The following table shows estimates from three different regressions, each based on data from the 1968 American National Election Study, where the dependent variable is individual opinions about desegregation on a scale of 1 (in favor of desegregation), 2 (in between), and 3 (in favor of strict segregation). Each model includes a different set of independent variables, with dashes indicating that a certain variable is not included in a given model. Underneath each estimate is its standard error in parentheses. Be sure to pay attention to which models are referenced in each part of the question below. (Note: Black is a dummy variable for Black respondents. Education variable scores between 1 and 10, with larger values implying higher education.) Model 1 Model 2 Model 3 (Intercept) 2.08 2.29 2.33 (0.04) (0.04) (0.04) Education 0.06 0.08 0.09 (0.01) (0.01) (0.01) Black - 0.69 0.88 (0.04) (0.09) Black Education - - 0.04 (0.02) Based on Model 2, if we imagined increasing someone's education by 1 unit, how much would we expect their opinion about segregation to change (and in what direction)