YSYi Song

Yi Song

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article

Previous Organizations / Institutes

The organizations / institutes from the scholar's other articles in the past ten years


Yi Song has a total of 43 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • AI Evaluation and Feedback to Support Middle-School Students’ Scientific Argumentation and Reasoning

    by Field M. Watts, Lei Liu, Teresa M. Ober, Yi Song, Euvelisse Jusino-Del Valle, Yun Wang, Xiaoming Zhai

  • Exploring the Relationship Between Plan Features and Argument Essay Performance

    by Yi Song, Paul Deane, Chen Li

    We conducted a post hoc analysis of 771 students’ argumentative writing plans and essays in the Criterion ® database, a digital writing tool, to explore the relations among plan features, essay quality, and writing traits. Students in the study were in Grades 5 to 10 from 68 schools. We found that older students produced writing plans that received higher scores and demonstrated greater genre-specific knowledge than younger students, but regardless of their grade, most students did not consider alternative perspectives or rebut counterarguments in their writing plans. We also found that students’ choice of plan templates was associated with the scores of their plans. Further, factor analysis showed that six of the seven plan feature scores hung together in a single factor (Factor 1) and correlated with multiple trait scores (Factor 2), accounting for most of the shared variance connecting plan scores with writing traits. The “both sides” plan feature loaded on a different factor by its own, suggesting that considering different perspectives is a challenging skill that students may need extra support to develop.


  • Open Access

    Improving argumentative writing and argumentative knowledge of English Language Learners and non-English Language Learners through dialogic instruction: a mixed-effects regression analysis

    by Ralph P. Ferretti, Sam Van Horne, Henry May, Hilary Mead, Haobai Zhang, Yi Song, John Sabatini

    Although rudimentary forms of argumentation emerge early in oral language, many students, especially English Language Learners (ELLs), struggle to construct evidence-based written arguments. These difficulties may stem from limited academic vocabulary, unfamiliarity with genre conventions, and fewer opportunities to use academic language involving argumentation. This study examined whether exposure to a dialogic instruction intervention that includes small-group discussions, vocabulary instruction, and analytical reading of argumentative texts was associated with improvements in argumentative writing and conceptual understanding of argumentative concepts among middle school ELLs and non-English Language Learners (non-ELLs). In total, 802 students in the 6th (34.4%), 7th (34.4%), and 8th grades (31.2%) from the mid-Atlantic region of the United States participated in the study. Mixed-effects regression models revealed that while non-ELL students entered the study with stronger pre-intervention reading and writing skills, specific indicators of instructional exposure (e.g., attendance and engagement with aligned tasks) significantly predicted posttest gains for both groups. Notably, ELLs with higher reading proficiency showed greater benefits from curriculum-aligned assignments, while ELLs with lower proficiency gained more from consistent attendance. These findings highlight the need for responsive instruction that aligns supports with students’ linguistic readiness and task demands. The activities associated with this project were approved by the University of Delaware’s Institutional Review Board. The authors have no conflict of interest to report.


  • Open Access

    Is exposure to dialogic instruction associated with improvements in middle-school students’ argumentative writing and knowledge of argument concepts? Evidence from propensity score analyses

    by Ralph P. Ferretti, Yi Song, Haobai Zhang, Sam Van Horne, Hilary Mead, Zachary Collier, John Sabatini, Henry May

    Observational and experimental studies demonstrate that young children possess an inchoate understanding of argumentation. Nevertheless, they often fail to adequately justify their viewpoints in their oral and written argumentation. The current study investigated whether exposure to a dialogic instruction intervention during the Covid pandemic was associated with improvements in middle-school students’ argumentative writing and their conceptual understanding of argumentative concepts. In total, 802 students in the 6th (34.4%), 7th (34.4%), and 8th grades (31.2%) participated in the study. Instructional exposure was operationalized using four indicators (attendance, instructional mode, teacher planning engagement, and student participation in teacher-assigned activities) and analyzed using Latent Profile Analysis (LPA) to identify distinct exposure groups. Missing data were managed through multiple imputation, and Propensity Score Analysis (PSA) was used to estimate associations between exposure profiles and student outcomes. We found that students in the highest exposure group outperformed their peers in lower exposure groups, suggesting that high exposure to the dialogic intervention was generally associated with practically significant improvements in the quality of middle-school students’ written arguments and their knowledge of concepts that represent argumentative discourse.


