Carolyn W. Keys’s research while affiliated with Georgia State University and other places

The purpose of this study was to investigate how collaborative peer discussion provided cognitive support for writing laboratory reports in ninth grade general science. The discourse of three pairs of students during collaborative writing sessions was analyzed to identify the types of social interactions in which the students engaged. These interactions were examined in relation to various intellectual demands of the laboratory reports and similarities and differences between the three pairs. Five types of interactions were identified and labeled as sounding board, peer teaching, supplies answer, debate, and incorporation. In general, student pairs used the constructive interactions of sounding board, peer teaching, and incorporation when writing difficult explanation and application sections of the reports. These interactions involved the processes of elaboration, explanation, and interpretation. Debate was used less frequently than the other interaction types, overall, and was hardly ever used by female pairs in the study. The results point to a high degree of interconnection between social interaction and cognitive processing.

Laboratory report writing should promote learners' ability to reason scientifically from data to conclusions, especially if opportunities for collaborative discussion are provided, yet few studies have examined the use of writing as a tool for the development of scientific reasoning skills. The purpose of this study was to investigate scientific reasoning skills used by students in a naturalistic collaborative writing context and changes in students' reasoning discourse over time. During the fall and winter of 1992-1993, three ninth grade general science classes taught by one teacher participated in a collaborative writing intervention. The students wrote ten collaborative laboratory reports over the course of the intervention. Report guidelines were designed by the teacher and the researcher to assist students with their writing. An interpretive research design was used to study the conversations of three target student pairs in depth. Eleven distinct, yet integrated scientific reasoning skills used to support report composition emerged from the data. The emergent scientific reasoning skills reflected the socially constructed nature of reasoning for collaborative writing. Patterns in the data revealed that all target students used reasoning skills in similar ways to: (a) assess prior models of scientific phenomena; (b) generate new models for scientific phenomena; (c) extend models to new situations; and (d) support other forms of reasoning. One of the target pairs improved their reasoning discourse over time by discussing conceptual meanings more fully and bringing specific information into their conversations. The other two target pairs did not change their reasoning discourse systematically throughout the intervention. The findings of this study suggest that an understanding of socially constructed scientific reasoning skills is important for designing meaningful collaborative activities and assessments in science. © 1995 John Wiley & Sons, Inc.

In this article we assert a potential research agenda for the teaching and learning of science as inquiry as part of the JRST series on reform in science education. Drawing on the theoretical frameworks of cognitive and sociocultural constructivism, cultural models of meaning, the dialogic function of language, and transformational models of teacher education, we propose that more research is needed in the areas of teachers' beliefs, knowledge, and practices of inquiry-based science, as well as, student learning. Because the efficacy of reform efforts rest largely with teachers, their voices need to be included in the design and implementation of inquiry-based curriculum. As we review the literature and pose future research questions, we propose that particular attention be paid to research on inquiry in diverse classrooms, and to modes of inquiry-based instruction that are designed by teachers. © 2001 John Wiley & Sons, Inc. J Res Sci Teach 38: 631–645, 2001

This study examined the thinking processes used by 16 eighth grade science writers during laboratory report writing and explored the possibility that writing can contribute directly to science learning. Using Bereiter and Scardamalia's (1987) knowledge-transformation model of writing as a theoretical lens, the study characterized specific content and rhetorical thinking engaged in by the students using think-aloud protocols and qualitative data analysis methodologies. Thinking aloud was also related to the quality of the students' written products. Five of the 16 students exhibited no mental reflection during writing, recording information straight from memory into the composition. Two students engaged primarily in rhetorical planning, specifying the sequencing and organization of their writing in advance. Nine students demonstrated scientific problem solving including hypothesis and evidence generation, examining patterns in the data, and making general knowledge claims in response to the need to generate content for writing, indicating that the act of report writing can stimulate science learning directly. However, thinking during writing was not necessary to compose a report that contained hypotheses and supporting evidence. © 2000 John Wiley & Sons, Inc. J Res Sci Teach 37: 676–690, 2000

