Order ID 6463784949 Type Essay Writer Level Masters Style APA/MLA/Harvard/Chicago Sources/References 6 Number of Pages 5-10 Pages Description/Paper Instructions Make sure to use Stata and include pictures
I included a sample submitted draft paper in the files below. You can use the Statiscal methods similar to that paper
Paper Assignment
ECMT 463
Structure of a Research Question
(This is not the same as the structure of the paper; this is how to go about structuring your thinking & research process.)
- What is your question to be tested?
“Does A cause B”? E.g. does a college degree increase income (holding other factors the same)?
- What is an “ideal” experiment that can be used to test and answer this question?
- What data are available to test this?
- How are you going to use methods and the data to approximate the “ideal” experiment?
Before thinking about each in detail, let’s look at a particular example of research…
- What is your question to be tested?
Are social networks on university campuses segregated, and if so, can university policies reduce segregation?
Motivation:
- Are universities really integrated?
- Social networks determine peer effects in college
- Social networks are important for the transmission of information.
E.g. do your peers help you get a job?
“While the frequency of alternative job-finding methods varies somewhat by sex and occupation, the following generalization seems fair: approximately 50% of all workers currently employed found their jobs through friends and relatives” (Montgomery 1991)
- What is your question to be tested?
If social networks on university campuses are not “very integrated”, can university policies increase integration?
What determines social networks?
Do individuals have School contact? Environment
Preferences Do individuals want / Tastes to be friends?
Social
network
What determines social networks?
Preferences / Tastes
- Parental background
- Political orientation
- Race
- Hometown
Social
network
School
Environment
- Composition of student
body
- Class room assignment •Dorm assignment
- Clubs / Activities
What determines social networks?
Preferences / Tastes
- Parental background
- Political orientation
- Race
- Hometown
Social
network
School
Environment
- Composition of student
body
- Class room assignment •Dorm assignment
- Clubs / Activities
University
Policy
- What is your question to be tested?
If social networks on university campuses are not “very integrated”, can university policies increase integration?
Can the university affect the composition of
social networks outside of the immediate environment it controls (e.g. dorm)?
- What is an “ideal” experiment that can be used to test and answer this question?
- Why not just survey who students’ friends are?
If two “purple” people in this room are friends, is it because
- (a) “purple” people like economics and econ majors meet in class and then form friendships [environment]
- (b) “purple” people just like other purple people [preferences]
- Regression on sample of purple people:
Non Purple Friendsi = 0 + 1 (Non – Purple People in Environment)i + ui
Correlated if you choose Preferences of friend type your environment. BAD!
- So….tell me the “ideal” experiment:
- Randomly assign students to roommates or dormmates or…
- What is an “ideal” experiment that can be used to test and answer this question?
Ideal experiment: what if students are randomly assigned more/less “different” peers?
What if the world did this experiment for us?
- Rice University – randomly assigns students to residential colleges
- So, some “dorms” will have disproportionately more minorities and others less
Research question: If I am in a dorm with disproportionately more people of a different race,
- Do I form more friendships with different race students in the dorm?
- Do I form more friendships with different race students outside the dorm?
- What data are available to test this?
From facebook
- All student profiles as of 1/17/05 for 10 universities in Texas
- 65,104 undergraduates
- (Self-reported) Demographics: year, birthdate, gender, high school, hometown, major, current courses, dating status, residence hall, political orientation, jobs, hobbies
- Social network: links to friends at own school
- Race – we classify based upon pictures
Summary Statistics
Texas Rice Baylor A&M U Texas Avg. Number of Friends 50.8 59.8 41.1 39.5
Avg. Degrees of Separation 2.30 2.62 2.95 3.00
1= friends 2= friends of friends 3= friends of friends of friends
- Small world
Cluster Coefficient (“cliquishness”) 0.24 0.19 0.17 0.20
The fraction of your friends who
are friends with each other
- Methods and Data that Approximate the “Ideal” Experiment
Following regression:
- Methods and Data that Approximate the “Ideal” Experiment
For Whites/Hispanics
❑ More blacks in dorm:
Increases number of black friends in dorm
Does NOT increase # black friends outside of dorm
❑ More Asians in dorm:
(same as above)
Increasing share of black dormmates from 5% to 10%
- Increase within-dorm black friend share 1.9 pct points (0.387*0.05=0.019)
Message: Suggests that a university has less ability to increase the diversity of interaction outside of environments it controls (in short run)
Economics: “irrational passion for dispassionate rationality”
- What is your question to be tested?
Is it interesting?
How might the answer be useful to “society” or “business”?
Give some background or context.
I want it to be a flavor of “Does A cause B?”
- What is an “ideal” experiment that can be used to test and answer this question?
What is the null hypothesis?
What data do you need? (What do you need to know?)
How will you hold all other factors that might affect the outcome constant while varying the “variable” of interest? I.e. What are your experimental controls?
(Ceteris Paribus)
What is the analytical method used to interpret the results?
- What data are available to test this?
What is name and source of the data used.
What is the nature of the data – randomized experiment, directly from real life,…
What variables are used and how are they defined? What are their units of measurement?
Summary Table or something similar of the data.
If done at all, what do you do to the data to clean it up or alter the sample composition?
- How are you going to use methods and the data to approximate the “ideal” experiment?
- What is the Null Hypothesis that you are testing?
- What is your estimation method (regression)?
- Can you say that it is “A causes B”? How does your data, choice of variables, and/or estimation method provide the same control of confounding factors as the controls in your “ideal” experiment?
randomized sample composition, not “B causes A” nor “C causes A and B”
- Are there any remaining problems with data and estimation that remain a concern but you cannot control for? (Readers will partially forgive you if you acknowledge remaining problems with potential “omitted” variables).
- Results Table
Interpret results
- Hypothesis rejected or not?
- Is it statistically significant?
- Is it economically significant? [relatively large coefficient(s)]
- Any surprises? If so, anything to be worried about?
- use p-values, confidence intervals to back up your explanation.
“Rules” on Writing the Paper
We will provide some datasets (in Stata)
- TAs will show you this week
- You can find your own data, if you choose
You will need to find a good question to address with the data available
- Obviously don’t just regress a misc. variable on other misc. variables
- Get a sense of what variables are in a dataset, step away, and think of what interesting questions exist with that type of information
(Later) TAs will talk about writing up your results
- See sample paper posted in eCampus
GET THIS PROJECT NOW BY CLICKING ON THIS LINK TO PLACE THE ORDER
CLICK ON THE LINK HERE: https://www.perfectacademic.com/orders/ordernow
You Have Any Other Essay/Assignment/Class Project/Homework Related to this? Click Here Now [CLICK ME] and Have It Done by Our PhD Qualified Writers!!