Vanidestine, Evelyn Savoy
(1986)
ERROR PATTERNS IN CLOZE MEASURES OF READING ACHIEVEMENT OF ADULT LEARNERS.
["eprint_fieldopt_thesis_type_phd" not defined] thesis, University of Maine at Orono.
Abstract
The purpose of the study was to examine, through the analysis of responses on a close passage, the error patterns which emerge from a comparison of adults at various reading achievement levels. Additionally, the close error patterns of male and female adult readers were compared. Subjects for the study were 174 adult students from ten adult education programs in Maine. Subjects were divided into three reading groups--low, middle, and high--on the basis of their scores on the Tests of Adult Basic Education. The close instrument was constructed by deleting every fifth word from a passage determined to be at the eighth grade reading level by the Dale-Chall Readability Formula. A system for classifying close responses was defined. Each clone response was placed in one of the following five mutually exclusive categories: (1) exact replacement (ER), (2) syntactically and semantically acceptable (SSA), (3) syntactically acceptable only (Syn A), (4) syntactically and semantically unacceptable (SSU), and (5) no response (NR). Analyses of variance were calculated to determine whether there were statistically significant differences among readers of varying reading achievement levels and by sex with respect to response type. Using Duncan's Multiple Range Tests, an analysis was performed to identify significant differences between cell means for variables exhibiting a significant F-ratio. Pearson product-moment correlations were computed for nine variables. The following significant differences were revealed: 1, High level readers made more ER's, more SSA' s, fewer SSTs, and fewer NR's than low level readers. 2. Middle level readers made fewer ER's than high level readers and more ER's and fewer NR's than low level readers.
3. Females made more SSA's than males. It was concluded that: 1. Low, middle, and high level adult readers differ in their usage of syntactic and semantic cues, In particular, very distinctive differences exist between high and low level adult readers. 2. Types of close responses made serve to verify the discriminating ability of the TABE. 3. Although a difference was found between male and female adult readers in their use of syntactic and semantic cues, the data available does not provide a clear explanation of this result.
Actions (login required)
 |
View Item |
