Date of Award

5-1982

Degree Type

Thesis

Degree Name

Master of Science in Education (MSEd)

Department

Education and Human Development

First Advisor

Dr. Arthur Smith

Abstract

The purpose of this study was to examine eighth grade subjects' abilities to resolve three types of referential anaphoric expressions across three different tie locations in expository discourse and to determine if the ability to resolve referential anaphoric expressions is significantly related to reading comprehension as measured by either a product-oriented or process-oriented test of reading achievement.

One of the chief means of creating cohesion within a text is through the use of anaphoric expressions which refer a reader back to concepts developed previously. Research at the elementary level has led a number of investigators to conclude that elementary children do not adequately comprehend anaphoric expressions. Other researchers have found that difficulty with pronominal referents extends into the junior-high-school range.

An anaphoric resolution test was developed by the investigator in which each of three types of referential anaphoric expressions were crossed with each of three tie locations, according to classifications established by Halliday and Hasan (1976). A two-way analysis of variance was used to provide insight into the influence that these factors have on the ability of eighth grade subjects to resolve referential anaphora. Scores from this instrument were then correlated with the Gates-MacGinitie Reading Test, Level E, a product-oriented measure of reading achievement, and the New York State Preliminary Competency Test in Reading, a process-oriented measure of reading achievement

It was found that the type of referential anaphoric expression did not make a significant difference in their resolution and that there was no significant interaction between the type of referential anaphoric expression and the tie location of the presupposed items. The tie location, however, did make a significant difference in the subjects' abilities to resolve referential anaphora. Further analysis demonstrated that there were significant differences between the immediate and both mediated and remote tie locations, but no significant difference between mediated and remote tie locations.

Correlation coefficients between the anaphoric resolution test and either measure of reading achievement were significant beyond the .01 level of confidence.

It was concluded that with average eighth grade subjects the ability to resolve referential anaphoric expressions in expository discourse is dependent to a significant degree upon the tie locations of the presupposed items when the presuppositions are nouns or noun phrases mentioned explicitly in the discourse. It was significantly easier for subjects to resolve these anaphoric expressions when their presuppositions were in the immediate tie locations than when they were in either the mediated or remote tie locations.

Mean percentages of correct responses were also computed which confirmed results of other studies and led to the conclusion that students in the junior-high-school range also do not adequately comprehend anaphoric expressions.

Further research is needed to: (1) establish the causative factor in the relationship between anaphoric resolution ability and reading comprehension, (2) determine if instruction in the resolution of anaphoric expressions would increase students' abilities to resolve them, (3) determine if there is a relationship between the ability to resolve anaphoric expressions and measures of intelligence, and (4) determine if other factors involved in anaphoric resolution ability are also significant.

Teachers need to be aware of the difficulty that students have resolving anaphoric expressions and attempt to help them in this area.

Authors and publishers of children's reading material, especially basal reader programs and content area textbooks, should be aware of the fact that popular readability formulas do not take anaphoric resolution ability into account.

Teachers, as well as authors and publishers, should be constantly alert to new research findings on anaphora.

Share

COinS