Research on the integration of ChatGPT revolutionizing English language learning in EFL contexts has been documented. Yet, research on the acceptance of ChatGPT from the perspective of a technology acceptance model (TAM) in the EFL writing context remains underexplored. This descriptive qualitative study aims to explore the students' acceptance of ChatGPT in assisting writing tasks using the TAM framework. The study involved six eleventh-grade students enrolled in the language program of a senior high school in Lamongan, Indonesia. Data were gathered through semi-structured interviews and classroom observation and then analyzed using a thematic pattern. The findings reveal a nuanced dynamic. EFL students view ChatGPT not merely as a correction tool in writing but as a cognitive scaffold that offers psychological safety through non-judgmental feedback, fostering self-regulated learning. Conversely, while the technical interface is intuitive, the students face significant semantic barriers when writing, requiring iterative "prompt engineering" to mitigate AI hallucinations. These results indicate that successful integration relies less on the tool's mechanics and more on the user's digital capabilities. The study implies a pedagogical shift towards critical AI literacy, encouraging educators to guide students in treating AI as a collaborative partner in hybrid intelligence rather than a replacement for critical thinking.
This study examines the legal and social paradox of marriage dispensation in Indonesia, particularly in cases of premarital pregnancy, where child protection objectives often conflict with the formal legitimacy of marriage. Although frequently justified as a protective legal mechanism, this article argues that marriage dispensation primarily functions as an emergency response, offering short-term legal relief while generating long-term social harm. Using a socio-legal and normative-empirical approach, the study applies the maqāṣid al-sharīʿah framework, particularly tarjīḥ maqāṣidī, to evaluate legal priorities in light of the higher objectives of Islamic law. Its novelty lies in measuring maṣlaḥah through multi-dimensional indicators scope, durability, precision, effectiveness, and temporal impact based on empirical data from Religious Courts and Women’s Empowerment and Child Protection agencies in East Java, Central Java, and West Java. The findings show that the benefits of marriage dispensation are limited to formal lineage recognition and stigma reduction. At the same time, its harms are extensive and enduring, including reproductive health risks, educational disruption, psychological vulnerability, and intergenerational poverty. From a maqāṣid perspective, this imbalance reflects a failure to protect life, intellect, and progeny. This study concludes that the marriage dispensation is normatively inconsistent and calls for prevention-oriented legal harmonization that prioritises child protection.
This study examines the effect of traditional game-based learning with Congklak on numeracy achievement among second-grade elementary students. The study addresses the limited availability of experimental evidence on culturally grounded, non-digital games for early numeracy instruction. A pre-test–post-test control-group design was employed with 60 students (aged 7–8 years) from a public elementary school in Indonesia. Participants were assigned to either an experimental group receiving Congklak-based numeracy instruction or a control group receiving conventional textbook-based instruction for four weeks. Numeracy achievement was measured using the Numeracy Achievement Test (NAT), consisting of 30 dichotomously scored items covering number recognition, counting skills, and logical reasoning (score range 0–100; reliability coefficient α = 0.82). Data were analyzed using paired-samples and independent-samples t-tests. Results indicated significant pre-test–post-test improvement in both groups (p < .001). However, the experimental group achieved higher post-test scores (M = 78.47, SD = 6.92) than the control group (M = 66.13, SD = 7.25), with a statistically significant between-group difference, t (58) = 4.96, p < .001, d = 1.27. These findings suggest that integrating traditional Congklak-based activities into mathematics instruction can support early numeracy development through culturally responsive and play-based learning experiences.
Waterlogging and household organic waste are environmental problems that require active community participation. Biopore infiltration holes can improve rainwater infiltration while providing a medium for organic waste decomposition. This community service program aimed to improve the knowledge, skills, and participation of residents of Turi Village, Lamongan Regency, in constructing and maintaining biopore infiltration holes. Activities included problem identification, partner coordination, socialization, hands-on training, mentoring, and evaluation. Thirty-two participants were involved. Evaluation used pre- and post-activity questionnaires, observation of practical skills, and monitoring of biopore conditions. The program produced 12 biopore infiltration holes. Average knowledge increased from 50% in the pretest to 90% in the posttest, while 50% of participants demonstrated correct construction and maintenance procedures. Organic waste added to the holes averaged approximately 1 kg per week. The program indicates that training combined with practice and mentoring can strengthen community capacity for rainwater and organic waste management. Continued implementation requires clear responsibility, routine maintenance, and periodic monitoring.
Mango (Mangifera indica L.) is one of the most widely cultivated horticultural crops due to its high economic value. However, leaf diseases remain one of the primary factors affecting crop productivity. Manual disease identification has several limitations, as it is time-consuming, highly dependent on the expertise of the observer, and inefficient for large-scale cultivation areas. This study aims to analyze the effects of epoch and batch size variations on the performance of the MobileNetV2 model for mango leaf disease classification using a transfer learning approach. The dataset consisted of 4,000 images representing eight classes: Healthy, Anthracnose, Bacterial Canker, Cutting Weevil, Die Back, Gall Midge, Powdery Mildew, and Sooty Mould. The dataset was divided into training, validation, and testing sets using an 80:10:10 ratio. Four training scenarios were then conducted using different combinations of batch sizes (16 and 32) and epochs (25 and 50 The experimental results demonstrated that the configuration with a batch size of 32 and 50 epochs achieved the best performance, obtaining an accuracy of 99.75%, while the precision, recall, and F1-score each reached 100%. These findings indicate that variations in batch size and number of epochs significantly influence the performance of the MobileNetV2 model, with the combination of a batch size of 32 and 50 epochs providing the most optimal classification performance for the mango leaf disease dataset.
