Pingfei Zeng

This paper explores Knowledge Tracing (KT) and its role in modeling students’ knowledge progression over time. While traditional approaches such as Bayesian Knowledge Tracing and Deep Knowledge Tracing have established a solid foundation in this field, previous research often overlooked the complex interactions between problem features and learner behaviors. To address this gap, this paper introduces the Knowledge Tracking Model for Integrating Question Information and Learning Behaviour (IQILB-KT), a novel framework that enhances the precision of knowledge tracing in online educational systems. By integrating the effects of question characteristics and student behaviour via a channel attention mechanism, the model captures deeper insights into both question information and the learning process. Furthermore, the IQILB-KT model employs a Bottleneck Attention Module (BAM) to calculate learning behaviour synergies, replacing the previous reliance on covariance statistics. This enhances the stability and accuracy of the model when dealing with learning behaviour synergy data. Finally, the efficacy of the IQILB-KT model was validated through rigorous testing on two real-world datasets from the ASSISTments educational platform. The results demonstrate that the model outperforms all existing methods in predicting students’ future performance and tracking their knowledge states in real time.

Globally, as we progress toward a more sustainable future, the sustainable development goals (SDGs) must be incorporated into various sectors, including education. This article explores the significant impact of inventive leadership styles on SDGs in education at the secondary school level in Pakistan. Based on this study, four hypotheses were developed using leadership variables such as empowerment of leadership, inspiring learners, resilient visions, and lead transformation. A quantitative survey method was utilized in this study. The researchers selected convenient sampling to collect data from 288 teachers, including heads of schools and administrators from 89 institutions. The data collection process included using a survey questionnaire to collect data on leadership styles and their relationship regarding SDGs within the field of education. In order to analyze the given data, two software programs were used, namely SPSS and Smart PLS 4.0. Several statistical methods were used to examine the collected data, such as confirmatory factor analysis (CFA) and structural equation modeling (SEM), which were implemented to confirm the measurement model. Additionally, path analysis was conducted to determine the relationship between independent and dependent variables. The findings revealed that overall leadership styles have a great positive impact on SDGs in education. Based on results, it has been found that it would be beneficial if leaders/teachers enhance their struggle using inventive leadership styles for achieving SDGs, fostering transformative education, and ensuring that learners and societies benefit from a more equitable and resilient future. Additionally, the need for more resources, resistance to change, and a need for leaders’/teachers’ professional development have been identified as potential obstacles. To overcome these challenges, a collaborative approach among stakeholders, policy support, and innovative practices are recommended.

In educational evaluations at home and abroad, polytomous scoring items are becoming increasingly important. They can provide richer and more valuable information, with unmatched advantages compared to binary (0-1 scoring) items. If used as a tool for teachers to diagnose and assess students in the classroom, CD-CAT (Cognitive Diagnostic Computerized Adaptive Testing) has significant implications for improving teaching effectiveness. However, in classroom teaching situations, it is not feasible to estimate item parameters accurately with a large sample, as in large-scale assessments. In such cases, non-parametric CD-CAT becomes the only viable option. Compared to parametric CD-CAT, non-parametric CD-CAT started later and is particularly lacking in research related to polytomous scoring. Item selection method is at the core of CD-CAT, so it is essential to develop a non-parametric item selection method suitable for polytomous scoring CD-CAT. This study proposes a non-parametric item selection method for polytomous scoring cognitive diagnostic computerized adaptive testing (PCD-CAT): the Manhattan Distance Non-parametric Difference index item selection method (MD-NDI). The results of simulation studies indicate: (1) MD-NDI item selection method is suitable for PCD-CAT scenarios and exhibits better performance when the item bank quality is poor or the sample size for estimating item parameters is limited. (2) MD-NDI does not require pre-testing of items and distributes item usage more evenly, effectively ensuring the security of the item bank. (3) Even in cases of incorrectly specified item bank of Qc-matrix, MD-NDI still shows higher pattern correct classification rates. (4) In the study of variable-length PCD-CAT, MD-NDI not only reduces the test length in most conditions but also has a higher pattern correct classification rates when reaching the test termination rule.

