This is going to be a very brief post, given my current time constraints.
We are setting up a large Ecological Momentary Assessment (EMA) study. I have some experience with EMA from running studies myself (e.g. on student mental health during COVID19) and collaborations, but the field is growing rapidly, and so we spent some time trying to obtain information on how to best move forward for a large-scale project.
Questions we have include 1) how to get a good overview of all the EMA services out there, 2) and whether we should opt for app-based vs web-based services. App-based means that participants install an app on their phone; web-based means that participants get an SMS or whatsapp message with a link, and they use their browser to complete the measurement.
These questions come up regularly and I have yet to find a centralized repository of information, so I wanted to collect some information here for future use. This is very much quick and dirty information, so if you know of things I can add here, please let me know (email@example.com).
First, Ruben Arslan has started a google document that you can contribute to. It provides an overview of EMA software. Since calls on Twitter for such comparisons have often resulted in app developers to respond, be mindful that it is not unthinkable that they may contribute to such information collection efforts, sometimes with financial interests in mind. So do verify before you make final decisions.
Second, I started a discussion for the question whether to use web-based vs app-based EMA. There were a lot of insightful responses I collect here. These are largely copy and pasted from many tweeps who contributed to this thread, not my own ideas — full credit to them. You can find the original authors in the link above. I edited some responses to make them easier to understand for non experts / spell out abbreviations.
advantages as +, disadvantages as –
- + requires no internet
- + pseudonymity easier to maintain (no phone number required for reminders)
- + possibility to link with smart watches and other wearable devices for e.g. digital phenotyping
- + Can be rebranded more easily (different app logo, different app name) for the purpose of a study
- + easier to guarantee notifications on time (no delay vs potential delay in e.g. SMS)
- + can potentially have specific triggers for starting survey (e.g. on phone unlock), which could increase compliance rates
- + integration with other phone apps (e.g. calendar)
- + when you are trying to identify ‘careless EMA responses’, the time taken to complete an item can be very useful and is somewhat easier to record in an app
- + lower attrition (one step less from reminder to filling in) 1
- – compatibility (e.g. >500 Android versions)
- – requires more maintenance & tech support
- – possibly harder to get through IRB (because it requires software installation on participants’ phones)
- – battery usage may be higher
- +/- much easier access to other device data (e.g., microphone, app use, location, accelerometer)
- + much easier to use across platforms, easier to maintain
- + user interface usually massively better
- + greater flexibility in survey design
- + easier to set up
- + easier to fill in using different devices
- + users rarely disable texts but can disable app notifications
- – often unable to link/port data across surveys
- – SMS signals can be blocked by carriers w/o any notification