Question: You are investigating the effect of income on family size in rural India. The relationship is proposed. The available data come from a survey of 100 villages, but include only village average family size and village average income, not observations of individual families. How-ever, you do know the number of families sampled in each village.
Family Size = β0 + β1 (Family income) + ε
a. What would be the consequences of using OLS to estimate the above model using the aver-aged data? (That is, how are your estimated coefficients and standard errors affected com-pared to the estimates that would be produced if individual family data were available'?)
b. Propose an alternative method for estimating the model, and explain why the alternative is better than OLS.