Search

A comparative longitudinal data analysis between two online non-thesis master's programs--natural resource management and environmental science--in a college of natural resources to determine the relationship between student characteristics and disenrollment risks. Risks varied between the two programs, with significance found to increase the risk of disenrollment due to cumulative GPA, gender, time between degrees, and the number of terms not enrolled. [For the full proceedings, see ED648717.]

Computer-mediated social learning contexts have become increasingly popular over the last few years; yet existing models of students' cognitive-affective states have been slower to adopt dyadic interaction data for predictions. Here, we explore the possibility of capitalizing on the inherently social component of collaborative learning by using keystroke log data to make predictions across conversational partners (i.e., using person A's data to make prediction about if person B is mind wandering). Log files from 33 dyads (total N = 66) were used to examine: (a) how mind wandering (defined here as task-unrelated thought) during computer-mediated conversations is related to critical outcomes of the conversation (trust, likability, agreement); (b) if task-unrelated thought can be predicted by the keystrokes of one's partner; and (c) how much data is needed to make predictions by testing various window-sizes of data preceding task-unrelated thought reports. Results indicated a negative relationship between task-unrelated thought and perceptions of the conversation, suggesting that attention is an important factor during computer mediated chat conversations. Finally, in line with our hypothesis, results from mixed effects models showed that one's level of task-unrelated thought was predicted by the keystroke patterns of their conversational partner, but only using small window sizes (5s worth of data). [For the complete proceedings, see ED630829.]

Problem decomposition into sub-problems or subgoals and recomposition of the solutions to the subgoals into one complete solution is a common strategy to reduce difficulties in structured problem solving. In this study, we use a datadriven graph-mining-based method to decompose historical student solutions of logic-proof problems into Chunks. We design a new problem type where we present these chunks in a Parsons Problem fashion and asked students to reconstruct the complete solution from the chunks. We incorporated these problems within an intelligent logic tutor and called them Chunky Parsons Problems (CPP). These problems demonstrate the process of problem decomposition to students and require them to pay attention to the decomposed solution while they reconstruct the complete solution. The aim of introducing CPP was to improve students' problem-solving skills and performance by improving their decomposition-recomposition skills without significantly increasing training difficulty. Our analysis showed that CPPs could be as easy as Worked Examples (WE). And, students who received CPP with simple explanations attached to the chunks had marginally higher scores than those who received CPPs without explanation or did not receive them. Also, the normalized learning gain of these students shifted more towards the positive side than other students. Finally, as we looked into their proof-construction traces in posttest problems, we observed them to form identifiable chunks aligned with those found in historical solutions with higher efficiency. [For the complete proceedings, see ED630829.]

Due to the precautionary measures during the COVID-19 pandemic many universities offered unproctored take-home exams. We propose methods to detect potential collusion between students and apply our approach on event log data from take-home exams during the pandemic. We find groups of students with suspiciously similar exams. In addition, we compare our findings to a proctored comparison group. By this, we establish a rule of thumb for evaluating which cases are "outstandingly similar", i.e., suspicious cases. [For the complete proceedings, see ED630829.]

Within the last decade, different educational data mining techniques, particularly quantitative methods such as clustering, and regression analysis are widely used to analyze the data from educational games. In this research, we implemented a quantitative data mining technique (clustering) to further investigate students' feedback. Students played educational games within a week on the educational games platform, Legends of Learning and after a week, we asked them to fulfill the feedback survey about their feelings on the use of this platform. To analyze the collected data from students, firstly, we prepared clusters and selected one prototype student closest to the centroid of each cluster to interview. Interviews were held to explain the clusters more and due to time and resource limitations, we were unable to interview all (N=60) students, thus only the most representative students were interviewed. In addition to the students, we conducted an interview with the teacher as well to get her detailed feedback and observations on the usage of educational games. We also asked students to take an exam before and after the research to see the impact of games on their grades. Our results depict that though educational games can increase students' motivation, they may negatively impact some students' grades. And even though playing games made students feel interested and fun, they would not like to play them on a daily basis. Hence, using educational games for a certain duration such as subject revision weeks may positively influence students' grades and motivation. [For the complete proceedings, see ED630829.]

