ECON10132/ECON20292: Further Statistics Project-
Project details-
The data for this project comes from the study of Hamermesh, D.S. and J.E. Biddle (1994), "Beauty and the Labor Market", American Economic Review 84: 1174-1194 (provided by Jeffrey Wooldridge Introductory Econometrics: A modern approach).
The data consists of 1, 260 working individuals and the variables included for the project are the following:
Variable name definition
wage hourly wage
looks 1 = homely, 2 = below average, 3 =average, 4 =above average, 5 = strikingly handsome
abvavg = 1 if looks >= 4, zero otherwise
belavg = 1 if looks <= 2, zero otherwise
educ years of schooling
black = 1 if black, zero otherwise
female = 1 if female, zero otherwise
service = 1 if service industry, zero otherwise
high educ = educ > 12, zero otherwise
The variable looks describe the physical appearance of the observation. The variables abvavg and belavg are defined in terms of the variable looks as described above.
The data is available in Stata and Excel formats on the course Blackboard site and you need to use a statistical software (i.e. Stata or Excel) to answer most of the following questions.
Note: You will probably only be able to fully comprehend and possibly do the analysis required to answer Question 4 once the course material has been covered.
1. What type of variables are wage, educ, and looks?
2. In the survey on which the data is based, the interviewer is asked to rate the respondents' physical appearance. The variable looks contains this information. What kind of problems may the construction of this variable create for analysing the data?
3. (a)What is the relative frequency of observations that looks "homely" (i.e. looks = 1)?
(b) What is the relative frequency of observations that looks "strikingly hand- some"(i.e. looks = 5)?
(c) What is the cumulative relative frequency (as a %) of observations that are looking "below average" (i.e. looks = 1 or looks = 2)?
(d) Considering your answers to 3a, 3b and 3c, what may be the benefits of using the variables abvavg or belavg?
(e) What is the average and median wage for individuals who are looking below average (belavg = 1) and for individuals who look average and above (belavg = 0)?
4. Considering your answer to 3e), are differences in average wages between in- dividuals who look below average and those who look average and above significant?
5. (a) What are the average wages for women and men?
(b) What is the ratio of average wages of men and women? Interpret this ratio.
(c) Does this ratio vary by abvavg?
(d) What can you conclude from your analysis in Question 5?
6. It could be that men and women work in different sectors and that in these sectors the physical appearance of the individual is of different importance. With this in mind, how could you use the data and variables at hand to shed some light on your findings in Question 5? What do you find?
7. (a) Do black people earn more or less than white people?
(b) On average, for an hour worked by a "strikingly handsome" black person, how much does a "strikingly handsome" white person earn?
(c) How does this compare to the average wages of "homely" black and "homely" white people? What are the average wages of black and white people who look above average (i.e. abvavg = 1)? Interpret your findings and comment on the number of observations in each of these groups.
(d) Compute the 90th and 10th percentiles, the standard deviation, the range, and the IQR of wages for black and white people. What can you say about the dispersion of the wage variable?
8. (a) What is the relative frequency of observations that is black and has more than 12 years of education and the relative frequency of those who are white and have more than 12 years of education? (b) What are the average earnings of those with more than 12 years of education who are black and those who have more than 12 years of education who are white?
Attachment:- Data.rar