Do Stress Patterns During Exams Reveal Hidden Insights into Student Performance?
"Imagine you're about to take a crucial exam or compete in a high-stakes game. Your heart races,
your palms sweat—stress is kicking in. But is this a bad thing?"
Stress significantly impacts student performance during exams.
This dynamic visualization explores how physiological stress markers, such as body movement, sweating, heart rate, and skin temperature, correlate with exam grades, both individually and in combination.
The dataset includes exam data from 10 students across three exams, with stress markers tracked using a smartwatch-like device worn on their non-dominant hand throughout the exam period.
Features / Markers: What Are These Mystery Abbreviations?
Euclidean Distance: 0.00
Acceleration (ACC)
Acceleration (ACC) in meters per second squared measures (m/s^2) body movement, where physical activity helps alleviate stress and muscle tension.
Electrodermal Activity (EDA)
Electrodermal Activity (EDA) in microsiemens (μS) measures skin electrical conductance, which increases due to stress-stimulated sweat gland activity.
BPM: 75
Heart Rate (HR)
Adjust the BPM slider to listen to the heart beat! (Scroll to stop)
Heart rate (HR) measured in beats per minute (bpm) increases as stress triggers the fight-or-flight response, releasing adrenaline.
Temperature (TEMP)
Temperature (TEMP) in Celsius (℃) measures wrist skin temperature, which increases due to stress-induced metabolic heat output.
Section 1: How Do Individual Physiological Factors Relate to Exam Performance?
"Many believe that stress is harmful, but in reality, moderate stress can enhance performance.
Elite athletes and top students often experience increased stress levels when they perform at their best."
Select different markers to compare their patterns across students and their Midterm 2 grades;
🔍 hover to see exact average values for comparison.
Although exam durations were fixed, physiological data varied in length, often extending beyond the exam period.
To standardize this, all start times were aligned, with a 1-hour buffer before and after each exam.
Since markers have different sampling rates, data was aggregated by exam time, with each point representing the average marker value per minute.
Section 2: How Do Physiological Factors Fluctuate Throughout Different Exams?
"However, not all stress is good. When stress levels spike unpredictably—like a student panicking mid-exam—performance declines.
It's not just stress itself, but how stable your stress levels are that matters."
Select a student and a marker to track changes over time during Midterm 1, Midterm 2, and the Final Exam;
🔍 hover to see exact average values for comparison
Although exam durations were fixed, physiological data varied in length, often extending beyond the exam period.
To standardize this, all start times were aligned, with a 1-hour buffer before and after each exam.
Since markers have different sampling rates, data was aggregated by exam time, with each point representing the average marker value per minute.
Section 3: How Do Multiple Factors Combine to Predict Exam Grades?
"Now that we've analyzed how stress impacts performance, what about your own stress levels?
If we could predict how fluctuations in stress affect your performance, we could take steps to manage it better."
"Our model doesn’t just show whether stress is good or bad—it helps identify when stress stability breaks down.
This means that by learning to regulate stress levels, you can optimize your performance instead of letting stress work against you."
This section explores how physiological markers interact to predict a student's exam grade.
Predictions are generated by a linear regression model trained on an aggregated dataset,
where each data point represents the mean value of each physiological marker per student per exam.
The model uses four physiological markers as features, with exam grade as the label.
Use the sliders below to adjust standard deviations for each feature and see how changes in stress markers impact grade predictions.
0000
Predicted Grade (%):
--
The Parallel Coordinates Plot below visualizes how physiological features interact.
Each vertical axis represents a marker (ACC, EDA, HR, TEMP) or the predicted grade, with each line corresponding to a student's data.
Color-coded lines indicate grade groups with Green representing predicted grade line:
Dark Green: High grades(Q1);
Light Green: Medium grades(Q2);
Light Pink: Low grades(Q3);
Dark Pink: Very low grades(Q4);
Follow a single line from left to right to see how a student's physiological responses influence their predicted grade.
Higher ACC, HR, and TEMP values tend to align with better performance, while elevated EDA often correlates with lower scores.
🎯 Key Takeaway 🎯 Moderate stress can boost academic performance! Slight and stable increases in heart rate ❤️, skin temperature 🌡️, and body movement ✍️ correlate with higher exam
scores 📈, while increased sweating 💦 correlates with lower scores 📉,
highlighting that some stress is not just normal, but beneficial for peak performance! — just don’t let it overwhelm you! 🧠✨
⚠️ This prediction is based on physiological data collected from students at the University of Houston.
Results may not generalize to all individuals or settings.
Source: The dataset used in this visualization is based on the work: Amin, M. R., Wickramasuriya, D., & Faghih, R. T. (2022).
A Wearable Exam Stress Dataset for Predicting Cognitive Performance in Real-World Settings (version 1.0.0).
PhysioNet. https://doi.org/10.13026/kvkb-aj90.
The abbreviation icons used on this website are from SVGREPO. The sound effects are from envato. We are very grateful for their free resources!
For more information about the dataset, please visit the PhysioNet website.
