#safeschools – A Data-driven Investigation into Seismic Safety Assessment of Italian Public Schools


By Elisabetta Tola, founder of formicablu.it and director of datajournalism.it.

May 2012: a series of earthquakes shakes Bologna, the city where I live, and the nearby provinces in the Emilia region. 27 people die. Hundreds are injured. Thousands have to leave their damaged homes and spend the summer in emergency camps. One of the strongest shocks is registered on May 29th, at 9 am. That’s when all children are in school. Including mine.

My country has one of the highest seismic risks in the world. But we are highly unprepared to cope with this risk. Most of our buildings are not safe and most people do not have the slightest clue about what to do when an earthquake strikes.

So, that morning, while I was helping the teachers to bring the children outside, I started reflecting on what I could do to improve the way we manage our coexistence with earthquakes. The trigger point came from a seismologist who said at a local radio station that laws and money had been put in place years before to assess the seismic safety of most Italian public buildings. Unfortunately, that was to no avail.

The story goes back to one of the most shameful pages of our recent history: the collapse of a primary school, due to an earthquake, in a small southern village, San Giuliano di Puglia, on October 31st 2002. That was the only building that came down, killing 27 children and one teacher. That school had just been renovated, but the work had been done in a completely illegal way. After that, our government approved a special law aimed to check all public buildings, schools included, within 10 years.

Guido Romeo, senior science editor of WiredIT, myself and two other colleagues set out looking for data on seismic risk assessment in schools. The magazine launched the #scuolesicure campaign (safeschools) to be run up to San Giuliano’s 10th anniversary.

And a real quest for data it was. Over 4 billion euros have been spent over 10 years (here is a timeline) but with very poor results. Less than 3000 public schools have been checked. This is more or less 1 in 10 schools out of the 22.800 considered at very high risk by the National Civil Protection and the National Institute of Geology. 8 million students and 1 million teachers who use these building everyday are deprived of the right to know whether they are safe or not.


We requested data from the National Civic Protection, the Ministry of Education, and various technical institutions, down to regional and local authorities. The more we digged the more we became disheartened. No one knows exactly how many schools there are in Italy: there is an open database enlisting them, but it is incomplete and inaccurate. The estimated number is of 42.000 public schools and 15.000 private one. Neither the Ministry of Education nor the National Civic Protection seemed to have a decently organised database describing the state of public buildings. And they were quite reluctant to disclose the (scarce) information they have.

We ended up scraping by hand a huge number of .pdf documents, passed onto us by a public officer. It was not secret information: it had even been published on the Official Gazette, although every person we interviewed seemed unaware of its existence! Basically, it consisted of tables detailing the money spent on checking seismic safety region by region.


Unfortunately, the data we were looking for was mixed with other data and there was no way to automatically clean these tables. We had to do it by hand, row by row, by merging our tables with the official list of schools published by the Ministry of Education.

That’s when we almost lost hope. In the two sets of documents, schools were mentioned with different names, locations were often uncertain and there was no unique coding system to identify a building.


Finally, we built our database using Excel, and, with the help of many great people (social activists, hackers, geomappers) we finally converted it into an interactive map, renamed our map of ignorance.

Most of our entries, in fact, lack a crucial piece of information: the result of the seismic risk assessment, the risk index. This index, ranging from 0 to 1, is meant to indicate how a building might respond to a strong earthquake: the closest to 0, the highest the risk. Only two regions out of 20, Lazio and Abruzzo, made this data available.

Our map works with a colour-code: each school is labelled according to the availability of information on risk. Pink, no information; yellow, checked but no risk index available; blue, checked and safe. Red, checked and at high risk.

#safeschools interactive map: the red label indicates a very high risk, with a risk index 0,28.

Next we looked into the financial as well as legal and technical details of how seismic risk assessment should be conducted. The entire set of articles has been summarised in Storify by Guido Romeo.

Our readers sent lots of comments and suggestions. The most significant one came from a mother living in Calabria, one of the regions with the highest seismic risk in the country. When enquiring about seismic safety of the local school, she got this as an answer: “You know, we are all in the hands of God”.

There it is, my main motivation. To fight this typical Italian fatalist attitude, the silly idea that we cannot do anything. Knowing the risk index does not solve the problem. But knowledge is undoubtedly the first step in the right direction. As a matter of fact, last year in Emilia we verified that some of the collapsed schools had been checked and rated at high risk. And yet, they were closed as a precaution only after the first strong tremor, which luckily happened at four in the morning, when no one was inside.

So, we won’t give up. I want to complete our map of ignorance with risk indicators and I’d like to share this work with colleagues who work in countries as earthquake prone as mine. Because we need to be alert in monitoring the safety and appropriateness of the buildings where we are educating and preparing our future generations.