This is a read only archive of pad.okfn.org. See the
- What kind of preconceptions do we have regarding ebola stories/narratives?
- How does the media provide information about ebola?
- What can we learn from the datasets we already have?
- Can we find more references from our datasources? Microdata?
- Do we have expertise/skills/technical language to describe what we're looking for?
- Should we use only verified/official data?
- Do crowdsources report ebola cases earlier than official ones?
- WHO response map x Cases map http://www.who.int/csr/disease/ebola/maps/en/
* Center for Disease Control: What is "contact tracing"? "Even one missed contact can keep the outbreak growing.": http://www.cdc.gov/vhf/ebola/outbreaks/what-is-contact-tracing.html
- On "contact tracing" methods:
- presentation: tiny.cc/ODC14
- Natalia Mazotte: email@example.com
Notes for blog post
Two sources found – CDC and WHO
Linking ebola outbreak to human development index of the countries and mortality rates -
political question: break down of political systems vs treatment, capacity of country governments to react to it
Cases under 100 people – how was it contained it so easily?
For some cases it was 80% fatality, when numbers went up, it's still high – still an awful way to die but definitely not a sure 'death sentence'
Eg. what's different about this political moment in international situation – why sensationalised coverage now?
Before – rural areas, now increased number of people travelling? It's now in cities...
MSF said it was out of control for the first time – what does this mean? Didn't know exactly which areas were contaminated – fear of the unknown.
Thought about questions – integrity of 'contact' tracing method – what kinds of preconceptions do we have on different data sources, crowd sourced data, media coverage. 'established' sources of data which might flow a bit more slowly...
What can we learn from data that does exist? What can we glean from the limitations of the data, where are the gaps? Simply: we don't have enough. What can we know, based on existing data sources, what is slipping through the cracks?
Scout work : we had some data sources, interesting maps – http://pad.okfn.org/p/ebolaexpedition
Information about contact tracing methodology: CDC puts guidance on it, but even from Nigeria – scientists saying that the contact tracing method was being applied very inconsistently. Where people get the disease, stay home, die without ever contacting the health system.
Risks of stigma: underreporting. If we were to report on data available
How much information presented as a map in a PDF – wow.
Incredibly messy data – eg. CDC, sometimes start to end, or months, years, asterisks, numbers of cases with an asterisk there – needs lots of time to clean.
Nobody worried about providing not just machine readable data, but clean data – presented for a human reader, not a machine.
Lots of data presented super simplified, so don't know what is getting lost.
Nigeria - * to say – we don't have good or consistent tools for this,
Trying to absorb yourself in the topic – before you can really start making any conclusions, is clear. Good research practice, looking at what can be triangulated and what can't.
Different interpretations of open data – transparency on research methods, judgements on quality, what you would want is to have every step documented, with syntax that took you from step A to step B – choose at what level you want to dive in.