Cultivating students' data science knowledge and skills is pressing and challenging, given its interdisciplinary nature, students' limited prior knowledge, and teachers' insufficient training. In data science learning, students may experience various academic emotions. Understanding what emotions students experience, how these emotions are associated with their perceived learning, and under what conditions they experience intensive emotions is critical to informing the design of data science programs and better supporting students. This study collected 839 emotion survey responses from 67 secondary school students in two cycles of a two-day out-of-school data science program. The program engaged students in collaborative inquiries on authentic problems through data science practices with the support of teachers, researchers and facilitators. We found that frustration, interest, surprise and happiness positively predicted students' perceived learning, whereas anxiety negatively predicted perceived learning. Students experienced peaks of positive emotions after an expert's enthusiastic introduction talk to data science in the first cycle and after one-to-one face-to-face consultations with data science experts in the second cycle. However, sharing their progress and challenges with the data science expert in the first cycle and preparing for presentations in both cycles made them experience intense negative emotions such as anxiety, frustration, and confusion. These findings provide implications for designing data science programs to elicit students' positive learning experiences and reduce intensive negative emotions.

GPT has demonstrated impressive capabilities in executing various natural language processing (NLP) and reasoning tasks, showcasing its potential for deductive coding in social annotations. This research explored the effectiveness of prompt engineering and fine-tuning approaches of GPT for deductive coding of context-dependent and context-independent dimensions. Coding context-dependent dimensions (i.e., Theorizing, Integration, Reflection) requires a contextualized understanding that connects the target comment with reading materials and previous comments, whereas coding context-independent dimensions (i.e., Appraisal, Questioning, Social, Curiosity, Surprise) relies more on the comment itself. Utilizing strategies such as prompt decomposition, multi-prompt learning, and a codebook-centered approach, we found that prompt engineering can achieve fair to substantial agreement with expert-labeled data across various coding dimensions. These results affirm GPT's potential for effective application in real-world coding tasks. Compared to context-independent coding, context-dependent dimensions had lower agreement with expert-labeled data. To enhance accuracy, GPT models were fine-tuned using 102 pieces of expert-labeled data, with an additional 102 cases used for validation. The fine-tuned models demonstrated substantial agreement with ground truth in context-independent dimensions and elevated the inter-rater reliability of context-dependent categories to moderate levels. This approach represents a promising path for significantly reducing human labor and time, especially with large unstructured datasets, without sacrificing the accuracy and reliability of deductive coding tasks in social annotation. The study marks a step toward optimizing and streamlining coding processes in social annotation. Our findings suggest the promise of using GPT to analyze qualitative data and provide detailed, immediate feedback for students to elicit deepening inquiries.

This study assessed the effectiveness of rehabilitating children with autism with a series of Leap Motion-based applications in a special school setting. Experiment was carried out according to an AB sequence. Experimental result showed that the two participants' fine motor skills improved significantly, and their recognition of colors and fruits after the intervention was 100%.

The advancement of mobile device is influencing the learning activities markedly. In this study, we attempt to use augmented reality (AR) technology to design and develop mobile-based English learning software for pre-school children in order to solve the problem of bored students and teachers' non-standard pronunciation, the mobile learning system is able to present the learning materials including virtual pictures, the meaning and pronunciation of words. A vivid picture will emerge when using mobile camera to identify an English word on card, which improves children's interests in learning. The 40 pre-school children who participated in this research were assigned to an experimental and a control group. From the pre-tests and post-tests as well as the interview of the English teacher, it was found that the students learning with mobile-based AR software had greater learning achievement than control group ones. Thus the teacher hold positive attitude to this software.