You just got an internship with the Bill and Melinda Gates Foundation. Your boss, Nalina Canagarajah, has long been interested in microcredit programs and wants you to look at a specific case to examine microcredit programs' effects. The Grameen Bank of Bangladesh is one of the best-known and oldest microcredit providers. The data set that Ms. Canagarajah has provided information about Grameen microcredit programs, participation, and household outcomes. The Grameen bank separates its programs into those for men and those for women. Some villages have no programs, some have only for men, some only for women, and some have both. Furthermore, only households with less than 50 decimals of land, equal to 0.5 acres, are officially eligible to participate in microcredit programs (the "target group").

The outcome Ms. Canagarajah is particularly interested in is household per capita expenditure (measured in Taka) because that captures whether microcredit can help households escape poverty. She wants you to examine the following major questions:

1. What is the effect of having a male-only microcredit program in the village?
2. What is the effect of having a female-only microcredit program in the village?
3. What is the effect of having a female from the household participate?
4. What is the effect of having a male from the household participate?

Before you can do these analyses, you have to set up a model that can explain a household's per capita consumption level. It is standard in the literature to use the log of per capita expenditure/consumption as the dependent variable. When deciding on variables, think about which variables either impact consumption/income levels or capture differences between different villages. Make sure you do not include variables that do not belong, such as other expenditure measures.

The data below are a subset of the Bangladesh Household Survey 1998/1999, conducted jointly by the Bangladesh Institute of Development Studies and the World Bank. Information was collected at the individual, household, and community levels. The variables and their definitions are listed on the last page.

To access the data, you should include the following line in your R script (requires an internet connection when you run the script): load(URL("https://www.dropbox.com/s/p6q5tl7ra5y1qva/micro_credit.rdata?raw=1"))