The dataset contains information on survey respondents' opinions about various public figures. These are "feeling thermometer" scores, which range from 0 (total dislike of the person) to 100 (total like). The relevant variables for this question are: white: a dummy variable indicating whether the respondent is white or not ideology: the respondent's ideology on a scale of 1 (most liberal) to 7 (most conservative) Biden: the respondent's "feeling thermometer" score for Joe Biden Following is the regression equation with Biden as the dependent variable and white and ideology as independent variables: (# is the error term. You do not need to care about it) Biden= + + + Regression results are following: Estimate Std. Error t value Pr(>|t|) (Intercept) 111.2059 2.8712 38.73 <2e-16 *** white -21.2645 2.1153 -10.05 <2e-16 ***ideology -8.7184 0.6144 -14.19 <2e-16 *** --- Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1 (a) What is the estimated intercept for white respondents? What about for non-white respondents?

(b) Regression including a term for the interaction between white and ideology: Biden= + + +()+ Estimate Std. Error t value Pr(>|t|) (Intercept) 90.373 4.646 19.452 < 2e-16 ***white 8.277 5.645 1.466 0.1431 ideology -3.061 1.171 -2.613 0.0092 ** I(white * ideology) -7.669 1.364 -5.623 2.84e-08 *** Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1 Is the relationship between ideology and Biden opinion stronger for white respondents, or non- white respondents? (Hint! Think only about the slope)

(c) Regression including a term for the interaction between white and ideology: Biden= + + +()+ Estimate Std. Error t value Pr(>|t|) (Intercept) 90.373 4.646 19.452 < 2e-16 ***white 8.277 5.645 1.466 0.1431 ideology -3.061 1.171 -2.613 0.0092 ** I(white * ideology) -7.669 1.364 -5.623 2.84e-08 *** Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1 What are the estimated intercept and slope of ideology among white respondents? What about among non-white respondents?