The Indian Institute of Science is proud to host the fully in-person sixteenth iteration of the International Conference on Educational Data Mining (EDM) during July 11-14, 2023. EDM is the annual flagship conference of the International Educational Data Mining Society. The theme of this year's conference is "Educational data mining for amplifying human potential." Not all students or seekers of knowledge receive the education necessary to help them realize their full potential, be it due to a lack of resources or lack of access to high quality teaching. The dearth in high-quality educational content, teaching aids, and methodologies, and non-availability of objective feedback on how they could become better teachers, deprive our teachers from achieving their full potential. The administrators and policy makers lack tools for making optimal decisions such as optimal class sizes, class composition, and course sequencing. All these handicap the nations, particularly the economically emergent ones, who recognize the centrality of education for their growth. EDM-2023 has striven to focus on concepts, principles, and techniques mined from educational data for amplifying the potential of all the stakeholders in the education system. The spotlights of EDM-2023 include: (1) Five keynote talks by outstanding researchers of eminence; (2) A plenary Test of Time award talk and a Banquet talk; (3) Five tutorials (foundational as well as advanced); (4) Four thought provoking panels on contemporary themes; (5) Peer reviewed technical paper and poster presentations; (6) Doctoral students consortium; and (7) An enchanting cultural programme. [Individual papers are indexed in ERIC.]

Assessing students' performance in online learning could be executed not only by the traditional forms of summative assessments such as using essays, assignments, and a final exam, etc. but also by more formative assessment approaches such as interaction activities, forum posts, etc. However, it is difficult for teachers to monitor and assess students' learning activities using the log data. To provide teachers with a more comprehensive view of students' distinct learning behaviour patterns, and to supply personalized interventions and support to meet the specific needs of each learning group, this study focuses on how to automatically acquire learning logs from Manaba, a Japanese commercial LMS, and how to cluster students' learning activities using the k-means algorithm. Firstly, we developed a program using Python to scrape students' learning activity log information from the Manaba web pages. We collected 56446 lines of clickstreams log data from 121 students in two computer literacy hybrid classes in the fall semester of 2022 (2022/9-2023/1). Secondly, we convert the raw logs into a structured dataset with 33 features which represent each student's learning activities. Then we extract and select 15 features representing three perspectives: raw activity, time on task, and learning frequency. Thirdly, we grouped students' learning activity patterns with the three perspectives into 5 clusters by the k-means clustering algorithm. As a result, this study identified five distinct learning activity patterns depending on how much, how long and how often the students learned online. For example, cluster 1 seldom learned but spent time on learning whom we considered the disengaged or struggling students, and cluster 5 had more learning activities with little time on each activity whom we considered the well-self-regulated students. The results of this study contribute to how to monitor students' learning activity in online learning and how to assess and support student's learning by their learning activity patterns. [For the full proceedings, see ED636095.]

The bibliographic science papers with high citations rates are often used as an indication of the science impact of an observatory. These high impact papers are presented as examples of the best science being done with an observatory's data. But, is the number of citations by itself a good indicator of the scientific impact of the paper, and is impact a good indicator of the scientific impact of the observatory? In this paper we will present results from a recent study of Chandra high impact papers and suggest some alternative methods for identifying such papers. This work has been supported by NASA under contract NAS 8-03060 to the Smithsonian Astrophysical Observatory for operation of the Chandra X-ray Center. [ABSTRACT FROM AUTHOR]

This paper addresses a key challenge in Educational Data Mining, namely to model student behavioral trajectories in order to provide a means for identifying students most at risk, with the goal of providing supportive interventions. While many forms of data including clickstream data or data from sensors have been used extensively in time series models for such purposes, in this paper we explore the use of textual data, which is sometimes available in the records of students at large, online universities. We propose a time series model that constructs an evolving student state representation using both clickstream data and a signal extracted from the textual notes recorded by human mentors assigned to each student. We explore how the addition of this textual data improves both the predictive power of student states for the purpose of identifying students at risk for course failure as well as for providing interpretable insights about student course engagement processes. [For the full proceedings, see ED599096.]

Peer assessment has proven to be a useful strategy for increasing the timeliness and quantity of formative feedback, as well as for promoting metacognitive thinking among students. Previous research has determined that reviews that contain suggestions can motivate students to revise and improve their work. This paper describes a method for automatically detecting suggestions in review text. The quantity of suggestions can be treated as a metric for the helpfulness of review text. Even before a review is submitted, the system can tell a reviewer when a review is lacking in suggestions and consequently advise that they be added. This paper presents several neural-network approaches for detecting suggestions and compares them against traditional natural language processing (NLP) methods such as rule-based techniques, as well as past machine-learning approaches. Our network-based classifiers outperformed rule-based classifiers in every experiment. Our neural-network classifiers attained F1- scores in the low 90% range, outperforming the support vector machine (SVM) classifier whose F1-score was 88%. The naïve Bayes (NB) classifier had an F1-score of 84% and the rule-based classifier had an F1-score of 80%. As in other domains such as determining sentiment, we found that neural-network models perform better than the likes of naïve Bayes and support vector machines when classifying suggestions in text. [For the full proceedings, see ED599096.]

