Predicting students' final performance from participation in on-line discussion forums

On-line discussion forums constitute communities of people learning from each other, which not only inform the students about their peers' doubts and problems but can also inform instructors about their students' knowledge of the course contents. In fact, nowadays there is increasing interest in the use of discussion forums as an indicator of student performance. In this respect, this paper proposes the use of different data mining approaches for improving prediction of students' final performance starting from participation indicators in both quantitative, qualitative and social network forums. Our objective is to determine how the selection of instances and attributes, the use of different classification algorithms and the date when data is gathered affect the accuracy and comprehensibility of the prediction. A new Moodle's module for gathering forum indicators was developed and different executions were carried out using real data from 114 university students during a first-year course in computer science. A representative set of traditional classification algorithms have been used and compared versus classification via clustering algorithms for predicting whether students will pass or fail the course on the basis of data about their forum usage. The results obtained indicate the suitability of performing both a final prediction at the end of the course and an early prediction before the end of the course; of applying clustering plus class association rules mining instead of traditional classification for obtaining highly interpretable student performance models; and of using a subset of attributes instead of all available attributes, and not all forum messages but only students' messages with content related to the subject of the course for improving classification accuracy.     

The relationships among group size,participation, and performance of programming language learning supported with online forums

This study examined the relationships among group size, participation, and learning performance factors when learning a programming language in a computer-supported collaborative learning (CSCL) context. An online forum was used as the CSCL environment for learning the Microsoft ASP.NET programming language. The collaborative-learning experiment was performed with one large group and 15 small groups.A total of 120 students participated in this experiment as part of a half-semester ASP.NET programming language course. The course contained an online forum for supporting the students' social activities and participation. This study used a participation-weighted rate for different participation types. A ‘learning score’ and a ‘learning satisfaction’ score were used to measure learning performance.The results of this study were as follows: (1) the online forum support aided collaborative learning, regardless of group size; (2) group sizes did not significantly influence learning scores directly but significantly influenced participation, and small groups had higher participation rates, which positively influenced learning scores; and (3) learning satisfaction using the online forum was higher than the average score. Small groups had higher learning satisfaction rates, and participation did not significantly influence learning satisfaction.Due to this study's results, we recommend that programs design instruction with small groups for teaching programming languages in online forums, support student-centered discussions, and encourage high levels of student participation to increase learning performance.     

A thematic analysis of patient communication in Parkinson’s disease online support group discussion forums

Online support groups have become a popular source of information, advice and support for individuals living with a range of health conditions. However, research has not commonly focused on patients living with Parkinson’s disease and their use of online support groups. Thus, the aim of this study was to gain an insight into the positive and negative aspects of online communication through an analysis of messages exchanged within Parkinson’s disease discussion forums. Data was collected from four forums and analysed using data-driven thematic analysis. The results revealed that participation in the forums allowed patients to share experiences and knowledge, form friendships, as well as helping them cope with the challenges of living with Parkinson’s disease. Conversely, a lack of replies, the experience of Parkinson’s disease symptoms, a lack of personal information, fragility of online relationships, misunderstandings and disagreements, all appeared to compromise the online experience. Practical implications and future research recommendations are proposed.     

Learning within incoherent structures: the space of online discussion forums

Abstract  Online discussion forums are an increasingly common use of new information and communication technologies in education. As a tool for promoting conversational modes of learning, it has been suggested that online discussion forums can lead to enhanced learning outcomes for students. However, there is a need to explore further the implications of the highly mediated nature of computer–based interaction on student learning within these virtual learning environments. This paper presents results from a detailed study of students' learning outcomes and patterns of interaction within an online discussion forum. The findings suggest that the typical nonlinear branching structure of online discussion may be insufficient for the realisation of truly conversational modes of learning. The paper discusses the implications of these findings in relation to students' learning.     

When to jump in: The role of the instructor in online discussion forums

As asynchronous discussion forums become more prevalent in online and flexible-delivery modes of teaching, understanding the role that instructors play in student learning in these forums becomes an important issue. Whether the instructor chooses to lead discussions or to keep a low profile can affect student participation in surprising ways. In this study, we investigate how instructor participation rates, the timing of instructor postings (during or at the end of a forum) and nature of their postings (questions, answers or a mix of the two) relate to student participation and perception.     

