GV900 Political Explanation, 2022/23
Instructor:
30 November, 2022
Copyright By PowCoder代写 加微信 powcoder
Homework assignment 2
Due Week 11 (On Faser)
9:45am (UK time)
Write an Rmarkdown code file (gv900-HW2.Rmd) to complete the following tasks.
• Submit 4 files, and 4 files only. That is, submit (1) the coversheet ( ESSAY COVERSHEET
2022-2023.docx, available on Moodle) and (2) your Rmarkdown code script
(gv900-HW2-yourID.Rmd). (3) the html document showing your code, your comments, results,
and graphics (gv900-HW2-yourID.html). (4) a pdf of your html so I can write comments on it
(gv900-HW2-yourID.pdf)(minus 5 points if fail to do so)
• Make sure that you delete your name from your Rmd code script and html document.
(minus 5 points if fail to do so)
• Execute everything before you submit, and m ake sure your file runs without an error. (minus
5 points if fail to do so)
• Your html output must have a proper header. (minus 5 points if fail to do so)
• In your Rmd file, add comments and annotations to everything you do. Don’t copy and paste
all the questions into your R md code file, but do show m e the question number for each
question. (minus 5 points if fail to do so)
1. You are going to use the following 6 R packages: ggplot2, gmodels, Hmisc, stargazer,
effects, and gridExtra. Load all the packages. [2 points]
2. Load Titanic passenger survival dataset available on Moodle (titanic.csv), and store it
as an object named td. [2 points]
3. The unit of observation is individual passengers. How many passengers does the dataset
have? That is, how many rows are there in the dataset? Provide a command to get the
answer. Also, write your answer in a comment line. [2 points]
4. The dataset contains various information on passengers, including their name (name) and
whether or not they survived (survived). The dummy variable survived is coded as 1
if a passenger survived and 0 otherwise. Create a simple frequency table of the survived
variable (that is, there is no need to change the column names or to obtain relative
percentages for this task; one line of command is sufficient) to see how many passengers
survived and how many did not make it. Provide a command to create a frequency table
of this variable (no need to write a comment). [2 points]
5. Calculate the survival percentage. Provide command(s) to calculate it. Also, write your
answer in a comment line. [2 points]
6. The data set contains a variable named pclass, which is coded as 1 (= passenger has a
1st class ticket; for those of you who don’t know, 1st class tickets are more expensive than
2nd class tickets, 2nd class more expensive than 3rd class), 2 (= 2nd class ticket), and 3
(= 3rd class ticket). This ordinal (ordered categorical) variable could be used as a proxy
for socio-economic class of passengers. Create a simple frequency table of this variable
(that is, there is no need to change the column names; one line of command is sufficient)
to see the distribution of this variable. Provide a command to create a frequency table of
this variable (no need to write a comment). [2 points]
7. In tasks 7–11, we will analyze the relationship between socio-economic class of passengers
and their survival using one of the three bivariate hypothesis testing methods you have
learned in Weeks 6 & 7. The question we ask here is: how does socio-economic class of
passengers influence the likelihood of passenger survival? Let’s say we hypothesize that
socio-economic class of passengers is positively associated with passenger survival. First,
what are the dependent and independent variables in our investigation? Provide your
answers in a comment line. [2 points]
8. Second, create a two-way frequency table (a.k.a, cross tabulation) of the two variables,
pclass and survived. Provide a command to create such a table (no need to write a
comment). Make sure that (1) values of the dependent variable are shown in rows and
the independent variable in columns, (2) your table shows column percentages but not
row percentages, cell percentages, or χ2 contributions, and (3) your table produces a χ2
test statistic. [6 points]
9. Read the table you produced the above and answer a few questions. (a) What is the
survival percentage among the 1st class passengers? (b) What is the survival percentage
among the 2nd class passengers? (c) What is the survival percentage among the 3rd class
passengers? Provide your answers in comment lines. [3 points]
10. Would you say that the relationship between survived and pclass is consistent with our
hypothesis I described in task 7? Why or why not? There is no need to comment on
statistical significance, but do comment on the pattern observed in the sample. Provide
your answers in a comment line. [3 points]
11. Fill in the blanks of the following statements that summarize the results. Provide your
answers in a comment line. You only need to write four options, such as (a), (b), (c), etc.,
in the correct order. [8 points]
Since the test statistic produces a p-value smaller than , we can the null
hypothesis of no association at % confidence level. We thus support for
our hypothesis.
