Open University Learning Analytics Dataset

Student Performance and Engagement Data at The Open University

By UCI [source]

About this dataset

> > This dataset provides an intimate look into student performance and engagement. It grants researchers access to numerous salient metrics of academic performance which illuminate a broad spectrum of student behaviors: how students interact with online learning material; quantitative indicators reflecting their academic outcomes; as well as demographic data such as age group, gender, prior education level among others. > > The main objective of this dataset is to enable analysts and educators alike with empirical insights underpinning individualized learning experiences - specifically in identifying cases when students may be 'at risk'. Given that preventive early interventions have been shown to significantly mitigate chances of course or program withdrawal among struggling students - having accurate predictive measures such as this can greatly steer pedagogical strategies towards being more success oriented. > > One unique feature about this dataset is its intricate detailing. Not only does it provide overarching summaries on a per-student basis for each presented courses but it also furnishes data related to assessments (scores & submission dates) along with information on individuals' interactions within VLEs (virtual learning environments) - spanning different types like forums, content pages etc... Such comprehensive collation across multiple contextual layers helps paint an encompassing portrayal of student experience that can guide better instructional design. > > Due credit must be given when utilizing this database for research purposes through citation. Specifically referencing (Kuzilek et al., 2015) OU Analyse: Analysing At-Risk Students at The Open University published in Learning Analytics Review is required due to its seminal work related groundings regarding analysis methodologies stem from there. > > Immaterial aspects aside - it is important to note that protection of student privacy is paramount within this dataset's terms and conditions. Stringent anonymization techniques have been implemented across sensitive variables - while detailed, profiles can't be traced back to original respondents. >

How to use the dataset

> > #### How To Use This Dataset: > > - Understanding Your Objectives: > Ideal objectives for using this dataset could be to identify at-risk students before they drop out of a class or program, improving course design by analyzing how assignments contribute to final grades, or simply examining relationships between different variables and student performance. > > - Set up your Analytical Environment: > Before starting any analysis make sure you have an analytical environment set up where you can load the CSV files included in this dataset. You can use Python notebooks (Jupyter), R Studio or Tableau based software in case you want visual representation as well. > > - Explore Data Individually: > There are seven separate datasets available: Assessments; Courses; Student Assessment; Student Info; Vle (Virtual Learning Environment); Student Registeration and Student Vle. > Load these CSVs separately into your environment and do an initial exploration > of each one: find out what kind of data they contain (numerical/categorical), if they have missing values etc. > > - Merge Datasets > As the core idea is to track a student’s journey through multiple courses over time, combining these datasets will provide insights from wider perspectives. > One way could be merging them using common key columns such as 'code_module', 'code_presentation', & 'id_student'. But make sure that merge should depend on what question you're trying to answer. > > > - Identify Key Metrics > Your key metrics will depend on your objectives but might include: overall grade averages per course or assessment type/student/region/gender/age group etc., number of clicks in virtual learning environment, student registration status etc. > > > - Run Your Analysis > Now you can run queries to analyze the data relevant to your objectives. Try questions like: What factors most strongly predict whether a student will fail an assessment? or How does course difficulty or the number of allotments per week change students' scores? > > - Visualization: > Visualizing your data can be crucial for understanding patterns and relationships between variables. Use graphs like bar plots, heatmaps, and histograms to represent different aspects of your analyses. > > - Actionable Insights: > The final step is interpreting these results in ways that are meaningful and actionable for

Research Ideas

> - Predicting Student Success: This dataset can be used to develop a predictive model that identifies students who are at risk of failing a course or dropping out of the university. By analyzing variables such as performance on assessments, level of engagement with the course materials, demographic factors and previous educational background, you could identify early warning signs and intervene before it's too late. > - Analyzing Impact of Course Structure: The dataset includes detailed information about different types of activities (or study materials) in each course module. An analysis could be carried out to understand how each type contributes to student success rates. Identification of key activities that have more influence could help reshape teaching strategies or course layouts. > - Personalized Learning Paths: Based on student's interaction history with various learning resources (such as forum posts or clicks), machine learning models can be built to recommend personalized learning paths for students enhancing their success rate in their courses. > > - Detecting Disparities in Student Outcomes: You might also use this data to investigate disparities in outcomes related to factors like region, age group, and disability status. > > - Online Learning Behavior Analysis : As it is digital learning dataset can also provide opportunity for research into online learner behaviors by studying clickstream data patterns against final results achieved by learners then suggesting enhancement areas respectively.

Acknowledgements

> If you use this dataset in your research, please credit the original authors. > Data Source > >

License

> > > See the dataset description for more information.

Columns

Acknowledgements

> If you use this dataset in your research, please credit the original authors. > If you use this dataset in your research, please credit UCI.