Soc. 215. Demographic Methods and materials Morgan (pmorgan@soc.duke.edu)

Assignment 2. Due February 7, 2002 1/24/2002

Number of events, Crude rates, age-specific rates, and standardization/decomposition. You MUST turn in a hard copy of the source documents. Also please include the web address.

Identify an issue of some interest to you. Pose a hypothetical question about the role played by age structure or some other aspect of population composition. Use already collected data and calculate relevant standardized rates. You could for instance, demonstrate

a) that real changes/differences in age structure have or do not have an important impact on crude rates and estimate what proportion of the difference in crude rates can be attributed to age composition, or

b) that reasonable hypothetical changes in age structure do or do not have an impact on crude rates.

Use data from the statistical abstract of the United States (http://www.census.gov/prod/www/statistical-abstract-us.html Or census and vital registration data available on the web. Examples of topics that you could choose:

1) A demonstration that changing age structure can (or cannot) account for some changes in arrest (or victimization) rates for (some or all) crimes. How has the age structure changed between 1960 and 1995? How have crude arrest or crime rates changed? If only age structure changed, would the rates have risen or fallen? Data needed to calculate arrest or victimization rates can be found in the Statistical Abstract of the United States, http://www.census.gov/prod/2/gen/96statab/law.pdf ).

2) Compare the crude fertility or mortality rates in two states that have substantially different rates. Assume the age-specific fertility or mortality rates for the entire U.S. applies to both states. Now apply the age structure of these states to the U.S. rates. How much of a difference does age structure make? Why?

(See Statistical Abstract of the United States : http://www.census.gov/prod/www/statistical-abstract-us.html.

Or compute the CDR for various North Carolina counties. Are the observed differences due to differing age-specific rates or to different age (and perhaps race) distributions? Find numerators for rates at: http://hermes.sches.ehnr.state.nc.us/SCHS/healthstats/deaths/dms/dms-pre.html . Find denominator (for 1990) at : http://venus.census.gov/cdrom/lookup/ You may assume that the 1990 Census numbers apply to the 1995 or 1996 deaths.

3) Chose two states (or counties) that have different levels of poverty. Determine whether these differences can be attributed to differences in age structure. Note that the unemployment rate is a prevalence measure (a ratio of unemployed to employed) not a standard occurrence-exposure demographic rate. However, the procedures in Chapter 2 can be applied to these types of ratios (rates).

4) Can recent changes in crude birth (or death) rates be explained by the changing U.S. age structure?

(http://www.census.gov/prod/2/gen/96statab/vitlstat.pdf; For example see Table 121 for time trend in crude death rates; See http://www.census.gov/prod/2/gen/96statab/pop.pdf for information on population distribution)