Procedures for Sampling Texts and Tasks in Kindergarten through Eighth Grade. Technical Report No. 30.

An understanding of how children interact with reading material in educational settings is crucial to an understanding of our successes and failures at teaching children to become skilled readers. A possible first step in attaining this knowledge is the collection of a corpus obtained by the systematic observation, description, and coding of reading activities in a sample of classrooms. In the present report we relate the development of a plan for collecting such a corpus, including a system for coding observation, procedures for sampling classrooms, and methods for indexing and accessing the corpus. Special emphasis is given to the development of the observational system. This system is intended to provide a unified, information-rich description of classroom reading activities and the reading materials involved. We have made a preliminary analysis of the changes in reading activities and reading materials which might be expected to occur between the lower (K-3) and middle (4-8) grades, and considered the feasibility of detecting such changes in a corpus of coded observations. Finally, after weighing the costs and benefits of collecting a representative corpus, we concluded that current needs, both theoretical and applied, might be better served by small studies focused on specific questions. Sampling Texts and Tasks 2 From its conception, the Center for the Study of Reading has been charged with spearheading a broadly based, multlfaceted attack on the problem of the development of skilled reading. One aspect of this mandate was to develop a plan for sampling the written discourse which children experience in the early and middle grades together with the different types of tasks associated with the discourse. In this report, we outline the development of an observational system which was designed for the collection of such a sample. We wish to stress at the outset that our major responsibility was the development and evaluation of a plan for sampling texts and tasks. Thus, the observational system described in this report is iot a fully developed instrument. We have, however, listed changes and modifications which would be required before Implementation. We discuss how the sample might be collected, compiled, and analyzed in order to detect changes in texts and tasks between the early and middle grades. Finally, we consider alternatives to our observational system and conclude that reading research would be better served by small studies focused on particular questions than by an omnibus sample of texts and tasks. We began with the intention of developing both observational and survey methods of sampling texts and tasks. However, as we came to a fuller understanding of the issues, we realized that the level of detail necessary and the diversity of texts and tasks to be found in the schools would make the use of survey methods impractical. We have therefore concentrated most of our efforts on the development and testing of an observational system. In this paper we recount this development, from Sampling Texts and Tasks