I spent years watching organizations make consequential decisions from data that could not support them. Programs reported participants without agreeing on what a participant was. Retention efforts targeted everyone equally because nothing told them where risk actually sat. Satisfaction surveys came back positive while people quietly left.
The available instruments were built for other settings. Most were normed on traditional undergraduates or on general corporate samples. Applied to a workforce program serving adults balancing jobs, children, and transportation, they asked about the wrong things and left out the ones that decided who finished.
So I built my own. Five instruments now, each grounded in established theory, each written for a population I have actually worked with, and each moving through formal research review rather than being published to a website and called validated.
Formal validation takes longer than self-publication. It is also the only path to a claim that holds up under scrutiny, which is the standard I want these instruments held to.
The five instruments look unrelated until you trace them back to the same question. My dissertation examined how social identity heuristics shape cognitive empathy, meaning how the categories we assign to people change how accurately we read them. That finding runs underneath everything I have built since. Belonging is a question of whether an institution reads you correctly. Leadership quality is a question of whether a manager reads their staff correctly. A merger succeeds or fails on whether two organizations read each other correctly. Different settings, one mechanism.
