Reference no: EM132276133
Problem Set - Questions
Question 1: RCT Design (500 wrods)
Instructions: In this question, you will be asked to design an experiment to answer a research question. You DO NOT need to collect any data for this exercise. Make assumptions about data that you need. These data don't necessarily have to exist. No more than 500 words total
There is an increasing emphasis on short- and medium-term social assistance for the needy in Singapore. The ComCare Scheme covers a significant portion of the assistance provided. To facilitate better delivery of ComCare assistance and a range of other services, the Ministry of Social and Family Development (MSF) established Social Service Offices (SSOs) in 2013. Background information can be found in the attached article "One-Stop Social Service" (Straits Times, March 7, 2015).
MSF has approached the students of LKYSPP to conduct an evaluation of SSOs. MSF has already made it clear that it cannot grant access to the beneficiary information it has collected. Therefore, no secondary data are available. However, it is open primary data collection as long as the survey team complies with the Personal Data Protection Act, 2012.
Suppose the program is still being rolled out across the island, design an experiment to learn about the effect of SSOs by answering the following questions:
1. Help MSF identify indicators that can measure the success of SSOs.
2. Based on your choice of indicators, describe your research question in one sentence.
3. Describe your experiment design and discuss data that need to be collected in order to answer the research question.
4. Discuss your empirical strategy (e.g. how to estimate the desired treatment effect using data collected).
5. Based on your experiment design and empirical strategy, what parameter will you be able to estimate and why? (e.g. ATE, ITT, LATE etc.).
Question 2: Regression Discontinuity
Discuss a scenario where you can use regression discontinuity to solve a policy problem (other than the examples discussed in class). In particular please specify the following:
a. What's the policy/research question you intend to answer?
b. What data do you need to perform the analysis?
c. What is the running variable? How do you assign control and treatment groups?
Question 3: Ground Water Access and Poverty
In many parts of India, groundwater irrigation is essential for agriculture production. Therefore access to groundwater and poverty are closely related. Please read the supplement reading "Wells, Water and Welfare: The Impact of Access to Groundwater on Rural Poverty and Conflict" and answer the following questions (you only need to read the required parts as outlined below to answer the questions.)
a. Read Introduction and Section I (pages 76-79) on the descriptions of groundwater access technologies. If we want to estimate the effect of groundwater access on poverty using RD design, what can be used as the running variable? Who should be assigned to the control group and who should be assigned to the treatment group?
b. Is it a Sharp or Fuzzy RD and why?
c. Using the groundwater level discontinuity as instrument for submersible pumps, the author estimates the effect of submersible pumps (which indicates more costly groundwater access) on poverty headcount (headcount is measured as fraction of individuals below poverty line) in the table below. What conclusions can you make from the table? Do you think this is a valid IV strategy, why/why not?
Attachment:- Assignment File.rar