The world has witnessed a revolution in the amount of data mined or generated from various sources. In the educational sector, there has been an adoption of improved learning techniques, and this has resulted in an upsurge in the data amount. An increase in data generation in this sector has stemmed from intelligent tutoring, the proliferation of online learning, and students’ demographic data. Besides, vast data amounts have been collected from different sources and stored in learning institutions’ educational forums and blogs (Wong, 2017). The vast amounts of data require specialized analytical techniques that can allow for their analysis and interpretation. The analysis of the data can result in the generation of valuable information and learning methods that can be used to improve the students’ retention and performance. One of the tools that can be used in the educational educator to collect and analyze the learners’ data is learning analytics. Learning analytics can be described as a process in which organizational data is collected, evaluated, analyzed, and reported for decision making (O’Farrell, 2017). In the educational sector, learning analytics customarily entails the utilization of big data analysis for the enhancement of the quality of education delivered to students. However, the primary challenge in the educational sector has been the inefficient utilization of data to address the students’ learning needs.
Whereas learning analytics has been utilized in various learning institutions, several studies have not analyzed its impacts on the higher education sector. It is for this reason that this paper provides a systematic review of the research studies that have been conducted in the recent past on learning analytics. Additionally, it addresses various problems in the higher education sector that can be solved through the use of learning analytics. It will also analyze the potential benefits of LA in this sector based on the review of various research and case studies.
The research paper aims to provide a brief overview of learning analytics, and how it can be utilized to improve the students’ learning outcomes in the higher education sector. Based on this, the research paper will include the following:
- An introduction to learning analytics
- The impacts or benefits of LA on the higher education sector
- The challenges faced by learning institutions in their attempt to implement LA tools
- What are some of the impacts of LA on the educational sector?
- What are some of the challenges that learning institutions are likely to face in their attempt to implement LA tools?
Introduction to Learning Analytics
In their research study, Dietz-Uhler and Hurn (2013) define learning analytics as the collection and analysis of students’ data to improve learning. They further note that LA allows learning institutions to predict, as well as enhance the students’ performance and retention rates. Additionally, Dietz-Uhler and Hurn (2013) note that through learning analytics, institutions and faculty members are usually able to make decisions that can optimize the students’ learning outcomes. Besides, learning analytics is an essential tool that can enable learning institutions to provide students with personalized learning (Dietz-Uhler & Hurn, 2013). Personalized learning would enable learners to have an enhanced and practical learning experience, a move that is likely to improve their performance levels. The use of data collected and analyzed through learning analytics will enable the faculty members to make effective decisions on how a learner can proceed with the course. Sclater, Peasgood, and Mullan (2016) in their study assert that LA often combines expertise from predictive modeling and data mining among other academic disciplines. They note that it has used to solve a plethora of challenges such as poor retention rates in different learning institutions.
O’Farrell (2017) asserts that data analytics constitutes an essential component of areas such as transport, retail industry, and the health sector. He also supports the claim that learning analytics allows learning institutions to improve the students’ learning outcomes. Through learning analytics, the higher education sector has taken advantage of the analyzed data to develop teaching strategies that can enhance the performance of students. The developed strategies have played an essential role in optimizing the students’ success and performance levels. O’Farrell (2017) in his study provides a typical example of the Irish educational sector that has adopted the use of learning analytics to improve the students’ success and retention rates.
Some authors have asserted that the primary aim of learning analytics is to generate actionable insights and valuable information through the exploration and the analysis of relevant data (O’Farrell, 2017; -Uhler & Hurn, 2013; and Sclater, Peasgood, & Mullan, 2016). Learning analytics also entails the aggregation and interpretation of essential data to provide an evidence base for the optimization of conditions that can enable students’ learning to flourish (O’Farrell, 2017). Before data from a plethora of sources is used in learning analytics, a learning institution has to ensure that complies with the data protection stipulations. Some of these sources include Wi-Fi logs, online forums, educational blogs, attendance data, library usage system, and student information systems. Notably, students’ associations within the virtual learning environment have been classified as the primary source of data used in LA. Virtual learning platforms comprise of online platforms such as Sakai, Moodle, and Blackboard. They have a wide range of resources like previous research and examination papers, quizzes, and lecture notes that teachers can give students to aid them in the learning process (O’Farrell, 2017). The first way in which LA can be used in learning institutions is to provide teachers with essential information on the types of resources that the students are relying on or using. Through this, the teacher will be able to assess the students’ concentration on the course resources.
