Part 2 -
2.1- Between the years of 2012 and 2015 there were 100 terrorist attacks in the United States (data from Global Terrorism Database). Of those 100 terrorist attacks, 8 were perpetrated by people who claimed to be motivated by the Islamic faith. The total U.S. population during this time was 318.9 million and the Muslim population in the United States is 1.8 million during those same years.
a) What is the probability of being Muslim given that you perpetrated a terrorist attack during this time period?
b) What is the overall probability that a person selected from the population is a terrorist?
c) Suppose that you board an airplane and see a Muslim passenger also aboard the plane. What is the probability that this person is a terrorist given that they are Muslim?
d) Suppose that on this same airplane there are 248 other non-Muslim passengers. What is the probability that at least one of these passengers is a terrorist given that they are not Muslim?
e) Who should cause you greater fear, the Muslim passenger, or the non- Muslim passengers, or neither?
2.2- The Human Development Index (HDI) is an index the United Nations uses to give a summary rating for each nation based on life expectancy at birth, education attainment, and income. In 2006, the top ten nations with the highest HDI rating, followed in parentheses by the percentage of the seats in parliament held by women and the HDI score were Norway (38, .94), Iceland (33, .90), Australia (28, .93) Ireland (14, .92), Sweden (45, .91), Canada (24, .91), Japan (11, .89), United States (15, .92), Switzerland (25, .93), Netherlands (34, .92). (i.e. the US had 15 percent of seats held by women and an HDI score of .92)
a) Calculate the mean and the standard deviation for the percentage of seats in parliament held by women in these states and interpret your results.
b) Calculate the mean and the standard deviation for HDI index and interpret your results.
c) Calculate the correlation between the HDI index score and the percentage of seats in the country's parliament held by women.
d) Calculate the coefficient on the HDI score of a regression in which the HDI score is the independent variable and the proportion of seats held by women is the dependent variable.
e) Calculate the r2 from this simple regression model.
f) Input these observations into Stata and show where the answers to (c), (d), and (e) appear in the regression output. (Here it is ok to past Stata output into your answer.)
g) Interpret the sign, statistical significance, and substantive significance of each coefficient from the model.
2.3- In a population μy = 100 and σy2 = 44. Use the central limit theorem to answer the following questions and show your work.
a) In a random sample of size n = 100, find Pr(Y ¯ ≤ 101).
b) In a random sample of size n = 165, find Pr(Y ¯ > 98).
c) In a random sample of size n = 64, find Pr(101 > Y ¯ > 103).
d) What sample size would be required so that the Pr(Y ¯ > 102) < .25?
2.4- A survey of 750 voters immediately after the 2016 election contained 350 voters without a college degree and 400 voters with a college degree. The survey showed that 52% of voters without a college degree voted for President Trump while 43% of voters with a college degree voted for Trump. Assume that the only two choices were Clinton or Trump. (Data for this question come from https://www.pewresearch.org/fact-tank/2016/11/09/behind-trumps-victory-divisions-by-race-gender-education/.)
a) Construct a 95% confidence interval around the estimates of Trump support for people with a college degree and for people without a college degree.
b) Can you reject the null hypothesis that the difference in average support for Trump among these two groups is zero at the 95% level? Why or why not?
Part 3 -
Research suggests that dehumanization stands at the heart of many group conflicts. As such, "humanizing the enemy"-i.e. getting members of a group to recognize the humanity of members of the group with which they are in conflict-stands at the heart of countless conflict resolution and coexistence programs worldwide.
However, research suggests that in contexts of protracted intergroup conflict (like the Arab-Israeli conflict), the standard approaches to humanizing an outgroup are not effective. These approaches typically present images of outgroup suffering, coupled with stories where outgroup members talk of their very human emotions, with the intention of creating empathy for the outgroup. Professor Gubler at BYU with other colleagues hypothesized that these approaches tend to fail in contexts of protracted intergroup conflict because to acknowledge outgroup suffering in these contexts often means an acknowledgement of ingroup blame, as ingroup members recognize their group's responsibility (at least in part) for the predicament of the outgroup.
