A/HRC/26/49/Add.1 2015) with the support of the World Bank, aimed at improving access to health and education for returnees and their sustainable integration through the installation of basic infrastructure (e.g. schools, health centres), support for agro-pastoralist activities, and capacity-building activities. 33. The Government reported that a number of returnees who had been removed from their jobs have been reinstated following their return, particularly at the local level, such as in Rosso, where local authorities stated that 95 per cent of education officers had been reinstated. However, victims and their families claim that many still await restoration into service or satisfactory compensation. In addition the Special Rapporteur heard from the victims he met that some of the children of those killed had been prevented from registering in the census as they were asked to produce a death certificate for their parents. C. Disaggregated data 34. The Special Rapporteur was concerned by the absence of data disaggregated by ethnicity and descent, which was underscored in all the meetings he held and confirmed in the various studies on the status of the population he was able to access, which did not include any disaggregation of data. All government interlocutors explained to the Special Rapporteur that this was a result of the need to respect people’s privacy and the difficulty individuals have in identifying with only one ethnicity or community. While acknowledging the rationale for this policy approach, the Special Rapporteur is of the view that the absence of appropriate tools to assess the spread of ethnic and descent-based discrimination, such as reliable statistics and data, is among the reasons for the lack of awareness as to the extent of these phenomena and understanding thereof. The mandate has consistently recalled that collection of ethnically disaggregated data and indicators is a prerequisite for identifying the groups affected by racial discrimination, gaining a better understanding of the nature and extent of the inequalities these groups and their members face and for designing targeted policies and measures to remedy such inequalities. 35. While acknowledging the risks that the collection of such data may entail, including aggravating tensions between individuals and groups and leading to further prejudice and stereotypes, and conflicting with individuals’ right to privacy, the mandate has indicated that these problems may be overcome if States abide by some key principles, including selfidentification (A/66/313 and A/65/295). D. Census 36. In May 2011 a nationwide census was launched, aimed at registering the population in a biometric database, systematizing national identity cards and finalizing the electoral lists. This is a welcome initiative, since in 2004 the Committee on the Elimination of All Forms of Racial Discrimination recommended that the State carry out a precise population census not limited to linguistic factors, and produce detailed indicators disaggregated by descent or ethnic origin. The Committee also recommended that the State party carry out targeted surveys, on the basis of voluntary self-identification, which will make it possible to determine the situation of the groups falling within the definition of article 1 of the Convention on the Elimination of All Forms of Racial Discrimination (see CERD/C/65/CO/5). 37. The authorities explained that this exercise was needed to safeguard the security of the country as well as that of its residents and citizens, through the establishment of a centralized national biometric database to replace the 13 uncoordinated regional population databases which were dismantled in May 2011. The census process started with the opening 10