Teaching Elementary Students Who Speak Black English Vernacular to Write in Standard English: Effects of Dialect Transformation Practice

59 pupils (per the What Works Clearinghouse evidence table)

3rd- and 4th-grade African-American pupils (roughly ages 8-10) whose written work showed Black English Vernacular features

Recorded because it is the only randomised study of teaching a body of USAGE conventions to children under about 11, and because the What Works Clearinghouse cites it as the sole support for the instruction "explicitly teach the conventions of written English". Six syntactic features differing between Black English Vernacular and Standard English were targeted. Three conditions: exposure to the SE features in stories; exposure plus explanation of the SE rules; and exposure plus rule instruction plus GUIDED PRACTICE transforming sentences from BEV into SE with corrective feedback. Two things bound how much weight it can carry. The dose is tiny - the WWC records it as whole-class delivery in 2 sessions totalling 60 minutes. And the outcomes were researcher-built measures of the six taught features, which the WWC judged did not meet its measure-eligibility standard at all; the only eligible outcome it could extract was writing output, at 0.27, not significant. Numbers here come from the abstract and the WWC evidence table, not from the paper.

The third condition - rule instruction PLUS guided transformation practice with feedback - "proved most effective in enabling students to translate BEV sentences into SE forms and to employ the targeted SE features in their free writing." Exposure alone was not enough, and explanation without practice was not enough: what moved usage was practising the transformation and being corrected. This is the archive's cleanest demonstration that a specific, finite set of usage conventions can be taught to 8-to-10-year-olds and will show up in their own writing - and it is one study, with 59 children, over an hour, with researcher-built outcomes, never replicated. The WWC extracted no eligible measure of the conventions outcome itself and a non-significant 0.27 on writing output.

Low for the internal comparison - children were assigned to three conditions and the contrast is between instructional conditions, not between children who did and did not receive something. Nothing here supports a claim about differences between dialect groups; the outcome is whether a taught form appears in writing.