I think behavioural research like this has value. But, once we're speaking in terms of nudges & clustering... I feel the terms are ambiguous for scientific lang

I think behavioural research like this has value. But, once we're speaking in terms of nudges & clustering... I feel the terms are ambiguous for scientific language. That is... This probably is useful to the gambling regulator, addiction counselors and such, especially if similar results are found in similar studies. OTOH, they may not hold true in a different country, or 7 years from now, etc. Conversely... I wouldn't be surprised to find many gambling ads surrounding loot box videos on youtube. The orthodox scientific approach is to seek generalities. Any clustering of loot box consumption & problem gambling exists on the plane of generalities that the UKGC or addiction charity operate in. It's an awkward place for science though. Academic pathologies notwithstanding, replication means something different in behavioural sciences... Either that or behavioural science has to be confined to a fairly small realm. Ultimately, what facebook do with their optimization algorithms (or any startup a/b testing) is similar to what gambleaware do. Companies have less need for formal publishing than NGOs, GOs and such. They need a way of informing decisions that convinces the boss, not the world. Not sure where I'm going with this.