Overview
In our research series on nonprofit donations, we’ve looked at how certain factors can affect a nonprofit’s chances of attracting scarce dollars from potential donors. This question has become all the more important as the number of nonprofits, and in turn the competition for scarce donation dollars, continues to increase.
In this study, we examine whether reporting “effective altruism” metrics can help nonprofits increase donations. Effective altruism is an evidenced-based approach to social good that focuses on finding the most effective ways to help those in need. Cost effectiveness metrics have become a popular example of this in the nonprofit space. The idea is that people should want to donate to the most cost-effective organizations, i.e., the nonprofits helping the most people per dollar spent.
Unfortunately, research suggests that most donors are not very effective altruists. Instead, we tend to give based on emotion (Berman, et al., 2018). But that doesn’t necessarily mean that donors don’t care about cost effectiveness. Holding emotional appeal constant, it’s possible that people might be more likely to donate to a nonprofit that reports effective altruism metrics.
In this study, we examine whether reporting “effective altruism” metrics can help nonprofits increase donations. Effective altruism is an evidenced-based approach to social good that focuses on finding the most effective ways to help those in need. Cost effectiveness metrics have become a popular example of this in the nonprofit space. The idea is that people should want to donate to the most cost-effective organizations, i.e., the nonprofits helping the most people per dollar spent.
Unfortunately, research suggests that most donors are not very effective altruists. Instead, we tend to give based on emotion (Berman, et al., 2018). But that doesn’t necessarily mean that donors don’t care about cost effectiveness. Holding emotional appeal constant, it’s possible that people might be more likely to donate to a nonprofit that reports effective altruism metrics.
The Experiment
We conducted a within-subjects experiment with 200 people on the online research platform Prolific to test whether people are more likely to donate to a nonprofit that reports effective altruism metrics relative to a nonprofit that just reports generic updates. Participants were instructed to read a few brief descriptions of nonprofits, including one that reported a cost effectiveness metric and one that did not, then answer a survey question asking how likely they’d be to donate to each nonprofit.
We used two hypothetical water conservation nonprofits for this experiment, randomizing which nonprofit reported cost effectiveness metrics (“[outcome] per $1 donated) while the other simply reported “updates provided to donors annually.” The two nonprofits we created for this activity included Rain On, “an organization that helps communities efficiently recycle rainwater,” and Conservation Water, “an organization that educates consumers on how to manage water use.” Below each of these descriptions was text stating either “(Updates provided to donors annually)” for the control condition, or “(50 gallons per $1 donated in 2020)” or “(50 educated per $1 donated in 2020)” for the effective altruism condition. We also randomized the order in which the two nonprofits were presented to each participant for counterbalancing.
Below is an example to help illustrate one combination that participants could have seen.
We used two hypothetical water conservation nonprofits for this experiment, randomizing which nonprofit reported cost effectiveness metrics (“[outcome] per $1 donated) while the other simply reported “updates provided to donors annually.” The two nonprofits we created for this activity included Rain On, “an organization that helps communities efficiently recycle rainwater,” and Conservation Water, “an organization that educates consumers on how to manage water use.” Below each of these descriptions was text stating either “(Updates provided to donors annually)” for the control condition, or “(50 gallons per $1 donated in 2020)” or “(50 educated per $1 donated in 2020)” for the effective altruism condition. We also randomized the order in which the two nonprofits were presented to each participant for counterbalancing.
Below is an example to help illustrate one combination that participants could have seen.
Please read the nonprofit description below.
Rain On
An organization that helps communities efficiently recycle rainwater.
(50 gallons per $1 donated in 2020)
Rain On
An organization that helps communities efficiently recycle rainwater.
(50 gallons per $1 donated in 2020)
Please read the nonprofit description below.
Conservation Water
An organization that educates consumers on how to manage water use.
(Updates provided to donors annually)
Conservation Water
An organization that educates consumers on how to manage water use.
(Updates provided to donors annually)
Following each nonprofit description, each participant was asked “How likely would you be to donate to this nonprofit? (1 = Very Unlikely, 7 = Very Likely)” on a 1-7 scale. This served as our outcome measure of interest, in line with prior studies (Lee, et al., 2014).
Given that participants’ self-reported likeliness to donate may differ from their actual giving, we also measured whether each participant gives more than $100 per year to charity or churches, using a brief demographic survey at the end of the activity.
Given that participants’ self-reported likeliness to donate may differ from their actual giving, we also measured whether each participant gives more than $100 per year to charity or churches, using a brief demographic survey at the end of the activity.
Results
A paired samples t-test revealed no significant difference in likeliness to donate when our nonprofits reported an effective altruism metric (avg. = 4.27) versus a generic statement about reporting an annual update (avg. = 4.09), (p = 0.108). The graph below presents this result visually for ease of interpretation.
Next, we tested whether the results differed based on whether participants donate to charity in real life, over $100 annually. Approximately 38% of participants reported doing so. However, we found no interaction effect on donation likeliness between participants’ actual giving habits and whether the nonprofit reported an effective altruism metric or a generic update statement (p = 0.919).
Lastly, we tested whether the results were affected by order (effective altruism first vs. generic update first) or specific nonprofit reporting the effective altruism metric (Rain On vs. Conservation Water). There was no order effect (p = 1.000). However, there was a significant interaction effect for the specific nonprofit reporting effective altruism vs. generic update (p < 0.0001). When Rain On reported the effective altruism metric (50 gallons of water per $1 donated in 2020), participants were 0.66 points more likely to donate to Rain On than Conservation Water (p < 0.0001). However, when Conservation Water reported the effective altruism metric (50 educated per $1 donated in 2020), participants were 0.31 points less likely to donate Conservation Water than Rain On (p = 0.046). The figure below graphically presents this interaction.
A possible explanation for this difference is that the actual outcomes reported in the effective altruism metrics matter. For Rain On, conserving 50 gallons of rainwater for $1 might sound quite impressive with climate change causing water shortages in places like the Western U.S. For Conservation Water, educating 50 people about water usage for that same $1 may seem less impressive given the low cost of disseminating knowledge via the internet and social media platforms.
Conclusion
Despite the allure of using effective altruism metrics to help as many people as possible, donors do not seem as interested as we might hope. But context may matter. The significant differences between the nonprofits in this study leads us to speculate that, perhaps, the actual effectiveness of each nonprofit matters more to donors than the overarching idea of simply reporting the metrics.
To put this hypothesis to the test, we conducted a follow-up experiment testing whether a nonprofit that reports high vs. low effectiveness metrics increases potential donors’ likeliness to give.
To put this hypothesis to the test, we conducted a follow-up experiment testing whether a nonprofit that reports high vs. low effectiveness metrics increases potential donors’ likeliness to give.
References
Berman, J. Z., Barasch, A., Levine, E. E., & Small, D. A. (2018). Impediments to Effective Altruism: The Role of Subjective Preferences in Charitable Giving. Psychological Science, 29(5), 834–844.
Lee, S., Winterich, K. P., Ross, W. T. (2014). I'm Moral, but I Won't Help You: The Distinct Roles of Empathy and Justice in Donations. Journal of Consumer Research. 41(3), 678–696.
Lee, S., Winterich, K. P., Ross, W. T. (2014). I'm Moral, but I Won't Help You: The Distinct Roles of Empathy and Justice in Donations. Journal of Consumer Research. 41(3), 678–696.