27.5 How Textbooks are Used

In this penultimate section I want to turn from discussing textbook design, where design is taken to be the equivalent of typography and layout, to considering the situation where design is taken to be synonymous with manipulation. In short, I am interested in how one might use the knowledge we have gained from the studies described in the earlier sections in order to manipulate text so that learners can read and use text more effectively.

One aspect of this research that is of interest here is to find out what readers actually value in different text design features, both separately and in combination. For example, Thompson and Maniam (two of my undergraduates) asked a group of university students to indicate their preferences for four various designs for tabular layouts presented by Ehrenberg (1977) and illustrated in Hartley (1994a). Strong support was found for Ehrenberg's personal judgments. In another study, Kim Little (another undergraduate) asked 87 adolescents aged between 12 and 16 for their preferences for various features of the design of textbooks. Table 27-2 indicates the results. Access structures were clearly appreciated, but devices that required work (tables, graphs, questions, and suggestions for further reading) were clearly not so popular. (See also Weiten, Guadagno & Beck, 1996.) Such findings have implications for textbook writers, particularly those who want to encourage deeper text processing and/or improve the quality of textbooks (e.g., Jones, 1988).

TABLE 27-2. The rank order of preferences for features in textbooks (respondents were asked to indicate whether they found the features listed as helpful, made no difference, or unhelpful in their reading of textbooks)
		% Endorsing (N = 87)
Rank	Feature	Helpful	No Difference	Unhelpful
1 2 3 4 5 = 5 = 7 8 9 10 11 12 13 14	Headings A section that tells you what the chapter is about Short chapters Cartoons Use of color to show important points Use of underlining to show important points Photographs Spacious layout Subheadings Tables A section to remind you what the chapter was about Graphs Questions at the end of each chapter Suggestions for further reading	71 69 68 68 63 63 63 54 53 48 47 44 41 40	28 29 29 25 34 34 30 39 41 40 45 43 46 47	1 2 3 7 2 2 7 7 6 11 6 14 13 13

There has been surprisingly little research on how students and teachers actually use textbooks, and on which features they appreciate. Newton (1984) describes some early British studies with university students and with teacher-trained college students reading science textbooks. In his 1984 report, he outlines the results he obtained from examining how 12th-grade pupils used textbooks in physics, chemistry, and biology. Basically Newton found that, in these British studies, it was rare for students to read the complete texts. It appeared that on average just over one-third of the physics text was read, slightly less than half of the chemistry text, and just over one-half of the biology text. In all cases, it was common for the texts to be read after the appropriate lessons rather then before them, and there were great variations in the amounts read by individual students. The main uses that the students made of the texts were to help them answer specific questions, to help them revise, and to provide supplementary reading. Newton concludes that the main role of the textbook in this study was to act as "a surrogate teacher" and a provider of supplementary reading.

There have been more detailed reports, with case histories, of how students use distance learning materials (e.g., see Marland et al., 1984, 1990, 1992). These studies have tended to focus on how such students allocate their time, what sections they read (or don't read),~ and in what order they carry out the assignments requested of diem. Marland et al. (1990) draw attention to the fact that different groups of students in their study paid attention to different features. Some focused on the course objectives, but others never looked at them; very few paid much attention to the headings. Tables, however, were inspected closely; some students were bemused by author-provided underlining; and there was little indication that any of the students sought to develop a broad, integrated understanding of the text (see also Chapter 13). Studies by Macdonald-Ross and Scott (1995) have exandned the reading skills of large samples of students entering the British Open University. The results suggest that many entering students have difficulties with academic text. Findings such as these have implications for text design, which will be discussed below.

Clearly, how teachers use textbooks in class is also an important consideration. Three. American studies (Alverman, 1989; Hinchman, 1987; and Zahorik, 1991) suggest that teachers, overall, appear to have three different ways of using textbooks in class. These are to provide:

• Authoritative content
• Basic material that they can embellish
• Material for discussion

Zahorik found that over 80% of the teachers in his sample said they would use a textbook when teaching a particular lesson, but over 40% said they would not have their pupils read it from cover to cover. Other investigators also provide more detailed accounts, with case histories, of how teachers use textbooks in class (e.g., DiGisi & Willett, 1995; Freeman & Porter, 1980; Garner & Alexander, 1994; Roth & Anderson, 1988).

As noted above, Newton (1984) indicates that textbooks in the United Kingdom are sometimes used as "surrogate teachers." He suggests that such a use tends to restrict the ways in which textbooks are written and designed. Authors, he writes, "can assume nothing" and "the expositional style adopted has tended to give the reader a passive role."

