Question

To test the effectiveness of a job training program on the subsequent wages of workers, We specify the model log(wage) = Bo + B,train + B2educ + Bzexper + u where train is a binary variable equal to one if a worker participated in the program. a. How do we interpret B, and B, in the model? b. Think of the error term u as containing unobserved worker ability. If less able workers have a greater chance of being selected for the program, and you use an OLS analysis, what can you say about the likely bias in the OLS estimator of B1 (i.e. is it biased upwards or downwards and why)?

Answer 1

(a) Bo is the log wage of the person who does not participate in the training program, has no education, and no experience.

On the other hand, keeping education and experience constant, if a person participated in the program, it changes the wages by 100*B1%. Thus, B1 is the impact of participation in the program on wages.

(b) There is omitted variable bias as the ability of workers should be a part of th regression. Now, ability is negatively related with participating in the program, and positvely related with the wages. Thus, the estimate of B1 has a neagtive bias (underestimated) as the negative impact of lower ability on wages is also now coming through the participation in program.