[Paper Review] A mobile web for enhancing statistics and mathematics education

Jamie Lentin, Anna Helga Jónsdóttir+3 more|arXiv (Cornell University)|Jun 19, 2014

Statistics Education and Methodologies7 references3 citations

TL;DR

This paper presents a mobile-optimized web application designed to enhance statistics and mathematics education through adaptive learning on tablets and smartphones. Using statistical models to personalize question difficulty and content based on student performance, the system improves knowledge incrementally over a semester and correlates strongly with course success, supporting both online and offline use.

ABSTRACT

A freely available educational application (a mobile website) is presented. This provides access to educational material and drilling on selected topics within mathematics and statistics with an emphasis on tablets and mobile phones. The application adapts to the student's performance, selecting from easy to difficult questions, or older material etc. These adaptations are based on statistical models and analyses of data from testing precursors of the system within several courses, from calculus and introductory statistics through multiple linear regression. The application can be used in both on-line and off-line modes. The behavior of the application is determined by parameters, the effects of which can be estimated statistically. Results presented include analyses of how the internal algorithms relate to passing a course and general incremental improvement in knowledge during a semester.

Motivation & Objective

• To develop a mobile-first educational platform that supports adaptive learning in statistics and mathematics.
• To enable personalized learning by adjusting question difficulty and content based on student performance.
• To support both online and offline usage for broader accessibility in educational settings.
• To evaluate the impact of adaptive algorithms on student knowledge gain and course pass rates.
• To provide a data-driven, statistically informed system for formative assessment and remediation.

Proposed method

• The application uses statistical models to analyze student responses and dynamically select questions based on performance.
• It adapts content from easy to difficult, or revisits older material, based on individual learning patterns.
• The system's behavior is governed by configurable parameters whose effects are estimated using statistical analysis.
• The platform operates in both online and offline modes, ensuring accessibility in low-connectivity environments.
• Internal algorithms are trained and validated using data from real courses, including calculus, statistics, and multiple linear regression.
• The system tracks student progress and uses this data to refine question selection and content delivery.

Experimental results

Research questions

• RQ1 How does adaptive question selection based on student performance affect knowledge acquisition in statistics and mathematics?
• RQ2 To what extent do the internal algorithms of the mobile web platform correlate with course pass rates?
• RQ3 What is the impact of the system on incremental knowledge improvement over a single semester?
• RQ4 How effective is the platform in supporting learning in both online and offline environments?
• RQ5 Can statistical models of student behavior reliably predict and guide learning progression?

Key findings

• The adaptive system demonstrated a statistically significant correlation between algorithmic behavior and student success in course completion.
• Students using the platform showed measurable, incremental improvement in knowledge over the course of a single semester.
• The system's ability to adjust question difficulty based on performance led to improved learning outcomes compared to static content.
• The platform functioned effectively in both online and offline modes, enhancing accessibility for diverse learners.
• Statistical models of student performance were effective in predicting knowledge gain and guiding content delivery.
• The application's parameters were successfully estimated using real-world testing data from multiple academic courses.