LA can also give students a chance to assess their levels of engagement with course materials. According to O’Farrell, real-time information has enabled students and their teachers to take informed actions where appropriate. Besides, learning analytics has been used to inform the students’ program and curriculum design. Through learning analytics, the teacher can have the opportunity of identifying those patterns that are likely to have negative impacts on the student’s learning process. By identifying these patterns, the learner will be able to record an improvement in his or her academic outcome. In the Irish educational sector, the use of LA has enabled teachers to prescribe those resources and activities that will improve students’ performance and success. Moreover, LA can be used in the educational sector to identify non-performing students and advise them to work hard. It can also be used to identify and monitor those students who have displayed rapid changes in engagement.
The Development of Learning Analytics
Data and big data analytics are essential concepts in many sectors over the years. Stakeholders in the higher education sector have been exposed to an environment characterized by large amounts of data from a host of digital platforms and VLES (O’Farrell, 2017). As a result, they have begun to recognize the essential role played by this data in addressing challenges that exist in the sector. Some of these challenges include the need to deliver effective learning services with fewer resources and the desire to provide students with a meaningful education. Based on this context, the concept of LA started to emanate in 2011. More recently, significant growth has been witnessed within the field of learning analytics. Due to the growth, the autonomous focus areas include predictive and social learning analytics. Predictive analytics often specializes in the analysis of data obtained from historical datasets. Its objective is to identify those cases and behaviors that can be used to predict the results of future and current events (O’Farrell, 2017). The data analyzed from predictive analytics is essential in helping a learning institution to identify those students at risk. On the other hand, social learning analytics primarily uses social networks to analyze the manner in which students interact with others. Some of these networks may include online forums and educational blogs that analyze how students learn from social interactions.
How to Use Learning Analytics
According to Dietz-Uhler and Hurn (2013), LA can ordinarily be utilized at the institutional and course level. A learning institution is likely to benefit from using LA at the curriculum and course level. At the course and curriculum level, LA has been used to improve the students’ retention rates and performance levels. LA plays an essential role in providing the course instructor with information concerning the learner on a real-time basis. With this information, the faculty or even the course instructor can analyze ways in which the learner can be helped to succeed. For instance, a course instructor may deduce that a student who has not read a discussion board needs immediate intervention. Likewise, an instructor can intervene and assess why a student who has been performing well has suddenly performed dismally on a particular assignment. Additionally, the instructor or the faculty can assess usage data in an LMS in situations where a student constantly asks questions concerning the course procedures (Dietz-Uhler & Hurn, 2013). Through the assessment, the instructor will be able to determine if the student has assessed the necessary LMS tools.
The use of learning analytics in the educational sector has also enabled faculty members and course instructors to improve the students’ learning opportunities. Faculty members will have the chance to assess and determine course areas that the student should improve on by monitoring the students’ academic performance, as well as their participation in a given course (Dietz-Uhler & Hurn, 2013). The course improvements allow for the enhancement of learning outcomes that a plethora of accrediting bodies has recommended. For instance, the IBM has suggested several ways in which schools and other learning institutions can assist in enhancing the students’ academic performance and achievement. Some of these ways include monitoring the students’ performance, preventing attrition from a given program, and “identifying outliers for early intervention” (Dietz-Uhler & Hurn, 2013). The IBM also suggests that there is the need to disaggregate the performance of a learner by selected attributes like ethnicity and year of study. The students’ academic achievement can also be enhanced through the development of effective instructional interventions and the evaluation of the learning curriculum.