There is some evidence to support these notions. Schallert, Alexander, and Goetz (1988) examined the strategies designed to help students process text that had been used by the authors of five popular introductory psychology and biology textbooks. The categories listed by Schallert et al. are shown in Table 27-3, together with approximate estimates of their amount of use. Schallert et al. concluded that, despite the presence of these cues, the authors generally required little effort and activity from their readers. Schallert et al. write:

Pictures and graphs were provided. Directed imagery, where an author might ask readers to imagine or construct a mental representation, was never used in our sample. Summaries were provided, but readers were not asked to summarize for themselves.... The most effort-demanding cues that were used with any substantial frequency were questions to be answered by the reader. These were usually found at the ends of chapters and may have been easily overlooked during studying.

The quality of these questions, too, may leave something to be desired (Armbruster & Ostertag, 1993; Turner, 1989).

TABLE 27-3. Strategies used by authors of five psychology and five biology introductory textbooks (Data based on Schallert, Alexander & Goetz, 1988)
Proportions of Use:
Psychology Textbooks	Biology Textbooks	Strategies used by Authors
45%	29%	Cues that direct the reader's attention (e.g., objectives, questions, boldfaces, italics)
25%	31%	Cues that signal content and organization (e.g., headings, summaries, overviews, outlines, intertextual features, text to graphic references)
10%	22%	Cues that help the reader to elaborate (e.g., examples, paraphrases, applications, marginal comments)
5%	11%	Cues that support the communication (e.g., tables, graphs, referenced drawings, photographs)
3%	5%	Cues that relate text material to familiar information (e.g., familiar quotes, allusions to common experiences and comparisons)
6%	2%	Cues that arouse and motivate the reader (e.g., humor, unreferenced illustrations, photographs)

Such a passive view of studying appears to be fairly commonplace among textbook authors. That view neglects the fact that readers vary enormously in their reasons for studying, in their ability and motivation, and in their methods of approach (see, e.g., Carbo, Dunn & Dunn, 1986; Lorch, Lorch & Klusewitz, 1993).

One particular distinction currently receiving much attention in Europe is that between "surface" and "deep" approaches to studying and reading (Entwistle, 1992; Marton & Saljo, 1984). Surface readers skim the text, retain isolated facts, and are not concerned with the overall structure or argument of the text. Deep readers, on the other hand, search for the underlying structure of the text, question it, relate ideas in the text to their own prior knowledge and experience, and so on. Table 27-4 suggests how these different study strategies may manifest themselves.

This distinction between deep and surface learning, of course, is only one of many similar ones. Whatever the terminology used, the question being raised here is how can one design instructional text encourage readers to take a deeper and a more active approach to reading? One answer, I think, is to identify successful learning strategies for reading, and to write the text in such a way that it encourages readers to practice them.

If, as Newton (1984) suggests, we consider a book as a device to think with, and if we consider that active participation is more likely to foster understanding than is a passive role, then we must consider how, as textbook designers, we might achieve this. Newton suggests, for example, that we can use self-test questions ("not necessarily difficult ones"), outlines, and advance organizers to help pupils enter into a dialogue with the author. Also, he suggests that pupils can be encouraged to use the materials provided in an active way (for example, by constructing tables and drawing diagrams). Marland et al. (1990), in their study of distance learning materials, similarly suggest that their findings have implications for text design. They write:

It may be helpful if writers were to: reduce the scope of the content to allow for more in-depth study of the text; be explicit about the expectations as to study strategies to be employed, level or quality of student response, and types of cognitive processes to be used when completing the in-text activity; structure the text in such a way that emphasizes a cumulative, interactive organic view of learning rather than a view of learning as the acquisition of isolated bits of knowledge; design assessment activities that require reinterpretation and integration of substantial chunks of content; use outcomes of in-text activities as prerequisite knowledge for further study; and make completion of some in-text activities compulsory.

In a paper I wrote in 1987, 1 listed 13 such strategies that writers, teachers, and students might use that would encourage deeper text processing. Jones (1988)*similarly describes a curriculum with such learning strategies embedded within it. Thus Newton, Marland, Jones, and I are arguing, along with others (e.g., Armbruster & Anderson, 1985; Rowntree, 1992), for what we call more coherent texts. Such texts are written for specific groups of readers; they use language with which the readers are familiar; they include experiences that readers share; they provide meaningful examples; they ask readers questions as they go along-not just in the headings or at the end (see Walczyk & Hall, 1989)-and they provide examples and problems that readers actually have to work through in order to follow the exposition. Such texts, too, can be supplemented by other kinds of reading materials (see Lapp, Flood & Ranck-Buhr, 1995).

I have provided elsewhere two chapters that illustrate how writers can use questions that readers have to answer in order to understand the following exposition (Hartley, 1985, 1986) and other examples of more complete texts in this vein, including Bransford's (1979) Human Cognition, Gagne's (1984) The Cognitive Psychology of School Learning, and Lockwoods' (1992) Activities in Self-Instructional Texts. Given the preponderance of textbooks in our schools, changing the ways in which we write them can make a major improvement in instructional practice.