The price we pay for the money we raise

The price we pay for the money we raise

The Development Engagement Lab’s recent experiment comparing the effects of positive and negative video campaign appeals shows little difference in the appeals’ ability to generate donations in the short-term, but reveals something startling: Some (but not all) of the emotions associated with negative appeals seem to act as a drain on respondents’ efficacy, or the feeling that they can make a difference, which we know from previous DEL work can lead to a dampening of engagement. The positive appeals, by contrast, boost efficacy. This signals the need for a mixed strategy for fundraising, one that balances the advantages and disadvantages of positive and negative imagery.

In part one of this blog series, we’ll explain what the experiment revealed about differences between positive and negative appeals in terms of donations and email capture – both on average and by respondents’ level of engagement – as well as the appeals’ impacts on efficacy, associated with giving and engagement. In part two, we’ll dive deeper into what the experiment reveals about the emotions engendered by positive and negative appeals, and how they impact engagement.

To read more about the methodology and the 'replicability' of the giving environment, click here.

The British public are split about whether development aid organisations are being honest when they claim that a donation can save a life. In a poll from mid-June, 30% said they believe the claim and 30% didn’t, with another 30% saying they neither agreed nor disagreed (the remainder said ’Don’t know’.) Less than half of those who already give in some way (‘Transactionally Engaged’) to these organisations believe the claim. Read that again: Less than half that already give in some way – be it donations or through purchasing goods or services – actually believe the claims of many of the organisations to which they donate.

If this seems like a precarious place for a significant portion of the sector’s funding to hang, consider too that our latest data shows yet another dip in those who say they’ve donated in the last year – the steepest since DEL began tracking in 2019 – from 19% in January to 13% in June.

Many ask whether the sector's fundraising tactics are to blame. Does the data bear this out?

What does it mean? Many point to an erosion of public trust in aid institutions. It’s true that trust in NGOs is low: In June 2021 29% said they trust development organisations, though this increased 3% from January figures. And, only 22% of people in Great Britain believe aid is ‘effective’ or ‘very effective.’

One popular theory asks whether the sector’s fundraising tactics might be to blame; namely the reliance on depictions of human suffering and extreme need. Research and years of trial and error have cast the reliance on negative tactics as tried and true for generating donations, but many are asking whether decades of exposure has begun to numb the public’s reflex to give. There are also questions about whether the single-minded use of these tactics over long periods of time has deprived the public of a sense of progress, masking the myriad strides to which overseas aid has contributed, thereby stunting the public’s feelings of ownership and pride toward successes. Could the sector be contributing to public’s feelings of stagnation, helplessness, and indifference in the face of global crises?

One of the most frequent questions about the results is: How does this actually compare with real-world giving?

We decided to test this theory. Our Partners volunteered 30-50-second clips which we categorised as either generally ‘positive’ or ‘negative.’ We selected appeals that focused on long-term development issues like hunger, nutrition, women and girls and water, sanitation and hygiene (WASH). In order to estimate the full impact of the positive and negative videos, respondents (N = 3,025; sample is nationally representative of GB adults) had to watch the full clip before moving on.

A note on replicating the ‘giving environment’

Among the questions most frequently asked to those researching donation and giving behaviour is this: How can the real-world giving environment truly be replicated in a ‘lab’ setting? The short version is that while no one would claim a one-to-one correspondence between reported intentions and behaviour, there is enough information to infer real patterns and tendencies. The longer version is here. But essentially, there is now a decent body of experimental research that has tested the difference real or hypothetical money makes or the intention-behavior gap (and how to close it), that means that we have a solid level of confidence in the validity of our results.


To test the impact of negative and positive video appeals on donations, respondents were assigned one out of ten total appeals. After watching the clip, respondents were asked how much, if any of £20 they would donate. The figure on the right shows donations by DEL engagement segment, which breaks down as follows:

• Negatively Engaged – take critical action against ending global poverty

• Totally Disengaged – take no action against global poverty

• Marginally Engaged – minimal action (read, watch, listen, discuss)

• Transactionally Engaged – donate or purchase/boycott

• Purposively Engaged – ‘big actions’ (volunteered, became member, MP, voice or march)

• Fully Engaged – 3 or more of big actions

As you can see in Figure 1, there was little difference between the positive and negative appeals in terms of respondents’ average donation. On average, negative appeals garnered slightly more – 6.25 compared with 6.04 to positive appeals – but the difference is not statistically meaningful. The implication of this is big: The evidence suggests that positive appeals can generate as much in donations as negative appeals.

