4. (.The following table is from DellaVigna and Kaplan's study of the impact of the introduction of FoxNews on voting behavior. (a) (10 marks) Consider the following Table, which shows the predictors of FoxNews availability in 2000. Each column refers to a different regression. The rows represent the variables being controlled for, as well as statistical tests and information related to them: TABLE III Dmnmmms or Fox News Avmsm, LINEAR PROBABILITY MODEL Availability of Fox News via cable in 2000 Dep. var. (1) (2) (3) (4) (5) Pres. republican vote share in 0.1436 0.6363 0.3902 0.0343 0.0442 1996 (0.1549) (0.2101)*\" (0.1566)" (0.0937) (0.1024) Pres. log turnout in 1996 0.1101 0.0909 0.0656 0.0139 0.0053 (0.0557)" (0.0348)**' (0.0278)\" (0.0124) (0.0173) Pres. Rep. vote share change 19981992 Control variables Census controls: 1990 and 2000 X X X X Cable system controls X X X U. S. House district xed X effects County xed effects X F-test: Census controls = 0 F = 3.54'\" F = 173*\" F = 1.11 F = 1.28 F-test: Cable controls : 0 F : 18.08'\" F : 2109*\" F : 18.61'\" R: 0.0281 0.0902 0.4093 0.6698 0.7683 N N = 9.256 N = 9,256 N = 9,256 N = 9,256 N = 9,256 For the authors' empirical design, why does it matter that FoxNews' entry in 2000 appears uncorrelated to political characteristics and demographic characteristics? (See the rst 2 rows and the row \"F-test: Census controls\" in the last two columns) (b) (15 marks) Restate the authors' main ndings that FoxNews appeared to change voting outcomes. How does it compare to Gentzkow's (2006) ndings about how TV decreases turnout