What did you think of the information we have received from Osakidetza during the pandemic?

I find the trend in recent days worrying. Over time, data should be given more easily, because they are creating dynamics, but we have seen the opposite. The publication of specific data has been decreasing. Previously, the data that appeared segmented now show us united, which makes the analyses difficult.

Urkullu said the other day that the public use of the data must be done with caution. The desirability of not publishing some data. And that worries me.

Lehendakari referred to the limitation of the information and noted that in some cases the data is misinterpreted in an ‘interested’ manner. Can it be better not to give data at times?

No. Quite clearly. I agree, sometimes the data is difficult to understand and can be used in an interested way, of course. If it is observed that the problem is in a position to understand this data, pedagogy should be provided. That they publish with all the necessary explanations, or that they empower the media, for example, to do that pedagogy. It does not seem to me that public data can never be given or reduced.

Among other things, because the option of not providing data can also be done ‘in an interested way’.

Of course. As the data are demarcated, this constraint is being carried out on the basis of interest. And if these interests are not published, the data is even less reliable.

Nothing escapes interest. When we talk about data, there's always a certain mindset that data is objective. Just say it's just data. And that's not true. The data is collected and disseminated according to interests that we are now seeing clearly: where the data is collected, what kind of tests are mixed, how the origins of those data are distributed, how they are published…

Behind all this lies a number of criteria which, if they are not made public and also more and more criteria are applied, without making this data public, the reading of the data becomes increasingly limited. The understanding of the situation is becoming more and more digested, which limits many possibilities for interpretation.

And how about the citizens with the data? Do you think these people are able to digest for themselves?

I've made my thesis about data journalism and mathematical anxiety, and from there I'll tell you my opinion. I think we have a huge asymmetry between data availability and people's overall capacity. We have far more data than we are able to understand and work as citizens. There is therefore a need to work on mathematical competence.

That said, if it's not realistic to bring it to that mathematical level that would require global citizenship, we should at least be empowering the media, which can play a very important role. As interpreters of what happens, you can walk a middle way.

Because we don't all have to be epidemiologists or mathematicians. The media can digest and somehow transmit this complexity of data, presenting as transparently as possible the criteria mentioned above. In this way, when receiving a reading, the citizen can know in what sense this information should be understood.

In this sense, are we sufficiently ‘empowered’ journalists?

Journalists traditionally have a frontier with mathematics. A tendency not to abuse mathematics. In my thesis, I've analyzed that a general profile of people who come to journalism has a bias against mathematics. A prejudice against his own ability to work mathematics, which makes it very difficult to do what I mentioned earlier.

What is the attitude that we should have towards data?

We should act more courageously, giving the data without fear, and with the necessary training, to make the interpretations as honest as possible. And I say this honestly, because it's impossible for it to be objective, even when we're talking about data.

There is always an interpretation, that is where we have to start. Having said that, how do we publish the data? Well, knowing and publishing their own limitations, so that each reader then understands the data as they can.

And by always publishing the source of the data so that whoever has the most capacity can access them.

What do you think is the indicator that best reflects the impact of the pandemic at this time?

The main difficulty we have at this moment is that of reporting the data. In fact, it is virtually impossible to know the actual amount of the positives and coronavirus related deaths. There is a discussion with the story, because it is not clear when a death occurs from the coronavirus or from some other cause. And if we get into the question of positives and tests, we also have a long debate.

However, the mortality data make it possible to overcome these difficulties in part, since the mortality figures of the past years are taken into account and compared with those of this year. Not only coronavirus related deaths, but all. And the increases reflected in it can serve to symbolize the impact of the pandemic.

At the international level, more and more MoMo data-based reports are being developed. The Financial Times, for example, uses this to give figures from all over the world, which were initially based on the positive ones and then on the dead.

The MoMo provides a more reliable approach, using statistical terminology, more robust, as the story issue does not influence this indicator. And as I have already said, that is precisely the biggest difficulty in measuring the pandemic.

Profile

Miren Berasategi (Donostia, 1982)

He is a professor and researcher at the University of Deusto. Graduate in Social Sciences, he attended the Master's Degree in Information and Knowledge Society and later specialized in statistical methods. Since 2014, he has analyzed the need to incorporate the journalistic training of data to the degrees of Communication, investigating the influence of mathematical anxiety in students. At the beginning of the year he presented his doctoral thesis entitled The Work of Data Journalism in Communication Grades: a proposal from the perspective of mathematical anxiety and its effect on performance.