Ethnographic Decision Modeling
Also known as: Ethnographic Decision Tree Modeling, EDTM, Decision Tree Ethnography, Ethnographic Decision Models
Ethnographic decision tree modeling is a method for building a formal, qualitative model of how people actually make a specific recurring decision — such as whether to plant a crop, seek treatment, or adopt a practice. Developed by Christina Gladwin and set out in her 1989 Sage monograph, it elicits the criteria and rules people use through ethnographic interviews, represents them as an if-then decision tree, and then tests the tree's ability to predict the choices of a fresh sample of decision-makers.
Key highlights
- Yields a testable, predictive model rather than a purely descriptive account of choice.
- Grounds the model in informants' own decision criteria, preserving cultural specificity.
- The explicit predictive-accuracy standard provides a clear, falsifiable validation criterion.
- Produces transparent, interpretable if-then rules that practitioners and informants can scrutinize.
Intuition
This section is available to Pro members. Upgrade to Pro
How it works
This section is available to Pro members. Upgrade to Pro
When to use it
Use ethnographic decision modeling when you want to understand and predict how members of a group make a specific, recurring, real-world choice, and you can interview people who make it. It suits applied questions in economic, agricultural, medical, and cognitive anthropology where the goal is both insight and prediction. It assumes the decision is discrete, repeated, and made on identifiable criteria. It is less appropriate for one-off or highly idiosyncratic decisions, for choices driven by factors informants cannot articulate, or when no behavioural data exist to test the model's predictions.
Strengths & limitations
- Yields a testable, predictive model rather than a purely descriptive account of choice.
- Grounds the model in informants' own decision criteria, preserving cultural specificity.
- The explicit predictive-accuracy standard provides a clear, falsifiable validation criterion.
- Produces transparent, interpretable if-then rules that practitioners and informants can scrutinize.
- Best suited to a single, discrete, recurring decision; complex or one-off choices fit poorly.
- Building and validating the tree is labor-intensive, requiring two or more interview samples.
- Assumes decisions follow articulable, largely sequential rules, which may oversimplify reasoning.
- Predictive success in one community does not guarantee transfer to a different cultural context.
Common pitfalls
This section is available to Pro members. Upgrade to Pro
Applications
This section is available to Pro members. Upgrade to Pro
Frequently asked
How is an ethnographic decision tree different from a statistical decision tree?
A statistical decision tree (as in machine learning) is induced automatically by an algorithm that splits data to maximize a numerical criterion. An ethnographic decision tree is built by hand from what informants say their decision rules are, in their own terms and order, and is then tested for predictive accuracy. The former optimizes fit to data; the latter models culturally real reasoning and validates it against behaviour.
What counts as a successful model?
The defining standard is out-of-sample prediction: the tree, built from one group of informants, should correctly predict a high proportion of the actual decisions made by a separate group. Gladwin's benchmark is typically around 85–90 percent or higher. Meeting that bar on fresh cases is what distinguishes a validated decision model from a mere restatement of the original interviews.
Why must criteria be ordered in the tree?
Because people apply considerations sequentially, and some are decisive constraints that short-circuit the rest. If a hard constraint such as affordability fails, the decision ends regardless of other preferences, so it must sit near the top of the tree. Getting the order right is essential: the same criteria in the wrong sequence will mispredict choices even when all the relevant factors are present.
Sources
- 1.Gladwin, C. H. (1989). Ethnographic Decision Tree Modeling. Qualitative Research Methods Series 19. Newbury Park, CA: Sage.ISBN 9780803934870
You have read it. What now?
Cite this page
ScholarGate. (2026, June 22). Ethnographic Decision Modeling. ScholarGate. https://scholargate.app/anthropology/ethnographic-decision-modeling