We conducted a randomized experiment to test how to use ICT to improve learning.
Student test scores improved when ICT is integrated into the teaching programme.
Both better and poorer performing students benefited when ICT is integrated.
No impact is found when ICT is not integrated into the teaching programme.