A little something I picked up...
 
EAST, JOAN ANNE KERNS.  THE DEVELOPMENT OF SPELLING
KNOWLEDGE IN A FIRST-GRADE CLASS: THE
SPELLING-READING CONNECTION (INVERTED SPELLING). Thesis
(PH.D.)--THE UNIVERSITY OF NORTH CAROLINA AT CHAPEL
HILL, 1993.
The purpose of this study was to document the development of spelling
knowledge in a class of first grade students over an eight month period. 
Of special interest was to determine if spelling, particularly invented 
spelling, was predictive of word recognition at the end of the year.
  The classroom teacher was the research investigator in this 
descriptive, observational study. Both quantitative and qualitative 
research methods were utilized in the study. The twenty-one students in 
the class engaged in everyday instructional activities; no interventions 
were carried out.
  In September students were assessed with a Letter Identification Test
(Clay, 1985, 1993) which measured knowledge of 54 letters. Measures of
conventional spelling and invented spelling were obtained in September,
November and January. Conventional spelling was measured with the
Writing Vocabulary Test (Clay, 1985, 1993); the score was the number of
self-generated conventionally spelled words in list format. An analysis
form and scoring system, based on Gentry's stages of invented spelling,
was devised to analyze the invented spelling which was obtained from
naturalistic writing samples. In May two subtests (Word Identification 
and Word Attack) from the Woodcock Reading Mastery Test-Revised (1987)
were administered; these were the criterion variables in the multiple
regression analyses.
  A descriptive account of four spellers in the class, two girls and two
boys, is also included. Their spelling growth is documented from the
beginning of the year to the end of the year.
  Findings include: (1) invented spelling as well as conventional 
spelling was predictive of word recognition ability; (2) spelling was a 
better predictor of word recognition than letter identification; (3) 
invented spelling can be reliably assessed using naturalistic writing 
samples; (4) knowing letter identities is important, but not sufficient, 
for spelling and reading; (5) children's spelling changed qualitatively 
and at different rates; (6) children did learn to spell in the absence 
of a traditional spelling program, as shown by the Writing Vocabulary 
Test results and naturalistic writing samples.
  Invented spelling reflects knowledge of alphabet letter identities and
phonemic awareness, particularly phonemic segmentation. Although 
invented spelling is usually discussed in reference to writing, this 
study suggests that young children's invented spelling can be a valuable 
source of information for the classroom teacher in planning early 
reading instruction. Analyzing young children's invented spelling can 
reveal what students understand about the structure of language and its 
representation in print and thus suggest what needs to be learned 
next.
Dan
--- WILDMAIL!/WC v4.12 
---------------