Computers in Schools: Why Governments Should Do Their Homework

Time and time again, governments and NGOs herald the purchase of ICT as a panacea for improving the quality of education. The recent plans of Gauteng in South Africa are a good example. This study from the Inter-American Development Bank (IDB) provides an useful summary of the research done on the impact of ICT in primary level classrooms.  Latin America and the IDB have been at the forefront om some high-profile “One Laptop per Child” projects such as the Plan Ceibal (Uruguay), Enlaces (Chile) and the OLPC Programme in Peru, on which I blogged before.

Some extracts:

The evidence so far is quite persuasive that programs that overlook teacher training and the development of software may yield low returns.

One promising avenue lies in the use of ICT to realise productivity gains in school management:

The collection, transmission, and analysis of data on enrollment, absenteeism, test scores, and infrastructure can help principals spot a problem in a given classroom, administrators spot an exemplary school, and policymakers track the performance of the educational system and the resources available. However, the gains in productivity seen in the business sector are rarely seen in the educational system, some have argued, because most education managers are not knowledgeable in the use of information management tools.

Studies that measured the impact of ICT, both of the access to computers and the use of computers, found more often than not no significant impact on learning outcomes – an overview is included in the report.  The authors note that it’s not sufficient for ICT investments to produce a positive impact, they should produce a positive impact compared to traditional instruction and, even better, to similar investments in other areas such as teacher training, smaller classes or libraries.

All other things being equal, the impact of ICT investments will be higher when the quality of teaching is low, as the potential for learning gains is higher. This underlines the risk of extrapolating findings from developed to developing country context.

Some recommendations from the report:

  • Given the high investments, the low number of decent impact studies is surprising.  The impact of ICT investments heavily depends on the context and on the implementation.  As such, results from impact studies cannot be generalized over different programmes.  Start on a limited scale and build impact evaluation into the programme design is important.
  • Important to keep the Total Cost of Operation (TOC) of ICT investment into account rather than the purchase price. This includes maintenance, training, connectivity and electricity costs.  Recurrent costs typically take up about 40-50% of the initial investment (in Latin America).  These are permanent costs, which imply savings elsewhere in the education system or an overall increase in expenditure. A large share of rural schools, high electricity and connectivity costs and high wages (as in South Africa) thus increase the share of recurrent costs.
  • Most successful ICT project implementation focus on honing ICT skills of learners and pursuing Computer-Aided Instruction (CAI), for example for maths.