Big Data and Digital Grading

 (Canva Images)

        The prospect of “big data” at once evokes optimistic views of an information-rich future and concerns about surveillance that adversely impacts our personal and private lives (Cope and Kalantzis, 2016). Big Data in Education is transforming ways to understand student performance and learning patterns by analyzing vast amounts of educational data. With this, educational Institutions can tailor learning experiences to individual needs (Roberts, 2023). 

       Big Data in Education involves collecting, processing, and analysing vast datasets generated by educational institutions. These include student demographics, academic performance, attendance, online interactions, and assessment results. Key characteristics of Big Data in Education are volume (large amounts of data), velocity (rapid data creation), and variety (diverse data types) (Roberts, 2023). Big data also focuses on advanced analytics and data mining to uncover patterns and trends, which help educators identify at-risk students, tailor teaching methods, and improve learning experiences. 

        To set the stage with a definition, “big data” in education is:

        1. the purposeful or incidental recording of activity and interactions in digitally mediated, network-interconnected learning environments—the volume of which is unprecedented in large part because the data points are smaller and the recording is continuous;

        2. the varied types of data that are recordable and analyzable;

   3. the accessibility and durability of these data, with potential to be (a) immediately available for formative assessment or adaptive instructional recalibration and (b) persistent for the purposes of developing learner profiles and longitudinal analyses; and

    4. data analytics, or syntheses and presentations based on the particular characteristics of these data for learner and teacher feedback, institutional accountability, educational software design, learning resource development, and educational research.

(Cope and Kalantzis, 2016)

        Applications of Big Data in Education extends far beyond data collection; it involves practical implementation that positively impacts various facets of the educational process. The following are some of its common applications: 

(theknowledgeacademy, Roberts, 2023). 

      On the other hand, critics worry about issues such as student privacy, the effects of profiling learners, the intensification of didactic pedagogies, test-driven teaching, and invasive teacher accountability regimes (Cope and Kalantzis, 2016). Other challenges include lack of data literacy skills for educators and educational administrators to effectively interpret Big Data, and students may not fully understand how their data is collected, used, or shared, leading to mistrust and resistance (Roberts, 2023). 

        There is also potential for bias in Big Data. For example, biased data may lead to unfair predictions or decisions related to student performance or behavior. The use of Big Data in Education also raises ethical questions regarding fairness, transparency, and accountability in decision-making processes. There is also the cost of implementing Big Data solutions, which requires significant financial investments in technology, training, and ongoing maintenance (Roberts, 2023). 

        However, there are some key benefits of Big Data in Education:

        1) Personalized Learning

        Big Data allows educators to tailor learning experiences to individual students. By analyzing data on student performance, learning styles, and engagement, teachers can create customized lesson plans that address each student’s strengths and weaknesses. 

        This personalized approach can enhance student outcomes and foster a more engaging learning environment.

        2) Improved Student Performance

        Big Data helps track and analyze student performance over time. Educators can identify patterns and trends, such as which subjects' students struggle with or excel in. This data allows for prompt interventions and support, aiding students to improve their academic performance and reach their educational objectives.

        3) Enhanced Decision-Making

        Educational institutions can leverage Big Data to make informed decisions. By analyzing data on enrollment trends, resource allocation, and student demographics, administrators can develop strategies to improve institutional efficiency and effectiveness. Data-driven decision-making can lead to better resource management and improved educational outcomes.

        4) Predictive Analytics

       Big Data facilitates predictive analytics, enabling educators to anticipate student outcomes and detect potential problems before they occur. For example, predictive models can help identify students at risk of dropping out, enabling early interventions to keep them on track. This proactive approach can highly reduce dropout rates and improve overall student retention.

        5) Enhanced Teaching Method

    By analyzing data on teaching methods and student feedback, educators can refine their instructional techniques. Big Data provides insights into which teaching strategies are most effective, allowing teachers to adapt their methods to better meet student needs. This continuous improvement process can lead to more effective teaching and better student engagement.

