EMIS and Data Sovereignty

The data sovereignty of Ghanaians is being violated in a big way right now and as a result, the livelihoods of hundreds of thousands of people are at risk.

Food sovereignty is important; there’s no denying that. If you don’t have control over your own food, then you don’t really have control over your life. After all, we all need food to live. Land sovereignty, too is vital, and much for the same reasons. Not having control over the land on which you live puts you in a seriously vulnerable position because it throws into question the stability required to be able to build something permanent. And not just a physical structure, either, but a life. It’s obvious that food and land sovereignty, and the stability that these things bring, are essential to develop and maintain a livelihood.

But, what about data sovereignty?

The only time I’ve heard this term used before is in describing the issues associated with cloud computing, but that’s not what I’m talking about right now. I’m giving this term a new use, and I’m doing so because I’m angry. I understand that this is an unusual way of expressing anger, but my hope is that someone will notice the new usage for this term and associate with it all of the vitality associated with land and food sovereignty.

Because right now the data sovereignty of Ghanaians is being violated and this violation is putting thousands of bright people at risk every year.

I cannot stress this point enough.

The culprit? EMIS, the very system that aims “…to improve planning, organizational efficiencies, data collection and analysis, information sharing and transparency in the education sector, especially as these activities may relate to helping meet objectives related to Education For All (EFA) and the Millennium Development Goals (MDGs)” (according to this InfoDev paper) and that is considered a “good practice in Africa“.

I’m going to ignore the issues I have with EFA and the MDGs (for now). Instead, allow me to explain how EMIS in Ghana works. Each year, a paper-based survey is created at the National Level and administered by officers at the districts and other data collection teams. The surveys are monsters that include over 150 pieces of data that must be collected from each school over a two-week school census period.

EMIS Officers at the district gather up all of the surveys for their district and enter the raw data into a Microsoft Access database designed at the national level of government.

From there, the data is locked away. It (sometimes) comes back to the district months later in the form of a pdf report which includes raw data and “analyses” of that data. Reports get to the regional level much later, if at all. The analyses are weak and only include a district-wide snapshot for one school year made up of UNESCO-suggested indicators like Gross Enrolment Rate (GER) and Gender Parity Index (GPI). Performance indicators are not included (but, hey, why should performance matter when everyone is getting a basic education?–I’m going to give my thoughts on this in my next post).

The entire process subverts the district EMIS/Education Statistics Officer and takes the data they collected about their district from them. District officers don’t have access to the raw data they enter into the EMIS database, so they can’t do their own analysis with it. Instead, they are at the whim of the decisions made by the national EMIS team in terms of what analyses should be done on the data. The analyses that are done are generally useless for district, or even regional, planning processes, but are great for placating donors who want to know that every child is in a classroom.

Like misplaced food or land sovereignty, misplaced data sovereignty has serious consequences.

The most obvious is a reduction in planning capacity at the district level. If education officers don’t have the data they need in order to be effective planners, how can they ensure that the key issues faced by their district are properly addressed? Simply put, they can’t. Beyond that, to address the issue of data availability to all levels of government who require it, parallel systems of collection which aim to get the required data have been created. The existence of parallel systems creates a lot of distrust, value dilution, and general strain on the processes that use them. Lastly, not having access to your own data is simply wrong. I struggle to elaborate on my last point because I find this violation to be appalling in a really visceral way and I’m not sure what else I have to say to get that across–it’s just wrong.

My initial statement probably seems pretty dramatic right about now, and I’ll admit that it is. But it’s also true. District offices are responsible for delivering effective education to the children in their district and in order to do that, they have to know what’s going on all the time. The Regional Education Office is responsible for monitoring and evaluating delivery by the district and also for planning resource disbursement to the districts. Once again, in order to do this effectively, they have to know what’s going on.

However, EMIS takes the control that district and regional offices need out of their hands. It puts a big kink in the planning process by limiting the scope of analyses and adhering to a timeline that doesn’t meet stakeholder needs and which isn’t constant from year to year. On a large scale, this means that education is suffering in Ghana, especially in the North. Literally hundreds of thousands of school-aged children either don’t have access to education, or have not made it through the pipeline because the system has failed them and districts often don’t have their own data with which to generate solutions to these issues.

Without education data sovereignty a poorly educated population is the best that can be achieved by districts, and the futures of hundreds of thousands of people are put into question; it’s as simple as that.



2 responses to “EMIS and Data Sovereignty

  1. Not just a problem in Ghana. Consider compulsory testing of kids at three grade levels in Ontario. Schools do get feedback fairly quickly but in a very general form.
    On the other hand, local data involves too small a population to have reliable results.

    • Thanks for the thoughtful response, grandma! I’ve been trying to get some info on information flow in Ontario education, but I was surprised to discover quite a lack of transparency, which I’m sure plays into the same issues faced by the Ontario education system. Still, though, they have access to basic information like enrollment data because the institutions collect it and store it well. Many schools don’t even have that in Northern Ghana, because they are only accountable to a system that asks them to give the data and then forget about it.
      The issue with small population sizes is definitely an existing one. The UNESCO indicators I mentioned all attempt to increase reliability by combining education and census data. Of course, that really only increases the unreliability, because census data and projections from census data aren’t the best. In the end, the issue of small populations for education data is an unavoidable one, but it can be mitigated by understanding that a small sample size leads to biases in the data.

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