213 This analysis revealed that the seven teaching programs showed huge differences in text length and segmentation. Some programs used texts with many short paragraphs of sometimes only a few sentences. Other programs used longer paragraphs, but fewer paragraphs per text. We also saw that in most programs the texts for grades 3/4 did not differ or hardly differed in length from the texts for grades 5/6. Not in all programs for grades 5/6 did the number of sentences per paragraph increase, nor did the sentences become longer compared to the materials for grades 3/4. In one case the paragraphs and sentences even became shorter. This is relevant in light of the usefulness of these teaching materials for text structure instruction. It is obvious that in very short text sections there is simply less room for combinations of text structures. We concluded that while the teaching materials certainly provide starting points for text structure instruction, they are not suitable for a step-by-step implementation and application of text structure knowledge and skills. In particular, these materials lack texts with a clear single structure that could serve as sample texts. We found examples at the paragraph level, but hardly any at the text level. Moreover, not all text structures were equally represented, nor did we see a clear increase in complexity in the transition from grades 3/4 to 5/6 in terms of combinations of text structures. Analysis of the assignments surfaced useful examples of assignments that could contribute to both subject knowledge and knowledge about text structure, but these examples were rare. Most of the assignments we were able to link to text structure were inference questions that required students to make a connection at the local level, between sentences. Often this involved a cause-and-effect relationship. 4.3 Designing integrated learning materials Teaching materials that integrate science education with reading and writing education are hardly available in Dutch primary education. In order to investigate whether the combination of text structure instruction and science education is workable in the context of primary education, we therefore designed teaching materials ourselves. We did this in a design team of primary school teachers and researchers, following the steps of design-based educational research (McKenney & Reeves, 2018). We formulated four design principles (DP’s) that guided the design process and also served as a framework for analyzing the study's outputs. DP1. Select concepts that enable hands-on activities that involve understanding processes and/or causal relationships and/or comparing and classifying content. DP2. Organize hands-on activities that help understand the subject concept and related connections. DP3. Design functional reading and writing activities focused on instruction on text structure, using graphic organizers where appropriate. DP4. Teach declarative, procedural and conditional knowledge of text structure in reading and writing activities and use model texts with clear structure. 214
RkJQdWJsaXNoZXIy MTk4NDMw