The Impacts of Learning Analytics on the Higher Education Sector
In their research study, Dietz-Uhler and Hurn (2013) assert that some of the impacts of learning analytics include improved accountability, increased retention rates, and improved student success. Apart from assisting individual students to improve their academic performance, learning analytics can also assist the administration to improve its decision-making process. Additionally, LA provides the administrators with the essential information they need to allocate resources efficiently. The use of LA in the higher education sector can also enable the administration to determine the institution’s challenges and successes, a move that is likely to enhance organizational productivity. Learning analytics has enabled the faculty members to identify those students who may be at risk during the learning process. Through this, the faculty members have been able to develop interventions and strategies of helping these students. Moreover, the faculty has used LA to transform pedagogical approaches and develop interventions aimed at assisting learners to gain insights into their performance and learning. The use of LA has also enabled higher learning institutions to build models that showcase the behaviors of successful students (Smith, Lange & Huston, 2012). The model may have information concerning the number of times a successful student accesses discussion board posts and the frequency at which they take quizzes among others. A faculty is likely to encourage students to engage in such behaviors if it builds a model that describes the behaviors of a successful student. It can also assess the behaviors of students who do not conform to this particular model.
Greller and Drachsler (2012) in their research study also note that the use of LA has enabled the faculty to identify gaps that exist in the knowledge and information acquired by learners. The awareness of such gaps will allow course instructors to concentrate on particular learners and assist them in improving their academic performance and success. Likewise, Zdrahal, Nikolov, and Pantucek in their 2013 research study assert that through LA, the higher learning institutions have been able to understand their students. They assert the main aim of using LA is to comprehend the learners’ behavior through the analysis of their data. The analysis of the learners’ data requires the faculty to assess the manner in which students interact with the VLE and the impacts of such interactions on a learner’s final performance. Adejo and Connolly (2017) also note that the use of LA in higher learning institutions has enabled course instructors and faculty members to identify the challenges that the learners may be facing during their learning process. Besides, they note that learning analytics has also assisted in the improvement of the students’ retention rates. The close monitoring of the learners’ persistence and learning can enable stakeholders in the higher education sector to detect undesirable emotional states and unruly behaviors in students. Consequently, the determination of such behaviors can enable faculty members to identify students who are at risk (Verbert, Manouselis, Drachsler & Duval, 2012). Additionally, factors that can motivate a student to drop out of school can be identified and addressed promptly. The faculty members, in turn, can develop strategies to assist those students who may have the motive of dropping out of the learning institution. These strategies or follow-up action may be in the form of counseling services among others.
The Challenges of Learning Analytics in Higher Education
Adejo and Connolly (2017) in their research study note that learning analytics aimed at providing significant benefits to both the learners and teachers in the learning institution. However, there are a plethora of challenges that have adversely affected the implementation and application of learning analytics in higher education. The first challenge that Adejo and Connolly (2017) highlight is database and data heterogeneity. Vast amounts of educational data tend to be in different formats and are normally spread around several heterogeneous databases. The large databases are often difficult to integrate into the learning institution. The second challenge is that learning analytics do not have the appropriate technological architecture. Such a situation poses a significant challenge to the implementation and application of LA technology in higher education. The third challenge is the issue of data ownership. The data collected from various databases of learning institutions and those gathered through the use of questionnaires and surveys usually have ownership issues. There is the need to clarify and address the issue of data ownership as it relates to the students’ data. The other challenge that Adejo and Connolly (2017) analyze in their research study is privacy issues. Issues such as data profiling and access need to be addressed before a learning institution decides to use learning analytics. In a learning institution, the educational administrator needs to develop guidelines and frameworks to monitor the use, as well as the access of the students’ data.
