Idiographic
Research at the level at which we also work: the patient before us.
The term idiographic is old — it comes from the German philosopher Wilhelm Windelband, who in 1894 placed it opposite nomothetic. But the tension to which it refers is fully alive.
Nomothetic research seeks general laws: what holds for people in general? It is the domain of randomised controlled trials, factor analyses on large samples, population parameters. Idiographic research studies the single case: one person, often measured, over time. It is the domain of case studies, n=1 designs, and intensive longitudinal data.
In mental health care we have, over the past decades, stacked almost exclusively nomothetic research. This has yielded much — we know which treatment forms work on average, for which diagnoses on average, in which trajectories on average. But that average often leaves us alone in the consulting room.
Two traditions, one application
The recognition is growing that we need both. Nomothetic research provides the hypotheses, idiographic research tests them in the context of a specific life. What in group studies is a correlation of .3 between sleep and mood may be, in your consulting room with this patient, a correlation of .8 — or zero, or negative. Only when we measure it with her do we know.
"The idiographic method shows what coheres in this life, at this moment — and how that shifts when something in the system moves."
What does idiographic research look like?
There are various ways to work idiographically. The most accessible is ESM (Experience Sampling Method): several short questionnaires per day, for weeks or months. With a few hundred measurements one obtains a time series showing how mood, behaviour, thoughts and context act upon one another in this person.
With such data one can apply techniques that make no sense at the group level: vector autoregression (which variable predicts which?), network analysis within the person, and the search for early warning signals of relapse. For researchers and clinicians who wish to work with these, accessible R-packages now exist — qgraph, mlVAR, esmpack — which make it feasible without one's own statistician.
What it does in the consulting room
The practical gain is manifold. For the clinician, what was previously only intuited becomes visible: yes, this patient feels worse after poor nights — or no, actually, in her life the relation runs the other way. For the patient, the pattern of her own life becomes legible. And for treatment, it means we can intervene on what truly coheres, not on what coheres on average.
The idiographic turn is not a flight from science, but a correction. It is science at the level at which we also act.