How often do we replicate previous research?

Surprisingly little – at least to my taste. Science is supposed to be cumulative: we should build knowledge on solid foundations. But how do we know what knowledge is solid? Through independent verification. If someone claims to have found a regularity, it is only through repeated testing of the same claim that we know that the result holds in other populations and contexts. It turns out that scientists rarely conduct such replications. Here are all the estimates of the prevalence of replications in various scientific disciplines that I could find in an extensive search on Google Scholar.

Some observations on the prevalence rates above.

The replication prevalence in psychology increased strongly after 2013. The detection of fraud by Diederik Stapel and the publication of a paper on ‘precognition’ by Daryl Bem may have contributed to this increase.

In marketing, in contrast, fraud cases such as those of Dirk Smeesters and Brian Wansink did not increase the prevalence of replications. In the period 2000-2011 the proportion of articles replicating previous research was lower than in the preceding period 1990-2004.

So called ‘top’ journals – publishing papers with a higher citation count on average – are more likely to publish replications in criminology. This also holds generally speaking in the social sciences – see the graph below.

There is no hierarchy of the sciences in the prevalence of replications. The “hard sciences” do not attempt replications more often than the social sciences – it’s the other way around. Publications in economics are far less likely to attempt replications than publications in marketing or in communication science. The prevalence of replications in ecology is devastatingly low at 0.03% (11 out of 38730 published papers).

If you know of a study I have missed that allows for an estimate of the prevalence of replications, please let me know. I could not find studies in public administration, political science, or sociology. The lack of studies documenting the prevalence of replication attempts in public administration and political science is surprising because in these disciplines several journals considered to be leading in the field have introduced data availability policies.

The table below and the ppt with the graphs above contains the data and references to the sources represented.

