"The new census will improve the quality of information on the most vulnerable population" 

"The census is already being compiled using mobile phone data to calculate population mobility indicators."

Spain will be the first country in Europe with a population with more than 16 million inhabitants which has decided, starting in 2021, to abandon the traditional census and adopt a new census incorporating Big Data.

Antonio Argüeso, Vice Director General of Sociodemographic Statistics at the Spanish National Statistics Institute (INE), main creator of this important change, participated a few days ago in a seminar organised by the Centre for Demographic Studies (CED) at the UAB campus, in which he analysed the evolution and future of the census in Spain, now that the INE has decided to pursue in the year 2021.

-How does this new "public statistical product" compare to what currently exists?

It will be a true census, in the sense that it will gather data on all of the population and therefore will be an exhaustive recount; but the information will not be obtained through door-to-door surveys, but rather be compiled through different administrative registries and that is what makes it so new.

-What are the main changes in how the data is obtained?

From our point of view, all changes to a traditional census is an improvement. We will not have less information; we will have at least the same information a traditional census through surveys offers us. We will additionally have new data (for example, a classification of households according to electrical consumption) and with a higher quality, quicker and with more frequency.

-Some experts doubt whether this new "census" will reflect the regional details and small populations which were earlier included, and that the social and demographic phenomena emerging in these areas will not be equally distinguished. Is that possible?

That is precisely the reason why these seminars were organised at the UAB. It is true that part of the academic community may be hesitant to accept this change in model for fear of having information being lost; we explain that this procedure will fundamentally introduce improvements, and I think we have been able to transmit that during these seminars. Of course, no regional details or information on the population will be lost; we can guarantee that.

-It is clear that the use of Big Data will multiply the possibilities of obtaining public censuses or statistics, but might it also be a handicap because it will be dealing with a larger amount of personal data, which could therefore compromise anonymity?

These new sources of data are a huge opportunity and the INE is determined to incorporate them, although we must be cautious in the process of doing so. We must understand the data, know how to explain it and how to treat it. It is not as easy as it seems and there will be an immediate revolution in our statistics. In the case of the census, we are already using mobile phone data to calculate population mobility indicators. Of course, we do this in the form of aggregate data, always respecting the privacy of individual information. In this sense, the use of big data does not represent a great change in how we act. The INE never offers information which can directly or indirectly reveal any personal data.

-To what extent is a public survey or census a true reflection of reality? In other words, isn't there an invisible part of the population (vulnerable sectors) that does not appear in these statistics? And, if so, how can this be solved to reflect these parts of society?

Of course, there is always a risk in any type of survey or statistical information. The most vulnerable groups are not necessarily the ones that fare worst in our statistics, it depends on the type. In fact, we find more difficulties in our door-to-door surveys to interview people of the upper classes. In any case, as was commented in the seminars organised by the CED, the future population census based on administrative data will improve the quality of the information on more vulnerable groups, since much of the data will come from administrative registries.

-Has the National Statistics Institute (INE) felt "used" politically, in general, in terms of having the data you offer be interpreted in different ways according to the interests of each party?

the data published by the INE are often a source of political debate, I would almost say that it is constantly so. But the INE stays out of this debate. We offer transparent figures to all of society, which can be equally accessed by all. That is why it is so important for a statistics institute, in Spain and in any of our neighbouring countries, to be an independent institution (the INE is an independent organisation) which publishes its statistics without any interferences.

-Regarding the global Covid-19 pandemic, population data and figures seem to have become more important than ever. Has this been managed correctly?

Generating statistics is more difficult than it seems. In any case, I cannot give a general opinion on that because I do not know how many of the indicators from other organisms have been created. The ones I am aware of, such as death figures, the INE offers the latest figures in this sense. They differ from what is shared on deaths due to COVID-19, but in our case we collect data on all deaths. This divergence also exists in many other countries in which only the deaths with a PCR are collected on one side and deaths for other causes on another side.

-A few days ago you participated in a debate organised by the CED-UAB on the evolution of the census in Spain. What were the main conclusions reached at the debate?

These meetings are very important for us and the Centre for Demographic Studies (CED) is a leading instiuttions in Spain. Population censuses are compiled with a methodology that represents a qualitative change and it is very important for these exchange forums to exist in order for the academic community to discover these updates first-hand and also for them to help us, since they are the main users of these tools and we must bear in mind their needs and the concerns from the beginning. I think we fully met all of our expectations. We are very satisfied.

-One of the issues that came up in the seminar wes a census as state-building tool...

Well, it is said that population and housing censuses are one of the elements that identify the consolidation of modern states, around the 18th century. It is then that these censuses of demographic interest (discovering the true number of inhabitants of a country) are first conducted; but not in feudal countries, in which other types of "censuses" were conducted, with the aim of collecting taxes for wars. The demographic censuses therefore are a sign of modernity. And as stated in the name of these seminars, we have one of the earliest modern censuses of the world, the population census ordered by the Count of Aranda in 1768.