The issue of consensus research framework is a significant factor that has adversely affected the application of LA in the education sector (Ferguson, 2012). Currently, there is the lack of a standardized framework that can aid in the implementation of learning analytics in the educational system. The other important factor that has posed a threat to the implementation and application of LA in the educational sector is the generalization of tools and application. Several high learning institutions have implemented and used LA in their system. Examples of these include the Purdue University, Open University Australia, and the University of Wollongong among others (Adejo and Connolly, 2017). However, it is worth noting that these LA technologies tend to be specific to the learning institutions. Notably, it is not possible for learning institutions to adapt those LA technologies that have been developed by other universities due to the geographical, economic, and social factors that affect such technologies. For LA to be effective and successful, it is important to address the adaptability challenge. The last challenge that has affected the application and implementation of LA technologies is the inadequate training of faculty members and course instructors on the application of LA tools and techniques.
Case Studies on the use of Learning Analytics in the Higher Education Sector
At Purdue University, the aim was to apply the concept of business intelligence to improve learners’ performance at the course level. Through this, the university aimed at enhancing the students’ graduation and retention rates. The signals system developed by Purdue University was developed to assist learners in understanding their academic progress early enough (O’Farrell, 2017). Based on this, the learners at this university would be able to seek assistance from the course instructors to improve their academic performance. The signals system ordinarily obtains data from the students’ grade book and the VLE to produce information showing how at risk each learner may be. The information will then enable the instructor to develop interventions that can assist those students who are at risk of performing poorly or dropping out of school (Sclater, Peasgood & Mullan, 2016). The predictive algorithm that course instructors at Purdue University used is ordinarily based on the students’ attributes, past academic history, and performance.
Apart from Purdue University, the University of Maryland is another typical example of a higher learning institution that has adopted the use of learning analytics. It is a public research university that has over 12,000 students (O’Farrell, 2017). In this institution, the high performing students used VLE more than those students who performed poorly. However, there is the need for more research on the impacts of VLE on the students’ retention and graduation rates. The New York Institute of Technology is another example of a higher learning institution that has implemented learning analytics (Sclater, Peasgood & Mullan, 2016). It has developed a predictive model in support with the counseling staff to identify those students who may be at risk of dropping out of school. The main objective of developing the model was to increase the learners’ retention rates. Additionally, it aimed at providing the counselors with information about each learner’s academic progress and performance (O’Farrell, 2017). After the identification of those students at risk, the counseling staff at the institute would then be required to provide the necessary support that can enable the learners to change their behaviors. The predictive model used at the institute was developed internally, and this makes the databases the counselors rely on be on the same platform.
In New York Institute of Technology, the predictive model customarily utilizes four data sources, which include financial data, surveys, registration and placement examinations data, and the admission application data. Financial data was used since in this institute; money is a factor that influences the students’ retention and completion rates. Apart from this institute, California State University is another higher learning institution that has in recently used learning analytics (O’Farrell, 2017). In this university, LA has been used to improve the academic performance and success of learners. The university argued that the learning analytics tools that several learning institutions relied on did not take into account vast amounts of data in the VLE (Sclater, Peasgood & Mullan, 2016). In this institution, the use of VLE as an analytic learning tool plays an important role in predicting students’ performance and academic success.