Field	Rate	Reference	Remarks
Social Sciences	1.3% (2/156)	Hardwicke T.E., Wallach, J.D., Kidwell, M.C., Bendixen,  T., Crüwell, S. & Ioannidis, J.P.A. (2020). An empirical assessment of transparency and reproducibility-related research practices in the social sciences (2014–2017). Royal Society Open Science, 7(2), 7190806190806. https://doi.org/10.1098/rsos.190806	Random sample of 250 articles, 2014-2017
Social sciences	2.8% (44/1559)	McNeeley, S., & Warner, J. J. (2015). Replication in criminology: A necessary practice. European Journal of Criminology, 12(5), 581-597 https://doi.org/10.1177/1477370815578197	5 ‘top’ journals, 2006-2010
Natural sciences	1.4% (94/6637)	McNeeley, S., & Warner, J. J. (2015). Replication in criminology: A necessary practice. European Journal of Criminology, 12(5), 581-597 https://doi.org/10.1177/1477370815578197	5 ‘top’ journals, 2006-2010
Discipline	Rate	Reference	Remarks
Psychology	5.32% (10/188)	Hardwicke, T. E., Thibault, R. T., Kosie, J. E., Wallach, J. D., Kidwell, M. C., & Ioannidis, J. P. (2022). Estimating the prevalence of transparency and reproducibility-related research practices in psychology (2014–2017). Perspectives on Psychological Science, 17(1), 239-251. https://doi.org/10.1177/1745691620979806	Random sample of 250 articles, 2014-2017
Communication science	3.7% (21/562)	Keating, D. M., & Totzkay, D. (2019). We do publish (conceptual) replications (sometimes): Publication trends in communication science, 2007–2016. Annals of the International Communication Association, 43(3), 225-239. https://doi.org/10.1080/23808985.2019.1632218	10 journals, 2007-2016
Advertising	2.87% (82/2856) interstudy replications	Park, J. H., Venger, O., Park, D. Y., & Reid, L. N. (2015). Replication in advertising research, 1980–2012: a longitudinal analysis of leading advertising journals. Journal of Current Issues & Research in Advertising, 36(2), 115-135. https://doi.org/10.1080/10641734.2015.1023874	4 ‘top’ journals, 1980-2012
Criminology	2.32% (16/691)	McNeeley, S., & Warner, J. J. (2015). Replication in criminology: A necessary practice. European Journal of Criminology, 12(5), 581-597 https://doi.org/10.1177/1477370815578197	5 ‘top’ journals, 2006-2010
Experimental linguistics	1.81% (153/8437)	Kobrock, K., & Roettger, T. B. (2023). Assessing the replication landscape in experimental linguistics. Glossa Psycholinguistics, 2(1). http://dx.doi.org/10.5070/G6011135	100 journals, 1945–2020
Marketing	1.01% (12/1185)	Kwon, E. S., Shan, Y., Lee, J. S., & Reid, L. N. (2017). Inter-study and intra-study replications in leading marketing journals: a longitudinal analysis. European Journal of Marketing, 51(1), 257-278. https://doi.org/10.1108/EJM-07-2015-0450	5 ‘top’ journals, 2000-2011
Marketing	1.70% (41/2409)	Evanschitzky, H., Baumgarth, C., Hubbard, R., Armstrong, J.S. (2007). Replication research’s disturbing trend. Journal of Business Research, 60, (4), 411-415. https://doi.org/10.1016/j.jbusres.2006.12.003	5 ‘top’ journals, 1990-2004
Business and management	1.48% (1235/83682)	Ryan, J. C., & Tipu, S. A. (2022). Business and management research: Low instances of replication studies and a lack of author independence in replications. Research Policy, 51(1), 104408. https://doi.org/10.1016/j.respol.2021.104408	All 121 journals, 2008-2017
Psychology	1.08% (347/321411)	Makel, M. C., Plucker, J. A., & Hegarty, B. (2012). Replications in psychology research: How often do they really occur?. Perspectives on Psychological Science, 7(6), 537-542. https://doi.org/10.1177/1745691612460688	Random sample of 500 articles, 1900-2012
Criminology	0.45% (178/39275)	Pridemore, W. A., Makel, M. C., & Plucker, J. A. (2018). Replication in criminology and the social sciences. Annual Review of Criminology, 1, 19-38. https://doi.org/10.1146/annurev-criminol-032317-091849	All journals, until 2014
Addiction Medicine	0.41% (1/244)	Adewumi, M. T., Vo, N., Tritz, D., Beaman, J., & Vassar, M. (2021). An evaluation of the practice of transparency and reproducibility in addiction medicine literature. Addictive Behaviors, 112, 106560. https://doi.org/10.1016/j.addbeh.2020.106560	All journals in PubMed, 2014-2018
Management	0.15% (240/159242)	Block et al. 2022 https://doi.org/10.1007/s11301-022-00269-6	56 top journals until 2020
Economics	0.10% (130/126505)	Müller-Langer et al. 2019 https://doi.org/10.1016/j.respol.2018.07.019	Top 50 journals, 1974-2014
Educational science	0.13% (221/164589)	Makel et al. 2014 https://doi.org/10.3102/0013189X14545513	Top 100 journals, until 2013
Ecology	0.02% (11/38730)	Kelly, C. D. (2019). Rate and success of study replication in ecology and evolution. PeerJ, 7, e7654. https://doi.org/10.7717/peerj.7654	All 160 journals, until 2017; Python code at https://doi.org/10.17605/OSF.IO/WR286

Inequality and philanthropy

In the public debate, the rise in inequality is linked to criticism of private philanthropy, not only as being a strategy to reduce feelings of guilt, but also as a way to evade taxes, buy goodwill, and favor causes that are benefiting the rich rather than society as a whole.

Rutger Bregman famously called for increased taxation of the rich, instead of praise for their philanthropy. The two are not mutually exclusive, as the graph below shows. In fact, there is no relationship at all between the volume of philanthropy in a country and the tax burden in that country.

