Highlights
QUESTION 1: The Impact of the 2015-16 Refugee Crisis on Public Opinion:
Many social scientists have asked whether contact with immigrants and refugees makes citizens more hostile to those groups. This is difficult to study with conventional cross-sectional data because migrants usually choose their destinations non-randomly. For instance, they may choose to move to places with existing immigrant communities, or where support for migrants is high. In this question, we’ll analyse a recent natural experiment that provides a better basis for measuring the impact of migrants. It uses data from the following paper, available in the ‘Essay’ folder on Moodle:
Dominik Hangartner, Elias Dinas, Moritz Marbach, Konstantinos Makatos and Dimitrios Xefteris (2019). “Does Exposure to the Refugee Crisis Make Natives More Hostile?” American Political Science Review 113 (2): 442-455.
To help answer this question, first read the paper. Your task is to re-estimate and re-examine some of the paper’s findings. The dataset for this question is contained in the file “2019essay_q1.Rda”. It contains part of the survey data from the original paper. Each row contains one survey respondent, with the following variables:
Answer the following questions:
a) Begin by examining the relevance of the continuous instrument distance
i. Estimate the first stage of the authors’ two-stage least-squares estimation, clustering the standard errors by a municipality. Give a precise interpretation of the result.
ii. Estimate an F statistic for this first-stage result1
iii. Using the results from (i) and (ii), how relevant do you think that the instrument is? Explain your answer.
b) On page 446, the authors confuse the exclusion restriction assumption with the randomisation assumption, failing to address the exclusion restriction assumption properly. You can do better! Briefly, explain:
i. What the randomisation assumption means in this study.
ii. What the exclusion restriction assumption really means in this study, and whether or not you think it is likely to hold.
c) Replicate the second-stage coefficients and standard errors for score_asylum and score_immig, reported on page 449 and in Figure 4 on page 450 (the first and seventh estimates in the figure, called “asylum-seeker component” and “immigration component”), clustering your standard errors by a municipality.
d) [8 points] Now, you will re-do the paper’s analysis of the impact of refugee arrivals on public opinion with the Wald Estimator. Use the binary instrumental variable low_distance and the outcome variable score_asylum.
i. Explain, in this case, what type of municipality is a complier and what type of municipality is an always-taker
ii. Calculate and report the proportion of compliers and the intent-to-treat effect
iii. Use your answers from (ii) to calculate and report the Complier Average Causal Effect (CACE) of receiving immigrants on hostility towards asylum seekers.
iv. Using an appropriate method, calculate and report the p-value for this CACE estimate.
QUESTION 2: A Simulated Experiment:
This question analyses a simulated experimental dataset contained in the file “2019essay_q2. Rda. "It includes 100 units and the following five variables:
For parts (a) to (d), we will assume that no missingness occurred, focusing on all 100 units. Answer the following questions:
a) Assuming no missingness, is it likely that randomisation failed in this experiment? Provide evidence from the dataset for your answer.
b) Again assuming no missingness, calculate and report
i. The true average treatment effect (ATE) for all units
ii. The average treatment effect from the experiment.
Then use your answers to explain whether selection bias is negative, zero, or positive in this experiment.
c) Explain the direction of the selection bias in (b): why is it negative, zero, or positive? Provide evidence from the dataset for your answer.
d) Still assuming no missingness, use an appropriate technique to adjust your answer from (b) (ii) to come as close as you can to recover the true ATE from the experiment. How close is your new estimate to the true ATE for all units?
e) Now we’ll see what happens when some units do not finish the experiment, assuming that missingness occurs as described by the variable r. Are units in this experiment missing at random? Provide evidence from the dataset for your answer.
f) Imagine that you ask a colleague for help in dealing with missing data in this experiment. Your colleague looks at your dataset and says:
i. “You cannot estimate an ATE for always-reporters with this data”
ii. “Attrition in this dataset is explained entirely by x”.
Is your colleague correct? Provide evidence from the dataset for your answers.
This POLS0012: Causal Analysis in Data Science Assignment has been solved by our Statistics Experts at My Uni Paper. Our Assignment Writing Experts are efficient to provide a fresh solution to this question. We are serving more than 10000+ Students in Australia, UK & US by helping them to score HD in their academics. Our experts are well trained to follow all marking rubrics & referencing style.
Be it a used or new solution, the quality of the work submitted by our assignment experts remains unhampered. You may continue to expect the same or even better quality with the used and new assignment solution files respectively. There’s one thing to be noticed that you could choose one between the two and acquire an HD either way. You could choose a new assignment solution file to get yourself an exclusive, plagiarism (with free Turnitin file), expert quality assignment or order an old solution file that was considered worthy of the highest distinction.
© Copyright 2026 My Uni Papers – Student Hustle Made Hassle Free. All rights reserved.