As such, in 2010, they set out to test a different approach to humanizing the outgroup in contexts of protracted intergroup conflict. Instead of asking individuals to feel empathy for outgroup suffering for which their group might be responsible, they conceived of an approach that requires one individual on one side of the conflict to express empathy for outgroup suffering that is not related directly to the intergroup conflict. This expression of empathy offers an opportunity to express empathy towards the outgroup without assuming blame. Thus, it presents a simpler, safer first step than that required in the other approaches, and as such may succeed at humanizing the outgroup where other approaches have failed.
To test their idea, they conducted a series of experiments with a large, nationally-diverse sample of Jewish-Israeli citizens of Israel, focusing on their views towards Palestinian citizens of Israel (PCI's). In these experiments, some participants were randomly assigned to an op-ed written by a PCI in which the author expressed empathy for Jewish suffering in the Holocaust (the humanization treatment condition), while others were assigned to the same op-ed, but with the expressions of empathy removed (control 2 condition). They hypothesized that this humanizing treatment would succeed in humanizing PCIs where the standard approaches (that showed PCI suffering, etc.) had not. In particular, they hypothesized that after the experimental manipulation, those in the treatment condition would 1) feel more empathy, and 2) report seeing PCIs as more human than those in the control condition.
The dataset, Study1.dta, contains a subset of the results from their first experiment. The corresponding codebook identifies how each variable was coded. In this experiment, to measure the degree of empathy participants felt after the treatment/control, participants were asked the following question used by Batson and colleagues to measure empathy (Batson et al., 1997, 2002): "On a scale from 1 (not at all) - 7 (very much), please indicate how much you felt each of the following emotions as you read this interview: sympathetic, compassionate, soft-hearted, warm, tender." To measure the degree of humanity participants assigned to the outgroup, they asked questions 11-22 in the dataset (see the codebook for the details on these questions). Question 11 and Question 19 are the two emotions in particular in this list that measure the degree of humanity; the others are to measure related concepts. Use these data to carefully test their hypotheses. As a hint to get you off on the right foot, you will want to create one variable for empathy (an average of the 5 empathy questions) and another variable to measure the degree of humanity ascribed to PCIs (an average of questions 11 and 19). Along the way, formally test the ceteris paribus assumption of the experiment.
President Trump has recently stated that he intends to solve the Israeli- Palestinian conflict by dispatching his son-in-law, whose main qualification for this task appears to be being Jewish, to the region. Write a memo to the president evaluating this particular strategy for humanizing the different groups involved in the conflict.
Part 4 -
In May,1978, Brink's Inc., was awarded a contract to collect coins from some 70,000 parking meters in New York City for delivery to the City Department of Finance. Sometime later the City became suspicious that not all of the money collected was being returned to the city. In 1983 the City presented evidence in court that Brink's employees had been stealing parking meter moneys-delivering to the city less than the total collections. The City sued Brink's for negligent supervision of its employees, seeking to recover the amount stolen. Five Brink's collectors were then arrested and charged with grand larceny.
The City's attorney provided the data found in the dataset parking1.dta. The variable con in the data file represents monthly parking meter collections by the principal contractor in New York City from May 1977 to March 1981. From May 1978 to April 1980 the contractor was Brink's (captured in our dataset with a dummy variable where brink = 1). You are working as a clerk (with statistical expertise) for the judge (without statistical expertise) hearing the case. Using whatever methods you think are appropriate, write a memo that supports the City's position. (Call your memo supporting the City's position Exhibit A.) Then, using the same data, write a memo supporting the position of Brink's, which will certainly argue that there was no theft, and if there was theft, that it was minimal. (Call this memo Exhibit B.) Conclude both memos with a statistically-justifiable answer to this question from the perspective of the City and Brink's, respectively: if Brink's is found guilty, what amount should the City be awarded from Brink's? In a third one-page memo following these two memos, discuss which of the two cases is more convincing and why, directly addressing what evidence there is that Brink's employees stole money. On this page, also briefly discuss the strengths and weaknesses of this approach to investigating this issue? What would you do (instead) to investigate this issue? (For purposes of this class, concern yourself with statistical arguments, not process or other legal arguments.)
Each of the three memos [worth 30 points each], should be no longer than one page. Include professionally presented graphs and tables (if necessary) in a professional appendix to each memo separately. Include your Stata files after this. Please turn in each of the three portions of Part 4 separately.
Attachment:- Assignment Files.rar