Racial disparities in federal disability benefits

Contemporary Economic Policy, Jan, 2007 by Erin M. Godtland, Michele Grgich, Carol Dawn Petersen, Douglas M. Sloane, Ann T. Walker

After estimating our initial or baseline model, we found several variables that did not represent criteria but that had a statistically significant influence on ALJ decisions. To investigate whether the effects of these variables on ALJ decisions differed by the claimant's race, we added interaction terms to our baseline model and tested their significance, both simultaneously and sequentially. Specifically, to test whether racial groups are treated differently when they are represented by attorneys, we included an interaction term between race and attorney representation. Similarly, we included an interaction term to test whether racial groups are treated differently when they are represented by persons other than attorneys. We also included interaction terms that allowed racial differences in decisions to vary by sex, earnings, translator, year of the decision, (13) and region. To test whether the award criteria are applied differently by race, we included interaction terms for severity and race, residual functional capacity and race, and occupational skill level and race.

A. Logistic Regression

We used logistic regression to estimate the model--an appropriate technique when the dependent variable is binary or has two categories, such as benefits being awarded or denied.

After testing various interaction terms, we found that the interaction term for race and attorney representation was the only statistically significant interaction term in the model. We removed the remaining insignificant interaction terms from the model to render our model somewhat more parsimonious and because removing them had little effect on our estimates of the variables left in the model. We did not, however, remove insignificant "main effects," or variables that were not interaction terms, from our models because our primary objective was to estimate the effect of race "net" of other factors that could influence the award decision, regardless of how small or statistically insignificant these other factors were.

The results of two of our models--our baseline model and our final model containing the significant interaction term--are presented in Table 3. The first numerical column in Table 3 shows the percentage of claimants in each category of every variable used in the analyses. The second and third columns present odds ratios that are estimated for each variable in our baseline model and final model, respectively. The interpretation of the odds ratio for a particular variable depends on whether the variable is a dummy variable or a categorical variable. For dummy variables, a statistically significant odds ratio that is greater/less than 1.00 indicates that claimants with that characteristic are more/less likely to be awarded than claimants without it. For categorical variables, a statistically significant odds ratio that is greater/less than 1.00 indicates that claimants in that category are more/less likely to be awarded than the claimants in the comparison category. (14)