2. Let kids denote the number of children ever born to a woman, and let educ denotes years of education of the woman. A simple regression model relating fertility to years of education is kids = Bo +...

Extracted text: 2. Let kids denote the number of children ever born to a woman, and let educ denotes years of education of the woman. A simple regression model relating fertility to years of education is kids = Bo + Bieduc + u, where u is the unobserved error. (1) What factors are contained in u (Please specify TWO factors)? How are these correlated with level of education? (2) Could you demonstrate, using ONE factor you answered in (1) as an example, that due to its correlation to education, the effect of education on number of kids born to a woman will be biasedly estimated. In particular, given the assumption you make on the correlation between that factor and educ as well as kids, will the coefficient of educ be overestimated or underestimated if directly using the regression model shown? (3) Again, using the factor you utilized in (2), show that what you could do in order to obtain an unbiased estimate of B1?