You can map the “where” and the “what”. These two methods cover both.
Now that we’ve collected over ten thousand stories and transcribed about 6000, I’m starting to try and make groups of stories about important subjects more digestible. This map shows where all the stories about rape are coming form:
For reference, here are locations for all stories:
Within theinteractive map, you can scroll down to read all 100 stories in a list, or click on any location marker to bring up a particular story.
Geo accuracy: Those 10 stories from Southwestern Kenya should be in Kibera (Kisumu Dogo is a village) – but otherwise BatchGeo.com worked okay. Overall, 2.3% of all stories were about rape, and over half these stories were told by men.
VAP was interested to see whether their stories (frequently about the problem of rape) were typical of Kenya as a whole. This data ought to be able to answer that question, but it is still hard to parse and compare groups of stories. It is surprisingly hard to calculate the frequency of stories about rape per region within Kenya. I’ll need to sort by city and leave the rest out of the equation.
Using SenseMaker(R) software licensed from Cognitive Edge, I imported other aspects of these 110 rape stories. (Every story mentions either rape or sita kimya.) Each story has more relevance to the idea or the people who benefited. The answers on our survey were: Good idea, succeeded; Good idea, failed; Bad Idea and Right people, Wrong People, Nobody. When you combine both answers SenseMaker lets you create a plot like this:
The dots represent where individual stories fall. Are they more about Good Ideas that succeeded and helped the right people (top), or are they Bad Ideas that benefited nobody (lower right)? I’ve moved these six labels around to get the combination that best parses the data into two major groups.
There are many other filter options on the left. Playing around with these 110 rape stories, I realize that the most represented organization is Sita Kimya and USAID (which funds this anti rape messaging campaign in Kibera, Nairobi).
This plot shows that 28 of 110 stories are related to Sita Kimya or USAID and the pattern is much like the whole set:
20 of the 29 stories from women are tagged as “NONE” or “None” – meaning the women did not identify any organization as the subject of their stories. Sita Kimya, as the USAID website explains, is clearly targeting men:
And they seem to be reaching their target demographic:
The above plot represents men who talked about Sita Kimya. 21 of the 78 stories about rape are about Sita Kimya specifically. Every single one of these men identify themselves as either an observer or an actor in the story they told. None are “affected by” the events in the stories.
So who is helping the women? In past stories, we saw that Box Girls International was teaching them self-defense skills, and I previously talked about how VAP tries to reach young women in Majengo with some straight talk about sex.
This kind of searching for patterns in story themes is much richer than the geomapping that is all the rage right now in big development agencies. But of course it is much harder to do successfully. How do you know when you’ve found the right pattern? There are multiple interpretations.
Here is a map of all the words from stories that mention Sita Kimya (in blue) and Mrembo (in red). The purple nodes are words that are shared between both stories about equally. It underscores just how different these two sets of stories are, in terms of what the storytellers emphasize:
- HIV and AIDS
- Parents, pregnancy, marriage, avoid, and body
- Police, hospital
- know, men
- Rape, sex, girl