Connecting preparation, lessons and follow-up through digital homework

Intelligent homework enhances the quality and efficiency of teaching  

Gao Yinli, Jie Wei, Zhao Mengying, Ye Jing  

Yinchuan No. 15 Secondary School  

Abstract:This case takes Yinchuan No. 15 Middle School as an example to explore the application of smart homework in enhancing the quality and efficiency of teaching. In response to the need for digital transformation in education and the drawbacks of traditional homework systems, the school introduced a smart homework system. Through the “two main and three-loop” model, it achieves intelligent homework management, precise pre-class preparation, immediate interaction during class, and personalized after-class assistance. The smart homework system not only improves teaching efficiency, but also promotes personalized learning for students, enhances teachers’ information literacy, and serves as a model for digital transformation in education, providing valuable experience for such transformations.    

Keywords:Smart homework; improving the quality and efficiency of teaching    

1. Background of the case  

At present, China is making every effort to advance the digital transformation of education. The “Action Plan for Deepening the Reform of Basic Education Curriculum and Teaching” issued by the Ministry of Education in 2023 clearly states that digital technologies should be utilized to enhance basic education and establish new models of teaching and learning, thereby providing a clear direction for innovative development in the field of education. However, traditional teaching methods have many drawbacks in the aspect of homework, showing a tendency of emphasizing teaching over learning, knowledge over skills, and results over the learning process. Teachers focus too much on the one-way transmission of knowledge, ignoring the learning process and the development of students’ abilities. Such a model cannot meet the urgent needs of the new era for personalized education and the cultivation of students’ higher-level thinking skills. At the same time, schools face numerous challenges in implementing precise teaching. The digital capabilities of teachers vary widely, and data collection and application are inconsistent. These issues significantly hinder the effective implementation of precise teaching. Nevertheless, homework data is easy to obtain and structured, making it a key opportunity for promoting teaching innovation. Smart homework, as an important application of educational digital transformation, offers multiple advantages such as data collection, intelligent analysis, and personalized recommendations. It provides strong support for solving problems related to traditional homework and achieving precise teaching. In this context, Yinchuan No. 15 Middle School has seized the opportunities of the times, actively responded to the digital education strategy, and has taken the exploration and application of smart homework as an important measure to improve teaching quality and promote the development of teachers and students. It has thus embarked on a new journey of teaching empowered by technology.  

II. Implementation strategies for using intelligent assignments to enhance teaching  

To effectively promote the intelligent homework project, Yinchuan No. 15 Middle School has carried out thorough top-level planning and systematic design. This ensures the deep integration of technology and teaching, and gradually establishes a sustainable intelligent homework system that suits the school’s actual conditions.  

(1) Overall design concept and theoretical framework  

Based on the core concept of precise teaching—that is, “precise goals, precise diagnosis, and precise intervention”—the school regards intelligent assignments as a crucial link in connecting various teaching stages and enabling data-driven learning. On this basis, No. 15 Middle School has innovatively introduced the “Double Leadership, Three Circles” classroom teaching model (as shown in Figure 1). “Double Leadership” means that teachers play a leading role while students are the main subjects; it emphasizes the collaboration and interaction between teachers and students. “Three Circles” refer to the organic integration of three stages: before class, during class, and after class, as well as the connection between these stages and data.  

Before class: Personalized preview tasks are assigned through intelligent homework, enabling a precise understanding of students’ learning situations, and laying the foundation for teaching based on students’ needs.  

In class: Implement hierarchical cooperative learning; use the intelligent homework system for real-time feedback on students’ learning progress. Combine online and offline teaching methods, so that more control over the class is given to the students.  

After class: Through the intelligent homework system, personalized tutoring and additional learning opportunities are provided, enabling intelligent grading and precise feedback on homework.  

The core of this model lies in using intelligent assignment systems to collect and analyze data throughout the entire teaching process, thereby creating a closed loop for teaching. This, in turn, enables more precise teaching practices and more personalized learning for students.  

