Tuesday, 11 July 2017

Experience sampling within psychology


Experience sampling allows for “real time,” in situ assessments of behaviour temporally close to the moment of enactment. Early attempts involved participants carrying specific devices, which were expensive and bulky, but the rise of smartphones means that this method can be deployed across a variety of research designs. For example, text messages can easily be sent using a specific account or a third-party automated system (e.g. here). 

More complex designs can also combine real-world data from smartphone sensors. This might include location via GPS, or health related data in the form of movement or heart rateExperience sampling can also can help reduce the temptation to provide social desirable responses. Most smartphones come equipped with a camera, and mobile phone apps allow participants to upload photos as supplemental data.


The changing face of experience sampling - from software running on expensive Personal Digital Assistants to smartphone applications that can be deployed to millions of smartphones.


However, this method doesn't feature as widely within psychology as you might expect and adds to my own frustrations on how slow social psychology is to adopt any new methodology. 
That said, it is unfair to level this criticism at the discipline in isolation. I also get the feeling that those who develop or propose new methods struggle to explain or convince others to adopt them. Many advances initially appear with great fanfare, but it often take several years until they become demystified/adequately explained to those who are most likely to use them. 

On a more positive note, there are moves to change this situation particularly when it comes to experience sampling with a number of groups developing open-source frameworks (e.g. http://www.experiencesampler.com/). On paper at least, these appear equally just as impressive as any expensive commercial solution.  


Self-reported anxiety levels measured twice over a 10-day period (N=51). This simple data-set reveals how even a simple measure of mood can vary considerably over several days.

Of course, experience sampling with smartphones has some limitations. Only participants with smartphones can participate (though most people today own such a device), and participants can still be strategic about what they report. Yet, experiencing sampling with smartphones will almost always improve the direction of travel when it comes to increasing the quality and validity, particularly when combining real-world behavioural markers (or digital traces) and self-report. 

Further Reading

Ellis, D. A. and Piwek, L. (2016). The future of... wearable technology. CREST Security Review, 1, 4-5. 

Piwek, L., Ellis, D. A. and Andrews, S. (2016). Can programming frameworks bring smartphones into the mainstream of psychological science? Frontiers in Psychology. 7:1252

Thai, S. and Page-Gould, E. (2017). ExperienceSampler: An Open-Source Scaffold for Building Smartphone Apps for Experience Sampling. Psychological Methods. Advance online publication. http://dx.doi.org/10.1037/met0000151