Ecological fallacy refers to a methodological error in which characteristics of a population as a whole are attributed to groups within that population without any real connection between them being demonstrated.
- The ecological fallacy is a logical error that can occur when individuals mistakenly infer information about individuals from aggregate data.
- The ecological fallacy can lead to false or inaccurate conclusions about social phenomena and the individuals within those phenomena.
- There are a number of ways to avoid the ecological fallacy, including careful study design, statistical methods, and critical thinking.
In This Article
The ecological fallacy is a mistaken conclusion drawn about individuals based on findings from groups to which they belong.
For example, if a university administrator found that the correlation between student math performance and having an engineering major was strong and positive, it would be an ecological fallacy to assume that the correlation would be the same for any particular student.
This would be an example of an ecological fallacy because it overlooks the fact that there is variation within each group of students.
Some engineering students may have lower grades than some liberal arts students, even though the average grade for engineering students is higher in mathematics.
The first person to describe ecological fallacies was the sociologist H. C. Selvin (1958). Selvin defined three different types of ecological fallacy:
The fallacy of composition: This occurs when it is assumed that what is true for the group must also be true for the individuals within that group.
The fallacy of division: This occurs when it is assumed that what is true for the individual must also be true for the groups to which they belong.
The fallacy of misplaced concreteness: This occurs when data from one level of analysis (for example, groups) is treated as if it were data from another level of analysis (for example, individuals).
Since Selvin first described ecological fallacies, they have been widely discussed in the literature on statistics and research methods, with implications for fields as varied as epidemiology, criminology, and economics.
Literacy rates and immigration
The ecological fallacy can apply to the relationship between large and smaller groups of individuals.
In the 1950s, sociologist William S. Robinson studied the correlation between literacy rates and the proportion of the population born outside of the United States.
He found that, in general, states with higher proportions of foreign-born residents had lower literacy rates.
However, he also found that this relationship differed when looking at specific groups within each state.
For example, in New York City, there was no relationship between the literacy rate of the foreign-born population and the overall literacy rate of the city (Robinson, 2009).
A study of crime rates may find, for example, that neighborhoods with a higher proportion of African American residents had higher crime rates.
However, when the data is analyzed at the individual level, it may show that African American individuals were no more likely to commit crimes than white individuals.
The erroneous relationship between breast cancer and fat consumption demonstrates one way that ecological fallacies can occur (Holmes et al., 1999).
A number of studies have found that there is a correlation between higher rates of breast cancer and higher levels of fat consumption at the population level.
However, the relationship disappears when these studies are analyzed at the individual level.
In other words, while it may be true that states or countries with higher levels of fat consumption have higher rates of breast cancer, this does not mean that individual women who eat a high-fat diet are more likely to develop breast cancer (Holmes et al., 1999).
Why They Happen
The term “ecological fallacy” is rooted in the concept of ecological correlation and individual correlation.
An individual correlation is one in which the statistical object or thing described is indivisible.
For example, the correlation between color and illiteracy for persons in the United States is an individual correlation because the kind of thing described is an indivisible unit, a person.
In an individual correlation, the variables are descriptive properties of individuals, such as height, income, eye color, or race, and not descriptive statistical constants, such as rates or means (Robinson, 2009).
Meanwhile, the statistical object is a group of persons in an ecological correlation. The correlation between the percentage of the population that is black and the percentage of the population, which is illiterate in each state in the United States, is an ecological correlation.
The thing described is the population of a state and not a single individual. The variables are percentages, descriptive properties of groups, and not descriptive properties of individuals.
The ecological fallacy occurs when the conclusions that can be drawn from ecological correlations are mistaken for individual correlations (Robinson, 2009).
There are a few reasons why ecological fallacies can occur. Firstly, Data from groups can be misleading when applied to individuals (Sedgewick, 2015). This is because group data averages out the differences between individuals within a group.
It is possible, in the aforementioned example about student math performance, that there may be some extreme outliers — say, a minority of engineering students who excel in mathematics or a minority of liberal arts students who have exceptional mathematical shortcomings — with most other students falling somewhere in between, regardless of major.
This could lead to a scenario, for example, where the mean mathematics score seems significantly higher for engineering students than those studying the liberal arts. Yet, the median marks for liberal arts students are the same as those of engineering students, if not higher.