(II) Focus on Practice: Exploration of the Deep Application of Intelligent Assignments in Various Aspects of Teaching  

Yinchuan No. 15 Middle School has integrated the intelligent homework system deeply into all aspects of teaching. From homework management, pre-class, in-class, and post-class tutoring, to teaching research and administrative management, precise, data-driven reforms have been achieved.  

1. Intelligent operation management: The foundation for improving quality and efficiency

The intelligent homework system has brought about revolutionary changes in homework management, freeing teachers from tedious administrative tasks.  

Firstly, there is the use of intelligent arrangements and innovative forms of homework. Teachers can utilize the school-based resources, teaching materials, and past exam questions provided by the system to selectively assign homework at different levels – such as basic questions, advanced problems, and extended tasks – in order to meet the needs of students at various levels. At the same time, innovative forms of homework are also introduced, such as English word cards, reading assignments, writing tasks, and project-based check-in assignments. These methods enrich the nature of homework and stimulate students’ interest in learning.  

Secondly, there is efficient grading and immediate feedback. Objective questions are automatically graded by AI, while subjective questions can be graded online by teachers, with the grading records retained. Alternatively, intelligent grading machines can be used to enable collaborative grading between humans and machines. As a result, the grading efficiency is greatly improved, and the time required for grading assignments is reduced by nearly 50%. The grading results are provided to students in real time, enabling them to make corrections and reflect promptly.  

Thirdly, it involves multi-dimensional data statistics and analysis. The system can generate multi-dimensional assignment reports in real time (as shown in Figure 2), including the completion rate of students’ assignments, average scores, score distribution, score rates for individual questions, option distribution, and a list of students who did not complete their assignments. These intuitive, visualized data provide teachers with great convenience in accurately understanding the learning situation of students.  

2. Pre-class preparation and accurate understanding of students’ learning situations: The premise of “teaching based on learning”

The application of intelligent homework in the pre-class preparation phase enables teachers to move away from traditional, experience-based lesson preparation and instead adopt data-based, well-prepared lessons.  

Firstly, prepare “survey”-type preview exercises. Teachers design a small number of key preview questions related to the new concepts that will be learned, and distribute them through the smart homework system. For example, before teaching the topic “The Conditions of Floating and Sinking of Objects”, it was found through the preview questions that “52% of the students in the class have the incorrect physical concept that ‘heavy objects sink, while light objects float’”.  

Secondly, it involves data-driven teaching design. The data obtained from students’ responses to preview assignments can clearly reveal their prior concepts, knowledge base, and potential difficulties. Based on this data, teachers can accurately determine the starting point of teaching, adjust teaching objectives and strategies, and optimize the design of teaching activities. In this way, teaching can be tailored to the students’ needs, making classroom teaching more targeted and effective. For example, based on the feedback from students’ preview responses regarding “The Conditions of Floating and Sinking of Objects,” teachers can design targeted experiments to help students develop the correct understanding of these concepts.  

3. Interaction during classes and immediate feedback: A key to dynamic regulation

The intelligent homework system extends into the classroom, becoming a powerful tool for providing immediate feedback and dynamic teaching adjustments.  

Firstly, there is the ability to conduct immediate assessments during class and display data promptly. In line with classroom learning objectives, teachers can use the Smart Homework system to conduct immediate evaluations during class. After students answer questions using their devices, the system can immediately count and display the results for the entire class, such as the accuracy rate and the number of students who selected each option (as shown in Figure 3). This approach overcomes the problems associated with traditional methods, such as the lack of representativeness of individual responses or the tendency for groups to give misleading answers.  

Secondly, there are adjustments to teaching strategies based on data. Teachers can quickly determine how well students have understood the learning objectives, using the immediate feedback provided, and then adjust their teaching methods accordingly: “Green-colored questions are no longer explained; orange-colored questions are discussed by the students; red-colored questions require a new approach to teaching.” This immediate feedback helps improve students’ learning efficiency by nearly 60%.

4. After-class tutoring and personalized learning support: A approach to teaching tailored to individual needs

The intelligent homework system breaks down data silos. By collecting and integrating data related to homework, in-class exercises, and exams, it creates a personalized learning profile for students, thereby enabling targeted tutoring and personalized learning.

