The Effects of Different Types of Unfocused Corrective Feedback on Complexity, Accuracy and Fluency in L2 English Academic Writing

Research on written corrective feedback in second language (L2) learning has made progress, answering the unknowns regarding its effectiveness. Currently, debate focuses on the most effective way of giving feedback. Controversy, however, remains and there is a scarcity of research on unfocused feedback. The present study examines the effects of unfocused direct, indirect and metalinguistic written corrective feedback (WCF) on the complexity, accuracy and fluency (CAF) of 139 L1 Arabic or Urdu – L2 English students' writing. The study also investigates if the moderating variables of aptitude, attitudes and proficiency affect the uptake of feedback. Students in four intact groups were designated as feedback groups, plus one control group. They wrote argument essays and were given four rounds of feedback and feedback support sessions over fourteen weeks; whereas learners in the control group received no feedback or support sessions. Students wrote both text revisions and new texts. Results showed that on text revisions, the direct and metalinguistic feedback groups had losses in fluency compared to the indirect and control groups. The indirect feedback group had significantly lower lexical diversity than the direct and metalinguistic groups. On new texts, there were no significant gains or losses from the unfocused feedback. The moderating variables of proficiency and aptitude had no significant relationships with CAF gains or losses, but positive attitudes towards feedback had a negative relationship with gains in complexity and fluency on text revisions. These results reveal that on text revisions, some forms of unfocused feedback have effects on fluency and lexical diversity, but on new texts there are no effects. Future work should examine if increasing the number of treatment sessions has positive effects on CAF, and discover at what point unfocused WCF may become too cognitively demanding. The results provide useful information for practitioners who could use a more blended approach between focused and unfocused WCF and increase the treatment sessions.