Guardian partners with Caribbean GIS to track chikungunya in the Caribbean

Screen Shot 2014-10-20 at 3.10.41 PM-2

Everything about chikungunya is painful. Even the virus’ name comes from a Kimakonde word describing the contortions of one suffering severe joint ache. Fever, rash, cramps, headache, nausea and fatigue are just some of the symptoms of the mosquito-borne illness.

Nor is tracking the spread of the disease across the Caribbean any easier. English-language reports on the virus’ transmission at the sub-regional level are put out by public health authorities, including the Caribbean Regional Public Health Agency (CARPHA), the Pan American Health Organisation (PAHO) and the Center for Disease Control (CDC).

But keeping up with information from all of these sources can be time-consuming, especially if you just want to keep an eye on the spread of the disease in your own country, or get a sense of the broader regional picture.

“It’s easy to point a finger and criticise but I thought it would be better to actually demonstrate that something better could be done,” said Vijay Datadin, founder and lead consultant at Guyana-based Caribbean GIS.

Datadin should know. He’s made a career of applying geographic information systems (GIS) to the complex interrelationships between human and natural resources.

“When I looked at the outputs of CARPHA, PAHO and even the CDC, I thought they could be enhanced. Specifically, PAHO is putting out data reports in PDF format, which is really less than ideal. Where there are maps, they could be made more informative and charts would help citizens understand the situation more easily. I felt it could be done better, because they’re still doing it in the old-fashioned way.”

Datadin is co-lead on a new open data project that aims to fix two major pain points associated with the “old-fashioned way” of sharing public health data. First, as a one-stop resource for official chikungunya numbers, the online tracker seeks to cut out the hassle of having to check multiple websites, in order to get the latest collated statistics on the spread of the virus.

The second pain point is how chikungunya transmission data are presented by the three leading public health agencies in the Americas. Modern public health sites like HealthMap aggregate news reports in real-time and push notifications to subscribers, filtering by relevance based on geolocation. They are built with responsive design to dynamically adapt to different form factors such as mobile devices, tablets and desktop screen displays. Plus they are mobile-optimised for lightweight browsing, and social-friendly for maximum user engagement.

By comparison, the regional websites are less impressive. The CDC website provides a static map showing countries where local transmission has been documented, and says that “chikungunya case counts are publicly released every Wednesday.” PAHO provides a weekly report every Friday afternoon of Chikungunya counts for most countries of the Americas and a static map showing countries with local and imported cases. CARPHA provides a weekly update of Chikungunya counts every Monday. The CARPHA site also has an interactive map with a useful timeline feature illustrating the progression of the disease through the region and mouse-over info boxes showing the number of cases in a country.

The region's public health services could learn from the open data approaches that are becoming the expected standard for providing public information, Datadin said.

“Around the world, public organisations are no longer simply publishing their data in PDF format or static maps but in open data formats and interactive maps. The value in doing it this way is that data scientists, researchers and other interested parties are then able to not just see the data but actually use it,” Datadin said.

His latest project, a joint initiative of Caribbean GIS and the T&T Guardian’s new media unit, brings to traditional public health reporting the transparency of open data formats and the interactivity of data visualisation. The end-product is an online map-based chikungunya tracker that makes it easy for anyone with Internet access to follow the regional diffusion of the disease, using public health data extracted from official sources. The tracker is online at http://www4.guardian.co.tt/map-chikungunya-caribbean.

Data for the map and charts on this page were extracted from PDF reports published by PAHO, reformatted and combined with a GIS base map. On the Caribbean GIS Health site the map is accompanied by other charts and timelines that provide historical context and make each country’s demographic situation easier to grasp at a glance.

The  improved Chikungunya dataset is also made freely available as a Fusion Table so that other researchers, students and citizen scientists can view, filter or merge with other data with just a browser, or download for further analysis.

Mapping Caribbean Crime

The joint project isn’t Datadin’s first foray into data journalism. In 2003, Datadin founded Red Spider, a small web development startup, which today maintains the Guyana Crime Reports,  an open data tracker for several categories of serious crime in Guyana. The website is part news aggregator, part crowd-sourced citizen journalism platform.

In 2013, Guyana Crime Reports collaborated with the now-defunct Bullet Points, an earlier open data journalism project involving Guardian’s new media desk, which tracked intentional homicide as well as deaths caused by shootings involving police officers. Datadin, who was at the time working on Guyana Crime, worked with Bullet Points to develop a GIS-powered map of 384 murders in its 2013 dataset.

“I feel that the Caribbean is better off when its citizens are better informed,” Datadin said.