The data on government expenditure are from the IMF. The data on philanthropy for 24 countries in this graph come from a report by the Charities Aid Foundation in the UK (CAF, 2016). These data are far from ideal, because they were gathered in different years (2010-2014), and using different methods. So for what it is worth: the correlation is zero (r= .00).The United States is a clear outlier, but even when we exclude it, the correlation remains zero (r= -.01).

The more reliable evidence we already have is on the proportion that gives to charity in different countries from the Gallup World Poll. This is currently the only source of data on engagement in philanthropy in a sizeable number of countries around the globe, even though the poll includes only one question to measure charitable giving. The surprising finding is that countries in which citizens pay higher taxes have a higher proportion of the population engaging in charitable giving. The correlation for 141 countries is r= .28. Within Europe, the association is even stronger, r= .49.

As you can see, and as others have noted (Salamon, Sokolowski & Haddock, 2017) the facts do not support the political ideology that keeping the state small makes people care for each other. In contrast: countries in which citizens are contributing less to public welfare through taxes are less involved in charity. To me, this positive relationship does not imply causation. I don’t see how paying taxes makes people more charitable, or vice versa. What it means is that the citizens of some countries are more prepared to give to charity and are also willing to pay more taxes.

Some further evidence on the relation between redistribution effort and philanthropy comes from an analysis of data from the Gallup World Poll and the OECD, collected for a grant proposal to conduct global comparative research on philanthropy.

The correlation between income inequality after taxes and the proportion of the population giving to charity is weakly negative, r = -.10 across 137 countries. In contrast, income inequality before taxes shows a weakly positive relation with the proportion of the population that gives to charity, r = .06.

This implies that in countries where the income distribution becomes more equal as a result of the income tax (‘redistribution effort’) a higher proportion of the population gives to charity. In countries that are more effectively reducing income inequality the proportion that gives to charity is higher. However, the correlation is not very strong (r = .20). The figure below visualizes the association.

The chart implies that countries in which the population is more engaged with charitable causes are more effectively reducing income inequality. My interpretation of that association is a political one. A stronger reduction of income inequality is the result of effort and effectiveness of progressive income taxation, a political choice ultimately supported by the preferences of voters. The same prosocial and inequality aversion preferences lead people to engage in charitable giving. Restoring justice and fairness in an unfair and mean world are important motivations for people to give. Countries in which a higher proportion of the electorate votes for reduction of income inequality are more charitable. 

Strictly speaking, the chart does not tell you whether income inequality causes giving to be lower. However, there is enough evidence supporting a negative causal influence of income inequality on generalized trust (Leigh, 2006; Gustafsson & Jordahl, 2008; Barone & Mocetti, 2016; Stephany, 2017; Hastings, 2018; Yang & Xin, 2020). Countries such as the UK and US in which political laissez-faire has allowed income inequality to rise have become markedly less trusting over time. Trust is an important precondition for giving – more about that in another post.

This post builds on Values of Philanthropy, a keynote address I gave at the ISTR Conference in Amsterdam on July 12, 2018. Thanks to Beth Breeze and Nicholas Duquette for conversations about these issues.

References

Barone, G., & Mocetti, S. (2016). Inequality and trust: new evidence from panel data. Economic Inquiry, 54(2), 794-809. https://doi.org/10.1111/ecin.12309

Bekkers, R. (2018). Values of Philanthropy. Keynote Address, ISTR Conference, July 12, 2018. Amsterdam: Vrije Universiteit Amsterdam.

