Filling the Great Void

Why We Should Bring Nonfiction into the Early-Grade Classroom

Data from several different sources converge on the point that informational text is scarce in primary-grade classrooms. One such source of data is the analyses of the text genres represented in basal reading series. The proportions we found reported in studies within the last two decades ranged from a high of 33.8 percent factual articles in eight basal reading series for grade 2 (Schmidt, Caul, Byers, & Buchman, 1984) to a low of 12 percent nonfiction in five basal reading series for grade 1 (Hoffman et al., 1994). In the most recent analysis of which we are aware, Moss and Newton (1998) examined six grade-2 basal reading series, copyright 1995 to 1997. They found a mean of 16 percent of selections that could be classified as informational literature.

One study (Flood & Lapp, 1986) examined the presence of expository text in three standardized reading tests, as well as in the tests and materials from basal reading programs (K–6). Flood and Lapp (1986) found that 38 percent of the selections in the standardized reading tests were expository (not reported by grade level), as compared with 16 percent of the selections in the basal readers and 38 percent of the selections in basal tests. The authors note the considerable discontinuity between the genres included in the basal readers themselves and the genres included in the tests children will take.

Another source of data regarding the scarcity of informational text in the primary grades are surveys about the presence or absence of informational text in primary-grade classrooms. In a recent survey of 126 primary-grade teachers, Yopp and Yopp (2000) found that only 14 percent of materials that primary-grade teachers reported reading aloud on a given day were informational. A survey of 83 primary-grade teachers conducted by Pressley, Rankin, and Yokoi (1996) indicated that only 6 percent of material read throughout the school day (not only read aloud) was expository. Notably, the latter sample was composed particularly of primary-grade teachers nominated by language-arts supervisors as the most effective in their jurisdictions. It is possible that such teachers use more or less expository text than a more typical group of primary-grade teachers.

Direct classroom observation has also been used as a source of data about the amount of informational text experience offered to children in early schooling. Kamberelis (1998) observed all assigned and self-selected reading and writing in three classrooms, one in each grade K–2, for a 4-month period. He found that science reports/books were read by the children far less often than stories and were also far less often the genre of assigned classroom writing. In raw terms, fewer than 20 science reports/books were used in language-arts instruction in each classroom; fewer than 10 science reports/books were assigned for writing. Duke (2000) conducted an observational study of 20 first-grade classrooms in both low- and high-socioeconomic-status (SES) school districts. The study revealed a scarcity of informational text not only in classroom written language activities, but also in classroom libraries and on classroom walls and other surfaces. Informational text was particularly scarce in classrooms in the low-SES school districts studied.

*  *  *

Notably, evidence suggests that informational text has not always been so scarce in primary-grade classrooms. On the contrary, analyses of reading materials available to young children throughout the history of American literacy instruction indicate that at some points in time informational text has enjoyed some prominence. The first such period occurred at and around the birth of the nation in 1776. Until this time, children's readers in this country contained primarily religious selections. After the American Revolution, readers included many selections intended to emphasize national pride, unification, and citizenship. Some informational selections were included for the purpose of instructing children about their environment, affairs of the state, and the workings of the new government (Smith, 1986).

With the rise of industrialism and the perceived link between education and worker productivity (Fifth Annual Report, 1842), instructional materials included more informational text. By 1870, McGuffey readers not only offered many childhood stories, but also included nature-based text and histories (Venezky, 1982). An examination of a midcentury reader, Wilson's School and Family series, shows a heavy emphasis on informational text (Smith, 1986). Moralistic, character-building stories were also prevalent, but attention to scientific study in particular is evident.

By the end of the 19th century, however, the tide turned away from inclusion of informational texts in instructional materials in reading. As part of a set of reforms intended to prompt a return to classicism, Charles Eliot, then president of Harvard University, called for the elimination of readers in favor of real literature. Eliot argued that children should be exposed to quality materials rather than watered-down offerings in readers. This influence, more than any other, shifted early reading materials toward narrative literature and away from expository text. The desire to appeal to the child's imagination became a driving force in textbook publishing (Venezky, 1982). By the 1920s, narrative, particularly realistic narrative, held absolute dominance in materials for reading instruction. Smith's (1986) analysis of 10 popular primers published in the 1920s shows that more than 80 percent of the pages were devoted to realistic fictional narrative; no pages were devoted to informational text.

