Table 4 Tests of the trust hypothesis (that a robot’s justifications, compared to Mere explanations, increase trust) in Experiment 1 (Hunger strike dilemma) and Experiment 2 (Hunger strike and DNR dilemma).
From: The power of justifications to repair human-robot trust, even under moral disagreement
Experiment 1 Hunger | Experiment 2 Hunger | Experiment 2 DNR | |
|---|---|---|---|
Total trust | \(\:{\eta}_{p}^{2}\) = 2.4% | \(\:{\eta}_{p}^{2}\) = 2.8% | \(\:{\eta}_{p}^{2}\) = 2.7% |
F(1,331) = 8.1, p =.005 | F(1,346) = 9.8, p =.002 | F(1,396) = 11.1, p <.001 | |
Moral trust | \(\:{\eta}_{p}^{2}\) = 4.9% | \(\:{\eta}_{p}^{2}\) = 1.9% | \(\:{\eta}_{p}^{2}\) = 4.8% |
F(1,331) = 17.2, p <.001 | F(1,346) = 6.6, p =.01 | F(1,397) = 19.9, p <.001 | |
Performance trust | \(\:{\eta}_{p}^{2}\) = 0.2% | \(\:{\eta}_{p}^{2}\) = 2.5% | \(\:{\eta}_{p}^{2}\) = 0.4% |
F(1,339) < 1, p =.381 | F(1,355) = 9.2, p =.003 | F(1,405) = 1.4, p =.229 | |
Moral trust, controlling for Performance trust | \(\:{\eta}_{p}^{2}\) = 5.7% | \(\:{\eta}_{p}^{2}\) = 0.2% | \(\:{\eta}_{p}^{2}\) = 5.1% |
F(1,330) = 19.8, p <.001 | F(1,345) < 1, p =.407 | F(1,395) = 21.1, p <.001 | |
Performance trust, controlling for Moral trust | \(\:{\eta}_{p}^{2}\) = 0.8% | \(\:{\eta}_{p}^{2}\) = 1.1% | \(\:{\eta}_{p}^{2}\) = 0.7% |
F(1,330) = 2.7, p =.103 | F(1,345) = 3.9, p =.049 | F(1,395) = 3.0, p =.086 |
