Magazine "Research" · Reading time approx. 9 min
Two professionals, the same family, two different genograms. What reliability and validity research says about it — and why that's not an argument against the genogram.
A Simple Question with an Uncomfortable Answer
Imagine two counselors independently conduct a genogram interview with the same family. How similar are the two results?
The question sounds academic. It isn't. If two professionals arrive at different pictures, the genogram isn't just capturing the family — it's also capturing the person who drew it.
The older research on this is sobering. In classic studies, agreement between different interviewers fell below the threshold that test theory considers usable. On validity — whether the genogram measures what it claims to measure — even less evidence exists.
What Actually Diverges
The differences don't arise from the hard facts. Birth years and numbers of siblings are usually undisputed.
They arise from everything else. Is this relationship "conflictual" or "distant"? Do you draw a wavy line or a dashed one? Does the deceased aunt belong in the picture or not? Do you record the miscarriage the client mentions in passing?
Every practitioner makes these calls differently. They depend on training, on one's own attentiveness, sometimes on time pressure on a given day.
There's a second source on top of that: the family itself doesn't tell the same story in two different conversations. Someone who stays silent about a brother in the first session might mention him in the second. That's not a flaw in the tool — it's a property of conversations.
The Attempt to Make It Measurable
In 2013, Curtis and colleagues presented a structured interview protocol in the Family Journal, the Family Genogram Interview. The idea: if the questions are fixed, the room for variation shrinks, and agreement rises.
It works. With a fixed protocol, the authors achieved markedly higher agreement scores than the classic, unstructured approach.
On construct validity — whether what's captured actually reflects what's intended — the results were mixed. Structure alone doesn't solve the problem.
Puhlman (2023) takes a different approach in the Journal of Family Theory & Review, treating the genogram not as a measurement instrument but as a qualitative analysis method with its own quality criteria. That's a fundamentally different angle — and arguably the more honest one for practice.
Measurement Instrument or Structuring Tool?
This is the crux. The methodological critique only lands if you understand the genogram as a measurement instrument.
A measurement instrument is supposed to produce the same values for the same object, regardless of who's measuring. A blood pressure monitor that shows different readings for two different doctors is broken.
A structuring tool serves a different purpose. It organizes a conversation, makes connections visible, and records what was discussed. That two professionals emphasize different things is not a defect — it's the nature of the thing.
Rempel, Neufeld, and Eastlick Kushner described in 2007, in the Journal of Family Nursing, how genograms and ecomaps can be used in research: not as measurement, but as an interactive elicitation method in which the act of creating them together is itself part of the data collection.
Hypothetical case example: At a clinic, a colleague takes over a case. During the handover, it becomes clear that her genogram looks different from her predecessor's: she has drawn one relationship line as conflictual, where her predecessor drew it as distant. Instead of settling who's right, the two of them talk through the difference. It turns out the family had described a conflict in the first session and a sense of silence between them in the second. Both genograms are correct — they capture two different moments.
What Follows for Practice
Date the genogram. A picture from March is not a picture from October. Noting the date prevents a snapshot from being read as a fixed fact.
Note the source. Who provided which piece of information? What the mother says is not the same as what's written in a file.
Treat relationship lines as hypotheses. Symbols for closeness and conflict are interpretations. They belong in the picture, but with the awareness that another practitioner might draw them differently.
Use a protocol for research. Anyone analyzing genograms systematically needs fixed questions and rules — otherwise they end up comparing interviewing styles instead of families.
Why This Critique Actually Helps the Tool
A tool whose limits you know gets used more safely. Practitioners who know that their genogram also reflects their own attention are more likely to ask themselves: Did I hear that, or did I interpret it? Did that come up in the conversation, or did I add it?
That question is uncomfortable and productive at the same time. It's what separates routine use from reflective use.
None of this damages the genogram — it sharpens what it actually is. It's not an instrument that measures families. It's a jointly created map — and maps are useful precisely because they are not the territory.
Sources
- Curtis, A. et al. (2013): The Family Genogram Interview — Reliability and Validity of a New Interview Protocol. The Family Journal.
- Puhlman, D. (2023): Qualitative genogram analysis — A methodology for theorizing family dynamics. Journal of Family Theory & Review.
- Rempel, G., Neufeld, A. & Eastlick Kushner, K. (2007): Interactive Use of Genograms and Ecomaps in Family Caregiving Research. Journal of Family Nursing.
- Joseph, Dickenson, McCall & Roga (2023): Exploring the Therapeutic Effectiveness of Genograms in Family Therapy: A Literature Review. The Family Journal.
- McGoldrick, M., Gerson, R. & Petry, S. (2020): Genograms — Assessment and Intervention. 4. Auflage, W. W. Norton.