Feature Focus : Geolocation in Trials
O Participant, Where Art Thou?
If you've ever uploaded a picture or video to social media and tried to tag a location, you will likely be familiar with geolocation data. Along with your data of interest - in this case a photo or video - the file itself will hold additional metadata which is captured by the device that you are using. This will include things like the device type itself (sometimes the operating system, too), the camera specification, location the file was created at and more.
This wealth of metadata can provide extra meaning and support to key data which might otherwise be overlooked. For example, if your study was only relevant for a particular location and you had specified this in your inclusion criteria, you might receive clues in geolocation data which suggest that participants might be in breach of the trial protocol or that information they provided was incorrect. In this case, these participants might be considered outliers and could be removed from statistical analysis.
Sub-grouping is also an important consideration in post-trial analysis. You could combine data with geolocation data to empower your outcomes. Are you perhaps looking at wellbeing and depression outcomes? Maybe participants scoring low on your scales are in areas where weather or community support is bad and crime and/or poverty might is high. Are your active and placebo groups accidentally very geographically disparate? Might this have an effect on the overall data? Using metadata in this way, you can better inform your study outcomes and bring to light important conversations which could help empower your future studies and study design. Is your treatment or intervention having a legitimate effect? or is it the result of otherwise unforeseen confounding factors? The more data at-hand at the table, the better.
While geolocation data is useful, some participants might be connected to alternative networks including virtual private networks, (VPNs). For this reason, on-the-spot data might not be as accurate, however, general overviews of a location (excluding the use of VPN) can be informative.
...What would you find geolocation data useful for in a research study or trial...?