Based on the graph, teachers comprehensively assess students’ knowledge mastery and learning quality, identify the root causes, and precisely locate each student’s weak points in learning. Starting from the students’ learning situation, teachers provide differentiated guidance to address specific gaps effectively.

Firstly, school teachers use “targeted classes” as a means to implement personalized teaching in depth. These targeted classes are mainly divided into those aimed at addressing issues related to the mastery of knowledge points, and those aimed at promoting students’ innovative learning.

In the targeted classes for mastering knowledge points, temporary targeted classes are established in response to specific mistakes made during homework or exams. Similar questions are provided to help students reinforce their understanding and promptly address any knowledge gaps. Phase-based targeted classes focus on students who have difficulty grasping certain aspects of the learning material. Teaching videos, course materials, and exercises are provided to help students improve their understanding; these materials are removed once students have mastered the content, thereby enabling them to keep up with the teaching progress. Long-term targeted classes aim to help students improve their overall academic performance by providing tailored content for those who need extra support, while also promoting their overall development. For example, if a teacher notices that students are struggling with the concept of “electricity and magnetism,” they can create a targeted class for those students, provide explanation videos, and thus increase the students’ score rate in that concept from 73% to 81.9%.

Innovative learning targeted classes are established based on students’ strengths and interests. For students with strong thinking abilities, innovative learning targeted classes are created to develop their innovative mindset and problem-solving skills; for those who are good at hands-on activities, experimental exploration targeted classes are offered to enhance their practical skills and scientific inquiry abilities.  

Secondly, it involves error correction and reinforcement activities led by teachers. The smart homework system automatically records all students’ errors (as shown in Figure 6). Teachers can, based on the principles of Ebbinghaus’ forgetting curve, prompt students to review their errors at various time points. For example, when assigning homework, the system can activate the “error review” function, and it will automatically send personalized errors from previous assignments. Teachers can also select common errors that occur frequently among the class and send those to students. Students can also download their own personalized error notebooks. The coverage rate of error correction is 76.7%, and the score improvement rate after re-reviewing errors is 80.4%.  

5. Data-driven precise teaching and research: The engine of professional development

The data generated by intelligent homework not only serves daily teaching, but also brings new vitality to teaching and research activities.  

Firstly, evidence-based group lesson preparation. The lesson preparation teams use data from smart homework assignments to accurately analyze students’ learning situations, and they jointly discuss common problems and solutions. For example, during the review period for ninth-grade students, some students had difficulty understanding the image-based methods. Through collective teaching discussions and the use of new teaching methods, the students’ score rates in knowledge points improved significantly. The school has standardized the process of group lesson preparation during the review phase (as shown in Figure 7); it emphasizes “analyzing students’ learning situations”. That is, the lesson preparation teams analyze students’ learning situations based on data, discuss common problems and solutions, and brainstorm new teaching strategies.  

Secondly, it involves optimizing school-based research and training. Schools use data from intelligent homework assignments to carry out thematic teaching activities, such as “Wendao Forums”. This helps to identify strengths and problems in teaching, enabling teachers to reflect on and improve their teaching methods based on the data, thereby enhancing their professional capabilities.  

6. Detailed teaching management: The basis for scientific decision-making

The data from the intelligent homework system also provides strong support for school teaching management. Through 32 indicators across 6 major categories, schools implement data-driven, detailed teaching management. Management personnel can easily access data reports from various perspectives: academic subject levels, teacher behaviors (such as the number of learning tasks, data on subjective question grading, data on resource creation, and data on the implementation of targeted classes), and student behaviors (such as data on video views of micro-lectures, homework completion rates, the rate of retesting of wrong answers, and score rates). This makes the teaching process more visual, data-driven, and transparent, thereby providing a scientific basis for teaching decisions and resource allocation.  

III. Stage Achievements  

Since the intelligent assignment system was put into use in the autumn semester of 2023, Yinchuan No. 15 Middle School has achieved encouraging results in various aspects of education and teaching. The data-driven educational reforms have not only improved students’ academic performance, but also promoted the professional development of teachers and the overall quality of the school’s operations.  

