```r > initial_lean <- "Harris supporter" > m0 <- feols(vote_chance_2 ~ conditionHarris * scale(vote_chance_1) * scale(topicZ), data = d0[lean == initial_lean]) > summ(m0) term result <char> <char> 1: (Intercept) b = 87.82 [87.11, 88.54], p < .001 2: conditionHarrisproHarris b = 2.13 [1.12, 3.14], p < .001 3: scale(vote_chance_1) b = 22.45 [21.77, 23.14], p < .001 4: scale(topicZ) b = -0.03 [-0.74, 0.69], p = .943 5: conditionHarrisproHarris × scale(vote_chance_1) b = -1.76 [-2.77, -0.75], p = .001 6: conditionHarrisproHarris × scale(topicZ) b = 0.16 [-0.85, 1.17], p = .752 7: scale(vote_chance_1) × scale(topicZ) b = 0.34 [-0.35, 1.03], p = .334 8: conditionHarrisproHarris × scale(vote_chance_1) × scale(topicZ) b = -1.39 [-2.40, -0.38], p = .007 ``` ```r > initial_lean <- "Trump supporter" > m0 <- feols(vote_chance_2 ~ condition * scale(vote_chance_1) * scale(topicZ), data = d0[lean == initial_lean]) > summ(m0) term result <char> <char> 1: (Intercept) b = 84.63 [83.74, 85.52], p < .001 2: conditionproTrump b = 2.21 [0.93, 3.49], p = .001 3: scale(vote_chance_1) b = 24.94 [24.01, 25.88], p < .001 4: scale(topicZ) b = -0.56 [-1.45, 0.32], p = .214 5: conditionproTrump × scale(vote_chance_1) b = -2.58 [-3.89, -1.28], p < .001 6: conditionproTrump × scale(topicZ) b = 0.74 [-0.54, 2.02], p = .257 7: scale(vote_chance_1) × scale(topicZ) b = 1.05 [0.11, 1.98], p = .028 8: conditionproTrump × scale(vote_chance_1) × scale(topicZ) b = -1.78 [-3.09, -0.48], p = .008 ``` binarized for plotting ```r condition topic pre_vote_likelihood_binary N <char> <char> <char> <int> 1: Pro-Harris AI Personality <98 218 2: Pro-Harris AI Personality >=98 366 3: Pro-Harris AI Policy <98 213 4: Pro-Harris AI Policy >=98 377 5: Pro-Trump AI Personality <98 208 6: Pro-Trump AI Personality >=98 351 7: Pro-Trump AI Policy <98 207 8: Pro-Trump AI Policy >=98 366 ``` ![[20241112190952 1.png]]