Together with our Faculty of Psychology we are currently preparing a lab study on experience sampling. Experience sampling is an method for collecting contextual data and has been heavily used for motivation research, health research, as well as in usability research for the past 30 years. This method is a way to get information about situated experiences and behaviour. For a method that is frequently used we expected to find tools that can be used with mobile devices. In this posting I briefly review a selection of ESM tools for mobile devices. 

The core principle of experience sampling is to ask the participants of a study to frequently repeat a very small survey. Typically, such a survey contains a few questions on the research topic and a few questions about the context of the participant. The underlying principle of repeated questioning and contextualising the answers makes experience sampling highly suitable for research on contextual and mobile learning. It had been already proven a reliable instrument in studies on ubiquitous computing and on media consumption. 

I have been looking around to see what mobile data enquiry tools are available for ESM and what is missing. My search showed that some work has done in this area for many years. 

The first hit brought me to the project homeage of CAES, which stands of Context-aware Experience Sampling. This project developed a small application for  for Windows based PDA. The original software of CAES is no longer maintained and has been integrated into the myexperience system. The CAES project is closely related to the National Experience Sampling Project and smart building projects at the MIT.

myexperinece is an open source project that has been initiated by Intel Research. The project develops a tool for mobile data collection under the BSD license. Just like CAES, myexperience is build for the windows platform and it heavily relies on the system internals of windows-based mobile systems. The interesting part about this project is that it has developed a XML format for scripting experience sampling studies based on sensor inputs.

A bit older is the Experience Sampling Programme (ESP) that runs on the Palm OS platform. This system is less flexible because it requires ESP studies to be scripted as part of the application that will run on the mobile device.

Far more recently the MobXamp application has been developed for the Apple Iphone. This close-source application has adapted more or less the same interaction principles of ESP to the Iphone. The project homepage is very firm on the technology but it seem that MobXamp is server driven and can assign studies to different user groups.

For Android driven smart-phones there is a tool called Affect Sampler. This tool is mainly designed for self-tracking and is available under the Apache License. There is not much information about this software. This software comes with a server component, which is written in as a set of python scripts for the Google App Engine.

All these solutions are tailored for a specific platform, but they show that there are a bunch of open source solutions to start analysing the requirements for our own project.