I think the first question is to define the goal of the therapy that you are trying to measure - e.g. is the therapy treating specific mental disorders or aimed at improving general mental health and wellbeing. There was so many different ways you could measure dependent on this

- child psychiatric scales

- child well-being scales

- parental report

- school report

- behavioural observation either global or against specific measures e.g. school refusal, aggression, self-harm etc

- self report against problems defined by the child (especially over 8s)

- eating, sleeping etc

- child's experience of the therapy

Finding measures is rarely the problem, but getting a tight definition of effectiveness and a good methodology is fundamental as it presenting a convincing link between the therapeutic intervention and the outcome rather than just correlational evidence (all engagement with any therapy can be a marker for improvement, so you might not show any specific effect from art therapy per se).

thank you for your response. I am looking for a way to look at school-based therapy as a pre-post intervention, with different kind of patients. therefore I'm looking for past research who might have done the same to see what kind of tools they might have used to examine the effectiveness of therapy.

You could consider using the Young Person CORE measure if the child is 11 years upwards - the CORE series of measures were developed to measure outcomes in psychotherapy