Premium
The metric matters: determining the extent of children's knowledge of morphological spelling regularities
Author(s) -
Deacon S. Hélène
Publication year - 2008
Publication title -
developmental science
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 2.801
H-Index - 127
eISSN - 1467-7687
pISSN - 1363-755X
DOI - 10.1111/j.1467-7687.2008.00684.x
Subject(s) - spelling , morpheme , psychology , orthography , linguistics , spell , coding (social sciences) , context (archaeology) , consistency (knowledge bases) , contrast (vision) , cognitive psychology , computer science , artificial intelligence , reading (process) , mathematics , paleontology , philosophy , statistics , sociology , anthropology , biology
All developmental research needs to carefully consider how children's knowledge is measured. The study of children's knowledge of spelling conventions, or the ways in which the English orthography encodes the roots and affixes and the sounds in words, is no exception. This experiment examined the extent of 7‐ to 9‐year‐old children's knowledge of the role of root morphemes in spelling words across different contexts and with different units of assessment. Different writing contexts did not appear to affect children's performance; children were better able to spell the first components of two‐ than of one‐morpheme words (e.g. only free in freely and freeze ), both when writing whole words and their first sections (e.g. completing__ or __ ly for freely ). A second analysis revealed that the unit of coding can influence conclusions. Children demonstrated similar abilities across ages 7 to 9 when only the first segments of words were coded; in contrast, there was evidence of age‐related differences when whole word spelling accuracy was assessed. In combination, these results suggest that children's knowledge of the principle of root consistency is remarkably robust to changes in writing context, but that coding is key when drawing conclusions. These findings remind us that the metric matters in studies of spelling, as in other domains, and they offer a manner to reconcile previously conflicting data on spelling development.