10.1 Consider the following simulated results from a multiple regression using state-level data in the United States. The dependent variable is the state’s smoking rate (measured as a percentage). The independent variables are the tax on cigarettes, measured in cents per pack, and a dummy variable indicating whether or not the state has a tobacco industry. The results are as follows:

SmokeRate = 15 − .5 × CigTax + 20 × TobaccoIndustry

Interpret the constant term, the coefficient of CigTax, and the coefficient of TobaccoIndustry. Make sure to state the units of each.

10.2 Consider a regression of medical costs on the independent variables Type1Diabetes and Type2Diabetes, described in Box 10.3. The unit of analysis is the patient.

What information does the coefficient of Type1Diabetes provide? What information does the coefficient of Type2Diabetes provide?

Suppose that, using data from many patients at many hospitals, the coefficient of Type1Diabetes is $300 and the coefficient of Type2Diabetes is $150. What is the expenditure difference between type 1 and type 2 diabetes predicted to be?

10.5 in the picture below

10.8 What multivariate method is suitable for the following research purposes?

Reducing a set of attitude questions in a survey to a smaller number of attitude dimensions.

Estimating the links between variables in a causal structure with several independent variables and several intervening variables.

Examining how school-level characteristics and classroom-level characteristics influence individual test scores of elementary school students.

Testing the relationship between increasing tobacco taxes and decreasing rates of smoking in the United States from 1960 to 2010.

Tracking a sample of 100 small start-up businesses regularly each year to see how differences in access to financing impact their survival over time.