How new runoff rules would have impacted 2007 polls

2007 Run off constituenciesIf the constitutional changes being proposed by Prime Minister Kamla Persad-Bissessar were in effect during the 2007 general election, former United National Congress (UNC) political leader Basdeo Panday would have been forced to face a runoff vote for his Couva North seat. And he would have had plenty company. On Monday, the PM proposed a series of changes to the electoral process, including term limits, the right to recall non-performing MPs and the introduction of a second ballot runoff vote system.

“A runoff poll is proposed so that each member of the House of Representatives will only become such a member if he obtains more than 50 per cent of the votes cast in a constituency,” the PM said. The map (Page A1) of the 2007 general elections results highlights the potential impact of that specific proposed constitutional change. Fourteen of the 41 electoral districts—more than 33 per cent of the available seats—would have required a runoff vote.

That election was won 26-15 by the People’s National Movement and interestingly, the winner is likely to have remained the same, as the runoffs would have been required mostly in the 15 constituencies won by the UNC. Including Panday, a total of nine UNC candidates would have gone back to the polls for runoffs, in Caroni Central, Couva North, Couva South, Cumuto/Manzanilla, Fyzabad, Mayaro, Princes Town North, Tabaquite and Vasant Bharath’s St Augustine.

Bharath’s runoff would have been against his current coalition counterpart Winston Dookeran, who was then political leader of the Congress of the People (COP). The remainder of the UNC wins—Caroni East, Jack Warner’s Chaguanas West, Naparima, Roodal Moonilal’s Oropouche East, Oropouche West and the Prime Minister’s Siparia constituency—would have been won outright. The People’s National Movement (PNM) would also have been affected, although to a lesser degree.

Of their 26 seats, five would have been returned to the polls—Barataria/San Juan, Chaguanas East, Pointe-a-Pierre, Princes Town South/Tableland and St Joseph.

In this scenario, the PNM would still have won 21 seats outright—Arima, Arouca/Maloney, D’Abadie/O’Meara, Diego Martin Central, Diego Martin North/East, Diego Martin West, La Brea, La Horquetta/Talparo, Laventille East/Morvant, Laventille West, Lopinot/Bon Air West, Point Fortin, Port-of-Spain North/St Ann’s West, Port-of-Spain South, San Fernando East, St Ann’s East, Tobago East, Tobago West, Toco/Sangre Grande and Tunapuna.

In the May 24, 2010 polls, none of the winning candidates got less than 50 per cent of the votes so no runoffs would have been needed. The People’s Partnership coalition government beat the PNM 29-12 in that election.

From T&T Guardian

Changing the game for data collection in the Caribbean

CAPTION: Dr Kenfield Griffith, CEO, mSurvey and Anil Ramnanan, PhD student at The University of the West Indies (UWI) Department of Computer Studies, at mSurvey’s workshop on Data Collection and Surveying using Mobile Technology, held at the Max Richards Building, Faculty of Engineering , UWI St Augustine on March 28, 2014. PHOTO: GERARD BEST A mobile SMS-based survey service from a Caribbean-based company could change the way data is collected and analysed in the region.

 

If Kenfield Griffith has anything to say about it, his company will soon be adding potent fuel to the digital revolution smouldering quietly throughout the islands of the Caribbean.

 

Born in Montserrat and of Barbadian extract, Griffith is the CEO of mSurvey, a mobile surveys company based in Kenya. Kristal Peters, Director of Business Development and Strategy, runs the company’s Trinidad and Tobago office.

 

"It's Friday morning. Let's create a survey together," Griffith says to a group of relative strangers gathered in a small room in the Max Richards Building at the Faculty of Engineering of The University of the West Indies, St Augustine for mSurvey’s workshop on data collection and surveying using mobile technology.

 

His confidence seems well placed. Within minutes, the demo survey is set up and sent to a pool of prospective participants located in Kenya and Trinidad and Tobago, who immediately start returning their responses via SMS technology. Soon, their data start streaming in to a dynamic web page, which aggregates and visualises the survey results in real-time. In no time at all, the roomful of workshop participants, about fifty in all, are analysing the fresh data.

 

The audience is an interesting mix of academics, researchers, policy makers, mobile carrier representatives, students and software developers, and many seem eager to learn more.

 

“Do you sell data to third parties?” asks one man seated toward the middle of the room.

 

"We have the technology that folks use to get other people's data. But we don't sell anyone's data to third parties," Griffith replied.

 

Moments later, he clarified his business model. The primary service that mSurvey provides is to help people, businesses and organisations to use mobile technology to get the precise data they need to make high-impact decisions quickly.

 

"We're trying to solve a problem here and that problem is getting data.”

