The New Art and Science of Teaching Reading. Robert J. Marzano

Before the 20th Century: Oral Reading Reigns

      Before 1900, oral reading was the primary focus of reading instruction. Early reading textbooks, such as William Holmes McGuffey’s (1879), were designed to improve a student’s oral reading, and in 1892, philosopher William James stated that one could judge the quality of a teacher by the quality of his or her students’ oral reading. One reason for this instructional emphasis was a lack of widespread literacy among the American populace; most households had only one person who could read (Hyatt, 1943; Smith, 1965). Therefore, reading in the home, whether for entertainment or information, was typically done aloud.

      Recitation lessons were the norm in schools; students listened to the teacher read and then tried to reproduce what they heard. Lyman Cobb (as cited in Smith, 1965) explained that students’ goals were:

      Distinct articulation of words pronounced in proper tones, suitably varied to the sense, and the emotions of the mind; with due attention to accent, to emphasis, in its several gradations; to rests or pauses of the voice, in proper places.

      Such practices prioritized elocution over comprehension (Hoffman, 1987; Hoffman & Segel, 1983), although teachers sometimes asked students to retell what they had read. Nevertheless, in 1891, Horace Mann claimed that 90 percent of students did not understand what they read.

       From 1900 to the 1970s: Meaning Resides in the Text

      As the 20th century began, the primacy of oral reading waned. According to P. David Pearson and Gina N. Cervetti (2017), rising literacy rates decreased the demand for oral reading in the home. Immigration, child labor prohibitions, and mandatory school attendance swelled public school enrollment, and an increased emphasis on scientific methods in education gave birth to the standardized testing movement. Silent reading rate and comprehension level moved to the education foreground. Pearson and Cervetti (2017) explain:

      Unlike oral reading, which had to be tested individually and required that teachers judge the quality of responses, silent reading comprehension (and rate) could be tested in group settings and scored without recourse to professional judgement; only stop watches and multiple-choice questions were needed. In modern parlance, we would say that they moved from a “high-inference” assessment tool (oral reading and retelling) to a “low-inference” tool (multiple-choice tests or timed readings). Thus, it fit the demands for efficiency (spawned by the move toward more universal education for all students) and objectivity (part of the emerging scientism of the period). (p. 15)

      Educators reasoned that if students could answer questions about a passage, they were able to read it and comprehend it.

      Early 20th century educators typically defined comprehension as accessing the true meaning of a text. That is, they viewed the meaning of a text as an inherent property of the text itself, and the job of the student (with the teacher’s assistance) was to figure it out. Reading “was viewed as a largely bottom-up process in which readers would visually analyze the features of letters, map the sounds onto letters and then onto strings of letters to pronounce words, and listen to the output to achieve understanding” (Pearson & Cervetti, 2017, p. 40). An examination of curriculum materials from this time—William H. Elson and William S. Gray’s (1936) well-known Dick and Jane series, for instance—reveals that the dominant approach to teaching reading involved asking questions about a text after students had read it.

       From the 1970s to the 1990s: Readers Construct Meaning

      Although a few prescient scholars, such as Edmund B. Huey (1908) and Edward L. Thorndike (1917), articulated sophisticated descriptions of reading comprehension in the early 20th century, widespread interest in readers’ cognitive processes did not occur until the 1970s. In what is often called the cognitive revolution (Anderson, 2010), reading theorists and researchers turned their attention to the reader (rather than the text) as the maker of meaning. They redefined comprehension as what happens when a reader, influenced and guided by his or her prior knowledge, interprets a text. This shift led to four major areas of inquiry, as shown in table 1.1.

Table 1.1: Areas of Inquiry for Reading Research During the Cognitive Revolution

Area of Inquiry	Question
Knowledge Structures	How is prior knowledge stored in the brain?
Text Structures	How is information stored in a text?
Comprehension Strategies	How do students combine prior knowledge and textual information to make meaning?
Teaching Techniques	How can educators improve students’ meaning-making skills?

      Here we briefly explain the major ideas associated with each area of inquiry.

       Knowledge Structures: Schema Theory

      With the acknowledgment that students bring prior knowledge to a text, scientists became interested in how students store that knowledge in their brains and how they access it when reading a text. One of the most prominent models was schema theory. Schema theory describes knowledge in the human brain as “little containers [called schemata] into which we deposit the particular traces of particular experiences, as well as the ‘ideas’ that derive from those experiences” (Pearson & Cervetti, 2017, p. 25). Schema theory became popular as a model of reading comprehension in part because it readily explained common phenomena, such as the human propensity to disagree about the interpretation of various texts—books, movies, newspaper articles, and the like. According to schema theory, disagreements arise because each individual approaches a text with a different, and unique, body of prior knowledge and experiences. Richard C. Anderson (1984) describes some of the additional phenomena that schema theory explained.

      • Learning and memory: For different types of texts, our minds have slots that we expect to fill with certain types of information; information that fits in the slots is that which we easily learn and remember.

      • Identifying important ideas: The slots in our schemata guide our search for what is important in a text, allowing us to separate main ideas and details.

      • Elaboration and inference: No text is ever completely explicit; our schemata help us to make educated guesses about how we should fill certain slots.

      In sum, schema theory posits that individuals store information about their experiences using particular structures; understanding those structures helps students access and apply knowledge when reading a text.

       Text Structures: Kintsch and Meyer

      A parallel line of research examined how texts store information, and how text structures interact with knowledge structures to create meaning. Researchers conceptualized structures in texts as “slots” that a writer fills with particular types of information. A reader who knows what to expect from the slots of a specific text structure will more effectively align his or her prior knowledge to the text and is more likely to comprehend it. According to Walter Kintsch (1983), whose model of text comprehension was, and still is, widely accepted:

      Comprehending a text involves the formation of two parallel, complementary structures, the textbase and the situational model. The textbase is a representation of the content of the text, consisting of the elementary propositions derived from the text which are organized into a coherent whole on the basis of some available knowledge structure, e.g. a schema. The situational model is developed in parallel with the text base proper, but it is a richer structure which contains not only the information derived from the text, but also the reader’s previous knowledge about the subject matter.