Nonexperimental studies in which the unit of observation is not the individual but a group of people in a particular environment, such as workers in a factory or inhabitants of a city or a country. These studies can be useful if information on individuals is not available; exposure is then an overall measure for the population under investigation. The outcome variable under investigation in ecological studies is often mortality. A well-known phenomenon occurring in this type of study is the so-called ecological fallacy. On comparing countries, it may be found that the higher the average level of a risk factor A for a country, the higher the level of mortality due to disease B, while within each country (based on individual measurements of A and B) risk factor A is negatively associated with disease B.