CAF (2016). Gross Domestic Philanthropy: An International Analysis of GDP, tax and giving. West Malling: Charities Aid Foundation. https://www.cafonline.org/docs/default-source/about-us-policy-and-campaigns/gross-domestic-philanthropy-feb-2016.pdf

Gustavsson, M., & Jordahl, H. (2008). Inequality and trust in Sweden: Some inequalities are more harmful than others. Journal of Public Economics, 92(1-2), 348-365. https://doi.org/10.1016/j.jpubeco.2007.06.010

Hastings, O.P. (2018). Less Equal, Less Trusting? Reexamining Longitudinal andCross-sectional Effects of Income Inequality on Trust in U.S. States, 1973–2012. Social Science Research, 74: 77-95. https://doi.org/10.1016/j.ssresearch.2018.04.005

Leigh, A. (2006). Trust, inequality and ethnic heterogeneity. Economic Record, 82(258), 268-280. https://doi.org/10.1111/j.1475-4932.2006.00339.x

OECD (2018). Tax Revenue, % of GDP, https://data.oecd.org/chart/5do5

Salamon, L.M., Sokolowski, S.W., & Haddock, M.A. (2017). Explaining civil society development. A social origins approach. Baltimore, MD: John Hopkins University Press. https://www.amazon.com/Explaining-Civil-Society-Development-Approach/dp/1421422980

Stephany, F. (2017). Who are your joneses? socio-specific income inequality and trust. Social Indicators Research, 134(3), 877-898. https://doi.org/10.1007/s11205-016-1460-9

Yang, Z., & Xin, Z. (2020). Income inequality and interpersonal trust in China. Asian Journal of Social Psychology, 23(3), 253-263. https://doi.org/10.1111/ajsp.12399

Altruism at a cost: how closing donation centers reduces blood donor loyalty

To what extent is blood donation motivated by altruism? Donating blood is ‘giving life’ and it is often seen as an act of sacrifice. In a new paper forthcoming in the journal Health & Place, co-authored with Tjeerd Piersma, Eva-Maria Merz and Wim de Kort, we checked whether blood donors continue to give blood when the sacrifice becomes more costly.

To see whether they do so, we tracked blood donors in the Netherlands between 2010 and 2018. The number of donation centers operated by Sanquin, the national blood collection agency in the Netherlands, decreased by 46%, from 252 in 2010 to 136 in 2018.

We found that donors who were used to give blood at a location that was closed were much more likely to stop giving blood. The difference was very large: donors for whom the nearest donation center was closed were 50% more likely to have lapsed in the year after the closure than donors for whom the nearest center remained open (15.3% vs. 10.2%).

The percentage of lapsed donors after closing the nearest donation center steadily increased with each extra kilometer distance to the new nearest donation center. Of the donors whose nearest donation center closed, 11.6% lapsed when the distance increased by less than one kilometer while 32.8% lapsed when the distance increased by more than nine kilometers.

Because the blood of O-negative donors can be used for transfusions to a larger number of other blood types, they are called ‘universal donors’. We expected a lower lapsing risk for universal donors as costs increase. This would be evidence of altruism.

First we thought that we had found such evidence. Universal donors are less likely to stop donating blood after the nearest donation center is closed than other donors. Also we found that universal donors were more committed to continue giving blood as the travel distance increased up to 5 kilometers. At longer distances, the pattern was less clear.

However, when we included the number of requests for donations as a covariate, the difference largely disappeared. This means that O-negative donors are more likely to continue to give blood because they receive more requests to donate from the blood bank. The sensitivity to these requests was very similar for universal and other donors.

We conducted mediation tests to establish that closing a donation center reduced donor loyalty because the travel distance to the nearest location increased, and to establish that universal donors were more loyal because they received more requests to donate blood.

One of the reviewers asked for a matching analysis. This was a good idea. It also provided a nice learning experience. I had never done such an analysis before. The results were pretty close to the regression results, by the way: no difference between universal and other donors matched on the number of requests. 

In sum, we found evidence that:

1. Blood donors are strongly sensitive to the costs of donating: closing a donation location increased the risk of lapse big time;
2. Blood donors are less likely to lapse when they receive more requests;
3. ‘Universal’, O-negative donors were less likely to lapse, because they received more requests to come donate blood;
4. Universal donors are equally sensitive to requests and to costs of donating as other donors.

