Our platform allows researchers to send custom push notifications to their participants. These notifications can be based on time, date and/or GPS location. One of the most popular and robust of these notifications is the Random Trigger. This part of the platform allows a researcher to control the randomization of when their notifications are sent out.
The paradox of this is that a true random algorithm cannot be mathematically defined, but in order to program randomness into the platform, it needs to be mathematically defined. What we have developed is a way to manipulate randomness by adding a certain level of control.
An example of this would be a researcher who wants to ask a participant a set of questions at some point between noon 9am and 5pm, like in the picture above. Now it is rather uncomplicated to build an algorithm that sends a single notification between these ranges. However, it becomes increasingly hard to maintain the random aspect when a researcher wants to set multiple random notifications with a minimal interval of time between prompts, something our platform allows for. In this example, the randomness is being manipulated. The researcher wants to control a buffer of time where a prompt goes off, but within this time block, it must fire at random.
In this case, the researcher is looking for a minimum of 35 minutes to pass before their participants receive the next notification, with 12 notifications in total to go off throughout the day. This ensures a random distribution of triggers so that participants do not have the possibility of getting all 12 notifications at once, or within short proximity.
Random triggers are very popular within the clinical psychology community and often used with the Experience Sampling Methodology (ESM), a technique that polls participants randomly and in-the-moment. Prior to this technology, researchers would program stop-watches to beep at random times, and participants would have to write down their experience manually. In other scenarios, researchers hacked together old Palm Pilots and used open-source, non-secure programs, which seem ancient in comparison to the abilities of smartphones.
The ability to manipulate the randomness of prompts is a great tool to have in your research arsenal. It’s a flexible and powerful tool to poll participants at the moment.