MAESTRO

Abstract

In this paper we present "MAESTRO" a novel concept to study behavior in realistic environments based on the experience sampling method (ESM). It constitutes a variation of ESM by giving researchers the possibility to dynamically and remotely "orchestrate" experience-sampling studies for evaluating usage behavior in various different contexts. MAESTRO allows user behavior driven and context triggered experience sampling by using a client-server architecture. So far a variation of different computerized ESM tools have been introduced capable of delivering questionnaires triggered by contextual information on mobile devices. We propose to enhance ESM by not only using context and user actions as a trigger for ESM questions, but also by exploiting long term user behavior and usage patterns for shaping personalized ESM questions to different types of users. Our concept lays the foundation of measuring user experience not only in a mobile but in a magnitude of different contexts. As a proof of concept we present an implementation of our approach on BlackBerry devices along with initial results of a case study validating this approach.