 

To have some idea of what Griffith means, you need only to have tried to get survey data quickly and reliably in the Caribbean context. For many organisations trying to use survey data to harvest meaningful insights and increase their ROI, the biggest stumbling block is the inability to gather data in the first place. Door-to-door surveys are costly and painfully slow. Open data sources like the World Bank are always just a click away but don't necessarily give the specific insights required for contextual decision-making. And commissioned online surveys are challenged by the limits of the local population's access and connectivity to the Internet. By some estimates, residential broadband Internet penetration in Trinidad and Tobago, for example, remains as low as 45%.

 

“Getting information in emerging markets is a pain point for most of us,” Griffith said.

 

But for every problem, a solution. Mobile penetration in Caribbean islands like Trinidad and Tobago can be as high as 140%. Everyone, statistically speaking, has a phone...or two. So the mSurvey platform allows the entire survey process to be completed over a regular mobile SMS plan at no cost to participants. Respondents don’t need mobile or Wi-Fi broadband Internet connection, nor even a smartphone.

 

For mSurvey, the ubiquity of the mobile phone has become the answer to one of the region's biggest obstacles to data collection.

Surveying the region: Changing the game for data collection in the Caribbean

A mobile SMS-based survey service from a Caribbean-based company could change the way data is collected and analysed in the region.

If Kenfield Griffith has anything to say about it, his company will soon be adding potent fuel to the digital revolution smouldering quietly throughout the islands of the Caribbean.

Born in Montserrat and of Barbadian extract, Griffith is the CEO of mSurvey, a mobile surveys company based in Kenya. Kristal Peters, Director of Business Development and Strategy, runs the company’s Trinidad and Tobago office.

"It's Friday morning. Let's create a survey together," Griffith says to a group of relative strangers gathered in a small room in the Max Richards Building at the Faculty of Engineering of The University of the West Indies, St Augustine for mSurvey’s workshop on data collection and surveying using mobile technology.

His confidence seems well placed. Within minutes, the demo survey is set up and sent to a pool of prospective participants located in Kenya and Trinidad and Tobago, who immediately start returning their responses via SMS technology. Soon, their data start streaming in to a dynamic web page, which aggregates and visualises the survey results in real-time. In no time at all, the roomful of workshop participants, about fifty in all, are analysing the fresh data.

The audience is an interesting mix of academics, researchers, policy makers, mobile carrier representatives, students and software developers, and many seem eager to learn more.

“Do you sell data to third parties?” asks one man seated toward the middle of the room.

"We have the technology that folks use to get other people's data. But we don't sell anyone's data to third parties," Griffith replied.

Moments later, he clarified his business model. The primary service that mSurvey provides is to help people, businesses and organisations to use mobile technology to get the precise data they need to make high-impact decisions quickly.

"We're trying to solve a problem here and that problem is getting data.”

To have some idea of what Griffith means, you need only to have tried to get survey data quickly and reliably in the Caribbean context. For many organisations trying to use survey data to harvest meaningful insights and increase their ROI, the biggest stumbling block is the inability to gather data in the first place. Door-to-door surveys are costly and painfully slow. Open data sources like the World Bank are always just a click away but don't necessarily give the specific insights required for contextual decision-making. And commissioned online surveys are challenged by the limits of the local population's access and connectivity to the Internet. By some estimates, residential broadband Internet penetration in Trinidad and Tobago, for example, remains as low as 45%.

“Getting information in emerging markets is a pain point for most of us,” Griffith said.

But for every problem, a solution. Mobile penetration in Caribbean islands like Trinidad and Tobago can be as high as 140%. Everyone, statistically speaking, has a phone...or two. So the mSurvey platform allows the entire survey process to be completed over a regular mobile SMS plan at no cost to participants. Respondents don’t need mobile or Wi-Fi broadband Internet connection, nor even a smartphone.

For mSurvey, the ubiquity of the mobile phone has become the answer to one of the region's biggest obstacles to data collection.

Tracking Trinidad and Tobago murders

I've blogged before about the open data journalism project called Bullet Points. Now I need your collective experience/expertise to help me with the project. Bullet Points tracks murders in Trinidad and Tobago. The murders are tracked on this Google doc.

The fatal incidents are listed chronologically and are numbered by victim (murder toll). The victims are announced on Twitter by their toll number. This system has some serious limitations.

PROBLEM Firstly, it is geared toward tracking the murder victims but is an inadequate way of tracking the actual killings themselves. For example, when there is a double murder, the toll goes up by two but the number of incidents goes up by only one.

