Teaching first graders to comprehend complex texts through read-alouds

Witte, P.G. (2016). Teaching first graders to comprehend complex texts through read-alouds. The Reading Teacher, 70(1), 29-38.


Priscilla Witte provides us with three useful tools to help teachers build young children’s text comprehension: 1) a matrix for determining the complexity of children’s texts, 2) a lesson framework for teaching children to analyze texts that are read to them, and 3) a rubric for assessing how well children analyze those texts.

All three of these tools are potentially useful in elementary classrooms, but for me, the Text Complexity Matrix (p. 31) looks the newest and most useful. Witte’s matrix includes the use of Lexile score ranges, but it also rates texts on several qualitative factors, including story structure, illustrations, point of view, vocabulary, language, theme, and knowledge demands. Scoring seems easy to do; the qualitative aspects do require a judgment call. I’m not sure how reliable and valid the matrix really is (a next step for Witte?) but the matrix could provide a starting point for teachers trying to determine text complexity. A limitation is that the matrix is clearly only for use with narrative text; I’d like to see Witte put together a similar matrix for informational text, which many teachers are focusing on more right now.

Witte’s Complex Text Analysis (CTA) lesson framework (p. 32) also seems useful, though not entirely new. It builds on the idea of gradual release of responsibility that we’ve seen before (e.g., Fisher and Frey’s GRR framework, which is cited in the article). At the heart of CTA is lots of teacher modeling, scaffolding of text analysis with a graphic organizer, and lots of practice, with growing independence as a goal.

Witte added to the value of her study by using increasingly complex texts with the CTA framework over an entire year. She developed an assessment tool called the Complex Text Analysis Rubric (p. 34), which she used to make baseline assessments and then periodic assessments as the year progressed. As with the Text Complexity Matrix, more reliability and validity work is needed; until that work is done, Witte’s reported “significant gains” (p. 35) must be viewed cautiously. Still, the work she has done has given us some potentially useful tools, and I’m considering trying them out myself and perhaps sharing them with others.