Is an intervention using computer software effective in literacy learning? A randomised controlled trial

Background: computer software is widely used to support literacy learning. There are few randomised trials to support its effectiveness. Therefore, there is an urgent need to rigorously evaluate computer software that supports literacy learning. Methods: we undertook a pragmatic randomised controlled trial among pupils aged 11–12 within a single state comprehensive school in the North of England. The pupils were randomised to receive 10 hours of literacy learning delivered via laptop computers or to act as controls. Both groups received normal literacy learning. A pre‐test and two post‐tests were given in spelling and literacy. The main pre‐defined outcome was improvements in spelling scores. Results: 155 pupils were randomly allocated, 77 to the ICT group and 78 to control. Four pupils left the school before post‐testing and 25 pupils did not have both pre‐ and post‐test data. Therefore, 63 and 67 pupils were included in the main analysis for the ICT and control groups respectively. After adjusting for pre‐test scores there was a slight increase in spelling scores, associated with the ICT intervention, but this was not statistically significant (0.954, 95% confidence interval (CI) – 1.83 to 3.74, p = 0.50). For reading scores there was a statistically significant decrease associated with the ICT intervention (−2.33, 95% CI −0.96 to −3.71, p = 0.001). Conclusions: we found no evidence of a statistically significant benefit on spelling outcomes using a computer program for literacy learning. For reading there seemed to be a reduction in reading scores associated with the use of the program. All new literacy software needs to be tested in a rigorous trial before it is used routinely in schools.