Another reason why ecological fallacies can occur is that people tend to assume that members of a certain group share characteristics with other members of that group.
This is called the “assumption of homogeneity,” and it can lead people to mistakenly believe that all members of a group are alike.
For example, someone may encounter a student from the class with the highest average Writing scores — perhaps measured by a standardized test — in their school district.
Assuming that this one student is a literary genius would be an ecological fallacy. Just because she comes from the class with the highest average doesn’t mean that she is automatically a high-scorer in writing.
She could be the lowest scorer in a class that otherwise consists of extremely high-scoring students (Sedgewick, 2015).
Lastly, ecological fallacies can stem from the tendency to often use stereotypes when judging others. Stereotypes are oversimplified and often inaccurate beliefs about the characteristics of members of a particular group.
For example, someone might hold the stereotype that all engineers are “nerdy” or “unemotional,” which could lead them to mistakenly believe that all engineers are smart but not social.
Although a group of engineers may, as a whole, be less socially competent than average, there may be individual members of the group who are extremely gregarious — breaking the engineer stereotype (Sedgewick, 2015).
Avoiding Ecological Fallacies
Ecological fallacies can be a problem in any type of research that uses data from groups, but sociologists are especially prone to them because they often study social phenomena at the group level, such as families, neighborhoods, or organizations (Piantadosi, Byar, & Green, 1988).
Although drawing conclusions about individuals based on data from groups is sometimes unavoidable in sociology research, being aware of the potential for ecological fallacies and taking steps to minimize or avoid them can lead to more accurate and verifiable conclusions at the individual level.
Among these ways to avoid ecological fallacies in sociological research are (Piantadosi, Byar, & Green, 1988):
When analyzing data from groups, examine data from individuals.
This can help to reveal patterns that might be obscured when looking at data from groups only.
Statistically, sociologists can use tools such as standard deviation and uncertainty (and tools such as chi-squared analysis) to measure the range, reliability, and potential variance of data among individuals.
Researchers can also become more aware of the potential for assumptions about the characteristics of group members and of which measures can be applied most equitably in measuring the characteristics across groups.
For example, stereotypes about the abilities of people from certain cultures can drive a disparity in standardized test scores to be left unquestioned ( confirmation bias ), when in reality, the performance disparity may have been due to the fact that the test asks questions that rely on the implicit knowledge of the social rules, customs, or cultural features of one culture.
For example, a question asking to create a verbal metaphor with moccasins may create disparities between testers from cultures where moccasins are fashionable and widely known and those where the shoes are unheard of.
Thus, the question may be a more apt measure of whether the test-taker knows what moccasins are than of their ability to construct abstract reasoning.
On a related and final note, researchers can prevent the ecological fallacy in part by being aware of their own biases and stereotypes when conducting research.
They can strive to examine data objectively and without preconceived notions about the results, even if this does not eliminate all individual biases.
Additionally and alternatively, a data analysis team may be composed of people from varying backgrounds.
Piantadosi, S., Byar, D. P., & Green, S. B. (1988). The ecological fallacy. American journal of epidemiology, 127(5), 893-904.
Sedgwick, P. (2015). Understanding the ecological fallacy. Bmj, 351.
Holmes, M. D., Hunter, D. J., Colditz, G. A., Stampfer, M. J., Hankinson, S. E., Speizer, F. E., … & Willett, W. C. (1999). Association of dietary intake of fat and fatty acids with risk of breast cancer. Jama, 281 (10), 914-920.
Kramer, G. H. (1983). The ecological fallacy revisited: Aggregate-versus individual-level findings on economics and elections, and sociotropic voting. American political science review, 77 (1), 92-111.
Piantadosi, S., Byar, D. P., & Green, S. B. (1988). The ecological fallacy. American journal of epidemiology, 127( 5), 893-904.
Robinson, W. S. (2009). Ecological correlations and the behavior of individuals. International Journal of Epidemiology, 38 (2), 337-341.
Sedgwick, P. (2015). Understanding the ecological fallacy. Bmj, 351.
Selvin, H. C. (1958). Durkheim’s suicide and problems of empirical research. American Journal of Sociology, 63 (6), 607-619.