A correlational study of spelling achievement between the Wide range achievement test and two forms of the reading spelling vocabulary proficiency test among subjects in grades two through fourteen

Date

1982

Journal Title

Journal ISSN

Volume Title

Publisher

Abstract

In 1978 Stetson used 14 previously compiled word frequency studies to assimilate a list of the 900 most frequently used words in speech, in print, and in writing. It is called the READING SPELLING VOCABULARY PROGRAM (RSVP). From the 900-word RSVP, 240 words were randomly selected, 4 words from each of 60 lessons, to create two equivalent forms (120 words each) of a placement test known as the READING SPELLING VOCABULARY PROFICIENCY TEST (RSVPT - Forms A and B). White (1980) correlated the two forms of the RSVPT with the reading subtest of the Wide Range Achievement Test (WRAT) using 266 students in grades one through nine and found correlations at or above .85 among the instruments. Even though Stetson claimed that the RSVP was appropriate for writing as well as for reading, no research was available on the RSVPT as a test of written spelling. The Problem and Rationale The purpose of the study was to compare the two forms of the RSVPT with the spelling subtests of the WRAT. Because most authorities agree that written spelling is the most accurate measure of spelling achievement (Weller & Broom, 1934 ; Nelson & Denny, 1936; et al), it was felt that validating the RSVPT would not only provide another test of written spelling, of which there are few, but provide teachers with an instrument that could estimate where spelling instruction should begin in a high frequency word list. Procedures and Data Analysis A total of 1,371 students enrolled in grades two through fourteen took a written dictated test on both forms of the RSVPT and the spelling subtest of the WHAT. A Pearson Product Moment Correlation was used to compare the scores on the three instruments. In addition, the Spearman Rank Order Correlation was used to compare the rank order of difficulty between the RSVPT when tested for reading (White, 1980) and when tested for written spelling. Findings. The r calculated among the three instruments was significant (p <.001) for all comparisons: r = .84 between the RSVPT - Form A and the WRAT; r = .83 between the RSVPT - Form B and the WRAT; r = .89 between RSVPT- Form A and RSVPT-Form B. The rank order correlation was significantly similar (p<.05) for reading and written spelling for five out of the twelve subtests on Form A for grades two through six, two out of the twelve subtests on Form A for grades seven through fourteen, five out of the twelve subtests on Form B for grades two through six, and two out of the twelve subtests on Form B for grades seven through fourteen. Therefore, (1) the reliability and validity of the RSVPT was established for written spelling while (2) the rank order of difficulty between reading and written spelling on both forms of the RSVPT was questionable.

Description

Keywords

English language--Orthography and spelling--Ability testing, Educational tests and measurements

Citation