This paper reflects on an attempt to introduce smartphones into a blended learning context and highlights several methodological considerations relevant to the collection of mobile data. While mixed methods are now common, using this approach for investigating the challenges of mobile data collection is not as common. This study employed a mixed method longitudinal exploratory multiple case study design. The study collected qualitative and quantitative data on student interactions within a yearlong series of collaborative language learning activities. This paper describes the methodological issues that emerged during the planning and implementation of the study. The participants were undergraduate students studying English translation at a four-year private university in Tokyo, Japan. This paper adds to the knowledge of employing mixed methodology design for mobile data collection and analysis. In particular the advantages of incorporating Multidimensional Scaling (MDS) analysis with qualitative data. The results suggest that separate forms of data collected at similar frequencies and times and then triangulated provided an effective methodology for studying collaborative learners in a highly mobile context. This can be seen as evidence for the inclusion of various data collection cycles within a single mobile learning research study. [For the complete proceedings, see ED601080.]

The contemporary Human activity utilizes huge volumes of digital data to solve efficiently multiple socio-economic, scientific and technical problems. Now the big data analysis is mainly oriented to the socioeconomic sphere with a goal to lift the profit. The science and technology to penetrate deeper in the nature of objects and systems under investigation prefer to limit the analysis area, concentrating, for example, only on properties of extra pure materials or isolated systems. In science the analysis should be convergent and the initial data may be and should be regularized to diminish the input data errors. The clusters now are considered as a new or still unknown state of matter. The big thermophysical data analysis appears as the most informative way to discover the properties of clusters in pure real gases, because the continuous spectrum of bound states in clusters prevents from the spectroscopic way for clusters' properties evaluation. The goal of the paper is to teach the main principles of the scientific regular data convergent analysis basing on the author's experience to extract clusters' properties in pure real gases from regularized experimental thermophysical big data. [For the full proceedings, see ED623149.]

A significant amount of research has gone into predicting student performance and many studies have been conducted to predict why students drop out. A variety of data including digital footprints, socio-economic data, financial data, and psychological aspects have been used to predict student performance at the test, course, or program level. Fairly good prediction results have been achieved using both traditional machine learning and more recently deep learning methods. While using diverse sets of data has achieved good results, this data is often difficult and expensive to collect, and may have privacy-related issues. This paper explores the extent to which only prior performance data readily available with registrars in most Universities can be used to predict student performance in future terms. Twenty term data from 789 students enrolled an engineering program at an American University were used to train long term short term (LSTM), Bi-directional LSTM and Gated Reference Units (GRU) models to predict student performance in future terms. The results are that all three types of models were able to reasonably predict the next term's performance (F1-score of about 0.70) regardless of the number of terms a student had spent the University. The models generally did not overfit. The prediction was reasonable until about trying to predict performance on seventh term in the future, but the performance dropped beyond this point primarily due to lack of sufficient data (F1-score of about 0.2). [For the full proceedings, see ED621108.]

When text datasets are very large, manually coding line by line becomes impractical. As a result, researchers sometimes try to use machine learning algorithms to automatically code text data. One of the most popular algorithms is topic modeling. For a given text dataset, a topic model provides probability distributions of words for a set of "topics" in the data, which researchers then use to interpret meaning of the topics. A topic model also gives each document in the dataset a score for each topic, which can be used as a non-binary coding for what proportion of a topic is in the document. Unfortunately, it is often difficult to interpret what the topics mean in a defensible way, or to validate document topic proportion scores as meaningful codes. In this study, we examine how keywords from codes developed by human experts were distributed in topics generated from topic modeling. The results show that (1) top keywords of a single topic often contain words from multiple human-generated codes; and conversely, (2) words from human-generated codes appear as high-probability keywords in multiple topic. These results explain why directly using topics from topic models as codes is problematic. However, they also imply that topic modeling makes it possible for researchers to discover codes from short word lists. [This paper was published in: "ICQE 2021," edited by A. R. Ruis and S. B. Lee, Springer Nature Switzerland AG, 2021, pp. 18-31.]

This paper presents a case study that is part of a research studying how university statistics teachers integrate digital tools in their practice. The case study is of a teacher who claims to follow constructionist ideas for her teaching of statistics, using the R programming language. The teacher developed a series of activities with R, aiming to promote the understanding of statistical concepts. We observed 11 of her classes, in a three-week period. These classes were analyzed using as framework the notion of Instrumental Orchestration and Papert's constructionist principles. We identified four ways in which this teacher orchestrates her class and the constructionist elements present in them. [For the complete proceedings, see ED629884.]

National failure rates seen in undergraduate introductory CS courses are quite high. In this paper, we develop a predictive model for student in-class performance in an introductory CS course. The model can serve as an early warning system, flagging struggling students who might benefit from additional support. We use a variety of features from the first few weeks of the course such as scores on assignments, interaction with the online textbook, and participation with the in-class polling system in order to train our models. We compare the performance of a number of machine learning algorithms on predicting final exam scores as well as final course grade. We find that the Support Vector Machine and AdaBoost are the most effective, and that we can achieve increasingly accurate predictions as we use data from further into the course. The regression coefficients give us insights into which features are most correlated with student success, suggesting that certain types of assignments are more indicative of learning than others. [For the full proceedings, see ED607784.]

