06.08.2013 Views

Projecting Fatalities in Crashes Involving Older Drivers, 2000-2025

Projecting Fatalities in Crashes Involving Older Drivers, 2000-2025

Projecting Fatalities in Crashes Involving Older Drivers, 2000-2025

SHOW MORE
SHOW LESS

Create successful ePaper yourself

Turn your PDF publications into a flip-book with our unique Google optimized e-Paper software.

projections are for all age groups. Because of confidentiality and disaggregation issues,<br />

however, we were unable to use the BEA <strong>in</strong>come projections.<br />

Census data sources conta<strong>in</strong> historical long-term <strong>in</strong>come data by various “types” of<br />

households (i.e., female head of household with no husband present, etc.), and by age groups.<br />

There are also historical data sets of median <strong>in</strong>come by states for a “typical” family.<br />

Unfortunately, <strong>in</strong>come data from the U.S. Census Bureau conta<strong>in</strong> limited disaggregation<br />

capabilities.<br />

The NHIS is a large (about 40,000 households), well-designed, cross-sectional study,<br />

conducted annually, of non-<strong>in</strong>stitutionalized <strong>in</strong>dividuals. We considered us<strong>in</strong>g NHIS to derive<br />

<strong>in</strong>come projections us<strong>in</strong>g either<br />

• Direct projection of elderly cohort-specific <strong>in</strong>come, or<br />

• Separate projections of (a) non-elderly cohort-specific work<strong>in</strong>g <strong>in</strong>come and<br />

(b) same cohort-specific decl<strong>in</strong>e <strong>in</strong> <strong>in</strong>come dur<strong>in</strong>g elderly years.<br />

Income data from the NHIS, however, had limited usage with<strong>in</strong> the model because<br />

NHIS is not a longitud<strong>in</strong>al study, and <strong>in</strong>come trends were difficult to project. Because we felt<br />

that projections of <strong>in</strong>come were a critical component <strong>in</strong> the model, ORNL issued a<br />

subcontract to a reputable firm (Standard & Poor’s DRI) hav<strong>in</strong>g experience <strong>in</strong> us<strong>in</strong>g<br />

population and <strong>in</strong>come files from the U.S. Census Bureau to obta<strong>in</strong> <strong>in</strong>come projections <strong>in</strong> the<br />

format and disaggregations desired. The DRI estimates and forecasts are expla<strong>in</strong>ed <strong>in</strong> further<br />

detail <strong>in</strong> Appendix B.2.1.1.<br />

GM Project G.6 4 - 3 October <strong>2000</strong>

Hooray! Your file is uploaded and ready to be published.

Saved successfully!

Ooh no, something went wrong!