        Overall, Big Data has brought tremendous potential to transforming learning and is becoming a crucial for successful implementation. 

(Canva Images)

Digital Grading

        Automated assessment applications have achieved better than human reliability and other methods of assisting assessment have opened up additional venues for utilization in the classroom and beyond. However, a lack of understanding of the differences between the different types of applications and their limitations has made selecting the appropriate application a difficult task (Aken, 2017). Computer-based text analysis (CBTA) has been referred to by many names in the literature including Automated Essay Scoring (AES), which is the most prevalent name particularly within education literature. 

    CBTA falls into two categories: automated assessment and machine-assisted analysis. Automated writing assessment requires no human intervention (subsequent to the initial configuration of the prompt) while machine-assisted analysis is dependent upon human interaction to provide an analysis of the text being analyzed (Aken, 2017). 

(Aken, 2017)

    Most CBTA programs are developed to assess written text to provide
a summary score (summative assessment). Some applications, however, are
primarily as a learning environment designed to assist students in learning how to write (formative assessment) as well as possibly providing summative assessment. Although most formative assessment applications also provide summative scores, the applications classified as summative assessment do not include formative assessment capabilities (Aken, 2017). 

        After Ohio started using American Institutes for Research in 2015 to provide and score state tests, Artificial Intelligence (AI) programs have increasingly taken over grading. Computers are now scoring the entire test for about 75 percent of Ohio students, State Superintendent Paolo DeMaria and state testing official Brian Roget told the state school board recently. The other 25 percent are scored by people to help verify the computer's work (O'Donnell, 2018). Multiple other testing organizations -  like Pearson (which handled the old PARCC tests), McGraw Hill and Educational Testing Service (which produces graduate school admissions tests) - have developed automated scoring systems that can quickly compare student essays to model answers humans provide.

        Just this past spring, the Texas Education Agency (TEA) used computers and artificial intelligence technology to grade students' open-ended questions on the State of Texas Assessment of Academic Readiness for Science, Social Studies, Reading, and Language Arts. According to a scoring report by the TEA, students' responses will be graded first by a computer. A hired human scorer will grade roughly 25 percent of those responses. If the computer has a "low confidence" score, it will also be re-scored by a human. Some tests are also up for review if the computer's programming catches unrecognizable responses like slang words, phrases, or languages other than English (Sessions, 2024). Students and parents who disagree with the computer and human scores can request a rescore for $50. The fee can be waived if the computer's or human's score is wrong. TEA guidelines state that the STAAR test measures how much a student has learned about a subject.

        While online grading saves time and money, I question its accuracy. Some argue that these algorithms may struggle to grasp nuances of language and could misinterpret students' work (Aken, 2017). Furthermore, the assessment criteria of algorithms are limited, potentially overlooking certain areas such as organizational skills or grammar usage.

         I think online grading should mainly be used a supplementary tool until it can be proven that these online and digital grading systems are more accurate, especially when it comes to writing. 

Future Reference: Superintendents question automated grading system, giving thousands of students zeros on a portion of STAAR test and Computers Scoring STARR Essays: Is Texas Sacrificing Quality for Efficiency? 

References

Aken, Andrew. An Evaluation of Assessment-Oriented Computer-Based Text Analysis Paradigms. Higher Education Research. Vol. 2, No. 4, 2017, pp. 111-116. doi: 10.11648/j.her.20170204.12

Cope, B., & Kalantzis, M. (2016). Big Data Comes to School. AERA Open, 2(2), 233285841664190. https://doi.org/10.1177/2332858416641907

Roberts, S. (2023, October 16). Big Data in Education: Applications, Limitations, and Future Scope. Theknowledgeacademy. Retrieved August 6, 2024, from https://www.theknowledgeacademy.com/blog/big-data-in-education/

Sessions, K. (2024, April 12). Details Emerge on Automated Grading of Texas' STAAR Tests. Government Technology. Retrieved August 6, 2024, from https://www.govtech.com/education/k-12/details-emerge-on-automated-grading-of-texas-staar-tests