Secondly, killings do not always occur in the same chronological order that the victims' bodies are discovered. Sometimes bodies of murder victims are discovered after several days. When this happens, the relevant killing must be inserted somewhere in the middle of the existing chronological list, which obviously changes the toll count of all subsequent victims. Unfortunately, because our current system relies on the toll count to identify the victim, it has now run into this serious limitation.

PROPOSED SOLUTION I think that what I need to do is add another column that contains a unique reference number for each incident. The unique reference number can follow a standard protocol such as [DATE][INTEGER]. For example, the most recent murder, which was discovered today, would have a reference number 2014013001. I think this will deal with the problem but I am not sure it is the best solution.

Thanks for taking the time to read this. If you think you can help, leave a comment below to share your perspective on this problem and proposed solution.

Mapping Caribbean crime: Guyana Crime Reports' data visualisations take fresh aim at crime

Crowdsourced, technology-driven and visually compelling. Not words you'd typically associate with crime reporting in the Caribbean. But that could soon change as one tech entrepreneur based in Georgetown, Guyana is taking a fresh approach to the country's crime problems.

Meet Vijay Datadin, the main player behind Guyana Crime Reports, the country's newest data journalism website, which adds a combination of GIS mapping and crowdsourcing to traditional crime news reporting.

"You see police in Guyana and across the region calling for public assistance in fighting crime. They've never particularly mentioned that this was the kind of help they wanted or needed, but this was what I knew I could do to help. So I decided to just do it," he said.

In 2003, Datadin founded Red Spider, a small web development startup, which today maintains the Guyana Crime Reports and its presence on FacebookGoogle plus and Twitter , @GuyanaCrime.

The website is part news aggregator, making it a good one-stop source for various crime reports related to Guyana, published in local and international news media. Citizens can also submit crime reports through a form on the website, although those reports are treated slightly differently from the ones aggregated from traditional media sources.

"Particularly when these reports are made anonymously, we follow up with some sort of verification exercise, especially if the report in question could damage to someone's name, reputation or character," Datadin said.

"But reports from the media have a certain amount of verification built into them because there's a journalist and/or editor involved in that publication process, so those are simply aggregated."

Enhancing Public Debate

Red Spider is considering forging informal relationships directly with journalists who share their interest in improving the way that crime is reported in Guyana. The aim, Datadin explained, is not to competewith old media companies but to enhance the essential news service that they provide.

"We're not a news service, and we will never be in the business of breaking news. At times we're first to break a story, and that's almost accidental."

For Datadin, old and new media share a common goal. They exist not just to distribute information but to help concerned readers make sense of large amounts of information over time. Ans he takes seriously the responsibility to help readers and followers to understand how local incidents of crime fit into a larger national picture.

"As a citizen of the country, it would be to my benefit if crime went down. So I'm doing this not for any immediate commercial benefit but because I think it needed to be done. There needs to be a public conversation about crime, a conversation based not only on opinion but on facts, one that affords a more reasoned and inclusive debate about factors that cause crime nad the policies that can help curb it," he said.

"I think that with a more informed public, we can have a better conversation about what we should do about crime," he added, conceding that the site has taken only early evolutionary steps toward that goal.

The ultimate objective, he says, is to have a positive impact not just on public discourse but on public policy.

"We made soft approaches to the Guyana Police Force and the Ministry of Home Affairs," he said.

Mapping Crime Data

Datadin holds a postgraduate Masters degree in Geographic Information Systems (GIS) from Edinburgh University, Scotland. Plus, the word 'data' is literally in his name! So the fact that Guyana Crime Reports relies heavily on maps to visually represent the spread and scope of crime incidence really shouldn't come as a surprise.

"Maps help readers to see exactly where one incident took place," he said.

By making the crime maps public, Guyana Crime Reports effectively creates an equal opportunity for anyone seeking to understand how crime is trending both in their area and nationally.

"Both the public and the Home Affairs officials can review the map and detect trends in a particular area or nationally. You can see not just what happened recently but what has been happening over time."

Building Digital Journalism

By using maps to visualise crime data, Guyana Crime Reports has already set a significant precedent for digital journalism in the region. Audiences across the region would benefit greatly if more Caribbean newsrooms added maps to their arsenal of storytelling tools.

Crime maps are a powerful communication tool, giving audiences a quick grasp of the bigger picture without requiring them to know the (sometimes gory) details of every incident.

Because maps impart understanding visually, they allow crime reporters can attract a different kind of reader, one who may be less inclined to...read. And any way you look at it, that's still good news.

Speaking of maps

One quick side note as a footer to this piece about maps. If you're wondering, Guyana is actually not an island at all but a South American country bordered by Venezuela, Brazil and Suriname. But it is a member of the Caribbean Community, Caricom.