Educational leadership is a multifaceted area of study. Unquestionably, leadership is the most deliberate field within the social sciences. Still, administrators have evaded the notions of leadership concept like a haunting tune. This study has focused particularly on the significance of varied leadership styles in teaching to sustain academic excellence at the secondary school level. The quantitative research method was used. Data was collected through the scale of diverse leadership styles (strategic, cultural, instructional leaderships and sustaining academic excellence) from 103 secondary schools in Punjab, Pakistan. The sample consisted of 540 teachers who were enacting as teachers presently. Based on research objectives and questions, two hypotheses were formed and tested using mean analysis to determine the average ranking of leadership styles. Pearson correlation to know the statistically significant relationship between each leadership styles, and overall scales with sustaining academic excellence. The results revealed that most teachers give preference strategic Leadership, then instructional leadership, and finally cultural leadership in their teaching to sustain academic excellence. Moreover, the findings also indicated that a statistically strong positive relationship among diverse leadership styles in teaching and sustaining academic excellence with the value (r = 0.752). Based on the findings, it has been concluded that when teachers increase their effort in the use of strategic, instructional, and cultural leadership styles, academic excellence may also sustain and improve.

Although much research has explored the personal, family, school, and social influences on adolescents' career expectations, little is known about the types of family-related career expectations of students and career expectation contexts in school. Using PISA 2018 test data, a multi-level latent profile analysis was conducted with indicators of parents' occupational status and children's career expectations. This research found that family-related career expectations of Estonian secondary school students can be divided into three types: enterprising, resilient, and disengaged. At the organizational level, contexts for career expectations in schools can be classified as enterprising, enterprising and resilient composite, and disengaged and resilient composite types. The research also found that science achievement and gender predicted latent profile memberships at the individual level. It is proposed that to value the science curriculum and its teaching is to invest in future human resources. It is suggested that families and schools should pay attention to stimulating male students' career aspirations. The career expectation types of secondary school students within and between schools are of great significance to the construction of national human resources, the development of schools, and the research of families. Keywords: career expectation types, multilevel latent profile analysis, PISA 2018 datasets, variations between schools, science achievement

At present, research on computational thinking in universities is gaining interest, and more attention is being paid to the cultivation and teaching of computational thinking. However, there is a lack of computational thinking assessment tools for college students, which makes it difficult to understand the current status and development of their computational thinking. In this study, computational thinking is regarded as the ability to solve practical problems. By analyzing the relevant literature, we identified five dimensions of computational thinking – decomposition, generalization, abstraction, algorithm and evaluation – and described their operational definitions. Referring to the Bebras and the problem situations in Google computational thinking education, we set up a life-based situation that college students are familiar with. Based on the life story situation, we developed a multidimensional assessment for college students’ computational thinking. This assessment tool contains 14 items, all of which are multiple-choice questions, and the structure and quality of the tool are verified by multidimensional item response theory. The results show that the assessment tool has good internal validity and can discriminate different disciplines of college students. The college students’ computational thinking test developed in this study can be used as an effective tool to assess college students’ computational thinking.

In educational measurement, exploring the method of generating multiple high-quality parallel tests has become a research hotspot. One purpose of this research is to construct parallel forms item by item according to a seed test, using two proposed item selection heuristic methods [minimum parameters–information–distance method (MPID) and minimum information–parameters–distance method (MIPD)]. Moreover, previous research addressing test assembly issues has been limited mainly to situations in which the information curve of the item pool or seed test has a normal or skewed distribution. However, in practice, the distributions of information curves for tests are diverse. These include multimodal distributions, the most common type of which is the bimodal distribution. Therefore, another main aim of this article is to extend the information curves of unimodal distributions to bimodal distributions. Thus, this study adopts simulation research to compare the results of two item, response, theory (IRT)-based item matching methods (MPID and MIPD) using different information curve distributions for item pools or seed tests. The results show that the MPID and MIPD methods yield rather good performance in terms of both two statistical targets when the information curve has a unimodal distribution, and two new methods yield better performance than two existing methods in terms of test information functions target when the information curve has a bimodal distribution.

A Q-matrix, which reflects how attributes are measured for each item, is necessary when applying a cognitive diagnosis model to an assessment. In most cases, the Q-matrix is constructed by experts in the field and may be subjective and incorrect. One efficient method to refine the Q-matrix is to employ a suitable statistic that is calculated using response data. However, this approach is limited by its need to estimate all items in the Q-matrix even if only some are incorrect. To address this challenge, this study proposes an item fit statistic root mean square error approximation (RMSEA) for validating a Q-matrix with the deterministic inputs, noisy, “and” (DINA) model. Using a search algorithm, two simulation studies were performed to evaluate the effectiveness and efficiency of the proposed method at recovering Q-matrices. Results showed that using RMSEA can help define attributes in a Q-matrix. A comparison with the existing Delta method and residual sum of squares (RSS) method revealed that the proposed method had higher mean recovery rates and can be used to identify and correct Q-matrix misspecifications. When no error exists in the Q-matrix, the proposed method does not modify the correct Q-matrix.