We just wrapped up two days of mapping strategy and TileMill training with a group of Romanian NGOs supporting the country’s Roma minority at the State Department’s TechCamp Bucharest. The NGOs strategized ways to share data and publish better data on the Roma population to leverage in discussions with local governments around social inclusion plans and better provision of public services.

But NGOs need more accurate data than what is currently available — even basic data like how many Roma are in the country is highly questioned. According to the 2002 Census, there are 535,140 Roma people in Romania, accounting for 2.5% of the country’s population. These estimates are thought to be wildly low because Roma’s generally do not declare their background in the Census. Several NGOs we worked with quoted wide ranging estimates from 1 to 2 million, with that taking into account large population decreases starting in 2007 after Romania joined the European Union and emigration increased. World Borders Dark + ndi_romania + World Blank Light Edit descriptionapi.tiles.mapbox.com

A look at the questionable 2002 Romanian census figures on the Roma population percentages in each county.

The main focus in breakout sessions was on getting better data to make better data driven decisions. “We are making decisions based on numbers that are not correct” emphasized Elena Mihalache, Public Policy and Advocacy Senior Advisor at the Roma Civic Alliance of Romania. This need was furthered by Ana Ivasiuc, Research Coordinator for Agentia Impeuna, who said “we need better data on housing and infrastructure in Roma communities. Without good data we can’t gage what better services are needed.” While the country just completed a new Census this year with official population figures expected in May 2012, ethnicity data is not expected to be released until 2013 and the NGOs we are working with expressed little confidence that the data will better count the Roma.

What makes this especially complicated is that the lack of data can be self perpetuating. “This false data problem is a symptom — we need to make people aware that being Roma is not a bad thing,” says Ana Ivasiuc. But getting over stereotypes requires good data to dispel folklore. NGOs are working to help more Roma students graduate high school — but it is hard to solve a problem when the causation of dropping out of school are based on data. For example, it is often suspected that Roma girls drop out of school because of early marriage, but this is based mostly on interviews with teachers. “The lack of data perpetuates the stereotypes” Elena added. “There is a little running around our tail in order to map the data we need to correct the data.”

TechCamp ended today with the Roma NGOs sketching out plans of how they will better share data together to help fill the wider data void. The group presented their action plan to TechCamp’s lead facilitator Noel Dickover and asked the State Department to help them fund a new data collection push that was all focused on open data. I look forward to see how their efforts work to improve data moving forward.