Talk:William S. Gray

Reading as skills by Gray

I have some questions about some details in this article. Here, it is said Gray and Leary identified 228 variables, an information that probably comes from Dubay(2007 : 144). However, in Klare (1963 : 48, 1988 : 15), Gray and Leary would rather have identified 289 factors instead. Other differences exists between these two highly reliable sources :

• the authors selected 80 (Dubay) / 82 (Klare) style variables.
• 17 (Dubay) / 20 (Klare) were significantly related to readability.

Does anyone know something about these differences ? Unfortunately, I was not able to get back to the source ?

• DUBAY, W. (2007). Unlocking Language - The Classic Readability Studies. Information Impact.
• KLARE, G. (1963). The Measurement of Readability. Iowa State University Press, Ames, IA.
• KLARE, G. (1988). The Formative Years. In ZAKALUK, B. et SAMUELS, S., éditeurs : Readability : It’s Past, Present, & Future, pages 14–34. International Reading Association, Newark, Delaware.

Fr endymion (talk) 10:15, 23 June 2010 (UTC)

Thank you for pointing that out. Sorry for the confusion.

Two of the discrepancies may have come from my citing the figures from Klare's earlier (Hermitage House 1954) collaboration with Byron Buck, "Know Your Reader: The Scientific Approach to Readability," which referred to the "228 variables they thought might be related to readability" (p. 90). The original Gray-Leary book says "The classified list of 289 factors as finally compiled..." (p. 26).

The 80 figure came from the next page of the Buck-Byron book, "Here, Gray and Leary listed 80 elements, 64 of which could be counted reliably." The charts in the original Gray-Leary book list 82 elements (pp 88-89).

The 17-20 discrepancy may have come from the discrepancies between the Klare and Buck book (17 on p. 95) and the Measurement of Readability (20 on p. 48). Both citations may be correct as they refer to a listing of the top (17 or 20) of the top 44 elements causing difficulty. The earlier Klare and Buck citation mentions the top 17 as all having correlations of .35 or above with difficulty.—98.237.159.70 (talk) 17:20, 16 July 2010 (UTC)

Thank you for the clarification you made. In the meantime, I have put my hand on an original from Gray and Leary (1935) and I was able to consult the pages you referred to.— Preceding unsigned comment added by Fr endymion (talk • contribs) 11:32, 4 August 2010 (UTC)

Primary sources

I've cleaned up the correlations a bit by putting them in a table, but the correlations and elements in the Readability Variables Section do not match the primary source at all. For example, as written, the text states:

Gray and Leary used five of the above variables, numbers 1, 5, 8, 15, and 17, to create a formula, which has a correlation of .645

However there is no mention of any correlation at .645 between pages 137 and 139. The R=.645 is for the 8-element equation on page 134. --Myqlarson (talk) 10:15, 27 November 2011 (UTC)