There are several ways to do testing.
For my part, I believe strongly in initial qualitative tests, which may take the form of discussion groups, cognitive tests, or one-on-one tests. There is a great deal of work that can be done at the cognitive level by using representative samples, among other things.
The big content test, in 2019, is also intended to be operational. It will allow us to know if we can implement these questions and apply them to the entire population.
There is therefore an iterative process, and we are putting time into it. We will perhaps do a second quantitative test; that is what we are examining. It is all woven into the schedule up to now.