Use your data set and Stata (or other econometrics packages) to answer the following questions.
1. (a) For each of the variables hrwage, union, and hours, what is the average, the standard deviation, the minimum value, and the maximum value in your data set?
(b) What is the number of observations in your data set?
2. (a) Generate a new variable lwage as the (natural) logarithm of hrwage. Label this variable.
(b) What percentage of individuals are union members? What is the average hourly wage for union members who are at least 40 and at most 49 years old?
(c) Generate a new variable, agegroup, which takes value 1 if the individual is younger than 25 years, value 2 for age group 25-34, 3 for age group 35-44, 4 for age group 45-54, and 5 for ages 55 and above. Tabulate the new variable. How many women are there in the age group 35-44?
3. (a) Examine graphically the relationship between age and union membership, using a bar graph. In this graph, each bar should show the average union membership status for one of the age categories defined in question 2c. Interpret your result.
(b) Calculate the correlation coefficient between hourly wage and years of education.Is there a strong relationship between these two variables?
4. (a) Run a regression of hourly wage on experience and education, and report the regression results.
(b) If education changes from 8 to 11 years, what is the predicted effect on hours worked?
(c) If education changes from 12 to 15 years, what is the predicted effect on hours worked?
5. (a) Generate a new variable for potential labor market experience using the formula exper = age - educ - 6. (In the following, we will refer to this variable as ‘experience’ or exper.)
(b) Run a regression of log hourly wages (lwage) on years of schooling and experience, and report the regression results.
(c) If education changes from 12 to 13 years, what is the predicted effect on the hourly wage?
(d) If education changes from 8 to 9 years, what is the predicted effect on the hourly wage?
6. (a) Generate a new variable, expersq, which is equal to the square of exper and thenrun a regression of log hourly wages on years of schooling, exper, and exper 2 , and report the regression results.
(b) Test if the coefficients on exper and expersq are jointly significant (at a significance level of 1%).
(c) Do you prefer the regression in part(a) to the regression in question 5b? Explain.
7. (a) Run a regression of lwage on union, educ, and exper, and report the regression results.
(b) Interpret the estimated coefficient on union, and then use your regression results from part (a) of this question to test if union has a statistically significant effect on log hourly wages (at a significance level of 10%).
(c) How much does the R2 increase by including union membership, compared to thespecification in question 5b?
(d) Based on your regression results from part (a) of this question, what change ineducation would be needed to offset a change from non-union to union in terms oflog hourly wages?
8. (a) Run a regression of lwage on union, exper, educ, and an interaction term betweenunion and experience (union × exper ). What does the coefficient on the interactionterm measure?
(b) Test if the effect of experience differs with union membership at the 5% significance level.
(c) What is the predicted log hourly wage for a union member with 10 years experience,and 12 years of education?
9. (a) Generate a new variable, ptwork, which equals 1 if hours is strictly less than 30 and equals 0 otheriwise. Then extend the model in question 7a by additionally accounting for ptwork. Write down the resulting model and estimate its coefficients.
(b) Based on your regression results from part (a) of this question, what is the predicted effect on hourly wages of working less than 30 hours per week? Is the effect statistically significant (at a 5% level of significance)?
(c) A researcher claims that unions help individuals with low labour market attachment to achieve better labour market outcomes. In particular, the claim is that individuals who work part-time, i.e. less than 30 hours per week, benefit more from union membership (in terms of log hourly wages) than individuals who work fulltime, i.e. at least 30 hours per week. Explain how you can investigate this claim in
the data, and report your findings. Explain your results.
10. (a) Prepare a table reporting the estimates (and corresponding standard errors) from the regressions in questions 5b, 6a, 7a, 8a and 9a. In this table, each column should report the estimates of one regression.
(b) After running all these regressions, summarize the main findings you have obtained. Are there any further variables or models you would like to include in the analysis? If yes, then explain why and state the models.