Hacker Newsnew | past | comments | ask | show | jobs | submitlogin

Thanks. To add to your quote above, from the paper:

> In Study 2, we extend our investigation by considering alternative explanations. In Study 1, it is possible that our con- trol condition decreased benevolence-based trust. In the control condition in Study 1, the confederate sent no message. This might have seemed impolite. To investigate this alternative explanation, we include two different comparison conditions in Study 2. We include a traditional apology (‘‘I’m sorry to interrupt.’’) and neutral greeting (‘‘How are you?’’).

> In addition, we extend our investigation by exploring superfluous apologies in a new domain. In Study 1, both the unfortunate circumstance (the random action of the computer) and the outcome (behavioral measure of trust) were related to the trust game. In Study 2, the apology (for a flight delay) is unrelated to our measure of trust (lending a cell phone). We expect that an apology for a flight delay will increase trust even when the subsequent interaction is unrelated to the flight.



So in Study 2, they do seem to include a non-superfluous apology.

Study 4 (the one about the rain, and the topic of the blog post) doesn't have the same controls. Why? I find this pretty odd.




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: