Making “Big Data” Work for Equality

This information gap is huge. Current estimates on disability indicate that EU governments lack comparable data on the life experiences of as many as 80 million people.

As the recent PRISM program scandal in the United States highlighted, corporations and governments can gather information of any kind about us. Your emails, the foods you like, where you travel, and your shopping preferences are all examples of personal data that can be mined for profiling purposes.

It’s ironic, then, that when discussing ethnic minorities or people with disabilities in Europe, “no data available” is a common excuse for not doing more to fight discrimination and inequality.

The Open Society Foundations have grappled with this excuse far too frequently in our grantmaking, advocacy, and litigation work. In 2013, we launched the Equality Data Initiative, which aims to compel European states to collect information such as employment situation, educational achievement, and housing distribution for groups at risk of discrimination. The pilot phase focuses on education in Bulgaria, Germany, Hungary, Ireland, Romania, and Sweden, and on public employment in France.

The collection of precise data on inequality is an essential element in identifying gaps in public policies, and implementing remedial steps. Lack of data masks inequalities and hinders the exposure and monitoring of these gaps by civil society. Government assertions about how well they are doing in closing the gaps cannot be verified.

Such benefits of collecting data on groups at risk of discrimination are well known to human rights bodies. The Council of Europe’s Commission on Racism and Intolerance, the Commissioner for Human Rights, and the UN Rapporteur on Contemporary Forms of Racism have all addressed multiple recommendations on the matter to European governments. In a 2007 report, the Commission on Racism and Intolerance described the “production of detailed statistics” as a “prime necessity for compliance with European laws.” And a report released this month by the Open Society Foundations cites a lack of equality data as a primary barrier to the implementation of the EU’s Race Equality Directive.

Under the UN Convention on the Rights of Persons with Disabilities, the EU and its member states have a responsibility to “collect appropriate information, including statistical and research data, to enable them to formulate and implement policies.” Though the European Commission, Fundamental Rights Agency, and Eurostat have attempted to take on this message, most EU member states have not.

This information gap is huge. Current estimates on disability indicate that EU governments lack comparable data on the life experiences of as many as 80 million people. Further, according to multiple Eurobarometer surveys, racial or ethnic discrimination has been widely perceived as the most widespread type of discrimination in the EU over the last 10 years. Despite this, no estimates are available on how many people experience inequality because of their racial or ethnic origin. Among EU member states, only the U.K. collects equality data on the life experiences of its racial minority population.

Why are most European governments reluctant to collect such valuable data? Two of the most common reasons given are privacy and the risk of further stereotyping minorities by categorizing them.

The privacy argument has been addressed by EU authorities. Since 1995, the EU’s data protection law has provided detailed guidelines for the collection of “special categories of data” such as racial or ethnic origin and health condition. The Council of Europe’s Strasbourg Court and Commission on Racism and Intolerance have affirmed that as long as governments comply with safeguards such as getting the consent of the respondents, enabling them to self-identify, and anonymizing the data, data collection is possible.

For its part, the Open Society Foundations’ Equality Data Initiative supports only data collection processes where individuals explicitly consent to provide sensitive information about themselves for their benefit and that of the larger community, where everyone has the right to choose how to define themselves in terms of racial or ethnic background or in light of a disability, and in which everyone’s own privacy is granted through the anonymized processing of the data.

The second argument, that data collection entrenches discrimination, has recently become a topic of intense debate throughout Europe. The argument against data collection has two major elements:

  • In continental Europe, discussion of race or ethnicity remains taboo because of lingering memories from World War II, when categorization of minorities led to human rights abuses.
  • Individuals sometimes fear stressing characteristics that may lead to stereotyping. Practices such as ethnic profiling by police or the segregation of children with disabilities substantiate these fears.

These fears are mitigated by the emergence of an increasing number of representative groups advocating for the collection of equality data, such as the Representative Council of Black Associations and the Collective against Islamophobia in France and the Disability Federation of Ireland.

When used correctly, equality data can do enormous good. For example, in England, the mission of the Equality Diversity Body of the Open University is to monitor, analyze and accommodate the needs of students and staff belonging to vulnerable groups. In order to perform its duties, the body asks students to fill in an anonymous form where everyone can decide to self-identify in terms of their special educational needs, physical disability, racial background, religion, and even sexual orientation. The argument used by the body is a simple one: “If you don’t fill in the form and then find that no one’s thought about YOUR needs well, you can hardly complain, can you?” Data collected through the form is used by the Open University to design measures to accommodate groups at risks of discrimination or underachievement.

It is high time that European governments pay more attention to the situation and the needs of their racial, ethnic, or religious minorities and disabled constituencies, starting with making them visible. Do you believe that European governments should collect data on groups at risk of discrimination? Share your views in the comment section below.

Learn More:



De fait les données ethniques, lorsque leur désignation est tabou, et donc non-contrôlée, ni même pensée, existent tout de même. Tabou, elles se transmettent alors parfois de manière inconsciente. L'héritage racial et colonial se transmet aussi de cette manière. En nommant la "race" nous pouvons l'interroger, la mettre en cause, la "travailler". En la taisant nous validons des stéréotypes qui continuent d'exister malgré (et grâce) à un non-dit apparent. Or la lutte contre les stéréotypes raciaux n'a rien d'une apparence, elle doit se travailler jusqu'au plus profond de nous-mêmes.

In the fight against ethnic profiling, data collection indicating the ethnic or national origin of persons stopped and searched has proven to be crucial to remedy the problem. Such data makes it possible to know if members of certain groups are being stopped disproportionately or not. This, in turn allows for debate about the issue based on objective facts and supports development of appropriate remedies.

gli esseri umani hanno un'attitudine unica nel contare e catalogare. Fondamentalmente ciò è dovuto al fatto che non possiamo parlare di umanità, in quanto nel mondo esistono i popoli che fanno gli interessi del proprio popolo. Se l'UE si concentrasse nell'aumentare i propri sforzi nel costruire una cittadinanza europea effettiva, dove le persone si sentano effettivamente cittadini europei, in quel momento non avremmo piu il bisogno di contare, schedare e catalogare le minoranze, in quanto non esisterebbero piu minoranze perchè sarebbero sostituite da una nuova identità: quella europea.

Thanks Marc, Lanna and Irvin for contributing to this debate. Your interventions show how the issue may be contentious in continental Europe: there are historical reasons on which states ground the refusal to name race, and, on the side of this, security policies (like ethnic profiling) for which states do not have a problem in resorting to race. Once we are able to count the stops and searches performed against racial minorities, this becomes evident. The ambition of some would be that of bypassing the concept of race and minority to directly claim European citizenship for all. But are we there yet? Or should we rather take account that differences among European citizens and especially in between European citizens and non-citizens living in Europe still exist and do something about it? Can we bypass the categories defining vulnerable groups if these categories are still evidently relevant to measure inequality and discrimination?

شكرا لكم
اتمني ان يكون مثل هذا الحراك في الدول العربية

Issues related to the identity of nationality and race are very sensitive. If they are not treated with much intelligence, they can create conflicts that can last for years or even decades. I take the case of my country. Côte d'Ivoire has suffered for more than ten years of an identity crisis. To return to your topic, I can say that a serious study that takes into account all these social aspects can provide reliable statistical data to the rulers. It must also place responsibility on two levels which are:

The first level is that governments must take into account in their governance to vulnerable people to say those from disadvantaged social strata, putting in place policies and strong institutions that allow professional insertion easier for those following, which will stop the violence, especially in the suburbs.

The second level is that young people and all those who are considered as vulnerable people must learn to fight to find a first job if only to ensure their survival and then pretend to a more elevated level of life. This is deplorable, especially in Africa, that we are still waiting for governments, without seeking to undertake.

To conclude, I would say that this unknown adverse population seeks first to be enterprising before expecting that governments will come in the set security.

Thank you

The rise of the extreme right in my country is a worrying & disturbing development. This has been enabled by lazy media coverage & politicians being able to encourage discrimination against people who are ill & in minority groups. If full & proper data were collected then there would be a full, fair & accurate way to demonstrate discrimination to counter unfair & unreasonable demonising of vulnerable & minority groups.

What is mentioned in the write up about gathering the group data and its purpose is also mentioned that the data serves for. I would like to take the side of the notion that gathering the group data has no harm and it will help the planners to plan for people and further, it will help to be vigilant on the bad things which are emerging up in the society.

As Olof Palme said, without a common security there is no real security. Inequality is a cause of insecurity. No fight against social unemployment and housing gaps will undermine goverments' legitimacy. So we need data and policy analysis to build real security.

While we in the USA cannot dictate what other governments do, it would be instructive to learn how US civil rights agencies use race, gender and ethnic data and how they protect vulnerable populations from the mis-use of such data. Both government and the private bar use data collected by such agencies as the Equal Employment Opportunity Commission and the U.S. Department of Labor as well as the Census Bureau to gauge whether there is evidence of job discrimination or discrimination in education as well as housing, health care, voting rights etc. If, for example, you do not know the race or ethnicity of your job applicants you cannot know if there is an indication that hiring practices have a direct or indirect impact on protected groups or if, in the workplace, women or minorities are clustered in certain job categories where others are not. It all begins with the data. We at the American Association for Affirmative Action would be pleased to work with groups in Europe and elsewhere to discuss the importance of collecting information regarding vulnerable populations.

Pre-incident data collection may be more harmful than beneficial. Example: knowing the race/ethnicity of a person assists someone else to mis-treat that person.

Post-incident data collection is crucial to establish justice and also to get a sense of "how large scale" the problem really is.

Incident here = an event where the subject (person) may have a potential to be wrongfully treated.

My 2c.


I personally feel that at this point in time, the European identity is not strong enough to overcome existing inequalities. For example the Roma, an European minority group by excellence, still suffer from discrimination and deep-rooted social exclusion on a daily basis throughout Europe. While there is always an inherent risk to data collection, provided that sufficient safeguards are in place, I am convinced - together with institutions at the European Union, Council of Europe and United Nations - that equality data collection could proof to be the missing link to eradicate discrimination and enhance equality. It has been proven that the measures that have been taken to far to tackle discrimination of vulnerable groups, including the Roma, are insufficient to make a tangible difference in practice. The lack of data on ethnicity has been identified on various occasions as an important shortcoming of the current systems and policies. It appears that many member states tend to hide against privacy and data protection legislation, which can be overcome. One has to keep in mind though that even if the mindset of policymakers in various member states would change and equality data collection would be put on the agenda, this by itself would not be sufficient to reach success. Whatever system of ethnic data collection is put into place, it needs to be accompanied by sensitization programs and campaigns. If not, there is be a real risk that the outcome of for example a census would be distorted because people are not convinced of the good that can come from such data collection and thus would not be willing to reveal their true identity.

First I want to tell you that in Togo the first association to collect data on disabled persons is ATPPEHP (an association for protection and promotion of handicap children). This association created in 1992 began collecting disable dada but because of lack of financial support its activities did not continue. NGO: ADET for minorities in general will tackle this mission in the following year if we get support. There are no problems to collect disable or minorities data if you decide to assist them and you must do it.

Governments who are unwilling to treat vulnerable groups with the consideration they deserve are less likely they will use the data to further "protect" these groups, but to further infringe their rights and liberties. Advocating for data collection in these cases is not a simply task.

E' ormai tempo che tutte le minoranze residenti in Europa abbiano il loro pieno riconoscimento a vivere con tutti i diritti civili e politici attribuiti ai cittadini delle nazioni e dei popoli del vecchio continente, secondo la Carta dell'ONU del 1949.

I and a 16 year old Roma in Florence created a censimento in which we listed individuals, age, how many children, what work skills (many), years of school (few), trauma such as widow, orphan, TB, dwelling in Romania owned or rented, no of rooms, no of windows, how many in that space. This created both positive information about work skills that could be utilized to rise from poverty, also what worsened the poverty. We continue to hold alphabet school with these families, all ages, now having the women and children in a space seperate from men not their husbands - otherwise they would not attend. We also preserve these families with work opportunity. Our families in Florence are forced to sleep in the streets in the rain, their shacks bulldozed by the police, and they are not allowed water.

Sadly unless you have confidence in the authorities people will hide their identities if they are minorities and are being targeted. Intelligence on demography and key community indicators is crucial but unlikely to be made available by communities at risk. We need to keep identifying injustices and takng our own and collective actions to fight all forms of persecution.

I believe with my experience that the authorities have collected infomation on minoritere.Förra police mästeren Gun gunmor and Maria Leissner in spetnen-.
Thought has been "good" from the beginning, but when the audience does not have the ability to influence or have inblink on issues related to group, then it can never be good
This has created divisions among the group, but also goat authorities were able to stop a group more. All chenges comes in from without the group, but with a few so-called Roma, but chosen by the majority peppol

I am an advocate for data collection because it is an excellent tool to prove the discrimination of ethnic/other minorities.
I would suggest to extend your work in Romania to the health sector. This intervention is so needed because Romanian authorities are saying "we are treating all citizens as equals", but the reality is contradicting them (reality - measured by good researches).

Dangers of misusing vulnerable people data should in no way undermine the benefits accruing for such an undertaking. If there is no information on their situation, then nothing can be done to treat them more fairly and in an equal manner. All in all: THE PROBLEM IS NOT IN COLLECTING DATA BUT THE WAY AND ULTIMATE GOAL FOR WHICH DATA ARE COLLECTED AND USED.

Yesterday the arrests of the fascist Golden Dawn parliament members, as conspirators for murders, attacks, extorsions, and, and, and..started to shake Greece. On this very important occasion of the rounding up of the fascist party and a proven, criminal network of Golden Down, I would like to add that even when data is available, it takes political will to become useful. Data existed, has existed on many levels for a long time, both scientifically and in terms of justice/ politicaly valuable, but nothing was done. The Golden Dawn had been even murdering migrants and remained quite untouched until the recent murder of a young Greek musician who voiced his objections with his verse to racism. This brutal, execution like, murder - finally-played a key role, it even put Greece to the risk of not taking the presidency of the European Community in January. This was one of the reasons the government decided to act with the decision to bring together all the criminal files which were gathering dust on shelves, regarding the criminal activities of the G.D. To conclude, in relation to the input from data which is necessary for change: the lack of data is the result of efforts by politicians to hide the true nature of racism which is not an opinion, it is a criminal practice. In education- speaking as a teacher- the efforts made to combat racism in schools has been a difficult struggle without state support. The G.D has infiltrated schools and since many years ago created youth rings, the 'Centaures'- you see they like to relate everything shamelessly to Ancient Greece- and they started with the age of 14, enrolling high school children, inoctrinated to pure blood theories, to humiliate and to beat up migrant children. Many of these crimes will go unpunished, because even the 'Centaures' were unprotected victims of propaganda themselves. Angry and frustrated, with unemployed parents, they were convinced that 'purity' was an asset to be proud of and shouted at the top of their voices, causing us chills, 'Blood, Honour, Golden Dawn', which rimes in the Greek Language.
As a teacher I had to put up with the indifference of the educational system, with attacks -physical as well as verbal- against sturents, with humiliating comments even by teachers. So, to continue my own private teaching efforts, two years ago, I created a program called 'Dialogues of Friendship' and I did not call it Antiracist education, I just presented the whole project as a tool to improve human rights application, conflict resolution and mediation, in order to create a better environment for the children. Soon the first thing that happened, complaints of girls towards a young student, GD supporter, reached the mediation team. He was telling the girls in his class "now, with all the migrants, your mothers will become unemployed and will end up as whores to support you". The issue was discussed with students, teachers, the boy's parents, and, what shocked me most was the offensive youth himself who told me"I am struggling for a better Greece", with a convinced attitude and an innocent face...We have to dig very deep to combat racism in Greece, it seems a feat under the present crisis situation, but there is no other way than to use all possible pressure towards politicians to restaure justice where it has failed to protect minorities and especially children, not just minority children who fall victims, but also all children, from the poison of racism.

As some already said, collecting datas is neutral (on the whole, because one may notice that collecting implies categorizations, which are never neutral) . What actually matters is the goal of such collections. The goals should be always transparent, which means under strict public scrutiny and/or shared between predefinite and therefore legimitate entities. I don't feel those statements are idealistic or uncommon and I expect our civil society and our representatives to adjust to such needs in the near future (it started already, somehow). Judicial powers must have their say on that obviously. So do we.

Of course i am in favour of collecting "private data" by state bodies in whatsever european country. But ONLY by them. No one else. And even so one can not be sure that these data won't be used for other purposes by them every time they think it would be useful and beneficial for them. See the latest scandal in the USA (Snowden case).
It has been mentioned that U.K. is the only european country that collects equality data on ethnic minorities. It's because this particular country has always respected and served human rights and minorities (religious or ethnic ones) before and after the WW2 in Europe. And, thus, it has not any fear of being accused of discrimination and misusing "private data".
As it has already been mentioned in your report, it is up to the use of those "private data", namely which body collects them and what for.
Conclusively, in a Europe that is being ravaged by recession and failed economic policies, human rights and minorities are easily ignored as not important issues by governments. Consequently, poverty and injustice are even more increased, leading to social turmoil at times, among others, and to creating of the so called "ghettos" in societies.

O problema está na indistinção de dados pessoais e dados íntimos. O meu nome, a minha idade, a minha residência, são dados pessoais e merecem protecção. Mas a verdade é que também são essenciais para a minha protecção, isto é, tenho interesse em que a minha comunidade saiba o suficiente sobre minha para que me possa ajudar e proteger em caso de necessidade.
De outra natureza são os dados respeitantes à minha intimidade; serão todos aqueles que não são necessários para o meu relacionamento pessoal e aberto numa sociedade aberta e ainda com as instituições jurídicas, securitárias e fiscais.

Os dados que sejam incontornáveis para o exercicio de direitos em relação à comunidade ou para a verificação do cumprimento dos deveres em relação a esta, devem, em minha opinião, ser colecionados pelas autoridades competentes ou bilateralmente pelas pessoas e entidades de que voluntariamente decida fazer parte. A atual lei de prtoecção desses dados parece ser suficiente.
É do interesse do individuo não passar necessidades, mas também é na mesma medida do interesse da sociedade, que nenhum dos seus membros passe necessidades ou seja esquecido.

Os dados da intimidade (os que posso ocultar sem comprometer a honestidade das minhas relações individuais, pessoais ou cocietárias), esses não devem ser nunca coligidos. Se eu os quiser livremente divulgar, escrevo um livro, um artigo, uma canção, faço um filme ou como exactamente agora, escrevo um post num blog.

Parabéns Opensociety Fundation.

I have gone through the matter published in this column. I agree in a democratic society it is the right of the individual to accept or discard. But in the dictatorial society it is difficult. In Nepal it is a transitional period. We have to analyze every phenomena carefully through the persee of the conscience, principle and legacy. I agree on bypassing the category of stratification of race, cast, ethnicity, developed or backward etc. These have become the historical measure upon which past regimes governed successfully and made burnig environment to be emerged such groups, sectors, community and society. The objective truth of the current situation is that who have been demarked as vulnerable must be well protected and preserved using by all means. The goals of all the international conventions seem to be so. In Nepal there has been seen the existence and pathetic condition of more than 27 such groups like Route, Kusunda, Chepang, Kissan, Hyiu, Bankariya, Mushahar, Lepchas, Dolpo, Manangwa. The states, NGOs, CBOs, and the private sectors must be collaborative to launch joint ventures to protect them by making project, plan and policies. They need support rather than blame and hate.

Participation and the Practice of Rights organisation is based in Belfast, Northern Ireland. Northern Ireland suffered 30 years of sectarian conflict where the majority of deaths were in North and West Belfast. These areas also suffered the long-standing social deprivation, lack of investment and chronic housing, health, and employment inequalities. We recently published a report entitled 'Equality Can't Wait' which details the longstanding religious inequality in social housing supply which disproportionately impacts the Catholic community in North Belfast. The public bodies in charge have misused statistics to attempt to obscure the reality of the inequality, and justify their failure to address it effectively. The report can be read at I would be very grateful to hear of other similar experiences, particularly in post-conflict areas.

Thank you

Add your voice