The Downside of Perseverance--Investigating and Moving Students Beyond Unproductive Persistence
Effective Years: 2015-2020
The project researches persistence in mathematics learning in a computer-based learning environment (CBLE). The research investigates how a CBLE can provide the supportive help and promote the self-regulatory strategies necessary for students to be not just persistent, but productively persistent math learners. The project will focus on the middle school years, an important and vulnerable point in the school trajectory, as mathematical concepts become increasingly difficult and abstract in the transition from arithmetic to algebra. Persistence is critical in CBLEs. Research shows has shown that changing student academic mindsets can have strong benefits for their persistence and achievement. Sustained effort is important but not always sufficient for learning and less attention has been paid to the downside of perseverance, called wheel-spinning, the time that struggling learners spend without making progress. The project will build instrumentation, theory, and intervention strategies to address the needs of struggling math students. For technology designers, the project will distill findings to develop research-based design principles for CBLEs. For educators, the project will translate findings into pedagogical approaches that can be used to enhance teacher practice. The research is designed to address three goals: (1) Develop automated detectors that can differentiate between wheel-spinning and productive persistence in real time; (2) Investigate student, teacher, and system factors that predict wheel-spinning and productive persistence; and (3) To design and test interventions to reduce wheel-spinning and promote productive persistence. The project will use detailed log data from thousands students who use an existing computer-based mathematics program. Learning analytic methods now provide the means to identify and study micro-level student behavioral engagement patterns. The project will use complementary data analytic, correlational, think-aloud, and experimental methods to examine productive persistence and its understudied yet common counterpart, unproductive persistence. The research will advance the state of the art in data mining-based measurement to provide instrumentation that can be used to enhance learning; extend empirical findings about the cognitive and self-regulatory processes that enable productive persistence in math learning; and extend the empirical base informing what supports in CBLEs or teacher practice are necessary for productive persistence. The project, supported through the EHR Core Research (ECR) program of fundamental research in STEM, will contribute important research findings regarding STEM learning and learning environments, which are important priorities of the ECR program.
50 Publications Related to This Project:
- Leveraging Auxiliary Data from Similar Problems to Improve Automatic Open Response Scoring
- Student Perception on the Effectiveness of on-Demand Assistance in Online Learning Platforms
- Exploring Common Trends in Online Educational Experiments
- Improving Automated Assessment and Feedback for Student Open-Responses in Mathematics
- Improved Automated Essay Scoring Using Gaussian Multi-Class Smote for Dataset Sampling
- Identifying Explanations Within Student-Tutor Chat Logs
- Automated Assessment in Math Education: A Comparative Analysis of LLMS for Open-Ended Responses
- Effect of Gamification on Gamers: Evaluating Interventions for Students Who Game the System
- Immediate Versus Delayed Feedback on Learning: Do People's Instincts Really Conflict with Reality?
- Precise Unbiased Estimation in Randomized Experiments Using Auxiliary Observational Data
- Contextual Factors Affecting Hint Utility
- Save Your Strokes: Chinese Handwriting Practice Makes for Ineffective Use of Instructional Time in Second Language Classrooms
- Facilitating Student Learning with a Chatbot in an Online Math Learning Platform
- Expert Features for a Student Support Recommendation Contextual Bandit Algorithm
- Multiple Choice vs. Fill-in Problems: The Trade-Off Between Scalability and Learning
- The Effect of Assistance on Gamers: Assessing the Impact of on-Demand Hints & Feedback Availability on Learning for Students Who Game the System
- Impact of Non-Cognitive Interventions on Student Learning Behaviors and Outcomes: An Analysis of Seven Large-Scale Experimental Inventions
- How to Open Science: Analyzing the Open Science Statement Compliance of the Learning @ Scale Conference
- How Common are Common Wrong Answers? Crowdsourcing Remediation at Scale
- Investigating the Impact of Skill-Related Videos on Online Learning
- A Bandit You Can Trust
- Automatic Interpretable Personalized Learning
- Examining Student Effort on Help Through Response Time Decomposition
- Using Past Data to Warm Start Active Machine Learning: Does Context Matter?
- Toward Personalizing Students' Education with Crowdsourced Tutoring
- Effectiveness of Crowd-Sourcing on-Demand Assistance from Teachers in Online Learning Platforms
- The Automated Grading of Student Open Responses in Mathematics
- Refusing to Try: Characterizing Early Stopout on Student Assignments
- Observing Personalizations in Learning: Identifying Heterogeneous Treatment Effects Using Causal Trees
- The Assessment of Learning Infrastructure (Ali): The Theory, Practice, and Scalability of Automated Assessment
- Assistments Dataset from Multiple Randomized Controlled Experiments
- Studying Learning at Scale with the Assistments Testbed
- Leveraging Natural Language Processing to Support Automated Assessment and Feedback for Student Open Responses in Mathematics
- Developing Early Detectors of Student Attrition and Wheel Spinning Using Deep Learning
- Forecasting Future Student Mastery
- Immediate Text-Based Feedback Timing on Foreign Language Online Assignments: How Immediate Should Immediate Feedback Be?
- The Future of Adaptive Learning: Does the Crowd Hold the Key?
- Testing the Validity and Reliability of Intrinsic Motivation Inventory Subscales Within Assistments
- Improving Sensor-Free Affect Detection Using Deep Learning
- Variations of Gaming Behaviors Across Populations of Students and Across Learning Environments
- The Effectiveness of AI Generated, on-Demand Assistance Within Online Learning Platforms
- Comparing Different Approaches To Generating Mathematics Explanations Using Large Language Models
- Toward Improving Effectiveness of Crowdsourced, on-Demand Assistance from Educators in Online Learning Platforms
- Exploring Fairness in Automated Grading and Feedback Generation of Open-Response Math Problems
- Deep Learning or Deep Ignorance? Comparing Untrained Recurrent Models in Educational Contexts
- Identifying Struggling Students by Comparing Online Tutor Clickstreams
- Effect of Immediate Feedback on Math Achievement at the High School Level
- Early Detection of Wheel-Spinning in Assistments
- Supporting Teacher Assessment in Chinese Language Learning Using Textual and Tonal Features
- Generalizability of Methods for Imputing Mathematical Skills Needed to Solve Problems from Texts
You are viewing a record from the ECR Projects database. Click here to search the ECR Projects Database or the ECR Publications Database.


