1. You are working for the Ministry of Labor in the People's Republic of China (PRC). You have access to the data found in labor.dta, on WebCT. It refers to a housing and labor development project in which 1400 villages participated over the course of nearly 10 years. The villages were divided into project villages (the treatment group) and non-project villages (the control group), and then monitored to see how they developed over time.
It is almost time to present some findings to your superiors regarding the project. Your supervisor has asked you to present evidence on whether or not project participation has had an effect on income per capita. You decide to go about answering his question by investigating the following factors, using the data you have (be sure to show your code and output above your written answer):
a. Test whether there is a significant difference between project villages in 1995 and project villages in 2004 on their income per capita. Show and explain the result.
b. Test whether there is a significant difference between non-project villages in 1995 and non-project villages in 2004 on their income per capita. Show and explain the result.
c. One of your fellow research assistants has regressed proj on cid (in Stata: reg proj cid) and claims that this provides evidence that the project counties were not chosen randomly. What is your response? (Note: there are at least TWO big problems with this statement, one you should see prior to running the regression and one you should see after running the regression. You only have to provide one reason though.)
d. When do you use ttest var1 == var 2 and when do you use ttest var1, by (var2)?
2. What do you call ordinary least squares when the Y variable is 0/1?
3. Explain at least 2 problems with the following [taken from an internet meme]:
WASHINGTON (AP) The results of a new survey conducted by pollsters suggest that, contrary to common scientific wisdom, correlation does in fact imply causation. The highly reputable source, Gallup Polls, Inc., surveyed 1009 Americans during the month of October and asked them, "Do you believe correlation implies causation?" An overwhelming 64% of American's answered "YES", while only 38% replied "NO". Another 8% were undecided. This result threatens to shake the foundations of both the scientific and mainstream community, as it provides proof that correlation is indeed causation.
4. Director Smith wants to raise standards for admission. She would like to automatically reject students who fall within the bottom 75% of a national exam. The exam is normally distributed with a mean of 1000 and a standard deviation of 100. What should she set for her cutoff value?
Assignment link -
https://www.dropbox.com/s/pewqshz7oylv26w/Assignment.rar?dl=0