Researchers in health economics, epidemiology, and public health

Researchers in health economics, epidemiology, and public health often conceptualize a person’s health as the outcome of a production process with many health-related input factors. Some of the inputs believed to be important in the production of health are: expenditures on health (healthcare expenditures, household income, etc.), health-related behaviors (smoking, drinking, etc.), and health awareness/knowledge (education).
The dataset statedata_hw2.sav (to be downloaded from BB) contains state-level data from 2004-2005 on several health outcomes (your set of dependent variables): heart disease death rate, percent of adults with disability, percent of adults with diabetes, percent of adults who are overweight, as well as on potential inputs in health production (your set of independent variables): healthcare expenditures per capita, state median household income, percent of smokers, percent of binge drinkers, and high school graduation rate. Using this state-level data, empirically examine the relationship between potential health inputs (independent variables) and health outcomes (dependent variables).
1) Using SPSS make two scatter plots of two distinct relationships between health outcome variables and health input variables. For example, one scatter plot of heart disease death rate and healthcare expenditures, and another scatter plot of percent of adults who are overweight or obese and median household income. Do NOT use the combination of heart disease death rate and percent of smokers since you will use that in part 4! Referring to the scatterplots, carefully comment on form, direction, and strength of each relationship.
[SPSS command: GraphsChartbuilderScatter/DotSimple Scatterchoose variables for Y axis and X axis, click “Titles” to put a title. (Alternative: GraphsLegacy Dialogs…)]
2) For the same two relationships that you looked at in part 1, determine the correlation coefficients. In each case, carefully comment on (i) what the correlation coefficient tells you (remember to interpret the sign also!) and (ii) whether its use is appropriate given evidence for or against a linear form from the scatterplot.
[SPSS command: AnalyzeCorrelateBivariatePearson (is the default) choose the two variablesclick “OK”.]
3) Find R-square and the least squares regression line (constant term and slope) for both relationships. For each relationship make a table of the relevant information and state the estimated equation (follow the example below). Carefully comment on what is the interpretation of the slope, the constant term, and R-square (r2) in each regression.
[SPSS command: AnalyzeRegressionLinearchoose variables as Dependent variable and Independent variableclick “OK”.]

"Looking for a Similar Assignment? Get Expert Help at an Amazing Discount!"