Skip to content

Statistics

January 14, 2011

We had two volunteer groups do a massive survey of slum households for us in Nov-Dec 2010. The data has been collated and is now being analysed, and I just had a 90-minute phone conference with another volunteer who is doing statistical analysis, looking for correlations and comparisons.

First off – I <3 these volunteers. Increasingly, we’re getting better at finding and working well with long-term volunteers. They take more work up front, but they become hugely helpful long-term. The funny part is that it’s next to impossible at the beginning to tell who’s going to turn out to be a great volunteer, and who will just fade away. I need to figure out how to hold on to great volunteers better.

The preliminary numbers are fascinating – way higher unemployment, far lower internal migration, big gender gaps at different ages. But figuring out what questions to ask is almost half the work.

Probably the most critical question in our survey (and right near the end, when the household was relaxed) was “Have you considered selling your child?” I expected no-one to say so, but 15 households did, and 3 refused to answer, with 232 families saying no. Over a third of the households, probably more (there’s some math stuff to do to get a reliable estimate) knew a family that had.

There are also families with kids working, violence, alcoholism, drugs and so on. We’re basically trying to figure out what factors are associated with that. Figuring it out in a way that’s accurate means figuring out very very carefully what you’re asking. A single adult household with X number of children and Y income/adult as a double-adult household with X children and Y income/adult, what distinguishes them? And flip that around, all my X-children households with single adults vs double-adults, what’s the difference in income then? It looks like the same question, but the second one is I think better because it’s more specific. And it’s still not an answer. Is the income lower because the children aren’t working outside because there’s no-one else to watch the younger kids but another kid? Is it lower because when the adult falls sick, they have to go for a loan more often, than a two-adult household where they are less likely to both be sick at the same time?

Our very very preliminary analysis was looking for regression correlation blah-blah-stats stuff on child trafficking and the number of children, the education levels of adults and income. And the link is weak, which puzzled the stats volunteer. Income was more important than the parent’s education – but so far all we’ve discovered is that there isn’t a smoking gun.

And that, I think is closer to the truth.

Getting from a giant spreadsheet of data to cute stats like “Living on less than $2 a day makes a child 40% more likely to be sold” isn’t just difficult. It’s pointless. You could maaaaybe get that answer, but it would ignore any other factor. We’re taking very messy human lives and trying to work out links between them, and how we frame the question shapes what we see in the numbers. Ask a vague question, get a useless answer.

It is really, really hard to get numbers on child sex workers, or even sex workers, in Cambodia. Child laborers as a whole, I’ve seen numbers from 350,000 to 1 million in Cambodia. Usually, when I’m pulling together a grant, I’ll just rely on the UN or IOM or another well-regarded acronym. When it’s something puzzling, I’ll dig through to the back and try and follow how they got there. A couple of times, especially with stats relating to sex work, the numbers make no sense because the questions or the analysis is bizarre. But you wouldn’t know that without looking under the covers.

For our survey, I wrote up a list of questions, and most of them are variations on: “This is a rule of thumb we use based on our experiences. Is it real or a bad guess?”

I’m so curious to find out – are single-adult households more vulnerable? Does a trauma make a family more likely to splinter? Are the families new to the slums better off than the long-term residents? Do the uneducated parents have a harder time with their kids and finding work?

And from that, we can hopefully figure out who and what we should be spending our time and money on.

But first, we have to ask the right questions.

Advertisements
No comments yet

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: