Related Experiment Video
Updated: Sep 19, 2025
05:59
Visualization of Intensity Levels to Reduce the Gap Between Self-Reported and Directly Measured Physical Activity
Published on: March 7, 2019
6.8K
Does feedback from physical activity measurement devices influence physical activity? Evidence from a randomized
Arie Kapteyn1, Htay-Wah Saw1,2, Jill Darling1
1Center for Economic and Social Research, University of Southern California, USA.
Survey Research Methods
|June 16, 2025
Summary
Wearable devices providing physical activity (PA) feedback increase user PA by 7%. This feedback is useful for interventions but may bias data collection in research studies.
Area of Science:
- Biomedical Engineering
- Human Movement Science
- Digital Health
Background:
- Wearable electronic devices are widely used for collecting physical activity (PA) data.
- Most wearables offer PA feedback, which can influence user behavior and measurement accuracy.
- Limited research has isolated the causal effect of PA feedback on measured activity levels.
Purpose of the Study:
- To analyze the causal effect of PA feedback from wearable devices on user behavior.
- To differentiate the impact of feedback on PA measurement versus PA promotion.
- To provide guidance on selecting appropriate wearable devices for different research objectives.
Main Methods:
- Conducted 4 field experiments over 8 months using probability-based internet panel recruitment.
- Participants wore PA devices with and without feedback for 7-day periods.
- Collected daily online survey data on participant experiences.
Main Results:
- Participants accumulated 7% more PA when using devices with feedback compared to those without (p<0.001).
- The feedback effect remained significant after accounting for other influencing factors.
- 81 out of 120 eligible participants provided complete data.
Conclusions:
- Wearable PA devices without feedback are suitable for population data collection requiring minimal measurement error.
- Wearables with PA feedback are effective for promoting physical activity in intervention studies.
- Awareness of potential measurement bias is crucial when using feedback-enabled wearables.