The analyses are based on register data from Sanquin on all blood donors (n = 259,172) and changes in geographical locations of blood donation centers in the Netherlands over the past decade. Because these data contain personal information we cannot share them for legal reasons. We do provide the complete Stata log file of all analyses at https://osf.io/58qzk/. The paper is available here: https://osf.io/preprints/socarxiv/na3ys/ This paper is part of the PhD dissertation by Tjeerd Piersma, available here.

We started with this project thinking that closures of donation centers may be natural experiments. But we soon found out that was not the case. Which donation centers were closed was decided by Sanquin after a cost-benefit calculation. Centers serving fewer donors were more likely to be closed.

As a result, the centers that were closed were located in less densely populated areas. It was a fortunate coincidence for the blood collection agency that donors in less densely populated areas were more loyal. They were willing to spend more time travelling to the next nearest donation center.

For our research, however, the lower lapsing risk for donors in areas where donation centers were more likely to be closed created a correlation between closure and our dependent variable, donor loyalty. End of story for the natural experiment.

We spent some time searching for instrumental variables, such as rent prices for offices where donation locations are housed. Many office locations in the Netherlands are vacant and available at reduced rent.

The number of empty offices in a municipality could reduce the costs of keeping a donation center open. However, we found that the percentage of empty offices in a municipality was not related to closure of donation centers. If you have thoughts on other potential IVs, let us know!

What Oxfam cancellations tell us about donor motivation

What can we learn from the drop in donations to Oxfam after the child abuse news broke? In the UK, about 7,000 donors cancelled, in the Netherlands 1,700, and in Hong Kong 715.

First, the drop does not tell us much about what makes people give. Most donors have continued to give. The 7,000 who cancelled in the UK represent 3.5% of income in the UK. The 1,700 donors who cancelled in the Netherlands are 0.5% of all donors. This means that defaults save lives. The default is to do nothing and continue to give. We’re seeing a small fraction go.

But for those discontinuing their gifts protest, we could say that we can tell why they were giving in the first place by looking at their reactions.

If they gave to Oxfam for altruistic reasons, they will find other charities to give to. They may find it hard to trust Oxfam now, and other charities named in the media.

There is the ‘one bad apple spoils the entire basket’ idea that donors will find faults with other charities as well once one gets bad publicity.

We’ll have to see how much that idea is worth. In previous episodes in the Netherlands, bad publicity about one charity usually did not spill over to other charities. In the Netherlands and Hong Kong it seems altogether more puzzling why donors stopped giving, as the abuse – as far as we know – did not involve the Netherlands or Hong Kong branch.

In my view the cancellations are a result of empathic anger. The more you care about children, the more angry you will be. While empathy has been heralded as an important factor in altruism, it also has a non-altruistic side. The emotion of anger itself and the cancellation may be viewed and communicated as a sign of caring. But it is not effective helping.

There is also a role for public relations. It may be that the abuse corrected an image that charity workers are holy superhumans. A charity that ‘paints itself as whiter than white’ reinforces that image. In times of PR crises like these such an image boomerangs donors away. If donors reckon with the possibility that a charity may attract bad apples as workers, they realize that one bad apple is not evidence of a disease, but of lax quality control.

Wat is normaal?

Geeft de gemiddelde Nederlander echt 559 euro per jaar aan goede doelen, zoals Arnon Grunberg gisteren schreef op de voorpagina van de Volkskrant?

Nee, dat is onwaarschijnlijk. Grunberg verwees naar een cijfer dat werd genoemd in het HUMAN televisieprogramma ‘Hoe normaal ben jij?’

Het cijfer klopt niet om twee redenen.

1. Het bedrag is veel hoger dan uit ander onderzoek naar filantropie naar voren komt. Het cijfer van Human komt uit een onderzoek dat waarschijnlijk niet representatief is voor alle Nederlanders. Human geeft geen informatie over de peiling die gehouden is, maar het is waarschijnlijk dat het een zogenaamde gelegenheidsgroep is: op de site kan iedereen deelnemen. Degenen die dat doen zijn bijna nooit representatief voor de Nederlandse bevolking.