Firstly, it enables high-frequency, regular use of the system. Since the implementation of the intelligent homework system, there have been positive feedback from teachers and students in our school. Taking the ninth grade as an example, during the entire 2024 academic year, 54 teachers used the system, and 1,032 students did so. A total of 21,612 assignments were assigned, involving over 170,000 questions. The submission rate of students’ assignments was 73.27%, and there were more than 5,000 targeted tasks. The process of grading and statistics of assignments saves approximately 50% of time compared to traditional methods. Students used the intelligent homework system to complete over 29,490 self-study tasks, and the rate of reworking wrong answers was 73.46%. A large amount of digital course resources have been created; these include video-based courses covering all subjects and the complete review for the ninth grade. The number of videos exceeds 1,100. The use of the intelligent homework platform has become regular, consistent, and widespread.  

Secondly, it involves improving teachers’ information literacy. The teachers in our school utilize various teaching technology tools to effectively integrate traditional and digital resources. They deepen the reform of school-based teaching and research, develop original test questions, and diversify the forms of homework, such as flexible homework, hierarchical homework, interdisciplinary homework, and project-based homework. High-quality school-based resource libraries are also established for different subjects, units, and classes. At the same time, teachers use information technology to collect and analyze students’ learning data, thereby accurately assessing teaching effectiveness. Instead of relying on experience, teaching becomes guided by data-based evidence. In 2024, the case study “Smart Education Enhances Teaching and Learning: Improving Quality and Efficiency to Support the ‘Double Reduction’ Policy” won the first prize in the district’s case competition. One digital-related project was approved as a regional research project, four projects were approved as Yinchuan City projects, and two projects were approved as Xingqing District projects. Teacher Shi Haibo from the Department of Mathematics shared his ideas on the design of smart homework for mathematics units in the whole city. The overall improvement in teachers’ information literacy not only stimulates innovation in education but also significantly enhances the efficiency of school management, aligning with the development trend of educational digitalization.  

Thirdly, it plays a role in providing examples. On May 15, 2025, Teacher Gao Yinli from our school presented a case study on digital empowerment in precise teaching at the parallel session of the World Education Digital Conference, titled “The Transformation of Teachers’ Roles and Their Skill Development in the Intelligent Era: From Experience to Expertise – The Professional Growth of Teachers”. On June 12, 2024, we participated in the regional educational digitalization promotion meeting and demonstrated the pilot projects of the school’s smart homework program; this was reported by Ningxia Daily. In September 2024, we shared our school’s experiences during regional business exchange activities focused on improving educational quality through digitalization. In July and November 2024, we organized training sessions on smart homework in Xingqing District, as well as observation activities at schools. We have carried out numerous exchange and demonstration activities both at the regional and school levels, creating a professional exchange platform. By presenting various typical lessons, experiences, and practices, we serve as a model school that can guide and influence others.  

IV. Problems and Prospects  

Looking to the future, Yinchuan No. 15 Middle School will continue to deepen the application of smart homework systems. It will also make improvements based on the shortcomings identified during implementation, striving to advance towards a higher level of smart education.

First, improve the data analysis model and the construction of teacher profiles. Further explore the integration of student assignment data with teacher classroom teaching behavior data and teaching research data, and attempt to create a map of teachers’ professional capability development, so as to provide more precise support for teachers’ personalized professional development.

Secondly, it is necessary to further enhance personalized learning path recommendations. Based on existing targeted classes, artificial intelligence and learning analysis technologies can be utilized to provide students with more intelligent and dynamic personalized learning paths and resources. This will enable adaptive learning that suits each student’s individual needs.

Thirdly, it is necessary to promote the integration of data from various sources and the development of comprehensive evaluations. We should actively explore ways to combine data related to smart learning activities, as well as data on students’ overall qualities and physical health monitoring, in order to create a more comprehensive system for assessing student development. This will help achieve the educational goal of integrating the five aspects of education. The aim is to establish a closed-loop evidence-based practice system that encompasses “design, implementation, and evaluation” and involves “interaction between three interconnected elements”.