Pixel International Conferences

In the last decade, one laptop per child projects were implemented in many countries. Such projects are usually intended to improve quality of instruction (although specific goals may differ between projects). Logically, follow-up investigations should determine how pupil competencies were influenced, as measured by standardized testing. This paper presents such an investigation, conducted within the ex-post evaluation study of the ‘Digital School’ project implemented in Poland in academic year 2012/2013. A unique trait of this study as compared with other similar work is the rare conjunction of three features. Firstly, use of a randomized project design to construct a valid control group, secondly, focus on the impact of the ‘Digital School’ on test score distribution, beyond the usual comparison of means and thirdly, using final exam results from two years, 2013 and 2014, allowing inquiry on how the effects of the intervention change over time. Monte Carlo permutation tests were applied to test for significance of differences in cumulative test score distributions. The results suggest that there might have been some benefits to some low-scoring pupils, but the effects seemed to be short term, very weak and limited to mathematical and logic skills.