Quantitative Methods in Policy Analysis

Fall 2007

Problem Set #7

1.

Use the welfare.xls dataset to answer the following questions about the participants in the Arkansas welfare-to-work experiment:

A.

What proportion of control-group participants received AFDC payments in the last quarter after random assignment? Give a 95% confidence interval.

B.

Repeat for the experimental group. Compare your answer to part A. What do you conclude?

C.

Did the experiments reduce the percentage of women receiving AFDC payments? Give a 95% confidence interval for the difference in the proporiton recieving AFDC payments in the last quarter. Compare your results to those of Gueron.

"Infant mortality" refers to the death of infants within the first year of life. The infant mortality rate is defined as the number of infant deaths within the first year of life per 1,000 live births. The U.S. average in 1999 was 7.1; in the 50 largest metropolitan areas, the average was 7.9 infant deaths per 1000 live births.

In 1999, there were 7,513 live births in the District of Columbia; 113 of these infants died in their first year of life. In 2000, 91 of 7,666 infants died.

A.

Compare the infant mortality rate in DC in 1999 to the average for the 50 largest metropolitan areas. State null and alternative hypotheses, test the null hypothesis, compute the corresponding p value, and give your conclusions in plain English.

B.

Compare the infant mortality rate in DC in 1999 to the the rate in DC in 2000. State null and alternative hypothesis, test the null hypotheses, compute the corresponding p value, and give your conclusions in plain English.