American Educational Research Journal

Using the observation procedures developed by Leinhardt, Zigmond, and Cooley (1981), we conducted a large-scale field study of reading instruction in special education resource room programs for fourth-, fifth-, and sixth-grade mildly handicapped students. Children in 23 resource rooms in one district and those from 5 resource rooms in another district were observed during reading instruction. Observations were also made in regular classrooms for a subset of handicapped students and their nonhandicapped peers. The research sought to answer questions regarding the standardization of resource room reading instruction, the nature and amount of reading instruction in these programs, factors associated with how students are scheduled for special reading instruction, reading instruction in resource rooms versus regular classrooms, and special education students’ reading instruction (resource room plus classroom) versus that of nonhandicapped peers. Process-product relations similar to those of the Leinhardt et al. research in learning disabilities classrooms were also tested. Results indicated considerable variability in reading instruction across programs and students that was not strongly linked to student characteristics. Overall, the amount of reading instruction was remarkably low, and instructional process variables in resource rooms were not predictive of achievement.

Ten classes were randomly assigned to a group that received feedback and assistance to maintain an average teacher wait time of between 3 and 5 seconds during a sequence of mathematics lessons. A control group of 10 teachers maintained a regular wait time and received placebo feedback. The study was replicated in a sequence of language arts lessons. The use of an average teacher wait time of between 3 and 5 seconds in whole class instructional settings was associated with higher mathematics achievement and improvements in the quality of teacher and student discourse. In extended wait time classes in the mathematics component of the study, decreases were observed in the number of utterances per unit time, the average length of teacher utterances, the frequency of teacher interruptions, failure of students to respond to teacher questions, the proportion of teacher reacting to student responses, and the proportion of teacher mimicking of student responses. Increases were observed in the average length of student responses, the proportion of teacher soliciting, the proportion of teacher structuring, and the proportion of probing after a student response. Similar results were obtained in the language arts component of the study. Changes in the types of teacher questions were observed also. In mathematics there was an increase in the proportion of questions requiring application of concepts, and in language arts there was an increase in the proportion of questions requiring student comprehension.

The teacher’s role as mediator of social play between severely or profoundly mentally retarded elementary school children and non-handicapped peers was investigated in the context of a structured peer-interaction program. Twenty-eight heterogeneous dyads from three integrated schools were given 2 weeks of either verbal instruction with modeling of cooperative play behavior (treatment) or noninstructional teacher attention (control), followed by 4 weeks of reduced teacher attention (maintenance). Results demonstrate the strong positive impact of the intervention upon the quality of relationships in the treatment group, as evidenced by both partners’ behaviors. Discovery of delayed or “sleeper” effects are discussed regarding the value of subsequent reduced teacher intrusion. Structured interaction programs are described as interim vehicles for normalizing heterogeneous relations, not as ends in themselves. The need for more ecologically valid research of naturally occurring relationships involving children with and without severe disabilities is highlighted.

The relationships between types of in service training activities and changes in teaching behavior were investigated. Three groups of junior high teachers (N = 19) attended five workshops on effective teaching. Group I received no extra activities, Group II participated in peer observations, and Group III was coached by the trainer. Pre- and posttraining observations, questionnaires, and interviews were used to assess behavior change and attitudes. Profiles of day-to-day variations in teaching practices were examined for stability of teaching behavior. Of the teachers needing to improve, three Group I teachers, six Group II teachers, and two Group III teachers approached or exceeded the criterion level of performance on academic interactions. Peer observation appeared to be more effective than trainer-provided coaching in boosting workshop effectiveness. Further study of the context features related to the effectiveness of peer observations in district-based programs is required.

Stimulated-recall data were gathered in interviews with 12 elementary school teachers describing the decisions they made during lessons conducted shortly before. Ideational units were identified in the protocols, and a grounded theory analysis was conducted. Categories and terminology were generated and, using the constant comparative technique, hypotheses grounded in the data were developed. The first hypothesis embeds these teachers’ interactive decisionmaking (IDM) in the structure of the lessons they were conducting, and subordinates IDM to teachers’ cognitive representations of the activity at hand. The second hypothesis identifies the central intention of IDM: to move a learning activity to completion according to the cognitive representation; and the third hypothesis specifies the role of decision rules and routines.

