This dissertation presents two lines of research examining how concerns about reputation influence our judgments and decisions. The first line of research explores why people fail to prioritize the cost-effectiveness of charities when donating, despite the potential to significantly increase the impact of their contributions. I examine whether this tendency arises in part due to reputational incentives that favor empathizing with donation recipients instead of deliberating about the cost-effectiveness of charities. Across eight studies, I found that “deliberators” were perceived as less moral and less desirable as social partners than “empathizers”. People correctly anticipated the negative reputational costs of deliberation and were more likely to donate to more empathic but less effective causes when reputational concerns were emphasized. However, a field experiment found that people were less likely to donate to an empathic cause when their donation choices were public rather than private, even though they believed that donating empathically would benefit their reputation. These results suggest that while there are disincentives for selecting charities by deliberating about their cost-effectiveness, the signaling dynamics are nuanced.The second line of research reevaluates prior evidence documenting systematic errors in social perception, such as the tendency for people to underestimate how much others like them. Across three studies, I investigate whether people’s responses to a range of paradigms examining estimated versus actual reactions are likely influenced by social desirability concerns. Participants were presented with prior paradigms from the perspective of “forecasters”, who predicted how an interaction partner would think or feel, or “evaluators”, who reported how they themselves would think or feel. I found that evaluators were more likely than forecasters to believe that exaggerating their positive evaluations was the socially desirable response and were more inclined to exaggerate their positive evaluations. These results were sizable, obtained across a range of paradigms, and testify to the likely influence of social desirability concerns in prior studies in this literature. I conclude by outlining several ways that researchers can mitigate these concerns.