The Marist College has also developed an alert solution that assists students during their learning process. The predictive model that this college developed has been transferred to other colleges and universities. It has assisted in the evaluation of solutions aimed at assisting students who are at risk of not completing their education (O’Farrell, 2017). The data used in the predictive model emanates from the learners’ demographic details, aptitude data, and different aspects of the students’ use of VLE. The models developed by Marist College have in the recent past been deployed to various state universities with learners with low retention rates. They were deployed in these universities with the aim of investigating their portability and assessing their effectiveness in addressing the needs of those students at risk of low retention rates. Researchers realized that the learners’ success in class was predicted by their current academic standing, grade point average, and the grades they have received in the course. Based on their analysis, the researchers noted that the VLE grade book was a valuable tool that could enable higher learning institutions to determine at-risk students. Through the VLE grade book, the course instructors would have the chance of developing interventions that can assist those students who are at risk of performing poorly in class or dropping out from school (Sclater, Peasgood & Mullan, 2016). The researchers at Marist College produced reports that showed the learners who were regarded to be at risk. These students were required to undergo two distinct intervention strategies. The researchers sent the first category of learners a message that informed them that they were likely not to complete their course. They also guided the group on the activities that they were to engage in to enhance their chances of completing the course. Conversely, the researchers directed the second group of students to the online support environment. In this environment, the learners were provided with study skills, stress reduction initiatives, and time management strategies. Additionally, the second group was subjected to mentoring programs from the university support staff. The third category of students was not subjected to any intervention since it was a control group (Sclater, Peasgood & Mullan, 2016). The results indicated that students in the groups that were subjected to interventions performed well then the students in the third group. The researchers concluded that the predictive model plays an essential role in providing learners with feedback on their learning process.
Edith Cowan University is another example of a learning institution that has in the recent past implemented tools for learning analytics. It has an LA initiative know as Connect for Success, which enables it to identify those students that require assistance. The C4S allows the university’s support staff to be in touch with several learners and manage various interventions for all of the students. The Edith Cowan University has invested in multiple research studies primarily aimed at identifying the learners’ variables that tend to be valuable in the prediction of attrition (Sclater, Peasgood & Mullan, 2016). Some of the factors that were identified include language skills, university entrance scores, and student grades. Through these factors, the university has been able to identify those learners who are likely to require assistance and support from the course instructors. The predictive model that the university uses customarily assigns two faculty members the responsibility of contacting those learners who may need support. The two faculty members will then send the students personalized emails that assist them (Sclater, Peasgood & Mullan, 2016). In case the students do not respond to the emails, the faculty members will call the students and inform them of the university’s decision to assist them in their studies. The support staff uses a dashboard to manage contacts, as well as support each learner. They can agree with learners and create an action plan that may ordinarily comprise a host of interventions such as appointments with the course instructor, referrals to skilled personnel, and personalized assistance among others (O’Farrell, 2017). Whereas the Edith Cowan University has experienced significant benefits from the use of C4S, it has experienced challenges which are mostly in the form of data integrity and privacy and student suspicion among others.
At the University of England, about 20 percent of the learners belong to the low-class. In this learning institution, there was the need to identify the struggling learners so that they could be provided with timely support (Sclater, Peasgood & Mullan, 2016). Moreover, the academic staff classified students at risk and in need of support if they failed to attend a class or submit assignments. The learning institution developed a learning analytics tool that could assist in capturing the learners’ learning wellbeing status. The Open University is another important higher learning institution that is currently investing in a learning analytics tool aimed at enhancing the students’ academic performance and success. Students’ retention has been an important issue for this learning institution. The institution has experienced challenges in its attempt to retain students due to the geographical differences. It has developed a learning analytics tool to enable the staff members to manage the interventions that the students receive. In this university, the students also have a dashboard that has allowed them to track their performance and academic progress. Through this dashboard, the learners have been able to make informed choices about their academic path. At the curriculum level, the university’s faculty members have used data analytics with the aim of assessing modules and informing students of any change to their course curriculum. The Open University regards the use of LA as an ethical process that ensures that the learner becomes the active participant (Sclater, Peasgood & Mullan, 2016). Apart from the Open University, the Nottingham Trent University has also used predictive analytics to enhance its learners’ academic experience. The predictive analytics tool used in this university has helped to facilitate dialogue between the faculty members and the learners.
