3. In 1973, the Indonesian government launched a major school construction program, the Seko lah Dasar INPRES program. Between 197374 and 197879, more than 61,000 primary schools were built. Duo (2001) evaluates the effect of this program on education and wages, using the fact the exposure to the school construction program varied by region and date of birth. First, children born in different regions were affected by the program differently because the government allocated more schools to regions where initial enrollment was low. Second, in Indonesia, children typically go to primary school between the ages of 7 and 12. Therefore, only individuals who were young enough to be in school when the program was launched were affected by the program. The following table reports the average educational attainment and earnings of two demo graphic groups (i.e., individuals aged 26 years old and 1217 years old as of 1972) residing in two different regions of Indonesia (i.e., \"high-construction\" region, where many schools were built, and \"low-construction\" region, where few schools were built). (a) Use the difference-in-differences strategy to calculate the impact of the program on edu- cation attainment and earnings of the targeted population. What is the meaning of the estimates? (6 points) (b) What is the identifying assumption of the differenceindifferences strategy? (4 points) 2 Years of Education Log Wages Aged 1217 Aged 26 Aged 1217 Aged 26 Highconstruction Region 8.02 8.49 6.87 6.61 Lowconstruction Region 9.40 9.76 7.02 6.73 (0) Can you suggest a way to test for the assumption? (5 points)