This study investigated the planning and debugging strategies and group processes that predicted learning of computer programming in small groups. Thirty students aged 11 to 14 learned Beginners All-Purpose Symbolic Instructional Code (BASIC) in 2-person groups. Students showed little advance planning; their style of planning and debugging was opportunistic. Specific planning and debugging variables that predicted learning included high-level planning of chunks of the program, and low-level planning and debugging of single statements. Group interaction variables related to learning included giving and receiving explanations, receiving responses to questions, and verbalizing input aloud while typing at the keyboard. Student demographic characteristics, aptitudes, and cognitive style measures did not predict planning, debugging, or verbal interaction.

This study examined the effects of reference maps on what children remember from written and aurally presented discourse. Subjects were presented one of three maps that varied in feature configuration and spatial distribution, and were asked to study these maps before reading or listening to a related story. Results of tests across both prose and maps conflicted with previous research in which learners appeared unable to maintain visual images while reading. We found essentially no differences in recall due to mode of text presentation, but the type of map that was presented profoundly influenced recall from both the text and the map. Maps that had pictorial features logically organized according to passage content greatly increased the learning of text that was related directly to map content. This indicates that maps serve a mnemonic-like function for remembering prose.

A set of skills is defined in the trigonometry of the right triangle and a learning hierarchy is validated. Examples of lateral and vertical transfer are identified within the hierarchy. Two instructional treatments identified as “achievement only” and “achievement with understanding” are used in two experiments with 90 eighth-grade students and 87 ninth-grade students in their ordinary mathematics classes. The achievement with understanding treatment is better for the development of lateral transfer for most students, and of vertical transfer for the more mathematically able students, whereas the differences between the treatment groups on tests of achievement and retention of taught skills are not significant. A small amount of additional instruction on vertical transfer items produces much better performance under both treatments.

A new method called Correlated Factors Modeling (CFM) is proposed as an alternative to LISREL and LISREL-like methods of estimating structural equations. CFM employs weak mathematics. It may be appropriate when (a) available theory is weak; (b) available data resources are weak in either sample size or sampling procedure, or both; (c) researcher knowledge of structural modeling methodology is weak; or (d) a rapid, easy, preliminary look at a model, such as can be obtained from a microcomputer, is desired. The factors produced by CFM are exact linear components of their indicator variates. Path models for hypothesized causal relationships among the factors can be computed on the reported correlations among the factors by the ordinary least squares (OLS) method of conventional path analysis. Heuristic goodness-of-fit indices are available to support comparisons of models. Two examples are reported.

The purpose of this investigation was to determine the functions of students’ reported inner speech during problem solving. Elementary school students in grades three to six who differed in mathematics achievement were randomly assigned to a low-task difficulty or high-task difficulty problem-solving condition. Immediately after working on assigned problems, and while anticipating more of the same, students were questioned about their inner speech during problem solving. Neither age nor ability differences were reflected in the students’ reports. Rather, reports indicated that students in general maintained an adaptive problem solving stance in both problem solving conditions. Adaptive behavior was characterized by strategic instructional statements focused on the task and positive affect statements focused on the self.

This study investigates the construct validity of three methods used to evaluate clinical competence in medicine: standardized test, supervisor performance ratings, and peer performance ratings. Three attributes of clinical competence are investigated: cognitive abilities, interpersonal skills, and professional qualities. Measures representing each attribute-method combination include: National Board of Medical Examiners (NBME) examination (standardized test of cognitive abilities); two scales derived from the California Psychological Inventory (CPI) (standardized test of interpersonal skills and professional qualities); and the three scales derived from the Resident Evaluation Form (REF) (peer and supervisor ratings of all three attributes). Scores for each attribute-method combination were obtained from a convenience sample of 166 resident physicians in three primary care specialties. These scores were cast into a multitrait-multimethod matrix design and analyzed using confirmatory factor analysis. Results suggest a lack of construct validity for the CPI and REF scales, moderate convergent validity for the NBME, and substantial method variance in the REF-derived ratings. Findings are discussed in terms of the implications for a theory of medical clinical competence, further research and development in clinical competence measurement, and current measurement practice in medical education.

This study identified two low-inference variables comprising lesson organization and determined their effect on student achievement. For each of the two main effect variables, two conditions were defined: proper sequence versus manipulated sequence, and explicit organization behaviors present versus explicit organization behaviors absent. Four treatment groups (equated on mathematical ability) were formed by the possible combinations of these conditions. A single lesson on numeration systems was taught to the four groups (67 university students), and achievement was evaluated using an investigator-designed posttest. Although the data indicated no significant effect for sequence, the study did yield a significant effect (p < .01) for explicit organization. No significant aptitude-treatment interaction was found, either for explicit organization or for sequence.