Ecological Fallacy

Robinson's classic analysis

Thorndike (1939) warned of it first, but Robinson (1950) systematized the fallacy. Using 1930 US census data, he computed the correlation between the share of foreign-born residents and the illiteracy rate across states. At the state level the correlation was about -0.53, so states with more immigrants had lower illiteracy. At the individual level, however, the foreign-born were slightly more likely to be illiterate, with a correlation of about 0.12. The difference arose because immigrants tended to settle in northern industrial states with high levels of education. Robinson criticized the common practice in sociology at the time of using group correlations as stand-ins for individual behavior, and the paper fundamentally changed how researchers think about units of analysis. It is closely related to Simpson's paradox, in which relationships reverse depending on the level of aggregation.

The Durkheim debate and ecological inference

Selvin (1958) examined Durkheim's (1897) Suicide as a case of the ecological fallacy, since it inferred the suicidal tendencies of individual Protestants from regional data showing higher suicide rates in areas with more Protestants. Alker (1969) built a typology of the fallacies that can arise between groups and individuals, or between groups and subgroups, and Hannan (1971) set out the problems of aggregation and disaggregation statistically. Solutions were also proposed. Goodman (1953) introduced ecological regression, which estimates individual-level rates from group data under certain assumptions, and Duncan and Davis (1953) offered a method for computing the bounds of possible values. King (1997) combined these into an ecological inference method that came into wide use in election analysis, but Freedman (1999) pointed out that every such method rests on assumptions that cannot be verified.

Areal unit problems and the opposite fallacy

Openshaw (1984) described the modifiable areal unit problem, in which correlations in the same data can change dramatically depending on how boundaries are drawn and at what scale areas are grouped. Gelman and colleagues (2007) showed that in the United States richer states vote more Democratic, while within each state richer individuals vote more Republican, revealing how easily the fallacy creeps into political analysis. Reanalyzing Robinson's data with multilevel models, Subramanian and colleagues (2009) stressed that the reverse error, the individualistic fallacy of looking only at individual-level relationships and ignoring group-level contextual effects, is just as dangerous. In the end, the key is to be clear about which unit of analysis a research question concerns, and alongside the distinction between correlation and causation, this is considered a basic principle of social science methodology.

FAQ

What is the ecological fallacy in simple terms?

It's the mistake of assuming a relationship seen in statistics for a whole region or country also holds for the individuals in it. Just because wealthier areas back a certain party doesn't mean wealthier people do.

Is there an opposite fallacy?

Yes. Looking only at individual-level relationships while ignoring the influence of group or regional context is called the individualistic fallacy. Both errors come from confusing units of analysis.

Sources reviewed: 2026-10-26

References

1. Durkheim, É. (1897). Le Suicide: Étude de sociologie. Paris: Félix Alcan.
2. Thorndike, E. L. (1939). On the fallacy of imputing the correlations found for groups to the individuals or smaller groups composing them. American Journal of Psychology, 52(1), 122–124.
3. Robinson, W. S. (1950). Ecological correlations and the behavior of individuals. American Sociological Review, 15(3), 351–357.
4. Goodman, L. A. (1953). Ecological regressions and behavior of individuals. American Sociological Review, 18(6), 663–664.
5. Duncan, O. D., & Davis, B. (1953). An alternative to ecological correlation. American Sociological Review, 18(6), 665–666.
6. Selvin, H. C. (1958). Durkheim's Suicide and problems of empirical research. American Journal of Sociology, 63(6), 607–619.
7. Alker, H. R. (1969). A typology of ecological fallacies. In M. Dogan & S. Rokkan (Eds.), Quantitative Ecological Analysis in the Social Sciences (pp. 69–86). Cambridge, MA: MIT Press.
8. Hannan, M. T. (1971). Aggregation and Disaggregation in Sociology. Lexington, MA: Lexington Books.
9. Openshaw, S. (1984). The Modifiable Areal Unit Problem (CATMOG 38). Norwich: Geo Books.
10. King, G. (1997). A Solution to the Ecological Inference Problem: Reconstructing Individual Behavior from Aggregate Data. Princeton: Princeton University Press.
11. Gelman, A., Shor, B., Bafumi, J., & Park, D. (2007). Rich state, poor state, red state, blue state: What's the matter with Connecticut? Quarterly Journal of Political Science, 2(4), 345–367.
12. Subramanian, S. V., Jones, K., Kaddour, A., & Krieger, N. (2009). Revisiting Robinson: The perils of individualistic and ecologic fallacy. International Journal of Epidemiology, 38(2), 342–360.
13. Freedman, D. A. (1999). Ecological inference and the ecological fallacy. In N. J. Smelser & P. B. Baltes (Eds.), International Encyclopedia of the Social and Behavioral Sciences (Vol. 6, pp. 4027–4030). Oxford: Elsevier.