Description: Reading literacy as the basis for understanding written messages is not based solely on the ability to decode words. In order to have access to the contents of a text and be capable of assuming a critical attitude toward them demands more than that. Reading literacy must also be coupled with the ability to see the context and possible messages contained within texts (Campbell, et al., 2001; PISA, 2003). National and international studies alike indicate that reading comprehension among Swedish pupils has declined on average since the early 1990s ((Fröjd, 2005; Skolverket Barns läskom-petens 2003; Skolverket, Nationella utvärderingen, 2003; Skolverket, Utbildningsinspek-tionen, 2004; Mullins, et al., International Report, 2003; Mullins, et al., Trends, 2003; PISA, 2003). Furthermore, Swedish studies indicate that the schools´ teaching methods have changed during the same period (Skolverket Nationella utvärderingen, 2003). Acquiring satisfactory reading literacy presupposes that the pupil encounters texts in a way that provides practice in that skill (Werquin 2006). Furthermore, there is increasing rhetorical agreement about the importance of teaching methods which embrace wholeness and context. Can this also be discerned in the pupils´ reading? The present paper presents some preliminary results from a pilot study whose purpose is to investigate the degree to which Swedish pupils actually read coherent texts. By "coherent" is meant full bodies of text rather than texts fragments taken from here, there and everywhere. Against the background of my results and the results of national and international studies in the field, I will discuss possible connections between pupils´ declining reading skills, reading habits in relation to educational texts, and the design of teaching methods. Methodology: This paper is based on results from a study conducted in questionnaire form among 368 pupils in grades 4 to 9 (10 to 16 years old) from eighteen schools. Schools located in cities, small towns and in the countryside are all represented in the study. Answers from 94 pupils in four different secondary school programs (three theoretical, one vocational) in three cities in dif-ferent counties were also included in the study. The questions concerned the reading of vari-ous sorts of classroom texts (textbooks, newspapers, texts downloaded from the Internet, etc) and reading in which the pupils read coherent texts comprising entire pages. In the question-naire, queries concerning the reading of fiction we kept separate from other types of reading material. Conclusions: The results showed that more than one-quarter of all the pupils in grades 4-6 fail to read a single coherent page of text as part of their schoolwork during a normal school day, and that very few pupils read more than six pages a day. The results are similar in grades 7-9. How-ever, here we have even more pupils, close to 30 %, who do not read a single coherent page of text in the classroom. In secondary school, the percentage of pupils who do not read coherent texts rises even higher, to nearly 40 %. Furthermore, the results reveal signs of a stronger po-larization in comparison with conditions in elementary schools, with more pupils who never read a coherent text and only a few who are exposed to a large amount of coherent text. The results of this study indicate that pupils receive little training in the reading of coherent texts in school, at the same time as international studies also indicate a regression of pupils´ leisure-time reading. (Mullins et al., Trends, 2003). In this light, the significance of studying and discussing the correlation between working methods and pupils´ reading of educational texts becomes clear. In a broader perspective we are discussing opportunities for schools to contribute to and maintain the reading skills of their pupils. This pilot study will be followed up by a similar but more comprehensive study, where the results will also be analyzed in rela-tion to the various school subjects.