Assignment - Health Care Spending in the United States
Topics covered: Comparison of location of subgroups.
Data File: health.dat
Health care is currently an important national issue in the United States. Several questions are being discussed and debated. These include the extent of coverage, sources of financing, and modes of delivery. One important question that has been raised concerns the variation of health care spending across the states. Is this variation very large, or could it just be the result of random chance? Is the variation accounted for by specific demographic properties of the states?
Data on per capita health care spending for all 50 states are available for 1991. Per capita spending figures were calculated by dividing the total health spending for a state by its population. The total health spending in a state includes amounts spent in the state by nonresidents, and excludes spending by residents who go to other states for their care. Along with the per capita health spending figures, data were collected on the percentage of per capita income that was spent on health care by each of the states. The corresponding figures for the United States as a whole are $1877 per person and 11.5% of per capita income, respectively. The original data sources were the Department of Health and Human Services and the Census Bureau. The present data come from The New York Times, October 15, 1993.
The Census Bureau divides the country into nine regions, with each state falling into one and only one region. These regions are: North East, Middle Atlantic, South Atlantic, East North Central, East South Central, West South Central, West North Central, Mountain and Pacific. Analyze these data for regional variation. Comment on any special features that you note.
Does it appear that health care spending differs across different regions of the country?
It has been suggested that the breakdown of the country into nine subregions is too fine, and that a grouping of six subregions is adequate. Group the states into six meaningful subregions to see if anything important emerges from the altered grouping. Are the conclusions drawn different from those for the data based on nine subgroups? Is there a different number of subgroups that you think might be better than either of these choices for these data?
Attachment:- Assignment Files.rar