  • Open Access

    A Framework for Designing an AI Chatbot to Support Scientific Argumentation

    by Field M. Watts, Lei Liu, Teresa M. Ober, Yi Song, Euvelisse Jusino-Del Valle, Xiaoming Zhai, Yun Wang, Ninghao Liu

    As large language models (LLMs) are increasingly used to support learning, there is a growing need for a principled framework to guide the design of LLM-based tools and resources that are pedagogically effective and contextually responsive. This study proposes a framework by examining how prompt engineering can enhance the quality of chatbot responses to support middle school students’ scientific reasoning and argumentation. Drawing on learning theories and established frameworks for scientific argumentation, we employed a design-based research approach to iteratively refine system prompts and evaluate LLM-generated responses across diverse student input scenarios. Our analysis highlights how different prompt configurations affect the relevance and explanatory depth of chatbot feedback. We report findings from the iterative refinement process, along with an analysis of the quality of responses generated by each version of the chatbot. The outcomes indicate how different prompt configurations influence the coherence, relevance, and explanatory processes of LLM responses. The study contributes a set of critical design principles for developing theory-aligned prompts that enable LLM-based chatbots to meaningfully support students in constructing and revising scientific arguments. These principles offer broader implications for designing LLM applications across varied educational domains.


  • Cultural validity: Promoting cultural responsiveness in classroom assessment

    by David Baidoo-Anu, Liu Lei, Dante Cisterna, Yi Song

    Drawing on a scoping-review methodology, this study aimed to map extant literature to conceptualize cultural validity in assessment. Three themes emerged from this review: (a) conceptualization of cultural validity, (b) promoting cultural validity in assessment, and (c) challenges to cultural validity in assessment. The review showed that culturally valid assessment ensures that assessment instruments and practices, assessment decision-making processes, and use of assessment scores in instructional and eligibility decisions are culturally responsive. Cultural validity should be considered as a core component of validity issues in the initial assessment-development process. To promote cultural validity in assessment, teachers/assessment developers should (a) develop awareness of cultural diversities, (b) leverage students’ everyday-life experiences, (c) allow students’ agency in assessment, (d) use collaborative-/participatory-assessment design, and (e) pay critical attention to language use. Standardization, lack of awareness of available research, paucity of resources on culturally responsive assessment practices, and limited professional-development-training opportunities on culturally responsive assessment were found to be the major threats to implementation of culturally responsive assessment practices. This review provides a foundation for better supporting teachers and assessment developers to effectively implement and engage in culturally responsive assessment.


  • Exploring the Role of Process Data Analysis in Understanding Student Performance and Interactive Behavior in a Game-Based Argument Task

    by Yi Song, Mengxiao Zhu, Jesse R. Sparks

    In this research, we use a process data analysis approach to gather additional evidence about students’ argumentation skills beyond their performance scores in a computer-based assessment. This game-enhanced scenario-based assessment (named Seaball) included five activities that require students to demonstrate their argumentation skills within a scenario about whether junk food should be sold to students. Our research sample included 104 middle school students. Process data analyses focused on an “Interview” activity in which students explored different locations and interviewed various characters to identify their opinions on the junk food issue and categorize each opinion as pro or con. Students could take various paths to complete the activity. Results indicated that the number of trials students made in the Interview activity predicted their performance on the Interview activity as well as the total Seaball scores. It was also found that most students improved their answers in the Interview activity after receiving automated feedback and making corresponding changes. Besides the connections between student activities and performance, results from analyzing the process data helped us to identify difficult items in the task. We conclude with implications for conducting process data analysis to better assess students’ argumentation skills and to inform task design.