Informational text never regained its foothold in emergent reading instruction in the 20th century, although it did enjoy fleeting periods of increased interest during this period. The Activity Movement of the 1930s led to a significant focus of attention on factual materials in some schools (Smith, 1986); after the first World War there was a brief emphasis on content reading throughout schooling (Venezky, 1982); and between 1950 and 1962, the Developmental Reading Series published by Bond and Fay broke from its peers in including attention to a wide variety of genres. But beyond limited exceptions such as these, informational text has remained rare in materials used for reading instruction.

*  *  *

At least three beliefs seem to underlie inattention to informational text in primary-grade or other early-childhood classrooms—that young children cannot handle informational text, that young children do not like informational text, and that young children should first learn to read and then (at about fourth grade) read to learn. In this section we demonstrate that none of these beliefs is supported by available research in this area.

Unsupported Belief 1: Young Children Cannot Handle Informational Text

The first unsupported belief is that young children cannot handle informational text, that narrative genres are the primary, if not the only, means by which young children can understand and communicate in the world around them. This view has certainly been held by a number of researchers and theorists (e.g., Britton, Burgess, Martin, McLeod, & Rosen, 1975; Egan, 1986, 1993; Moffett, 1968; Sawyer & Watson, 1987). The extent to which it is held by classroom teachers, publishers, or other groups has not been well documented, though it perhaps can be inferred by the textual choices made by these groups.

We could locate no research to support the assertion that young children are unable to handle informational text, nor could we find evidence in support of the primacy of narrative genres for young children. However, we located several studies offering evidence that young children can learn from and about informational text if given opportunities to interact with such forms. In a landmark study in this area, Pappas (1991a, 1991b, 1993) asked kindergarten-aged children, on three occasions, to pretend to read information books and storybooks that had been read to them immediately before. For both the informational books and the storybooks, children's pretend readings showed increasing similarity to the books read to them in terms of a number of language features. This suggested that the children were able to learn about information book language, given exposure to informational texts.

In a related study, Duke and Kays (1998) asked kindergarten-aged children to pretend to read an unfamiliar information book before and after a 3-month period of exposure to other information books through teacher read-alouds. Children's pretend readings on the second occasion reflected greater knowledge of several important features of informational text, such as the use of timeless-present-tense verb constructions and generic noun structures (Firefighters fight fires versus The firefighter is fighting a fire). Again, these results suggest that young children can learn the language of informational text and reflect that knowledge in a pretend-reading context.

Some data from the Duke and Kays (1998) study and a study by Moss (1997) suggest that young children can learn content, as well as language, from informational texts. In the Duke and Kays study, children reflected content knowledge derived from informational texts in their journals. For example, after hearing the book Potato (Watts, 1988) read aloud, one child drew a detailed picture of a potato plant sprouting and explained the process in some depth. After hearing the book Earthworm (Soutter-Perrot, 1993), another child drew a picture of earthworms, depicting the segmented structure of their bodies. Characteristics of all-terrain vehicles, facts about spiders and their prey, and different types of ocean animals are other examples of the content found in children's journals following information book read-alouds. In another study, Moss examined first-grade children's retellings of an informational text (Selsam, 1973) read to them. Eighteen of the 20 first-graders she studied produced retellings scoring 3 or higher (out of 5) on a modified version of the Richness of Retelling Scale (Irwin & Mitchell, 1983). Again, young children appeared to learn content from informational texts.

Studies examining children's discussions around informational text also suggest that they are capable of interacting successfully with such texts. Hicks (1995) documented the ways in which children in first grade participated in sophisticated discussions of informational text in the context of a classroom that included many texts in this genre. Oyler and Barry (1996) showed how students in a first-grade classroom initiated intertextual connections among information books when given opportunity to do so.

Another source of data regarding young children's ability to interact with informational text is teachers' own reports. Several teachers have published accounts of using informational text in their early childhood classrooms (e.g., Dalton & Mallett, 1995; Duthie, 1994, 1996; Fisher, 1994; Guillaume, 1998; Kamil & Lane, 1997a, 1997b; Read, 2001; Smith 1992). Although these accounts differ substantially in their scope and focus, all indicate that students were successful with, and indeed benefited from, inclusion of informational text in the classroom.

