Researchers have written that poor readers and good readers behave differently. In an early review of clinically observed characteristics of poor readers, Robinson (1946) included restlessness, introversive or withdrawal tendencies, inadequate school relations, and conscious self-control bordering on rigidity. Harris and Sipay (1985) cited expressed hostility, negative emotional response to reading, lack of effort, passivity, distractibility or restlessness, and lack of attentive concentration as characteristics of poor readers.

Poor readers, in general, although not in every case, have been characterized as tending to demonstrate maladaptive behavior (Gentile & McMillen, 1987; Jorm, Share, Matthews, & MacClean, 1985). Based on a substantial review of research, Gentile and McMillan characterized the behavior of poor readers as ranging from anger and aggression to avoidance and apprehension.

Classroom behavior has been shown to be highly related to reading achievement among first- and second-grade children (Jorm et al., 1985; McMichael, 1979; Swanson, 1984). On the other hand, Zigmond, Kerr, and Schaeffer (1988) found that the classroom behavior of learning-disabled adolescents enrolled in Grades 1 through 11 is not significantly different from the behavior of their non-learning-disabled peers. Among the behaviors studied were on-task behavior, disruptive behavior, and volunteering comments.

Confusion arises about behavioral characteristics of good and poor readers when clinically derived subjective descriptions are compared with objectively measured classroom behavior and when the behavior of primary children is compared with that of adolescents.

The present research attempts to provide consistency by investigating a single set of objectively observable behaviors of both good and poor readers in classroom settings from Grades 1 through 11.

Subjects. The subjects were 108 students enrolled in regular classes from Grades 1, 3, 5, 7, 9, and 11. Classrooms were selected from public schools in a mid-sized (population 65,000) city in the north central United States. We chose subjects who were the 3 best and the 3 worst readers in each of three classrooms at each of six grade levels. The relative standing of students was determined by examining the latest standardized reading achievement test scores for each student in each class, except for first-grade students, for whom kindergarten teachers' ratings were used as the basis for selection. The 3 students with the highest and the 3 students with the lowest reading achievement scores became subjects. The fourth highest and the fourth lowest students became alternates if any of the original choices were absent on the first day of observation.

At each of six grade levels, we chose a total of 9 good readers and 9 poor readers from three classrooms. Selection thus resulted in the total of 108 students from 18 classrooms, 54 categorized as good and 54 categorized as poor readers.

Procedures. Based on a review of literature that specified classroom behaviors associated with good as opposed to poor readers (Gentile & McNfillan, 1987; Jorm et al., 1985; McMichael, 1979; Zigmond, Kerr, & Schaeffer, 1988), discussion with classroom teachers, and review of methods for objectively observing student behavior in classroom settings (Deno, 1980; Grambrell, Wilson, & Gantt, 1981; Hoge, 1985; Hoge & Luce, 1979), we chose specific behaviors that seemed likely to differentiate good from poor readers. We observed these specific behaviors in classrooms on a trial basis to ensure that they could be consistently identified. From the original set of specific behaviors, we chose two that could be consistently identified and precisely defined. Behavioral definitions were refined through pilot sessions performed in classrooms not used in the actual research. Pilot sessions continued until a reliability of 90% was attained by independent observers recording the behavior of the same students at the same time. The behaviors and definitions used in the research follow:

1. Seconds to start–number of seconds from the beginning of an activity, as indicated by the teacher, until the student is first on task. Duration of recording: 5-min maximum. First on task–materials are out and the student is in place, listening to the teacher, making eye contact with the appropriate stimuli, and writing, or has pencil poised, ready to write. The student is not on task when looking for materials.

2. Off-task–any movement off a prescribed activity that does falls into one of three categories--looking around, staring into space, doodling–or any observable movement off the task at hand. Off-task behavior was measured by counting the number of off-task incidents a child evidenced during a 30 min observation interval.

Research data was gathered by observing each classroom for 30 min a day for 10 days. To prevent experimenter bias, we were not told which students were poor readers, but only which 6 students to observe.

Reading classes were not observed because of a lack of secondary-level reading classes and because the behavior that characterizes poor readers, according to the literature, is more general than a simple reaction to a reading class. Instead, social studies classes, which require students to apply reading skills, were selected for observation. When certain elementary classroom teachers did not teach clearly defined social studies lessons, we substituted language arts lessons.

Behavioral observation began at the start of each day's lesson. Prior to the start of the lesson, the classroom teacher supplied the trained observer with a list of materials that the children needed for the lesson. The teacher also indicated to the observer when the lesson began. The observer than measured the length of time until each student was first on task. The maximum time allotted was 5 min. After 5 min elapsed, the observer recorded materials missing, that is, materials the student did not have that were required for the lesson. Length of time until each student was first on task and materials missing were recorded for each student each day.