How effective is sharing information with potential donors about previous donors’ donation amounts (that is social information) in affecting the donation behavior? Social information informs individuals about the behavior of others which can be used to increase individual donation amounts. If social information increases donations, why is that? In which situations is the effect of social information stronger? Previous research shows that donors tend to mirror other donors’ donation amounts. The idea that potential donors would adjust donation amounts based on information about the donation behavior of others is referred to as the social information hypothesis. Based on this hypothesis, showing potential donors that the average donation amount is for instance €15 would increase the likelihood that they give €15. Using social information to deliberately guide human decision-making about donations is increasingly popular. Practitioners implement social information hoping to increase donation behavior. Charities, for instance, depend on mechanisms that can stimulate donations. With practitioners depending on social information to stimulate donations, we need to have a clear understanding of how social information can increase the effectiveness of charitable stimulants.

In a review of the literature, I argue that it is difficult to infer from the cur- rent literature on social information effects to what extent social information affects donation amounts. This is because this literature reports positive, null and negative effects. Based on this, I conclude that a coherent view of social information effects is currently missing. In addition, previous studies have provided insufficient explanations to account for the effects they found (or lack thereof). As the literature provides only an incomplete view of the effects of social information, characterized by inconsistent findings, prac- titioners cannot safely apply social information as a stimulant for donation behavior. With the studies in this dissertation, I therefore aim to shed light on when and why social information affects donation behavior.

In three empirical studies, I use the average donation amounts of earlier donors in an attempt to affect participants’ donation behavior. If the donation amount is higher than the intended donation amount, the effect of social information is positive. By mentioning the average donation amount, I take advantage of the fact that giving is right-skewed; the average is often much higher than the median amount donated by previous donors.

Based on the empirical studies discussed in this dissertation, I conclude that social information modestly increases donation amounts, but it does not increase the propensity to give. More specifically, I conclude that social information mostly has a positive effect: it results in higher donation amounts without decreasing the number of donors, while also decreasing the number of low donations. Thus, free riding did not occur; as this would have implied that some donors would have used the public good without contributing. That is, if free riding had occurred, social information would have reduced the number of donations. In addition, I found no evidence to support the crowding-out effect, in which one donor’s donation amount results in the decrease of other donations. Pure altruism would stimulate people who learned about the average donation amount of €15 to reduce their own donation with €15.

Earlier studies suggest that social information effects are interpreted by potential donors as social norms (regarding how others behave or should behave). I found no support for a mediation by social norms based on ob- servational data and mediation analyses. If social information would imply a social norm and therefore affect giving, social information would result in a conformity effect, in which people would donate as others do. Considerthe example of the average of €15. If donors conform, they would copy the suggestion of €15. However, this was not the case: social information did not make the mentioned amount more popular, nor did it decrease the variation among the group presented with social information. Thus, social informa- tion did not stimulate conforming behavior. Instead, the results suggest that donors were more likely to donate amounts above the suggestion amount. We found no support for a mediation effect. While both perceived injunctive (how you ought to behave) and descriptive (how others are behaving) social norms influenced donation behavior, the manipulation of social information did not affect these norms. I conclude that social information effects are not mediated by perceived social norms.

My research encourages future studies to investigate whether and how quality signals and perceived awareness of need mediate relations betweensocial information and donation behavior. Quality signals provide informa- tion about the degree of excellence of one charity compared with other charities. The perceived awareness of need refers to the perceived necessi- ty for help, in this case a donation. First, I argue that social information could function as a quality signal, to determine if a charity or project provides good quality. I expect that people prefer to donate to projects and practitioners of higher quality, as low-quality projects are less likely to provide utility to people. If other donors contribute high amounts, donors may perceive the donation as a signal that the charity provides good quality, and a higher donation is a good investment. Second, I argue that social information could increase awareness of need. Awareness of need is essential for giving and can be increased if beneficiaries clearly communicate this need to potential donors. I argue that if others contribute high amounts, donors may perceive the donation as a signal that there really is need for help and a higher donation is needed.

In addition, social information enhanced a donor’s mood after donating. A common criticism of the use of nudges (subtle pushes towards desired behavior) such as social information is that it comes at a price: most people dislike being “nudged” because it reduces their freedom, and this could lead to an aversion to giving. However, my results of an online experiment con- ducted in the UK do not support this idea. The study demonstrated a pos- itive, rather than a negative, effect of social information on donors’ moods. The results suggest that social information does not lead to a perception of decreased control or freedom, because a positive perception of control is associated with a positive mood effect.

Furthermore, I examined the social information effect on donations in a crowdfunding context, where funding is assembled online. This is relevant,because the effect has not previously been systematically measured in such a setting. The empirical results of my thesis are relevant for practitioners in the crowdfunding business as they struggle to raise significant amounts of money. In crowdfunding projects, the empirical chapters of the dissertation suggest that social information affects giving in an online context: donors to a crowdfunding project donated higher amounts when presented with the average donation amount of previous donors. In addition, I found that the effect of social information depends on the fundraising stage. The re- sults indicate that social information is most effective in increasing donation amounts at the beginning of a campaign: at this stage, social information increased donations by 20%, whereas social information had no effect at the end of a campaign.