What counts as evidence for the long-term realisation of whole-class scaffolding?

To promote whole-class scaffolding of mathematical language, a teacher was encouraged to employ a repertoire of seven strategies (e.g., reformulating) in a multilingual primary classroom (22 pupils; aged 10-12). This paper investigates whether the enactment of these strategies has led to long-term whole-class scaffolding as identifiable by its key characteristics: diagnosis, responsiveness and handover. Comparison of pupils' pre- and post-test scores on three linguistic key elements all yielded statistically significant differences with large effect sizes, thus confirmed handover. A statistically significant shift from high-support to low-support strategies revealed responsiveness to pupils' levels over nine lessons. A qualitative analysis showed interrelatedness of performed strategies and scaffolding characteristics (e.g., diagnosis). The results provide empirical evidence of the long-term realisation of whole-class scaffolding. (C) 2013 Elsevier Ltd. All rights reserved.