A political scientist is interested in the factors that determined support for Barack Obama during the 2016 election. Using data from the 2016 American National Election Survey, the political scientist develops and estimates the following model of Obama feeling thermometer scores:

Obama feeling thermometer = a + b1(Partisan identification) + b2(Ideological conservatism) + b3(Gender) + b4(Black) + b5(Family income)

where Obama feeling thermometer is the feeling thermometer score for Barack Obama, ranging from 0 (for those with very negative feelings toward Obama) to 50 (for those who are neutral) to 100 (for those with very favorable feelings toward Obama); partisan identification is a seven-point scale of partisan identification, ranging from 0 (for strong Democrats) to 6 (for strong Republicans); ideological conservatism is a seven-point scale, ranging from 0 (for strong liberals) to 6 (for strong conservatives); gender is a dichotomous variable, coded 1 for women and 0 for men; black is a dichotomous variable, coded 1 for black respondents and 0 for all others; and family income is a 28-point scale measuring respondents' family income, ranging from 0 (lowest income) to 27 (highest income). Based on his estimation of this model using OLS regression, the following results (with expected direction of the coefficients in parentheses) are obtained:

Variable b se prob. Partisan identification [-] -9.056 0.251 -36.13 0.000 Ideological conservatism [-] -3.912 0.351 -11.14 0.000 Gender [+] 2.965 0.815 3.64 0.000 Black [+] 13.990 1.453 9.63 0.000 Household income [-] 0.065 0.052 1.27 0.204 Intercept 78.097 1.379 56.63 0.000 F ratio = 780.80 Prob (F) = 0.0001 R-squared = 0.535 N = 3,402\f