Het standaard onderzoek naar filantropie, Geven in Nederland (GIN), voert de Vrije Universiteit Amsterdam uit sinds 1995. Het geeft een navolgbaar representatief beeld. Gemiddeld geven huishoudens 341 euro, zo blijkt uit de laatste editie van het GIN onderzoek uit 2017.

2. Het cijfer gaat over een gemiddelde, en dat is niet normaal. Als je het rekenkundig gemiddelde berekent over alle Nederlandse huishoudens, dan zie je niet goed wat de typische Nederlander geeft. De helft van de Nederlandse huishoudens geeft namelijk minder dan 60 euro, blijkt uit GIN. Het gemiddelde wordt sterk beïnvloed door een klein aantal huishoudens dat heel veel geeft. De grafiek kun je gebruiken om te zien hoe normaal je bent: geef je tussen de €150-€200 per jaar, dan hoor je in het derde kwartiel, de groep van ongeveer een kwart van de bevolking die meer geeft dan helft van de Nederlanders. Het kwart meest gevende Nederlanders geeft vaak meer dan €1.000.

Gevonden: Student assistent voor het onderzoek Geven in Nederland

De werkgroep Filantropische Studies van de Faculteit Sociale Wetenschappen aan de Vrije Universiteit Amsterdam is het expertisecentrum op het gebied van onderzoek naar filantropie in Nederland. De werkgroep houdt zich bezig met vragen zoals: Waarom geven mensen vrijwillig geld aan goede doelen? Waarom verrichten mensen vrijwilligerswerk? Hoeveel geld gaat er om in de filantropische sector? Voor het onderzoek Geven in Nederland heeft de werkgroep Suzanne Felix aangenomen als onderzoeksassistent.

 

Werkzaamheden

Geven in Nederland is een van de belangrijkste onderzoeksprojecten van de werkgroep. Sinds 1995 wordt het geefgedrag van huishoudens, individuen, fondsen, bedrijven en goededoelenloterijen elke twee jaar in kaart gebracht en samengevoegd tot een macro-economisch overzicht. De werkgroep Filantropische Studies brengt de resultaten van het onderzoek tweejaarlijks uit in het boek ‘Geven in Nederland’. Felix werkt mee aan het onderzoek naar nalatenschappen en giften door vermogensfondsen en huishoudens.
update: 3 september 2016

Brief guide to understand fMRI studies

RQ: Which regions of the brain are active when task X is performed?

Results: Activity in some regions Y is higher than in others.

Heeft de culturele sector de cultuuromslag naar ondernemerschap gemaakt?

Presentatie rapport ‘Culturele instellingen in Nederland’

Werkgroep Filantropische Studies Vrije Universiteit Amsterdam

 

Vrijdag 10 juni 2016, Theater Griffioen, Uilenstede 106, 1183 AM, Amstelveen

 

In 2012 werd de Geefwet ingevoerd met een multiplier die de aftrekbaarheid van giften aan culturele instellingen verhoogde. Bovendien kregen culturele instellingen meer mogelijkheden eigen inkomsten te genereren uit commerciële activiteiten. Tegelijk kregen veel instellingen te maken met bezuinigingen en de vraag om meer ondernemerschap. Hoe hebben Nederlandse particulieren en bedrijven met een hart voor cultuur gereageerd op de verhoogde aftrekbaarheid van giften aan cultuur? Zijn zij ook inderdaad meer gaan geven? En hoe hebben de culturele instellingen gereageerd op de bezuinigingen enerzijds en de multiplier anderzijds? Wat voor instellingen hebben de omslag naar ondernemerschap wel kunnen maken en wat voor instellingen niet?

Deze vragen stonden centraal in een onderzoek dat de werkgroep Filantropische Studies heeft uitgevoerd op verzoek van het ministerie van OCW naar de effecten van de Geefwet op het genereren van inkomsten door culturele instellingen. Het onderzoek verschaft inzicht in de stand van zaken van de culturele sector op dit gebied en de mate waarin de Geefwet bijdraagt aan de versterking van de culturele sector door stimulering van giften aan cultuur.