Figure 1: Negative & positive appeals earn roughly the same amount in donations

When we looked at the donations based on levels of engagement with global poverty (Figure 2 below) – ranging from ‘Negatively Engaged’ to ‘Fully-Engaged’ – we found that while the Fully Engaged group (about 3 percent of the British public), on average donated a full dollar more to positive campaigns than negative ones, the result was not statistically different. In other words, even among the most engaged audiences, positive and negative appeals play similarly.

Figure 2: Donations to positive and negative appeals by engagement

Offered the opportunity to sign up for direct debit donations, we also found that there was no difference between positive and negative appeals for likelihood to sign up: About 2.3% opted for direct debit for both appeal types. And when we looked at respondents’ level of engagement, the result was the same: No statistical difference. Finally, when asked if they would donate again in the future, the respondents once again proved there is no difference between positive and negative appeals.

For those making the transition to more positive imagery – out of concern for donor fatigue or the sometimes racist or imperialist framings of negative imagery – the news should be comforting: Our data shows no difference in the potential for positive imagery to generate donations.

Email capture

There’s good news for those trying to balance the emotional impact of their fundraising appeals: positive videos are associated with a greater likelihood among respondents to share their email address for future engagement.

Figure 3: Willingness to sign up for more info

On average, respondents who saw positive appeals were slightly more likely to give their email address – 2.595% compared with 2.519. While this doesn’t seem particularly large - the difference between gaining 260 and 250 sign-ups in a campaign going to 10,000 people – it’s still statistically significant. The contrast becomes more meaningful when we look at the breakdown of respondents by their engagement habits.

Figure 4: Email sign-up by engagement

Efficacy: The golden ticket to engagement

DEL research has established that personal efficacy – or the belief that one can make a difference – is positively correlated with donations, concern for poverty in poor countries and feelings of ‘global interconnectedness.’ In other words, people who believe they can make a difference in the world are more likely to donate, feel concerned about poor countries and feel interconnected.

By evaluating respondents’ feelings of efficacy after watching appeal clips, we found that positive appeals significantly increased personal efficacy compared to negative appeals. On a scale of 0-10 (‘Can’t make any difference at all’ to ‘Can make a great deal of difference’) those who saw the positive appeal rated their efficacy at 5.2, compared with those who watched the negative appeal, who scored 4.6.

Figure 5: Positive appeals increase efficacy

Breaking this down by audience engagement, it’s also worth noting that positive imagery significantly increased efficacy for the Marginally Engaged.

The case for balance

So what does this mean for our comparison of positive and negative appeals? Our data shows that while positive and negative appeals have similar effects in the short-term for elicitng donations, positive appeals have the added benefit of increasing efficacy, which has been shown to increase the likelihood to donate, care more about poor countries and facilitate a sense of global interconnectedness. We can’t say yet whether negative appeals have an opposite effect, numbing the reflex to donate, but the anecdotal evidence is mounting. At the very least, there’s a case to be made for incorporating more positive imagery as an insurance policy, making the most out of the powerhouse potential of efficacy. For organisations looking to improve outcomes, it’s not a zero-sum game: both types of appeals likely have a place in the toolbox.

In our next blog in this series, we’ll elaborate on this balance by exploring the results of the emotion testing element of the experiment, which looks at how certain emotions triggered by appeals have unique impacts on the public’s willingness to give and engage.

Written by

Molly Anders

Molly Anders

Research Insights and Engagement Lead at the Development Engagement Lab

Jennifer Hudson

Jennifer Hudson

Professor of Political Behaviour at University College London (UCL)

Soomin Oh

Soomin Oh

Soomin Oh is a postdoctoral research fellow at Development Engagement Lab. Her research focuses on the political economy of development and looks at the patterns of international and domestic redistribution. She has worked on impact evaluations of development projects in collaboration with USAID and the World Bank in the West Bank, Nepal, and Tajikistan. Soomin received her PhD in Political Science from Duke University, specialising in Political Economy.

David Hudson

David Hudson

Professor of Politics and Development in the University of Birmingham

Write for us

[email protected]

Related Blogposts