Create a do file where you start a log. Make sure to document your syntax add a brief explanation of every piece of code you write.

First Midterm Take-Home Section

Part 1. Using the 2016 Living Standard Survey of Nicaragua available on BrightSpace (under Assignments 1st Midterm Take-Home Section), please work on the following:

1. Create a do file where you start a log. Make sure to document your syntax (add a brief explanation of every piece of code you write and use to answer specific questions).

2. In the do file create a list of variables named “Nic_HHVar” including the household characteristics such as: household income per capita, total members living in the hh, being a poor household or not, Total hh assets, Total utilities in the hh, Living in the urban area, Violence index.

3. Also create second list of variables named “Nic_HeadHHVar” including the head of the household characteristics: age, schooling, literacy, internet user, gender, marital status, life satisfaction, etc.

4. Using the two lists of variables, describe the characteristics of the households and the characteristics of the head of the households for the whole sample. Make sure to describe what you see in the tables.

5. Is the Nicaraguan dataset a cross-sectional or a panel dataset? Explain.

6. The dataset includes a variable “satisfaction index” measuring the life satisfaction self- reported by the head of the household. This index is given in a 5-likert scale where 0 is poor, 1 fair, 2 average, 3 good, and 4 excellent life satisfaction. Describe how many head of the households reported having a poor life satisfaction and how many reported having good to excellent life satisfaction.

7. The dataset also includes the variable “Treatment” equals to 1 for households receiving remittances (money sent by their relatives living abroad) and equals to 0 for households not receiving remittances. How many households are in the treatment and comparison groups? How many head of the households are in the dataset?

8. Compare the characteristics by each group. Create a table showing the basic statistics (household and head of the hh characteristics) of each group and describe their similarities and differences.

9. Test whether the treatment and comparison groups have similar characteristics. Explain your findings. Based on your findings, would you use these groups to evaluate the impact of remittances on the poverty condition of households in Nicaragua? Explain your argument; why or why not?

Students are asked to create and submit a do file, a log file, and a written word document where you will include your answers to Q1-Q9. Please use the following filenames: Nic_test1.do; Nic_test1.smcl and Nic_text1.doc and upload them on BrightSpace.

Part2. Use the BB_2waves.dta dataset for the home-take test1. This dataset is part of a randomized intervention Vanderbilt did to assess whether educational books given to first-time mothers of low income households were an effective method for reducing post-partum depression. First time mothers (eligible) females were randomly assigned to one of two groups:

1)  the intervention group, who received free educational children books so mothers could read to their children while learning in the process, 2) the comparison group, who did NOT receive any educational books. Women were personally interviewed at their homes every 2 months since the time their babies were 2 to 18 months of age. At every personal interview mothers of the intervention group received a new book that was related to the baby’s physical and emotional development age. The BB_2waves.dta dataset is only a sub-sample of a much larger dataset. This subsample includes only data collected at baseline and 12months after the baby was born. Participants came from different hospitals. The main research question is whether the intervention (educational books) improved mother’s depression. Your job is to do the following:

Q1.Create a do file where you start a log. Make sure to document your syntax (add a brief explanation of every piece of code you write and use to answer specific questions). 

Q2.Describe the given dataset. Is this a cross-sectional or panel dataset? Who is the unit of analysis?

The outcome of analysis is mothers’ total depression. You will find two measures of the outcome of analysis:

1) awcess, a continuous Total Depression Score, where higher score means more depressed mothers;

2) awcesdep, a binary variable created to identify mothers whose depression score is in the clinical range, 1 when awcess>=10, and 0 no-depressed. Describing the dataset includes:

a. Create a clean dataset where all variables have their corresponding labels and value labels.

b. If you create any new variable make sure to add corresponding label and value label (if need it).

c. How many mothers are included in the evaluation sample? How many were assigned to the intervention and the comparison groups? Describe the mother’s characteristics by group condition. Some of the characteristics collected and included in the dataset interest are mothers’ marital status, their education, household family income, mother’s employment status (working or not), health condition, race, etc. Create a table showing the characteristics of both groups and describe their similarities and differences.

d. The variable AnyPost1 identifies the mothers who dropped from the evaluation sample after baseline. How many mothers dropped from the evaluation sample in total? How many mothers remained in the study? Can you claim that the decision of stopping participation after baseline was done at random? Explain whether this is a problem or not.

e.       At baseline, does depression correlates to mom’s age, education, being married, whether the pregnancy was planned, mother’s income, receiving public assistance, mother’s health condition, employment situation, drinking alcohol, # times changing addresses, race, or group assignment? Explain.