U bent van harte welkom op een symposium waarop de onderzoekers de resultaten presenteren aan de culturele sector. U kunt zich hier aanmelden.

 

Programma

15.30    Aanmelden

16.00    Presentatie onderzoek door prof. dr. René Bekkers

16.30    Annabelle Birnie, Drents Museum

16.45    Marielle Hendriks, Boekmanstichting

17.00    Drankje

 

 

Locatie

Theater Griffioen, Uilenstede 106, 1183 AM, Amstelveen

Routebeschrijving – klik hier

 

 

Meer informatie

Meer informatie over het onderzoek vindt u op www.cultuursector.nl

Conditional Review Acceptance Policy (R1)

“Thank you for your invitation to review. Did the authors provide the data and the code they have used to produce the paper? Will the paper be published in open access mode? If twice yes, I will consider reviewing the paper.”

This is my new automatic reply to requests for review journal articles that I receive from editors and their assistants. In june 2014, I introduced a conditional review acceptance policy (CRAP). The policy was to review only those articles that the journal agrees to publish in a Free Open Access mode – making the article publicly available, without charging any fees for it from universities, authors, or readers. The revised policy now also includes the question whether the data and code will be publicly available, as proposed by the Peer Reviewers’ Openness (PRO) initiative. The revision rewards open science.

 

The Future of Foundation Support for Research and Innovation

Recently the EUFORI Study was published, in which a network of experts coordinated from our Center for Philanthropic Studies at the Vrije Universiteit Amsterdam mapped the support from foundations for research and innovation in Europe. Read the synthesis report here. The research was based on an extensive survey of 1 591 foundations supporting R&I in Europe and a qualitative analysis of 29 different country reports. We concluded that foundations contribute a significant amount of money to R&I: annually at least €5 billion, out of an asset base worth at least €127 billion. These are lower bound estimations, because it was impossible to estimate contributions by foundations that did not participate in the study.

The final chapter of the synthesis report presents six recommendations. The main objective of the recommendations made in this final chapter is to increase the potential of R&I foundations in Europe. Considering the underlying potential, actions towards greater support by foundations for research and innovation should and must involve engaging all actors: national governments, EU institutions, the foundations themselves, the corporate sector, universities and other research institutes, and the public at large.

Recommendation 1: Increase the visibility of R&I foundations

This recommendation is addressed to foundations, national governments, the EC and EU administration, businesses and the public at large. It relates to the current fragmented landscape of R&I foundations in Europe. The landscape of foundations in Europe is characterised by a few well-established foundations and many smaller foundations with modest resources mainly operating in the background. Growing visibility will enhance the impact of existing funding. If foundations become more aware of each other’s activities, the effects and impact of their contributions can be increased. Moreover, the other stakeholders involved such as the business community and research policy-makers will become knowledgeable about the foundations’ activities. From the perspective of the beneficiaries, research institutes, universities and researchers will more easily find their way to foundations. Visibility will lower the transaction costs for all the parties involved. For foundations, governments and businesses it will increase their knowledge about ongoing research/new research funded and vice versa. For grantmaking foundations it will facilitate the review process of research proposals and submissions; it is to be expected that more visibility will reduce the amount of incorrect applications. For the beneficiaries of the foundations’ support (research institutes, universities and researchers) – the grantseekers – it will increase their funding opportunities, they will more easily find their way to foundations, and it will facilitate submission processes. For potential (major) donors it will offer visible causes to benefit. Increasing the visibility of R&I foundations could have a positive effect on potential (major) donors as it could encourage them to support a research foundation. Increasing the visibility of and information about R&I foundations was already addressed by an expert group in 2005. They argued: ‘.. foundations and their donors would be more aware of the foundation landscape (increasing collaborative working and, possibly, giving), foundations’ contribution to various sectors could be properly assessed and the information could inform policy-making in this area. It is in fact a prerequisite to other actions’. The present EUFORI study is a step forward. A lot of information is now available. Next to this synthesis report, 29 country-reports, new data, an active network of researchers and the EUFORI website can contribute to the profiling of the R&I foundation sector in Europe.