Building and especially improving a classification kernel represents a challenging task. The works presented in this paper continue an already developed semi-supervised classification approach that aimed at labelling transcripts from educational videos. We questioned whether the size of the ground-truth data-set (Wikipedia articles) or the quality of the keywords used in the semi-supervised labelling have a significant impact on the accuracy metrics of the final obtained data model. Experimental results took into consideration three Wikipedia data-sets of "Small," "Medium" and "Large" sizes. For each data-set there were used three sets of keywords: offered by video authors, determined by "rake-nltk" on available transcripts and determined by "rake-nltk" on Wikipedia articles that serve as training and testing data for the LDA [latent Dirichlet allocation] model that determine keywords on the transcripts. Experiments show that the size of the data-set has little importance, while the quality of the keywords has a more significant impact. Therefore, an improved version of the previously developed classifier has been obtained by improving the quality of the keywords involved in semi-supervised training. This result paves the way towards further improvements that may finally be deployed as within a recommender system of educational videos at the Universitat Politècnica de València. [For the full proceedings, see ED607784.]

Massive online Open Courses (MOOCs) make extensive use of videos. Students interact with them by pausing, seeking forward or backward, replaying segments, etc. We can reasonably assume that students have different patterns of video interactions, but it remains hard to compare student video interactions. Some methods were developed, such as Markov Chain and Edit Distance. However, these methods have caveats as we show with prototypical examples. This paper proposes a new methodology of comparing video sequences of interaction based both on time spent in each state and the succession of states by computing the distance between the transition matrices of the video interaction sequences. Results show the proposed methodology can better characterize video interaction in a task to discriminate which student is interacting with a video, or which video a student is interacting with. [For the full proceedings, see ED607784.]

Feature engineering, the construction of contextual and relevant features from system log data, is a crucial component of developing robust and interpretable models in educational data mining contexts. The practice of feature engineering depends on domain experts and system developers working in tandem in order to creatively identify actions and behaviors of interest. In this paper we outline a method of iterative feature engineering using the misclassifications of earlier models. By selecting cases where earlier models and ground truth disagree, we can focus attention on specific behaviors, or patterns of behavior, that a model is not using in its predictions. We show that iterative feature engineering on cases of false positives and false negatives improved a model predicting quitting in an educational video game by 15%. We close by discussing applications of this method for addressing model performance gaps across different classes of learners, as well as precautions against model overfitting with using this method of feature engineering. [For the full proceedings, see ED607784.]

Network analysis in educational research has primarily relied on self-reported relationships or connections inferred from online learning environments, such as discussion forums. However, a large part of students' social connections through day-to-day on-campus encounters has remained underexplored. The paper examines spatial-temporal student networks using campus WiFi log data throughout a semester, and their relations to the student demographics and academic performance. A tie in the spatial-temporal network was inferred when two individuals connected to the same WiFi access point at the same time intervals at the 'beyond chance' frequency. Our findings revealed that students were more likely to co-locate with the individuals of similar gender, ethnic group identity, family income, and grades. Analysis of homophily over the semester showed that students of the same gender were more likely to co-locate as the semester progressed. However, co-location of the students similar on ethnic minority identity, family income, and grades remained consistent throughout the semester. Mixed-effect regression models demonstrated that features derived from spatial-temporal networks, such as degree, the grade of the most frequently co-located peer, and average grade of five most frequently co-located peers were positively associated with academic performance. This study offers a unique exploration of the potential use of WiFi log data in understanding of student relationships integral to the quality of college experience. [For the full proceedings, see ED607784.]

Asking questions about your data is a constant application of all business organizations. To facilitate decision making and improve business performance, a business intelligence application must be an integral part of everyday management practices. Microsoft Excel added PowerPivot and PowerPivot officially to facilitate this process with minimum cost, knowing that many business people are already familiar with MS Excel. This paper will design an introductory class to business intelligence (BI) using Excel PowerPivot. If an educator decides to adopt this paper for teaching an introductory BI class, students should have previous familiarity with Excel's functions and formulas. This paper will focus on four significant phases all students need to complete in a three-credit class. First, students must understand the process of achieving small database normalization and how to bring these tables to Excel or develop them directly within Excel PowerPivot. This paper will walk the reader through these steps to complete the task of creating the normalization, along with the linking and bringing the tables and their relationships to excel. Second, an introduction to Data Analysis Expression (DAX) will be discussed. [For the full proceedings, see ED592847.]