It would be appropriate to assume the volume of younger users (with potentially more apps) outweigh older users. Here's what we've found:
The 1,500 apps that gather data span all types of interest categories to ensure there are broad age, demographic, and behavioral characteristics encompassed in the data set. Additionally, we count the number of unique devices in a data set, so a heavier user of apps (like a younger adult) isn't over-counted relative to older users. For these reasons, throughout our analysis and deep work with this data, we haven't seen any bias toward a single user group.
At the same time, we are always looking at ways to make our data more insightful and we've just licensed an expanded set of metadata around the data we receive. For a subset of users, this will include age ranges to allow us to include this information in our evaluation and normalization processes.
Need more help? Talk to a real human.