With the exception of some large and well-established foundations in Europe, there is a lack of a common research identity among the foundations supporting R&I in most countries. Research and innovation are often not seen as a purpose/field in itself but are instead used as an instrument for other purposes and areas in which foundations specialise (such as health, technology, society). This is reflected by a lack of dialogue between the foundations supporting R&I (occasionally between foundations that deal with similar topics, e.g. foundations supporting cancer research). Bringing foundations together at a European level and following the recommendations of the expert group from 2005, the European Foundation Center (EFC) created the European Forum of Research Foundations. This forum provides a platform for a group of large and well-known R&I foundations in Europe. In order to increase the visibility of foundations supporting R&I at a national level, the encouragement of the creation of national forums of research foundations is recommended as the next step. The opportunities and mutual benefits for foundations supporting R&I at a national level should be explored. The next step: Explore the opportunities and mutual benefits of the creation of national forums of research foundations.

Recommendation 2: Explore synergies through collaboration

Unity in diversity is one of the main challenges for all the players involved in the R&I domain. These players can be distinguished in the domain of research (governments, business, foundations and research institutes/researchers), each with their own distinctive role. Together these groups can make a difference in increasing the potential for R&I. They can create synergy through collaboration, which should be interpreted in the broadest sense, varying from information sharing, networking, co-funding and partnerships. Mutual advantages can be derived from pooling expertise, sharing infrastructure, expanding activities, pooling money due to a lack of necessary funds, avoiding the duplication of efforts and creating economies of scale.

Get to know each other, meet and see where to reinforce each other’s efforts

Based on the conclusions of the EUFORI Study there is an indication for the need for improved dialogue, information exchange, networking and cooperation between the foundations supporting R&I, as well as between foundations, governments, business and research institutes (researchers). The needs, opportunities, mutual benefits and barriers for collaboration should be further explored, including mutual responsibilities when cooperating. The creation of national forums or networks of foundations supporting research and innovation, regular meetings between the foundations and other stakeholders involved (national government, EU government, research institutes and business) could bring these groups together.

An EU-wide study is recommended on the needs, opportunities, mutual benefits and barriers for collaboration between foundations, national governments, the European Commission, the business sector and research institutes. A network of national experts (mostly members from ERNOP) built for the EUFORI study can be of added value for this study and can facilitate the collaborative relations between the EC/ RTD, the R&I foundation sector and other stakeholders in Europe. It would be well-advised to set up an independent expert group before the start of this study with selected experts and stakeholder representatives in the field of foundations, the business sector, research institutes and public authorities at a national and European level. The expert group should provide input for the design of the study and could adopt an advisory role. Subsequently, it is recommended that the study will be finished by a follow-up conference for all the players involved aimed to discuss the implementation of the outcomes of the Collaboration Infrastructure Study. In this call for collaboration we have to consider two possible, interrelated pitfalls; namely the danger of ‘substitution’ and the danger of threatening the independence of foundations. Foundations, and civil initiatives in general, make their own choices and preferences and are based on social democracy. Governments, on the other hand, have their own responsibility based on political democracy. Businesses have their own market-driven values. Sometimes they reinforce each other, sometimes they may act as opponents. It concerns different worlds, differing in terms of constitution, values, legitimacy and organisation style. The independence of private R&I foundations should be respected. Foundations derive their legitimacy from many contacts with the ‘capillaries’ in society, thus offering them the opportunity to function ‘as the eyes and ears’ for innovation. This grass-roots connection represents the philanthropic tradition in Europe: ‘voluntary action to serve the public good’. The next step: Launch a Collaboration Infrastructure Study.

Continue reading the recommendations from the EUFORI study here.