The research paper’s primary aim is to analyze ways in which learning analytics has been utilized in universities and other higher learning institutions. In this study, we conducted a meta-analysis to analyze and synthesize the various outcomes of different research and case studies. Relevant research and cases studies were collected from different databases using the keywords “learning analytics” for the period from 2012 to 2018. The first selection criterion used to include the studies was that the research study had a report on the use of learning analytics in accredited universities and learning institutions. The other selection criterion was that the study had the description and objectives of LA and its implementation in the higher learning institution. The selected studies were required to have the results or outcomes of using LA in higher learning institutions. Whereas the initial search yielded several results, only 40 research studies met the inclusion criteria after screening. The studies which were analyzed in this research paper included the appropriate quantitative data analysis. 20 studies met the final inclusion criteria.
From the analysis of various research studies, it was evident that learning analytics had a host of benefits for the students, staff members, and higher learning institutions. Some of the studies reported that the use of LA in learning institutions results in improved student retention (Jones, 2012; Olmos & Corrin, 2012; Picciano, 2012; and Shum & Ferguson, 2012). According to these studies, the close monitoring and surveillance of students’ persistence and learning behavior enable staff members to detect learners’ unusual and unruly behaviors (Adejo & Connolly, 2017). Through the use of LA in learning institutions, institutions and faculty members can identify and become aware of the various factors that can motivate students to leave or dropout from school. LA enables course instructors to identify those students who are at risk of engaging in a plethora of negative behaviors. Follow-up action such as counseling can be developed with the motive of providing such students with the necessary support that they need. Therefore, the results of the analyzed and reviewed studies indicate that the use of LA in higher learning institutions can result in the enhancement of the students’ retention rates.
The review of the studies also indicates that the use of LA supports informed decision-making (Abdous, He & Yen, 2012; Dyckhoff, Zielke, Bultmann, Chatti & Shroeder, 2012; and Picciano, 2012). These studies indicate that the use of LA provides learning institutions with information that they can use to make effective and informed decisions. The decisions aimed at improving the students’ learning outcomes. The data generated from learning analytics can also enable stakeholders in the higher education sector to make informed decisions about resource allocation. For example, they can allocate resources to faculties based on the courses’ popularity within the institution. Apart from enabling learning institutions to make informed decisions, the use of LA also increases cost-effectiveness (Johnson, Adams & Cummins, 2012; Jones, 2012; Mattingly, Rice & Berge, 2012 and Olmos & Corrin, 2012). In most cases, learning analytics can be incorporated into the learning management system among other platforms. In turn, course instructors and tutors can support learners and provide them with course feedback on online platforms. Feedback and assessment reports on the students’ academic performance can be cost-effectively delivered to various stakeholders.
The analysis of various research studies also indicates that the use of LA assists course instructors in understanding the students’ behavior. Notably, the analysis of different data sources provides faculty members and course instructors with information that can enable them to understand the learners’ behaviors and ways in which they interact with others (Wong, 2017). Additionally, it can enable faculty members to determine how students utilize learning resources and materials. With this information, faculty staff and course instructors can have the chance of evaluating the effectiveness and significance of instructional designs, as well as pedagogies for academic improvement. The use of LA in higher institutions of learning allows faculty members to determine students’ preferences and learning behaviors. The assessment of students’ preferences and behavior plays an essential role in allowing the faculty staff to design course materials that address the learners’ diverse needs. The other result that was obtained is that the use of LA in universities and other higher learning institutions provides personalized assistance for the learners. LA generates vital information about the learners’ learning patterns and attributes, which can help in making the students’ academic experiences engaging and personal. Making the students’ learning experiences has been shown to improve their learning outcomes and educational achievements. Through LA, the course instructor can generate and send alerts to those learners who have failed to meet the academic threshold. The students’ success in higher learning institutions can be significantly improved in situations where the learners are advised to engage in personalized learning activities.
The review of various research studies and case studies indicate that the use of LA in higher learning institutions has resulted in timely intervention and feedback. The use of LA provides course instructors and tutors with essential information about the progress and academic performance of the learners. The holistic information that LA generates can enable the instructors to develop individualized interventions that can assist in enhancing the learners’ academic achievement. Learners are likely to form a sense of belonging with the course instructor if they receive personalized feedback about their academic performance. Instructors who use social network analytics understand the learners’ needs and challenges. Additionally, such networks allow instructors to determine those learners whose academic performance is below par. After the determination of such students, the course instructor can isolate and provide them with a more personalized learning experience.