(a) 173 (b) 172 (c) 128 (d) 127 (e) 99.9 (f) 99
(g) 95 (h) 90 (i) 0.1 (j) 0.05 (k) 0.01 (l) 0.001
(m) accept (n) reject (o) find (p) do not find
12. The dataset contains a variable named fare, which is the price of the ticket each passenger
has. It is shown in pre-1970 GBP. (Note: £ 1 in 1911 is equivalent in purchasing power
to about £ 112 in 2018.) Do female passengers tend to have a more expensive ticket
compared with male passengers? Explore the relationship between the female variable
(coded as “Female” for female passengers and “Male” for male passengers) and fare.
Choose an appropriate bivariate statistical testing method for these two variables from
the three methods you have learned in Weeks 6 & 7, and perform the test. Provide
command(s) to perform the analysis. (Hint: I am not asking you to run a regression.) [5
13. Interpret the results of the bivariate test you performed above and answer the question
(do female passengers tend to have a more expensive ticket?). Comment on the observed
pattern in the sample as well as the statistical significance, and draw a conclusion (i.e.,
answer the question posed here). Your answers must have up to three sentences. [6 points]
14. The dataset contains a variable named age (age of the passenger). Do older passengers
tend to have a more expensive ticket compared with younger passengers? Explore the
relationship between age and fare graphically. That is, create a plot that shows the
relationship between these two variables using the ggplot function. Provide commands
to create the plot. [5 points]
15. Perform an appropriate bivariate statistical test (again, choose one from the three methods
covered in Weeks 6 & 7) to explore the relationship between age and fare. Provide
command(s) to perform the analysis. (Hint: I am not asking you to run a regression.) [5
16. Interpret the results of the bivariate test you performed above and answer the question
(do older passengers tend to have a more expensive ticket?). Comment on the observed
pattern in the sample as well as the statistical significance, and draw a conclusion (i.e.,
answer the question posed here). Your written answers must have up to three sentences.
[6 points]
17. Regress fare on age, and produce a regression table using the stargazer function. [4
18. Create a plot that illustrates the estimated effect of age on fare based on the model you
estimated above. [3 points]
19. Judging from the numerical and graphical results of the regression analysis, would you
say that age and fare are positively related? Comment on the observed pattern in the
sample as well as the statistical significance, and draw a conclusion. [6 points]
20. Graphically explore the relationship between age and fare, holding constant the female
variable. That is, create a plot using the ggplot function that shows the relationship
between age and fare for female and male passengers separately. Try to have one plot
that has two panels (one for female and one for male). [5 points]
21. Regress fare on age and female, and produce a regression table using the stargazer
function that summarizes the results of this model and the model you estimated in task
17. [4 points]
22. Which one of the two regression models performs better? Fill in the blanks of the following
statement. Provide your answers in a comment line. You only need to write four options,
such as (a), (b), (c), etc., in the correct order. [8 points]
Since is for the model, the model fits the data better.
(a) standard error (b) p value (c) R2 (d) Adjusted R2 (e) the number of stars
(f) smaller (g) greater (h) first (i) second
23. Create a plot that illustrates the estimated effect of female on fare based on the second
regression model you estimated. [3 points]
24. Create a plot that illustrates the estimated effect of age on fare for male and female
passengers separately based on the second regression model you estimated. Try to have
one plot that has two panels (one for female and one for male). [6 points]
End of file
程序代写 CS代考 加微信: powcoder QQ: 1823890830 Email: powcoder@163.com