For this study an attempt was made to bridge the disciplines of linguistics and composition in order to examine factors contributing to textual coherence. Pairs of descriptive texts written by fifty college students were examined in order to identify the factors which differentiate quality and topic. Students were asked to compose a descriptive paragraph on the topic of fall. They were then encouraged to use their five senses, given leaves, and asked to compose a paragraph describing the leaves. The pairs of texts thus elicited were evaluated for preference by readers. The ANOVA revealed a significant difference (p=.001) between the two topics with fall texts preferred over the more specific leaves texts. Results suggest that encouraging students to use their five senses does not improve their writing. It may be more important to move through various levels of abstraction than to merely focus on sensory detail. The texts were also scored holistically by two trained evaluators. Results of this grading were used to choose five high- and five low-coherence texts on each of the two topics. These 20 texts were then analyzed in terms of the tagmemic referential hierarchy. A MANOVA was done to examine the dependent variables of Slot (location in time or space), Role (purpose or reason), and Cohesion (sociocultural context) in relation to quality and topic for these texts. Slot was found to be significant for both quality (p=.025) and topic (p=.004). Role was significant only for quality (p=.001). Cohesion was nonsignificant for either quality or topic. These results suggest that students should: (a) be encouraged to locate their texts in time and space in order to develop an adequate context for readers; (b) be encouraged to include purposes and reasons for the statements they make; and (c) be encouraged, where appropriate, to include a focus on humans and their activities (either literal or figurative) since such reference is a distinguishing feature of high-coherence texts.

Our society requires an ability for us to interact and cooperate with the people in our surroundings. This applies to both grown-ups as well as young people. In school, the teachers have a responsibility to prepare pupils for the society and using cooperative learning could be one way to do this. The syllabus for the subject of Swedish invites for teachers to work with cooperative learning by formulating that the pupils should be given the opportunities to write and process texts individually and together with others, this does not appear in the syllabus for the subject of history. Hence our interest in examining cooperative learning in both subjects. The aim of this study is to examine whether cooperative teaching can foster pupils' knowledge development in the subjects’ history and Swedish for grades 4 - 6. The study will analyze the differences between teaching through traditional teaching methods and cooperative learning. To fulfill this aim the following questions were formulated: How does pupils' understanding of concepts in history differ after cooperative learning compared to after traditional teaching?How does pupils’ descriptive writing in Swedish differ after cooperative learning compared to after traditional teaching?The study is an intervention where pupils received teaching through traditional teaching as well as through teaching using cooperative learning. After each lesson, the pupils did a test and to enable the analysis these were collected to examine the differences. The results show, in accordance with previous research, that cooperative teaching did develop the pupils’ results in both subjects.