Meta-analysis of the impacts of interventions on the learners’ academic performance
The review of the different case and research studies has indicated that the primary role of LA in the higher education sector is to identify those students who are at risk of performing dismally in class. The course instructor will then have the chance of delivery early interventions to such students with the view of enhancing their graduation and retention rates and improving their academic performance. In several research studies, the main approach that was used to provide interventions for learners at risk was to gather and evaluate data on their learning patterns, preferences, and activities. A computational model was then utilized to identify and prioritize those learners who were deemed to be at risk of attaining poor grades and dropping out of school. The outcomes of this predictive model would then be used to analyze the interventions and strategies that can be used to reduce such risks. A common practice that was observed from the reviewed case studies was to charge an academic staff with the mandate of contacting those learners at risk and providing them with personalized learning assistance. Significantly, the approach was hailed as an effective strategy of improving the learners’ success and performance as measured by indicators like the graduation rate, retention rate, and academic progress. Overall, the meta-analysis indicated that learning analytics is imperative in assisting higher learning institutions to make decisions and develop interventions that can help in improving the learners’ academic performance and success.
From the analysis, it is evident that various research and case studies have reported positive results. Based on the outcomes, higher learning institutions should apply and implement learning analytics to analyze data about students’ learning outcomes, interactions, preferences, and experiences. The implementation of LA in higher learning institutions is likely to result in a host of positive benefits to the academic staff, learners, and the learning institution. LA is likely to benefit the learning institutions in operational areas like learners’ support and quality assurance. In this research paper, we reviewed predictive models that could assist in improving the learners’ academic performance and success. The models were primarily developed with the intention of validating and prioritizing learners who may be at risk of dropping out of school or performing poorly in school (Bichsel, 2012). Notably, the quantitative analyses revealed and confirmed that the learners’ performances and academic achievement improved after being exposed to the LA-based interventions. Thus, learning analytics is an important strategy that higher learning institutions can exploit to improve the students’ learning experiences and outcomes.
The outcomes of the reviewed and analyzed studies suggest that the learners’ behavior can be changed through the use of the learning analytics tools. The meta-analysis indicated that only a small proportion of research and case studies correlated with the application and implementation of learning analytics in the higher learning institutions provided quantitative analysis data. Such a limitation results from the fact that the concept of learning analytics is still at its infancy stage. Thus, there is the need to conduct more research studies on the theories and models that can be used by higher learning institutions to apply and implement LA (Ferguson, 2012). Additionally, it is essential to establish, as well as innovative test interventions that are learning analytics support. Current LA-bases solutions, as analyzed in this research study, were customarily based on the association that exists between course instructors and learners. It is worth noting that whereas these solutions tend to be effective; their efficiency often differs among various learners.
Recommendations and Conclusion
From the analysis and review of the various case and research studies, it is evident that LA is gaining momentum in most higher learning institutions. Due to the significant benefits of LA to these institutions, it is vital for universities and colleges to adopt and implement LA tools and initiatives. They need to take advantage of the data available in their databases to improve students’ retention rates and enhance their performance. The use of the students’ data will have a profound impact on the learners’ success, performance, and academic achievement. Besides, there is a wide range of challenges that affect the successful implementation of LA by various higher learning institutions. Due to these drawbacks, it is expedient that higher learning institutions in different parts of the world make necessary resolutions that can assist in the enhancement of the students’ data security and privacy. Notably, this research study recommends that it is imperative to have an established standard in the utilization of learning analytics in higher learning institutions. Such standards will play an essential role in reducing legal risks associated with the breach of the students’ and the faculty members’ privacy. Furthermore, the education administrator should be in the frontline in advocating for the formulation of strategies that promote the training of course instructors and faculty members on the utilization of various tools and initiatives of learning analytics. In the future, it would be beneficial for learning institutions to have in place a dashboard that can enable course instructors to have a glance at information about students at-risk so that an appropriate solution can be developed. Lastly, there is the need for more empirical research that can measure the acceptance rate of LA tools by higher learning institutions.
