Methodology for Indiana State and County Population Projections by Race and Hispanic Origin, 2005 to 2040
The Indiana population projections by race and Hispanic origin provide an extension to the total projections by age and sex for the state and its counties that were released in 2007. Many of the assumptions on future fertility, mortality, and migration trends introduced into those projections are also implemented here. Click here to view the 2007 projections methodology. Assumptions that are unique to these projections are detailed below. Furthermore, the 2007 projection totals for the state are used as control totals for these projections by race and Hispanic origin.
The Indiana state-level population projections by race and Hispanic origin were produced using the cohort-component method. This method projects individual age, sex, race, and Hispanic origin cohorts forward through time while applying specific mortality and migration rates. In addition, fertility rates were applied to the appropriate cohorts to generate the number of births during each projection period. These projections follow the standard format of five-year age cohorts carried over five-year projection periods. The five-year cohorts begin with the 0-4 age group and extend through 80-84, with the final age group including all individuals 85 and above.
The Native Hawaiian and other Pacific Islander population was combined with the larger Asian population due to this race’s small numbers in Indiana.
Projections are provided at the county level for any county that has a total population of at least 2,000 in any given race or ethnic group. White populations in each county are projected since all meet the 2,000 population threshold. However, far fewer counties have minority populations that are sizeable enough to be projected with sufficient confidence. All counties with fewer than than 2,000 people in a particular group are pooled, projected together under the label “remainder.”
The U.S. Census Bureau’s 2005 population estimates by age, sex, race, and Hispanic origin for Indiana and its counties were used as the benchmark population. Unlike decennial population counts, annual population estimates are subject to periodic revision as estimates are released in subsequent years. However, to remain consistent with the total population projections, a modified version of the “vintage” 2005 population estimates (released in 2006) were used rather than the most recently available data. Therefore, it is likely that the 2005 benchmark population reported in these projections will differ somewhat from the revised 2005 population estimate published in later years.
There is always uncertainty associated with population projections. Future events and human behavior can be difficult to predict. The complexities associated with population data by race and ethnicity, however, provide additional sources of uncertainty. For instance, race data from census counts or vital statistics are based on self-identification of the respondent. Self-identification of persons with a multiracial background (a growing population) can shift over time.
Population projections are susceptible to two other sources of error which should be noted. First, projections for less populous areas will be subject to a greater relative margin of error than more populous areas. Predictions for Indiana, for instance, will be more reliable than projections for one of its smaller counties. Second, long-term projections will be more unreliable than near-term projections. It’s advised that data users give greater weight to projections up to 2025 or 2030.
Finally, it is important to note that these projections are strictly demographic, meaning that they rely exclusively on the recent trends and the assumptions discussed above. No economic, land use, or environmental assumptions were introduced.
Age-specific fertility rates (ASFR) for the age groups 10-14 through 45-49 were calculated for the state and each county by race and ethnicity using data from the Indiana State Department of Health for the years 2003 to 2005. The ASFRs for each year were averaged for the state and for each county by race in order to smooth year-to-year fluctuations.
The fertility assumption introduced in the total population projections that fertility rates will rise slightly of the next 35 years (based on U.S. Census Bureau projections) is applied to each sub-population with the exception of the Hispanic population. Birth records in recent years indicate that fertility rates among Indiana’s Hispanic population are much higher than the state’s on whole. The assumption is implemented that, over time, Indiana’s Hispanic population will exhibit fertility rates that are similar to the rest of the state as this population becomes more established. Therefore, the Hispanic fertility rate is converged towards the total state fertility rate over the projection period so that in 2035 they are identical.Special steps are also required to determine fertility for the “two or more” race group. A direct fertility rate cannot be derived for this population so the rate is an average of the white, black, and Asian rates. Also, the age 0-4 multiracial population in each projection cycle is a combination of births directly to parents of a multiracial background and an estimated number of births from parents of different races. Therefore, a small percentage of births in each single race group are allocated to the two or more race category. Records from recent years indicate that the number of births to parents of different races is increasing. It is assumed that this increase will continue and the percent of single race births allocated to the multiracial group will increase slightly over the projection period.
Survival rates for each age-sex group were calculated by constructing abridged life tables with five-year age groups except for the 0-1, 1-4, and 85+ population segments using 2005 vital statistics provided by the Indiana State Department of Health along with 2005 population figures. Rather than calculate unique survival rates for each race and ethnic group by county, two statewide fertility rates are calculated and applied to the state and counties. One survival rate, which is slightly higher than the other, is applied to the white, Asian, and Hispanic groups. The second rate is applied to the black and American Indian groups. An average of the two is applied to the multiracial group. These different rates are based on the reality that mortality patterns in the United States tend to vary by race.
With life expectancy expected to improve, it is assumed that survival rates for all groups will increase throughout the projection period. As with the total population projections, the increase in survival rates is based on projections from the U.S. Census Bureau.
Net migration rates for the state and each county were developed by averaging two sets of net migration rates calculated for each age, sex, race, and ethnic group. The first set of net migration rates was calculated using data from the 2000 Census on age, sex, and race specific in- and out-migration from 1995 to 2000. These data include a “some other race” classification that is not used in these projections or post-census population estimates. The migration listed for “some other race” was distributed proportionally among other race groups in order to make these data compatible.
The second set of rates were derived using a residual method which applied survival probabilities to the 2000 census population estimates by age, sex, race, and ethnicity in order to determine the expected survived population in 2005. These expected population figures were compared with the 2005 census population estimates with the difference assumed to be net migration.
Migration is the most unstable component of population projections as patterns can be subject to large swings with little warning. Indiana’s migration trends have shifted several times in the past 25 years. With this uncertainty in mind, a method was introduced to reduce migration rates after the initial projection cycle. This reduction involved converging all rates, positive or negative, toward zero in a non-linear trend that causes larger increase or decreases to occur in early projection cycles before leveling off in later cycles. This assumption is applied in the same manner and magnitude as in the 2007 total population projections.
Prepared by Matt Kinghorn, Demographer
Indiana Business Research Center
Kelley School of Business