Student interactions in online discussion forums: their perception on learning with business simulation games

Digital technology offers new teaching methods with controversial results over learning. They allow students to develop a more active participation in their learning process although it does not always drive to unequivocal better learning outcomes. This study aims to offer additional evidence on the contribution of business simulation games to students’ learning outcomes, considering student interactions in online discussion forums. We conducted a qualitative research with the online discussion forums of 5 different courses at bachelor and master levels, which involves 41 students’ teams. The final sample was composed of 3681 messages posted by the students. The results reveal that some generic and specific managerial skills exert a positive influence on learning outcomes. Students mostly highlighted teamwork, decision-making, information processing, reaching agreements, and dealing with uncertainty as the most relevant contributions of the game towards their learning. These results have instructional and pedagogical implications for determining the best way to enhance students’ motivation and learning outcomes when using digital technology methods, which involves recommendations that affect their design and monitoring.     

Elementary students' participation style in synchronous online communication and collaboration

This research investigated how differently elementary school students participate in synchronous online collaborative learning by analysing their discussions with their partners. Two hundred and seventy-eight Taiwanese students, ranging in age from 11 to 12 years old, were involved in this study. The students were randomly arranged within-class into three-member groups. Each group used a collaborative learning system to discuss and accomplish a group assignment (creating a shared concept map). The textual discussions by each individual student were collected, categorised and accumulated. Cluster analysis was adopted to statistically classify students based on their discussion characteristics. We found that a student, while participating in online collaboration and communication, exhibited a distinct style: less contributing, coordination emphasising, communicative or task-oriented. In addition, students exhibiting the task-oriented pattern and students predominantly showing communicative behaviours were found to have better learning performance and retain more knowledge than students observed to be coordination emphasising or less contributing within group collaboration.     

Towards identifying unresolved discussions in student online forums

Online discussion is a popular form of web-based computer-mediated communication and is a dominant medium for cyber communities in areas of information sharing, customer support and distributed education. Automatic tools for analyzing online discussions are highly desirable for better information management and assistance. For example, a summary of student Q&A discussions or unresolved questions can help the instructor assess student dialogue efficiently, which can lead to better instructor guidance for student learning by discussion. This paper presents an approach for classifying student discussions according to a set of discourse structures, and identifying discussions with confusion or unanswered questions. Inspired by the existing spoken dialogue analysis approaches, we first define a set of forum “speech acts” (F-SAs) that represent roles that individual messages play in threaded Q&A discussions, such as questions, raising issues, and answers. We then model discourse structures in discussion threads using the F-SAs, such as whether a question was replied to with an answer. Finally, we use such discourse structures in classifying and identifying discussions with unanswered questions or unresolved issues. We performed an analysis of the discussion thread classifiers and the system showed accuracies from 0.79 to 0.87 on several discussion classification problems. This analysis of human conversation via online discussions provides a basis for development of future information extraction and intelligent assistance techniques for online discussions.     

Influence of group configuration on online discourse writing

This study examines the impact of different strategies for grouping students in online, discourse-intensive distance education courses. The mixed methods research focused on note writing productivity (based on 366 participants in 25 classes) and participants' perceptions (12 graduate students and 10 instructors) relating to three different class configurations (large whole class, small whole class, large class divided into subgroups). Each configuration exhibited advantages and disadvantages in terms of supporting note writing, however, the data analyses suggested that the advantages for writing in subgroups outweighed those of the other two conditions. Splitting larger classes into smaller subgroups appears to reduce information overload and encourages more focused, in-depth discussions. The research concludes with a list of pedagogical recommendations and suggests new software features that may help learning within specific group configurations. This study may have implications for both practitioners and researchers who wish to promote more fruitful online discussions.     

Influence of group configuration on online discourse reading

Online discourse reading plays a very important role in collaborative discussions. However, not many studies have examined the influence of group configuration on online discourse note reading. The current study examined note reading workloads and participants' perceptions of the three group configurations (large whole class, small whole class, large with subgroups) in online graduate-level courses from one institute. In this mixed-methods study, we analyzed tracking logs from 25 graduate-level online courses (25 instructors and 341 students) and interviews from 10 instructors and 12 graduate students with diverse backgrounds. Findings suggest that all three configurations had their own advantages and disadvantages in fostering online discourse reading. However, our analysis suggests that the advantages of subgroup discussions in supporting note reading outweigh those of the Small and Large configurations. The overload effects in information reading due to large class sizes can be minimized by dividing students into small groups for discussion purposes. Group configuration into proper-size groups may reduce students' reading loads. Interviewees felt that the waving of small groups into large classes benefited their collaborative discussions. We conclude this paper with a list of pedagogical recommendations and new software features that may help group configuration and enhance learning in online courses. This study may have implications for both practitioners and researchers to seek optimal group configurations to achieve more fruitful online discussions through note reading.     

