments—and explored the influence of socioeconomic factors on these outcomes. A total of 1,168 students from second grade to the beginning of secondary school were assessed using standardized instruments to evaluate psycholinguistic precursors of reading (phonological awareness, rapid automatized naming, short-term verbal memory, and alphabetic knowledge), oral language skills (vocabulary, grammatical comprehension, and oral comprehension), and reading performance (fluency and comprehension). Results revealed significant performance differences between public and private school students across nearly all variables, with medium to large effect sizes favoring the private system, particularly in reading fluency. Linear regression analyses identified family income as the most consistent predictor of reading fluency, vocabulary, and comprehension, while parental education also showed specific effects by grade level. Controlling for socioeconomic status (SES) substantially reduced or eliminated most differences between systems, especially in comprehension and linguistic skills, suggesting that SES explains much of the educational gap linked to structural inequalities. However, medium effect sizes persisted in decoding fluency, indicating a possible residual impact of the educational system. Findings underscore the urgent need to strengthen early reading instruction and to ensure universal access to high-quality pre-primary education, as 40% of public-school students lack functional reading ability by the end of 2nd grade. These results highlight the importance of educational policies aimed at mitigating structural inequities and promoting equitable literacy development in the Dominican Republic.

In order to describe and compare the school literacy environment and teaching practices among second and fourth grade Spanish teachers in public and private schools in the provinces of Panama and Panama Oeste, a total of 71 Spanish class observations were conducted, utilizing the ELLCO K-3 rubric to quantify the quality of literacy teaching, along with field notes for deeper qualitative understanding of the teaching process. Researchers found limited evidence of a high quality classroom environment or effective educational practices recommended by the literature. In second grade, scores were similar between private and public schools, slightly below 2 on a scale of 1 to 5. However, in fourth grade there was a larger difference between both systems, with scores of 1.85 and 2.7 respectively. Teaching practices aligned with the effective development of language and literacy were mostly absent in both types of schools, with scores below 1.4 in second grade and below 1.6 in fourth grade, with a slight advantage towards private schools. Results suggest the urgent need for interventions that address teachers’ current educational strategies through initial and continuous training, and a renewed commitment to integrate effective approaches in the teaching of literacy in Panama.

El modelo de la concepción simple de la lectura postula que la comprensión lectora es el resultado de la conjunción de dos factores: la comprensión del lenguaje oral y la destreza en la decodificación, que a su vezse asienta sobre precursorescognitivos como la conciencia fonológica y la velocidad de nombrado. Diversas pruebas internacionales han mostrado en estudiantes de Panamá un bajo nivel lector previo a la apariciónde la pandemia porSARS-CoV-2. Este estudio se realizó tras un año de suspensión de la educación presencial en el país. Su finalidad fueevaluar de forma virtual el estado del desarrollo lectorde una muestra representativa de 339 estudiantes de 2°, 4° y 6° de primaria de la región metropolitana de Panamá, estudiar la contribución a la lectura de determinadas capacidades psicolingüísticasy comprobar el ajuste del modelo de la concepción simple de la lectura en un contexto de bajo desempeño lector. Los resultados mostraron un desarrollo precario de la lectura en 2° grado. Por otra parte,el aporte a la lecturade la velocidad de nombrado fue creciendo a lo largo de la primaria y el de la conciencia fonológica resultó más persistente de lo esperado. Por último, se encontró un buen ajuste de la concepción simple de la lecturaal contexto de estudio, aunque con los rasgos distintivos deun aporte persistente del factor de decodificación a la comprensión lectora y una mayor contribución del componente de precisión quedel de fluidez al factor de decodificación.