Learnability of a Configurator Empowering End Users to Create Mobile Data Collection Instruments: Usability Study

Area of Science:

• Computer Science
• Human-Computer Interaction
• Software Engineering

Background:

• Traditional paper-based data collection in research has significant drawbacks.
• The QuestionSys framework enables users without programming skills to create custom mobile data collection apps.

Purpose of the Study:

• To validate the feasibility of a model-driven, end-user programming approach for mobile data collection app development.
• To assess the learning effects and performance of users with varying levels of expertise.

Main Methods:

• A study involving 80 participants who modeled 10 data collection instruments using the QuestionSys configurator over two sessions.
• Performance metrics including time, operations, and errors were evaluated.
• Participants were categorized as novices or experts based on prior process modeling knowledge.

Main Results:

• Novices demonstrated significant learning improvements in errors, operations, and time.
• Experts showed learning in operations and time, with minimal initial errors.
• Novices' performance in time and operations surpassed experts' initial performance, and their error rates became comparable to experts' initial rates.

Conclusions:

• Novices effectively utilized the QuestionSys configurator, exhibiting fast but initially unstable learning.
• Novices' performance, particularly in time and operations, reached expert levels with limited experience.
• The QuestionSys framework empowers researchers and clinicians to independently develop mobile data collection applications.