Gill Ragsdale gathers answers from 30 years of phonics research

Use of phonics can improve learning to read English as a second language, according to a review of 30 years’ research by Dennis Odo, of Pusan National University, South Korea.

Odo conducted a comprehensive statistical analysis (a meta-analysis) pooling the overall results of 46 studies from 1990 to 2019 looking at the effect of using phonological awareness and phonics on learners’ reading skills.

Phonics, the system of linking spoken sounds to one or more written letters to help learners decode and learn new words, is a common strategy with young learners in the primary English L1 classroom, but less commonly used with L2 learners. Even in the L1 classroom, however, phonics remains controversial, especially as phonics does not map as reliably onto English letter clusters as they do when applied to more orderly languages, such as Spanish or German. In the L2 classroom, phonics is often considered to over-complicate the learning process.

While phonics is the applied system, it’s underpinned by the skill of phonological awareness, ie, the ability to recognise and identify spoken sounds. This is a skill that can be harnessed to support reading without referring to the system of phonics.

Debate has been heated (even called ‘Reading Wars’) as to whether phonics is helpful or harmful when teaching reading, the alternative being teaching whole words. Nevertheless, in recent years, phonics has become the mainstay of early years education in English primary schools, and thought to be especially useful with weaker readers and those whose native language is not English – suggesting untapped potential for EFL/ESL tuition.

Seeming to support this, Odo’s meta- analysis found a significant, moderate effect of phonics instruction on improving reading skills (Hedge’s g=0.53) overall. However, the effect size varied considerably across the included studies. Some of this variability could be accounted for by differences in study design, the way phonics was taught, how long phonics instruction was used, the first language of the learners and the education stage.

Strong experimental designs with a comparison control group tended to have lower effect sizes, as did those studies using standardised assessments, as opposed to those prepared by the researcher. The differences in effects sizes due to study design were substantial: 0.26-0.33 vs 0.71-1.68. This doesn’t promote confidence in the overall effect size. However, understanding these moderating influences, ie, what makes the teaching of phonics more or less effective, is valuable in itself.

At first glance, the larger effect when using researcher-designed assessment may suggest researcher bias but it’s just as likely to reflect the nature of standardised assessments. If using phonics really does improve reading and this can be reliably measured, why is it not being measured using standardised assessments?

When using phonics, strategies ranged from combining phonological awareness and phonics (g= -0.03 – so basically no effect at all), to those relying on phonological awareness alone (g=0.46 – moderate) and to those focusing on phonics (g=0.76 – borderline large effect) suggesting using the phonics system, rather than just phonological awareness is instrumental in supporting reading skills.

The learning context was also a factor with greater effects in the EFL compared to the ESL classroom (g=0.26, a small effect vs g=0.83, a large effect). Odo suggests that this may be due to EFL teachers spending more time on practising decoding activities, such as decoding pseudo-words, while ESL teachers are necessarily more focused on real content vocabulary. (There was also some evidence that the L1 writing system was a factor: phonics had a larger impact on learners whose first languages were alphasyllabaries (abugidas), eg, Amharic or Hindi, compared to those whose languages were alphabetic, or logographic, ie, Chinese.)

With regard to the stage of education, while the use of phonics with primary and elementary students showed moderate effects (g=0.43-0.56), the effect at middle school was very much larger (g=1.72).

A quality meta-analysis such as Odo’s is no small job, but all that hard work ultimately rests on the quality of the studies included. I have yet to read any such analysis that does not lament on this issue, and that the disparity between studies is like comparing apples, some of which are rotten, to oranges.

Overall, then, is this a final verdict on the use of phonics?Since this is a statistic al analysis, we are helpfully suppled with confidence intervals around the effect size to help us make up our minds. In this case, the overall moderate effect size of g=0.53 has a confidence interval of 0.27- 0.79. This means that we can be 95%confident that the true effect size is somewhere in that range. That is, between small and large.

It seems that, further, more robustly designed studies are needed to truly resolve this issue with more confidence – but this meta-analysis and the influencing factors described will be an excellent resource for future research.

RESEARCH

• Odo, D M (2021), ‘A meta-analysis of the effect of phonological awareness and/or phonics instruction on word and pseudo word reading of English as an L2’, SAGE Open. doi: 1177 /21582440211059168.