Editing assistance tool validation for English language learners

Abstract

PurposeEditing assistance software programs are computer-based tools that check and make suggestions for the grammar, spelling and style of a piece of writing. These tools are becoming more popular as recommendations for students who struggle with written expression, such as English language learners (ELLs). The purpose of the present study is to compare the performance of four different programs with embedded editing assistance tools in their ability to identify errors in the writing of ELLs.Design/methodology/approachRepeated measures ANOVAs were conducted to determine whether there were differences in the number of errors (i.e. spelling, grammar, punctuation and errors that change the meaning of the text) identified by editing assistance programs (i.e. Grammarly, Ginger, Microsoft Word, Google Docs and human raters) for writing by ELLs.FindingsThe results of the present study indicate that the four programs did not differ in their identification of spelling errors. None of the editing assistance programs identified as many errors as the human raters; therefore, editing assistance cannot yet replace effective human editing for ELLs.Research limitations/implicationsLimitations with the present study include manual verification of errors flagged by editing programs, multiple raters, a small sample size and a young sample of students.Practical implicationsThe paper includes practical factors to consider when integrating editing assistance software into the classroom, including the development needs of students, the impact of students' first language and student training on the technology.Originality/valueThis paper provides school psychologists, teachers and other professionals working with students with specific, evidence-based recommendations for implementation of editing assistance AT.