Abdous, M., He, W., & Yen, C. (2012). Using data mining for predicting relationships between
online question theme and final grade. Educational Technology & Society, 15(3), 77-88
Adejo, O. & Connolly, T. (2017). Learning analytics in higher education development: A
roadmap. Journal of Education and Practice, 8(15), 156-163
Bichsel, J. (2012). Analytics in higher education: Benefits, barriers, progress, and
recommendations. EDUCAUSE Center for Applied Research.
Dietz-Uhler, B. & Hurn, J.E. (2013). Using learning analytics to predict (and improve) student
success: A faculty perspective. Journal of Interactive Online Learning, 12(1), 17-26
Dyckhoff, A. L., Zielke, D., Bultmann, M., Chatti, M. A., & Schroeder, U. (2012). Design and
implementation of a learning analytics toolkit for teachers. Educational Technology & Society, 15(3), 58-76.
Dziuban, C., Moskal, P., Cavanaugh, T., & Watts, A. (2012). Analytics that inform the
university: Using data you already have. Journal of Asynchronous Learning Networks, 16(3), 21-38.
Ferguson, R., 2012. Learning analytics: drivers, developments and challenges. International
Journal of Technology Enhanced Learning, 4(5-6), 304-317
Greller, W. & Drachslrer, H. (2012). Translating learning into numbers: A generic framework for
learning analytics. Educational Technology & Society, 15(3), 42-57.
Johnson, L., Adams, S., and Cummins, M. (2012). The NMC Horizon Report: 2012 Higher
Education Edition. Austin, Texas: The New Media Consortium
Jones, S. J. (2012). Technology review: The possibilities of learning analytics to improve learner
centered decision making. The Community College Enterprise, 89-92.
Mattingly, K. D., Rice, M. C., & Berge, Z. L. (2012). Learning analytics as a tool for closing the
assessment loop in higher education. Knowledge Management & E-Learning: An International Journal, 4(3), 236-247
O’Farrell, L. (2017). Using learning analytics to support the enhancement of teaching and
learning in higher education. National Forum. Retrieved from www.teachingandlearning.ie/wp-content/uploads/2018/01/Final_LA-Briefing-Paper_Web-with-doi.pdf
Olmos, M. & Corrin, L. (2012). Learning analytics: A case study of the process of design of
visualizations. Journal of Asynchronous Learning Networks, 16(3), 39-49.
Picciano, A. G. (2012). The evolution of big data and learning analytics in American higher
education. Journal of Asynchronous Learning Networks, 16(3), 9-20.
Shum, S. B. & Ferguson, R. (2012). Social learning analytics. Educational Technology &
Slater, N., Peasgood, A. & Mullan, J. (2016). Learning analytics in higher education: A review
of UK and international practice. JISC. Retrieved from www.jisc.ac.uk/sites/default/files/learning-analytics-in-he-v2_0.pdf
Smith, V. C., Lange, A., & Huston, D. R. (2012). Predictive modeling to forecast student
outcomes and drive effective interventions in online community college courses. Journal of Asynchronous Learning Networks, 16(3), 51-61.
Verbert, K., Manouselis, N., Drachsler, H., & Duval, E. (2012). Dataset-driven research to
support learning and knowledge analytics. Educational Technology & Society, 15(3), 133-148.
Wong, B.T.M. (2017). Learning analytics in higher education: an analysis of case studies. Asian
Association of Open Universities Journal, 12(1)
Zdrahal, A., Nikolov, A. & Pantucek, M. (2013). Improving retention: predicting at-risk students
by analyzing clicking behavior in a virtual learning environment. ACM