Assume that we trying to explain infant mortality as a function of GDP per capita in a sample of countries from around the world. Which of the following factors could most likely lead to endogeneity?
A) Whether the country has been at war
B) Population size
C) (Geographic) size of country
D) Use of proportional representation system

Question

Quizplus · Accepted Answer

Endogeneity refers to a situation in which an explanatory variable is correlated with the error term in a regression model. This correlation can arise due to omitted variable bias, measurement error, simultaneity, or reverse causality. When trying to explain infant mortality as a function of GDP per capita, endogeneity can be introduced if there are factors that both affect infant mortality and are correlated with GDP per capita, but are not included in the model.
Let's consider the factors listed:
a. Whether the country has been at war: Being at war can have a significant impact on a country's infrastructure, healthcare system, and overall economic performance. Wars can lead to destruction of healthcare facilities, disruption of services, and economic instability, all of which can increase infant mortality. At the same time, wars can also affect a country's GDP per capita, as resources are diverted from productive uses to military spending, and economic activity may be disrupted. If the fact that a country has been at war is not included in the model, it could lead to endogeneity because it is likely to be correlated with both infant mortality and GDP per capita.
b. Population size: Larger populations might have different healthcare challenges and economies of scale that could affect infant mortality rates. However, population size is less likely to be correlated with GDP per capita in a way that would introduce endogeneity. GDP per capita is a measure that already accounts for population size by dividing the GDP by the number of people in the country.
c. (Geographic) size of the country: The geographic size of a country can influence the distribution of healthcare services and infrastructure, which can affect infant mortality. However, geographic size is not typically directly correlated with GDP per capita, and thus it is less likely to introduce endogeneity in the relationship between GDP per capita and infant mortality.
d. Use of proportional representation system: The type of political system, such as the use of proportional representation, can influence policy decisions, including healthcare policies that might affect infant mortality. However, the link between the use of proportional representation and GDP per capita is tenuous and less direct. It is less likely to be a source of endogeneity compared to other factors that have a more direct impact on both infant mortality and economic performance.
Among the factors listed, (a) Whether the country has been at war is the most likely to lead to endogeneity in the model explaining infant mortality as a function of GDP per capita. This is because war can directly affect both infant mortality rates and GDP per capita, and if not included in the model, it could bias the results.

Assume That We Trying to Explain Infant Mortality as a Function