Government Social Research
Analysis of 12 relevant Randomised Controlled Trials (RCTs) showed that systematic phonics instruction taught as part of a wide literacy curriculum had a statistically significant positive effect on reading accuracy for both normally developing children and children at risk of reading failure.
In terms of reading comprehension and spelling, no statistically significant effect of phonics was found. In addition, no statistically significant difference in effect was found between synthetic and analytic phonics; however, only three relevant randomised controlled trials were found on this question so this conclusion was based on weak evidence.
The reviewers were unable to draw any conclusions about the proportion of literacy teaching that should be based on phonics from the available RCT evidence. Overall, the strength of RCT evidence on phonics use was found to be moderate on the question of reading accuracy, but weak on all other questions.
> A systematic review of the research literature on the use of phonics in the teaching of reading and spelling (pdf) [DfES website]