13. For the next set of questions, consider the following results from regressing the rating of Obama as president on 13 independent variables: Table 2. Predictors of Obama rating Standard t-statistic Coefficient Error Republican -21.42 2.15 Democrat 23.19 2.15 Ideology (conservative =high) -10.23 0.25 Perceptions of the US -8.30 0.47 economy (better) Perceptions of family finances -3.13 0.31 (better) Dog -4.95 0.60 Cat -2.26 0.62 Horse -5.04 2.24 Reptile 0.46 1.77 Ferret/Gerbil -1.76 1.52 Bird 0.71 1.35 Fish -0.29 0.86 Other pet 3.0 1.54 Constant/Intercept 67.87 2.46 N = 17,092, R2 = 0.753 The unit of analysis is randomly selected individuals responding to a national survey Variable key Obama rating: 0 to 100 rating of how the individual feels about Obama as president Republican: coded 1 if the individual identifies as a Republican and 0 otherwise. Democrat: coded 1 if the individual identifies as a Democrat and 0 otherwise. Ideology: the individual's identified ideology, scored as: 1 Extremely liberal; 2 Liberal; 3 Slightly liberal; 4 Moderate, middle of the road; 5 Slightly conservative; 6 Conservative; 7 Extremely conservative Perceptions of the US economy (better): coded 1 if the individual felt the US economy had gotten better over the past year and 0 otherwise. Perceptions of family finances (better): coded 1 if the individual felt their family finances had gotten better over the past year and 0 otherwise. Dog through Other pet: each variable is coded 1 if the individual owns such a pet (e.g. a dog) and 0 otherwise i. Calculate the t-statistic for each estimated coefficient in the table. [6]ii. If t-statistic of 2 or greater produces a P-value of less than 0.05, what can you conclude regarding the null hypothesis that dog ownership has no effect on an individual's rating of Obama? [3] iii. What is the magnitude of the effect of owning a cat versus not owning a cat? [4] iv. What can you conclude about how well this model (Table 2) fits the data? [3] 14. For this question, consider the following cross tabulation from a survey of black and white Americans comparing their racial identity to their partisan identification: Democrat, Republican or Independent. Table 3. Racial and Partisan Identities Candidate Black White Row total Democrat 579 27% 34% Independent 29% 45% 43% Republican 4% 28% 23% Column total 100% 100% 100% V = 2196; x2 statistic: 230.5; p-value