Topic correlation and individual influence analysis in online forums

Over the last few years, online forums have gained massive popularity and have become one of the most influential web social media in our times. The forum document corpus can be seemed to be composed of various topics evolved over time, and every topics is reflected on a volume of keywords and social actors. In this paper, we attempt to study the interesting problem: for the evolving topics, were there any correlation between them? We propose a method for discovering the dependency relationship between the topics of documents in adjacent time stamps based on the knowledge of content semantic similarity and social interactions of authors and repliers. We introduce mutual information measure to estimate the correlation between the topics. Applied to the realistic forum data, we show how topics are related and which postings can be recommended to another as similar topics. We also show how the authors impact the topics and propose a new way for evaluating author impact.     

Influence of group member familiarity on online collaborative learning

This study investigated the effects of group member familiarity during computer-supported collaborative learning. Familiarity may have an impact on online collaboration, because it may help group members to progress more quickly through the stages of group development, and may lead to higher group cohesion. It was therefore hypothesized that increased familiarity would lead to (a) more critical and exploratory group norms, (b) more positive perceptions of online communication and collaboration, (c) more efficient and positive collaboration, and (d) better group performance. To investigate these hypotheses, 105 secondary education students collaborated in groups of three. The results of this study indicate that higher familiarity led to more critical and exploratory group norm perceptions, and more positive perceptions of online communication and collaboration. Furthermore, in familiar groups students needed to devote less time to regulating their task-related activities. The expectation that familiarity would lead to better group performance was not confirmed. These findings imply that online educators pay attention to the effects group member familiarity has on online collaborative learning.     

A system for extracting and comparing memes in online forums

From their origins in the sociological field, memes have recently become of interest in the context of `viral' transmission of basic information units (memes) in online social networks. However, much work still needs to be done in terms of metrics and practical data processing issues. In this paper we define a theoretical basis and processing system for extracting and matching memes from free format text. The system facilitates the work of a text analyst in extracting this type of data structures from online text corpuses and n performing empirical experiments in a controlled manner. The general aspects related to the solution are the automatic processing of unstructured text without need for preprocessing (such as labelling and tagging), identification of co-occurences of concepts and corresponding relations, construction of semantic networks and selecting the top memes. The system integrates these processes which are generally separate in other state of the art systems. The proposed system is important because unstructured online text content is growing at a greater rate than other content (e.g. semi-structured, structured) and integrated and automated systems for knowledge extraction from this content will be increasingly important in the future. To illustrate the method and metrics we process several real online discussion forums, extracting the principal concepts and relations, building the memes and then identifying the key memes for each document corpus using a sophisticated matching process. The results show that our method can automatically extract coherent key knowledge from free text, which is corroborated by benchmarking with a set of other text analysis approaches, as well as a user study evaluation.     

A study of the relationships among learning styles, participation types, and performance in programming language learning supported by online forums

This study is focused on the relationships among learning styles, participation types, and learning performance for programming language learning supported by an online forum. Kolb’s learning style inventory was used in this study to determine a learner’s learning type: ‘Diverger’, ‘Assimilator’, ‘Converger’, and ‘Accommodator’. Social Learning Theory was also used to define four participation types. These types in turn were used to describe the learning associated with the use of online forums: ‘Replier’, ‘Asker’, ‘Watcher’, and ‘No activity’.A total of 144 students participated in this experiment as part of a half semester ASP.NET programming language learning courses. The course contained an online forum for supporting the students’ social activities and participation. In this study, ‘learning score’ and ‘satisfaction’ were used to measure learning performance.The results of this study were the following: (1) different learning styles were associated with significantly different learning scores and that the ‘Accommodator’ style was associated with superior learning scores; (2) participation types were also associated with significantly different learning scores and that the ‘Replier’ type is associated with superior learning scores; (3) learning satisfaction is not significantly different among the different learning styles or different participation types, but the average is significantly higher than average values (3.5) of 7-point Likert scale; (4) there is no significant association between learning styles and participation types. Explanations and discussions of these results are offered.Based on the results of this study, we propose that programming language learning, supported with online forums and students’ active participation, increases learning performance as measured by student learning scores.     

