Date of Award


Degree Type


Degree Name

Master of Science in Education (MSEd)


Education and Human Development

First Advisor

Dr. Frances Moroney


The purpose of this study was to investigate the effect of a teacher directed approach versus an independent approach of teaching students to use context clues to determine word meaning. A randomized, two treatment, control group posttest design was used for the study. The sample consisted of 30 students (three classes) taught by the same team of instructors and equated in terms of age, reading level and pretreatment use of context clues.

Eighty words were selected from the social studies units to be studied during the eight week treatment period. The 80 words were divided into five word blocks—one block for each treatment session. During each session both treatment groups read the identical five passages. Each passage, varying in length from one to three sentences, contained one vocabulary word in context and defined that word using one of Deighton's (1959) categories of context clues (example, restatement, definition and modifier). Deighton's fifth category, inference, was not included.

One treatment group (teacher directed) read the passage orally and through questioning and discussion the experimenter elicited from the students the context category present and the meaning of the vocabulary word.

A second treatment group (independent group) read the five passages silently and answered a question following each passage which required the students to apply the meaning of the vocabulary word to their own experiences.

The control group received no treatment.

Upon completion of the eight week treatment period students were posttested on their ability to use context clues. The posttest consisted of 26 items, each containing a nonsense word in context.

Each item defined the nonsense word using one of Deighton's context clue categories. Following each item the student had to choose the correct meaning for the nonsense word.

An analysis of variance and Lindquist's (1956) test of critical differences were used to analyze the data at a .05 level of significance. The results indicated the teacher directed method of teaching students to use context clues was significantly better than the independent method and both methods were significantly better than no instruction. Recommendations for classroom application as well as suggestions for future research were given.