This article presents and discusses preliminary research on a new heuristic tool for learning from laboratory activities in secondary science. The tool, called the science writing heuristic, can be used by teachers as a framework from which to design classroom activities. Theoretically, the science writing heuristic represents a bridge between traditional laboratory reports and types of writing that promote personal construction of meaning. Two eighth-grade classes participated in using the science writing heuristic during an 8-week stream study. The teacher and one of the researchers collaboratively developed activities based on the science writing heuristic that the teacher implemented. Nineteen target students were studied in depth. Characteristics of report writing and students' understanding of the nature of science were investigated, using interpretive techniques. There is evidence that use of the science writing heuristic facilitated students to generate meaning from data, make connections among procedures, data, evidence, and claims, and engage in metacognition. Students' vague understandings of the nature of science at the beginning of the study were modified to more complex, rich, and specific understandings. The implications of the study for writing in science classrooms is discussed. © 1999 John Wiley & Sons, Inc. J Res Sci Teach 36: 1065–1084, 1999

The purposes of this study were to (a) investigate how one group of middle school students generated meanings for scientific data and expressed them in writing, and (b) develop a methodology for assessing the relationship between students' written text and their scientific thinking. Previous research on writing to learn has focused on the reformulation of content material supplied by the teacher, rather than authentic inquiry data that provide opportunities for meaningful interpretation. The research design was a content analysis of documents produced in a naturalistic setting. Data analysis focused on the frequency, placement, and purpose of meaningful inferences embedded in the reports, as well as coding for the elaboration, extension, and enhancement of science ideas. Results indicated that many student reports included a minimal number of written inferences, expressing only vague meanings for the data. However, same students integrated inference and data, using inference to (a) explain specific meanings for data points in context, and (b) pose new hypotheses to explain data. An analysis of expansion characteristics revealed that some students were able to elaborate richly upon their initial ideas through the use of language, thereby generating new meaning for the investigations. (C) 1999 John Wiley & Sons, inc.

The purposes of this study were to (a) investigate how one group of middle school students generated meanings for scientific data and expressed them in writing, and (b) develop a methodology for assessing the relationship between students' written text and their scientific thinking. Previous research on writing to learn has focused on the reformulation of content material supplied by the teacher, rather than authentic inquiry data that provide opportunities for meaningful interpretation. The research design was a content analysis of documents produced in a naturalistic setting. Data analysis focused on the frequency, placement, and purpose of meaningful inferences embedded in the reports, as well as coding for the elaboration, extension, and enhancement of science ideas. Results indicated that many student reports included a minimal number of written inferences, expressing only vague meanings for the data. However, some students integrated inference and data, using inference to (a) explain specific meanings for data points in context, and (b) pose new hypotheses to explain data. An analysis of expansion characteristics revealed that some students were able to elaborate richly upon their initial ideas through the use of language, thereby generating new meaning for the investigations. © 1999 John Wiley & Sons, Inc. J Res Sci Teach 36: 1044–1061, 1999

This article explores the history and theoretical paradigms associated with writing to learn in science, including the debate surrounding the teaching of traditional scientific genres that has received attention in Australia and the United Kingdom. It is asserted that unique features of writing in traditional scientific genres, such as experiment, explanation, and report, promote reflection and the production of new knowledge, especially through the formation of meaningful inferences for data. The author presents sample data illustrating the potential for meaningful learning associated with writing in communicative genres, considers the limited potential of creative writing for developing scientific understandings, and recasts a description of scientific genres in light of contemporary classroom practices. (C) 1999 John Wiley & Sons, Inc.

This article explores the history and theoretical paradigms associated with writing to learn in science, including the debate surrounding the teaching of traditional scientific genres that has received attention in Australia and the United Kingdom. It is asserted that unique features of writing in traditional scientific genres, such as experiment, explanation, and report, promote reflection and the production of new knowledge, especially through the formation of meaningful inferences for data. The author presents sample data illustrating the potential for meaningful learning associated with writing in communicative genres, considers the limited potential of creative writing for developing scientific understandings, and recasts a description of scientific genres in light of contemporary classroom practices. ©1999 John Wiley & Sons, Inc. Sci Ed83: 115–130, 1999.