If the dialogue after the course is important, why cut it of - that way, you might miss the most important effect of your intervention(e.g.: which courses encourage participants to engage socially, which do the opposite). I would maybe standardize the time for social interaction after the class and ask the teachers to document the experience (if need be, also in a standardised way).  (not an expert, but I have a little experience with intervention research)

thanks for the answer, Richard :)

if we start with e.g. 4 intervention groups, and randomly, in 2 the social engagement would become high, and in 2 the group would remain unorganized, uninterested emotionally, than it is a scientific challenge to control these effects (that is why we first had decided to prevent the instructor to initiate dialogues).

if anyone has concrete ideas how to control this effect (measure the experienced social support in the group), share us please :)

Hi Benedek,

I would be very careful with a narrow notion of "measuring" - especcially if you have a small number of comparable  interventions  The very core of intervention research is to leave controlled lab conditions and to understand effects in real-life conditions. Therefore, you will get complex interactions that defy any attempt to controll or measure with any precision - often, you can miss the point completely. On the other hand, you can understand very much abot the mechanisms of your intervention (ant therefroe about their practical potential and shortcomings) by gathering and analysing  a broad spectrum of related data (interviews with single participants, focus group interviews, also some mearurements). I like using "constellation aanalysis" as a tool to map out the notions different people connect to an issue/an intervention , but ohter methods might work just as well, as long as they help you understand, rather than measure, the effects.

good luck with your interventions,

Richard