Big data analytics in education: Opportunities and challenges
Big data analytics refers to the process of examining and analysing large, varied, complex, and dynamic data sets to reveal useful information and improve decision-making process. Such kind of analysis has changed the way decisions are made in recent years within the private, public, and science sectors, including in education. As technology is increasingly used in education, the huge amount of data about students and the educational institutions is continuously generated. Despite some controversies and challenges, big data analytics offers possibilities for teachers, institutions, and other education stakeholders to explore educational phenomenon more efficiently as well as improve educational programmes more effectively at three different levels of education; micro (learning), meso (institutional), and macro (inter-institutional).
The opportunities for big data analytics in education
At the micro level, the analysis of big data can help institutions to improve the quality of learning and teaching while streamlining processes and reducing administrative workload [1]. The use of virtual learning environment, academic information system, library management system, internet of things, or other digital tools in educational institutions produces vast amount of data, including students’ demographic background, attendance, activities in the virtual learning environment (e.g., numbers of logins, mouse clicks, forum participation), and test results. More advanced machine learning techniques such as text mining and sentiment analysis enable automatic information extraction from non-structured textual data. These techniques can be used to identify students’ emotion and quality of thinking through analysing learners’ comments, messages, forum posts, or blog contents.
Not only does the analysis of these data enable educators to better monitor students’ learning experience and behaviour, it also can be used to predict students’ performance and identify potential problems through the application of learning analytics. Besides it has the potential for the implementation of early alert systems, this also helps teachers to evaluate the instructional methods, give better feedback, and take necessary interventions in a timely manner compared to using traditional formative assessment methods. Moreover, providing learners with timely information about their learning performance increases students’ motivation, confidence, and academic success [2]. Further implementation of learning analytics also opens possibilities to enrich learning experience by facilitating personalisation of the learning process through automatically adapting the virtual learning environment, delivery format, contents, or learning path according to each individual’s needs and preferences [3, 4].
On the meso or institutional level, big data analytics helps educational institutions to evaluate the performance of institutional programmes and design evidence-based programmes. This includes the evaluation of teachers’ and non-academic staff’s performance, financial processes, as well as human resources and facilities procurement and management processes by analysing the data from platforms such as academic information system, human resources information system, and facility usage logs. Using data from academic information system, furthermore, enables institutions to measure and predict the improvement in graduation rates and students’ satisfaction along with their contributing factors [1].
Big data analytics can also be used on the macro level to utilise available data to provide a broad view of institutional performance and accountability and identify areas that need particular attention through continuous report to multiple stakeholders, such as government, public, and others [1]. Besides its potential in enabling policymakers to monitor institutions’ performance and educational policy implementation [5], data accountability also opens possibilities for inter-institutional collaborations. The availability of open-access data (e.g., PISA survey results) enables a school to understand how they compare with other schools and learn from each other to improve. Collaboration between school and government also opens possibilities to address societal issues, for example in maximising school enrolment rate or minimising unemployment rate within a district. Furthermore, big data enables the co-creation of governing structures and delivery of better policies and strategies used in education. An example for this is using text analytics and sentiment analytics to analyse internet users’ comments on a given topic by extracting tweets or social media posts on specific keywords or hashtags or education blog posts. If these data are acquired, processed, and analysed in real time, policymakers will be able to monitor public opinion on educational policy, evaluate policy implementation, and use crowdsourcing to engage the public in problem solving [5].
Challenges of big data analytics in education
On the other hand, there are also several issues with the use of big data in education. Big data analytics demands the digitalisation of educational and institutional processes. This digitalisation involves the availability of different software and hardware which requires a tremendous cost in the beginning. There is also a significant cost as the process of collecting, storing, and analysis of data produced requires access to a high-speed computational infrastructure and capable of handling a massive amount of data [6]. Another problem is that the collection of data from different system and different departments often comes in different types and format, causing interoperability issue, while there is a risk of data-loss within the data cleansing and data integration process [6].
Another main issue regarding data analytics in education is related to data privacy. The use of learning analytics which tracks students’ private information and learning data might be misused and cause harm to students in the long run as they move along in educational system and enter the workplace [5]. There is also a security threat of storing a large amount of data in one database, while this makes big data analytics easier. Big data analytics also needs a specific security technology as common security technologies are inefficient in handling big data, which is large and dynamic in nature. Additionally, accessing personal digital data without informed consent ignites the debate over the evolving ethical boundaries of conducting research involving humans in the era of big data [5]. Often, users are not fully informed that their activities in the online platforms are tracked for a purpose. For this issue, it is suggested for institutions to consider creating data governance models and data protection policies, as well as the context in which the data will be [6]. It is also recommended for institutions to deter any potentially wrongful use of data by implementing good learning management, reliable data warehousing and management, flexible and transparent data mining and extraction, as well as accurate and responsible reporting [6].
In conclusion, the use of big data in education offers many possibilities to improve learning by providing instructors and students with better information regarding learners’ learning process and opportunity in implementing personalised learning by automating learning content and environment adjustment based on learners’ data. On the institutional level, big data analytics also helps institutions to measure programme performances and foster evidence-based institutional policy-making, while on the macro level, big data enables policymakers to have a broader view of institutions’ performance and provides inter-institutional collaboration opportunities, as well as co-creates a new governing structures by involving public in problem-solving. However, there are also some issues regarding cost, infrastructure, data privacy, security, and ethics that also need to be addressed in implementing big data analytics in education.
As sharing educational data within the institution and between institutions increases in the use of big data in education, it is important to develop national and international standards of data sharing to address data interoperability, privacy, and security. Furthermore, it is also recommended for educators and institutions to focus the use of big data for learning improvement, rather than for research, as repurposing this data to research might not be ethical [6].
References
[1] Daniel, B. K. (2016). Overview of big data and analytics in higher education. In B. K. Daniel (Ed.). Big Data and Learning Analytics in Higher Education (pp. 1–4). Springer, Cham. DOI: https://doi.org/10.1007/978–3–319–06520–5_1
[2] Anirban, S. (2014). Big data analytics in the education sector: Needs, opportunities and challenges. International Journal of Research in Computer and Communication Technology, 3(11), 1425–1428. Retrieved from: http://www.ijrcct.org/index.php/ojs/article/view/935/pdf
[3] Avella, J., Kebritchi, M., Nunn, S., & Kanai, T. (2016). Learning analytics methods, benefits, and challenges in higher education: a systematic literature review. Online Learning, 20(2), 1–17. DOI: 10.24059/olj.v20i2.790
[4] Gibson, D. (2017). Big data in higher education: Research methods and analytics supporting the learning journey. Tech Know Learn, 22(3), 237–241. DOI: https://doi.org/10.1007/s10758-017-9331-2
[5] Wang, Y. (2016). Big opportunities and big concerns of big data in education. TechTrends, 60(4), 381–384. DOI: https://doi.org/10.1007/s11528-016-0072-1
[6] Daniel, B. K. (2017). Big data and data science: A critical review of issues for educational research. British Journal of Educational Technology. DOI: 10.1111/bjet.12595