A theory of online learning as online participation

In this paper, an initial theory of online learning as online participation is suggested. It is argued that online learner participation (1) is a complex process of taking part and maintaining relations with others, (2) is supported by physical and psychological tools, (3) is not synonymous with talking or writing, and (4) is supported by all kinds of engaging activities. Participation and learning are argued to be inseparable and jointly constituting. The implication of the theory is straightforward: If we want to enhance online learning, we need to enhance online learner participation.     

Using discussion logic in analyzing online group discussions: A text mining approach

We propose a Discussion Logic-based Text Analytics (DiLTA) framework, which combines theories developed in social science and text mining fields. The framework extracts features that uncover discussion logic and uses these features in analyzing online discussions. A series of models are proposed including conversation disentanglement, coherence analysis, and visualization. Validation experiments showed that DiLTA achieved significantly superior performance over existing text analytics methods in reconstructing internal structure. A case study using DiLTA-enabled visualization on a healthcare forum illustrates the great potential of DiLTA in assisting comprehension of the internal linkage, structure, and logic of online group discussions.     

Students’ participation intention in an online discussion forum: Why is computer-mediated interaction attractive?

Anecdotal evidence indicates that an online discussion forum may not be utilized to its full potential in enhancing the effectiveness and efficiency of teaching due to a lower than expected student participation rate. This paper seeks to identify the motivational behavioral factors influencing students’ intention to participate in an online discussion forums (ODF). Drawing on the literature on social psychology and applying the theory of reasoned action, we develop a conceptual model of intention to participate in an online discussion forum and empirically test the hypotheses in a cross-sectional quantitative survey. The findings indicate that expectancy on hedonic outcome and utilitarian outcome and peer pressure positively influence the participation intention of students. Also, the perceived importance of learning positively moderates the relationship between utilitarian outcome expectancy and participation intention. Theoretical and practical implications of the findings are discussed.     

Investigating students' learning approaches,perceptions of online discussions,and students' online and academic performance

The main purpose of this study was to understand the relationships between students' approaches to learning, their perceptions of online discussions, students' contributions in asynchronous discussions, and their academic performance. Two sets of questionnaires were used for understanding students' approaches of learning and perceptions of online discussions. The online postings from seven weeks of discussions were coded into three major categories: Initiation, Elaborated Response (ER), and Response with Resources (RWR). The results showed, first, some aspects of students' perceptions influenced the numbers of ERs and RWRs. Secondly, students' contributions to Initiation messages and RWR significantly related to deep motivation and deep strategies; however, the numbers of these two types of messages were negatively correlated to surface strategies. Finally, cluster analysis revealed three distinct groups who scored significantly different in almost all aspects of approaches to learning and perceptions of online discussions. Students in the cluster who adopted deep approaches and scored highest in the perception scales outperformed students in the other two clusters, both in terms of the number of ER messages and academic performance. Pedagogical implications for teaching with online discussions are discussed in this study.     

Influence of participation, facilitator styles, and metacognitive reflection on knowledge building in online university courses

Understanding how to foster knowledge building in online and blended learning environments is a key for computer-supported collaborative learning research. Knowledge building is a deeply constructivist pedagogy and kind of inquiry learning focused on theory building. A strong indicator of engagement in knowledge building activity is the socio-cognitive dynamic of epistemic agency, in which students exercise a higher level of agency for setting forth their ideas and negotiating fit with those of others rather than relying on their teacher. The purpose of this paper is to investigate the influence of (a) levels of participation, (b) facilitator styles and (c) metacognitive reflection on knowledge building in two blended, post-secondary education contexts. A study of a total of 67 undergraduate students suggest that high levels of participation, a supportive facilitator style, and ample opportunities for metacognitive reflection on the students’ own participation strategies are most conducive for fostering epistemic agency for knowledge building. Implications of these results for research and instructional design of online courses are discussed.