Fork me on GitHub

Project Team 12: Data Within Education, DS 4200 F20

Taeyoon Jin, Paul Vicks, Haicheng Liu

Service-Learning Course Project as part of DS 4200 F20: Information Visualization, taught by Prof. Cody Dunne, Data Visualization @ Khoury, Northeastern University.

Abstract

One of the biggest dilemmas within the current use of technology for education is the fact that massive amounts of student data is being left untapped at the expense of the opportunity to aid instructors. Keo LMS, a learning management system company, has provided this team with key student data ranging from click rates to student submission cycles. Leveraging this meaningful student performance data, data will be visualized to provide a platform to help online educators transform the way they digest their students’ online patterns. Keywords: Educators, Students, Performance

Visualization

Below are two visualization designs. One includes a table and scatterplot. The other includes a table and line chart.

The objective of the first scatterplot visualization is for the end user to quickly identify grade trends as it correlates to time spent on a quiz assignment. Many institutions that exclusively utilize mock tests as a tool find such visualization especially useful as it can be directly used to improve student time management skills.

The second linegraph visualization aims at providing the end user with data to show a student performance over time. Such tools can be utilized to quickly identify sudden changes in average grades to provide further help.

The data within this current example does not reflect real student data, structure of classes and grades along with data structure from existing classes and the Keo platform was taken to generate a sanitised dataset.


Class Timed Quiz Grade Distribution

Note: The data below does not reflect any real data containing private information.

Interact with the scatter plot or table through dragging or clicking to isolate and zoom in to the scatter plot. Double click the scatter plot to zoom out and load in all data from the dataset.



Student Average Grade Over Time

Note: The data below does not reflect any real data containing private information.

Select a student from the drop down below to get started. The line graph and table both dynamically updates according to the student selection.

Selected Student:

Demo Video

Video below shows the different levels of interaction the above tool is capable of along with possible use cases for the real world.


Presentation

Click the presentation below to be directed to the online presentation.

W3Schools.com

Visualization explanation

The scatterplot and table visualization accomplishes domain tasks two and three. These domain tasks were position of time spent by overall grade and visualization of classwork grades, respectively. The line chart and table visualization satisfy domain tasks one and four. These domain tasks were visualization of class grades and average grades per student.

In terms of interactivity, the scatterplot and table visualization utilize brushing and linking, zooming, and also details on demand. Details on demand are provided when a group of scatter points are brushed over and the table correspondingly adjusts to only include the datapoints that have been brushed. The table provides more details than just the scatterplot can inform the user. Double clicking on the scatterplot resets the zoom level to normal and resets the table to include all original data. For the line chart and table, the visualization leverages linking and details on demand (details are provided for each student after a student is selected from the drop-down arrow).

Color schemes for both charts were chosen from primary color pallets as colors were used to encode categorical variables like class type. For the scatterplot, selected circles were filled with light grey so that the colored outline of each circle still could stick out. When table rows are selected, the circles that are not correspondingly selected have a reduced opacity so that they allow the selected circles to pop out (which remain at normal opacity). Selected rows from the first table are highlighted in a pink shade.

Acknowledgments