Dependent Variable | Loyal Democratsa | Loyal Republicansa | Undecideda | |||
---|---|---|---|---|---|---|
Change to More Liberal (n = 192) | Change to More Conservative (n = 93) | Change to More Liberal (n = 56) | Change to More Conservative (n = 154) | Change to More Liberal (n = 175) | Change to More Conservative (n = 235) | |
Parents or close relatives | 0.029 | 0.090 | –0.131 | –0.056 | 0.110 | 0.078 |
(–0.080) | (–0.013) | (–0.017) | (–0.060) | (0.025) | (0.009) | |
Close friends or “significant others” | 0.116 | 0.074 | –0.237 | –0.024 | 0.077 | 0.079 |
(0.050) | (–0.028) | (–0.161) | (–0.047) | (–0.113) | (–0.047) | |
Other peers or coworkers | 0.149* | 0.150 | 0.112 | 0.062 | 0.220** | 0.197** |
(0.055) | (–0.081) | (0.108) | (0.026) | (0.072) | (0.146*) | |
Newspapers or magazines | 0.056 | 0.146 | 0.145 | 0.065 | 0.216** | 0.119 |
(0.074) | (0.124) | (0.047) | (–0.081) | (–0.017) | (0.048) | |
Television or radio programs | 0.021 | 0.108 | 0.113 | 0.182* | 0.283*** | 0.076 |
(–0.067) | (0.069) | (0.044) | (0.145) | (0.093) | (–0.086) | |
Social media (e.g., Facebook, Instagram, Twitter, WhatsApp) | 0.188** | 0.433*** | 0.162 | 0.114 | 0.273*** | 0.138* |
(0.068) | (0.332**) | (–0.100) | (0.047) | (0.173*) | (0.047) | |
Information shown on the Internet | 0.098 | 0.026 | 0.207 | 0.076 | 0.140 | 0.097 |
(–0.069) | (–0.319**) | (0.125) | (–0.045) | (–0.171*) | (–0.006) | |
Formal or informal meetings | 0.124 | 0.350** | –0.041 | 0.139 | 0.263*** | 0.172** |
(0.002) | (0.314**) | (–0.103) | (0.133) | (0.080) | (0.098) | |
Credibility or likability of politicians | 0.164* | 0.087 | –0.084 | 0.048 | 0.177* | 0.046 |
(0.135) | (–0.148) | (–0.001) | (–0.015) | (0.030) | (–0.051) | |
R2 (for all nine sources of information as independent variables) | 0.071 | 0.303*** | 0.119 | 0.060 | 0.133** | 0.060 |
Note: Table entries are Pearson correlations. Partial correlations, controlling for all other sources of information, age, and income, are in parentheses.
↵* p < 0.05;
↵** p < 0.01;
↵*** p < 0.001.
↵a The total number (n) in the two subsamples exceed the total number of those in the group because change to a liberal or conservative direction was measured regardless of group affiliation and included those who did not report change.