Miguel Brechner of Plan CEIBAL didn't dive deep into XO cost calculations during his TEDxBuenosAires talk, but these seem like awfully low numbers:
How much did it cost us? We invested around one hundred million dollars. So that we do not delve too much into figures, each computer cost us around $188. Sixty dollars was the rest of the cost: servers, networks, antennas, tech support, parts, logistics, delivery... everything else. This was all accomplished with public funds, both domestic and foreign.
If we calculate four years of effective life per machine, it will cost us about $75 per year, of which $48 is the computer and $27 the rest of the servicing a project of this magnitude requires. To give you an idea: in the deployment phase that's less than 5% of the educational budget, and less than one two-thousandth of the gross domestic product.
I am curious to see how they are controlling the costs - perhaps Internet access is affordable due to a competitive marketplace (wish we had one of those in the United States) or existing subsidies for educational access.
Do these costs include Ministry-level overhead and teacher training, or have those been rolled into existing budgets? I wonder not so much as a criticism of their cost calculations - clearly CEIBAL is a shining star in both OLPC distributions and educational technology projects - but rather as a best-practices interest.
100 million invested into 380,000 laptops comes out to 263/laptop, less $188 for the raw cost leaves $75 (a rounding or inflationary error larger than the claimed $60) towards the rollout of servers, and a laundry list of what appears to be mostly logistical, support, and setup costs, but could also include teacher training.
CEIBAL calculates a four year lifespan of the computer, amortizing the hardware cost across the years as $48/year (which sums to $192 instead of $188), and claiming $27 annually per computer in administration and other expenses.
Boring TCO quibbles aside, the most fascinating piece I got from this talk was how they were framing their successes (Yes, Measurement and Evaluation is my exciting topic here):
"A lot of the data we gathered points to one thing: it was worth it. It was worth it because kids are more motivated when they go to school. It was worth it because they are more motivated to do homework. It was worth it because they are not repeating grades where we have been able to measure. It was worth it because we gave thousands of children identification documents since we did not give a laptop unless they had some sort of national ID, or at least the parents' ID. So in that sense the children were properly identified.
It was worth it because it increased self esteem in a lot of children. A lot of children learned about photography, about film, about music... At last, it was worth it because we have transformed a privilege, which was to own a computer in the year 2006, to a right."
Notice that nowhere in that list was "test grades have increased" or "literacy has improved by n percentage points." Early on in OLPC's life, 1CC pushed back strongly from any measurement and evaluation work, as they were rightfully dubious of the ability of standardized testing to measure the impact of working with the OLPC. These however are great meta-measures of student engagement, which are (almost) as valuable as any test score.
I remember questioning Negroponte himself on the potential value of these meta indicators back when he spoke in 2006 on the GWU campus (when he was still carrying around a fake prototype OLPC XO for the show-and-tell portion) and having them be thrown in with test scores as indicative of a lack of commitment and belief in the XO's transformative powers.
While some circumstances have an intense need to show test score improvements, the reality is that even in the best of circumstances, laptops rarely improve test scores. You have to balance that reality with the more common-sense feeling that computer literacy and access to ICTs matter.
Still, there is some programmatic need to show some return on the massive investment that laptop programs take. Increased attendance is at least a weak initial proxy to help support the final toll - does a loan-funded laptop investment pay for itself?