Journal of Applied Statistics and Information Technology

Development of a Digital Game-Based Learning Tool to Foster Environmental Conservation Awareness

Nuttapong Paitoon — 2025-12-24

This research aimed to design and develop a digital game to foster environmental conservation awareness among upper primary school students. The game, titled “Plant More Trees,” was created using Scratch 3.0 and consisted of 12 progressive levels involving tasks such as watering, fertilizing, and protecting trees from pests. The objective was to integrate environmental knowledge and promote sustainable behavior through interactive gameplay. The game was tested with 71 students from Grades 4 to 6. The results showed a statistically significant increase in environmental knowledge after gameplay, with high levels of satisfaction reported by the participants (average satisfaction scores of 4.19 - 4.30). The findings indicate that Plant More Trees is an effective learning tool for instilling environmental awareness in children and holds potential for use in modern educational contexts.

Cognitive and Behavioral Factors Affecting Thai Farmers’ Intention to Use IoT in Smart Agriculture

Techit Paisanthanik — 2025-12-24

The objective of this study is to examine the cognitive and behavioral factors that affect Thai Farmers’ Intention to use IoT in Smart Agriculture. Data were gathered from a simple random sample of 400 Thai Farmers who were familiar with IoT technologies but had not yet implemented them in their agricultural practices. Structural Equation Modeling was used to assess both direct and indirect effects of various factors on IoT adoption. The results show that cognitive factors -- namely perceived security, self-efficacy, cost-effectiveness, enjoyment, and perceived control -- significantly influence farmers' intention to use IoT through their perceived usefulness. In particular, the perceived enjoyment and perceived control have a significant direct influence on perceived usefulness, while other cognitive factors have only indirect effects. For behavioral factors, the intention to use IoT technology was influenced directly and indirectly by cognitive factors through perceived usefulness and perceived ease of use. These findings provide empirical insights for government agencies in designing effective and sustainable policies to promote IoT adoption in the agricultural sector. Moreover, developers of IoT devices and platforms can utilize these insights to design products that align with the needs and behaviors of Thai farmers.

An Investigation of the Word Recall: Cognitive Assessment in Aged Person from The Health, Aging and Retirement in Thailand (HART) Project

Chidapha Worapukdee — 2025-12-24

This study aims to evaluate the efficacy of the ten-word recall test which is an effective tool in cognition assessment. Data from the Survey from HART project, Panel 2022-2023, Wave 4 have been used for in-depth analysis. Our analysis first focuses on understanding difficulty level, discriminating power and reliability of the four sets of the ten-word recall test by utilizing descriptive statistics. We then compare difficulty level and discriminating power among the four sets of the ten-word recall test by one-way ANOVA.

Our findings reveal that all four sets can be classified as moderate with regards to their difficulty level with eight out of ten words. Moreover, only the first set has very good discriminating power with eight out of ten words whereas the other three sets also have very good discriminating power with six out of ten words. Moreover, both discriminating power and difficulty level have a statistically nonsignificant difference means among the score from four sets of ten-word recall test. Besides, improving reliability is necessary.

A 3-Phase Quality Inspection Model for Improving Delivery Efficiency and Quality in Government Software Development Projects

Praphan Surapapha — 2025-12-24

This research presents a model for efficient, high-quality delivery of government software development projects using a 3-phase quality inspection method integrated into the project management process to address delays and reduce post-delivery rework. Inspections are systematically applied at three key checkpoints throughout the project lifecycle, verifying that project deliverables are complete, accurate and consistent with (1) the Scope of Work (SOW), (2) the Terms of Reference (TOR) and (3) the ISO/IEC 29110 standard.

A quasi-experimental study compared an experimental group of 91 projects (2015–2024) using the 3-phase model with a control group of 31 projects (2011–2014) that followed only the standard ISO/IEC 29110 process. Both groups delivered all project milestones on time, but the experimental group achieved much higher delivery efficiency. Over half of its milestones (53.9%) were completed ahead of schedule, compared to only 15.3% for the control group. Delivery quality was also slightly higher in the experimental group: 99.3% of its milestones required no post-delivery rework, versus 93.9% in the control group. A Mann–Whitney U test confirmed that the experimental group’s superior delivery efficiency and quality were statistically significant (p < .01). These results indicate that the 3-phase inspection mechanism enables defects to be detected and corrected early, significantly reducing delays and the need for post-delivery rework. A path analysis further confirmed that all three inspection phases have a significant positive impact on both delivery efficiency and quality. These findings underscore the value of proactive quality control from a project’s beginning to its end. Incorporating periodic inspections into project management processes can facilitate project delivery that is "faster and of higher quality." This preventive approach can be practically applied to address late delivery issues and significantly enhance the success of government software development projects.

A Poisson-Generalized Mixture Exponential Distribution

Chadarat Tapan — 2025-12-24

In this study, a new Poisson-generalized exponential distribution is proposed. Its statistical properties, including the cumulative distribution function, probability generating function, moment generating function, survival function, and index of dispersion, are derived. The proposed distribution has three parameters and exhibits a right-skewed, unimodal shape. Parameter estimation is performed using the maximum likelihood method. The distribution is then applied to two real overdispersed count datasets and compared with the Poisson and negative binomial distributions. The results indicate that the Poisson-generalized exponential distribution provides a better fit to both datasets based on the Akaike information criterion, Bayesian information criterion, negative log-likelihood, and the Kolmogorov–Smirnov goodness-of-fit test. These findings suggest that it is a viable alternative for modeling overdispersed count data. Future research may explore its extension to regression frameworks and other applied contexts.