Assignment Estimating the returns to education

docx

School

University of California, Los Angeles *

*We aren’t endorsed by this school

Course

133

Subject

Economics

Date

Jan 9, 2024

Type

docx

Pages

3

Uploaded by ElderIceHornet17

Report
Adriana Lleras-Muney Economics Department Office Hours: Tuesdays 1-3pm Bunche 9373 Intergenerational Poverty in America Assignment: Estimating the returns to education Your job is to download and use data from the American Community Survey conducted in 2021 to investigate whether it pays to attend college. Present your results in the form of a memo which will include only the numbers and figures that you found most useful. 1-Obtain data from the 2021 ACS available at https://usa.ipums.org/usa/index.shtml . Download a STATA/R data set. Before you download the data think carefully about the following questions: Sample: Who do we want to study? o Restrict attention to individuals from California o Make other restrictions that seem appropriate. For instance, explain whether or not you want to include foreign citizens, or individuals over the age of 45, etc. Variables: What do we want to study? o Extract annual earnings, education, age, gender, race/ethnicity, cognitive difficulty, ambulatory difficulty. Why? o Select another 5 variables to study thinking carefully why you need the information and how you will use it on your analysis. 2-Use these data to describe and investigate how education affects labor market earnings. a. Plot a histogram of the distribution of education. Describe what you see: what is the fraction of individuals without a high school degree? What fraction has (at least) a college degree? b. Summarize earnings patterns: what is the mean, median, 10 th percentile, 90 th percentiles of earnings? c. For each category of education, plot average earnings. What is the average increase in earnings for each year of school? d. What is the difference in average earnings of those with and without a college degree? Is the gap similar for men and women? For individuals under 40 and those over 40? For those with a cognitive difficulty? e. What other graphs or statistics would you like to report? Justify your choices. For instance, plot the distribution of earnings for those with college and
those without. Does everyone that goes to college do better than everyone that does not? 3-Use these data to estimate the “returns to college”. a. Regress log (earnings) on education dummies, age, age squared, a dummy for female. Report your findings: what does the coefficient on college degree mean? The coefficient on female? b. Add dummies for race and ethnicity. Interpret the coefficient for the Hispanic dummy. Does the college coefficient change when add these controls? Why are we including controls? c. What else would you like to report? Justify your choices. For example you might want to estimate regressions separately by gender or race. Or you might want to include other controls in your regressions. 4-Going to college. a. We want to understand if going to college leads to greater earnings. Are the estimated increases in earnings observed among those who go to college causal? Why or why not? b. Do these average changes apply to everyone? Discussion: What is the average cost of attending college? Can you conclude based on your analysis that it pays to attend college? How would you determine this? Explain. Grading Your memo will be 3 pages long at most (1-inch margins, use single spacing at 12-point font), excluding references and appendices (back up materials). All facts should either be computed by your group or be accompanied by appropriate citations. Late assignments get a grade of 0. This is a data assignment. Original figures, statistics and regression analysis are expected. Using figures and results from other people’s work is NOT acceptable. Points will be deducted for not following instructions: o Follow formatting requirements. o Make sure you include citations when relevant. o Figures and tables should include labels, have titles and be self- explanatory. o Unformatted computer output is not acceptable. o Reading of the existing literature on the topic is strongly encouraged. o Come to class ready to discuss your findings. Students who volunteer to present their findings in class will receive extra credit.
Your preview ends here
Eager to read complete document? Join bartleby learn and gain access to the full version
  • Access to all documents
  • Unlimited textbook solutions
  • 24/7 expert homework help