# Small Area Health Insurance Estimates (SAHIE)

## SAHIE Age Model Methodology 2000: Model Details

The model is multiplicative; that is, we model the proportion of people with insurance as the product of a series of predictors that are mostly rates, and we model the unknown errors. To estimate the coefficients in the model, we take logarithms of the dependent and all predictor variables, except for the region indicator variables, which have the value of 1 for counties in the region and 0 otherwise. Another advantage of a multiplicative model is that it makes it plausible to maintain that the (unobserved) errors for every county, no matter how large or small, are drawn from a normal distribution, which is how they are modeled. The regression predictions are, in effect, combined with the direct CPS ASEC sample estimates. Finally, we control the county estimates to the national CPS ASEC estimates and form the state-level estimates.

### Model for the proportion of people with health insurance coverage

Dependent variable:

• log of the proportion insured in each county as measured by the 3-year average of values from the CPS ASEC.

Predictor variables:

• log of the proportion of people with family Income to Poverty Ratios (IPRs) between 200% and 300%, as estimated from tax returns;
• mean of the log IPR, as estimated from tax returns;
• variance of the log IPR, as estimated from tax returns;
• log proportions of persons under age 18 who are participants in the Medicaid program;
• log proportions of persons age 35-64 years who are participants in the Medicaid program;
• log proportion of the population who are receiving Supplemental Nutrition Assistance Program (SNAP) formerly known as the Food Stamp program;
• indicator for the West Census region;
• log of the proportion of people of Hispanic origin from demographic population estimates;
• product of the indicator variable for the South Census region and the log proportion Hispanic;
• log of the proportion of people who are American Indian or Alaska Native from demographic population estimates; and
• log proportion who are 65 or more years old from demographic population estimates.

### Model for the proportion of children under age 18 with health insurance coverage

Dependent variable:

• log of the proportion insured under age 18 in each county as measured by the 3-year average of values from the CPS ASEC.

Predictor variables:

• log of the proportion of people with family Income to Poverty Ratios (IPRs) between 200% and 300%, as estimated from tax returns;
• mean of the log IPR, as estimated from tax returns;
• variance of the log IPR, as estimated from tax returns;
• log proportions of persons under age 18 who are participants in the Medicaid program;
• log proportions of persons age 35-64 who are participants in the Medicaid program;
• log proportion of the population who are receiving SNAP;
• indicator for the West Census region;
• indicator for the South Census region; and
• product of the indicator variable for the South Census region and the log proportion Hispanic.

For further information on these variables, see information about data inputs.

Source: U.S. Census Bureau | Small Area Health Insurance Estimates |  Last Revised: August 29, 2012