Unsupported Belief 2: Young Children Do Not Like Informational Text, or at Least Prefer Other Forms of Text

A second unsupported belief that may underlie inattention to informational text in the early grades is that, whether or not they can handle informational texts, young children do not like them, or at least prefer other genres. One piece of evidence indicating that this belief exists was provided in a study by Kletzien and Szabo (1998). As part of this study, the researchers asked six teachers in grades 1 through 3 to predict which text their students would prefer to read between forced choices clearly reflective of informational and narrative genres, such as "All About Soccer" (informational) and "Chris Makes the Team" (narrative). The teachers in the study predicted that their students would prefer the narrative titles much of the time, yet in actuality, the students preferred the informational titles at least half the time. Just how widespread this belief is, however, is not well established in the existing research literature.

Available evidence does not support the notion that young children do not like informational text or even that they prefer other text forms. In most cases, available research simply is not relevant to the question. For example, many of the studies on reading interests and preferences have not included young (pre-K–2) children, and many have investigated preferences for particular topics, rather than genres (Monson & Sebesta, 1991). (Genre preferences cannot be easily inferred from topic preferences—books about animals, for example, can be narrative or informational in form.)

Of the studies that have investigated genre preferences among pre-K–2 children, results are mixed. Robinson, Larsen, Haupt, and Mohlman (1997) found that kindergarten and prekindergarten children chose modern and traditional fantasy narrative more often than informational text when given the option of several different genres of text. But Kletzien and Szabo (1998) found that grades 1-3 children preferred information books at least as often as narratives, with boys generally choosing information books more often than girls. The variation in these and other studies of reading preferences may be explained by a host of factors, including differences in methodologies used, age of subjects, consideration or lack of consideration of gender effects, subjects' familiarity or lack of familiarity with the texts assessed, and so on (see Kletzien, 1999, for a review).

A study by Horowitz and Freeman (1995) suggests that the ways in which texts are used in the classrooms studied may have an impact on children's attitudes toward them. In their study, a second-grade class, in which discussion followed a read-aloud, preferred an informational science book to a narrative science book; in a second-grade class with no discussion following the read-aloud, the narrative science book was preferred.

Unsupported Belief 3: Young Children Should First Learn To Read and Then (at About Fourth Grade) Read To Learn

The two unsupported beliefs discussed previously may feed into a third unsupported belief—that children must learn to read before they can read to learn. To our knowledge, the popular articulation of this belief derives from Jeanne Chall's (1983) classic work Stages of Reading Development. (To read more about Chall's stages, see "The Classic Study on Poor Children's Fourth-Grade Slump.") However, the intent in Chall's work was more to describe the stages children go through in reading development than to argue that these are the stages children should go through in reading development. There may be an empirical basis for describing stages of reading development as such—at the time of Chall's writing, as now, informational text was scarce among beginning reading materials—but there is no empirical basis for saying that it therefore should be such.

Research on what happens when more informational text is included in primary-grade classrooms is beginning to emerge. A study in first-grade classrooms in low-SES school districts indicates that there are benefits—even as early as the end of first grade—to including more informational text in classroom activities and the classroom environment (Duke, Martineau, Frank, & Bennett-Armistead, 2003). Children in classrooms with more informational text had the same levels of overall reading and writing achievement as children in comparison classrooms and were better writers of informational text. These students also did not show the decline in attitudes toward recreational reading that was found among students in the comparison classrooms. Moreover, children in the classrooms with more informational text who entered first grade with low sound-letter knowledge had higher reading comprehension and writing achievement by the end of first grade than comparable children in the other classrooms. Overall then, including more informational text in first-grade classrooms had positive effects on reading and writing achievement, as well as motivation for at least some groups of students, and no negative effects for any group by the end of first grade.

*  *  *

The most obvious argument for greater attention to informational text in the early grades is that it will make children better readers and writers of informational text (e.g. Christie, 1984, 1987). Fundamental to this argument is the supposition that more and earlier exposure to informational text will result in greater abilities to read and write informational text (Duke, 2000). Studies in which children's knowledge of informational text has appeared to develop following exposure (e.g., Duke & Kays, 1998; Pappas, 1993) seem to support this supposition, as do general patterns in which children seem to read and write better those forms of text to which they have had ongoing exposure (e.g., Kamberelis, 1998; Purcell-Gates, 1988; Purcell-Gates, McIntyre, & Freppon, 1995). Sometimes cited along with this argument are statistics on American students' relatively poor informational reading and writing abilities, coupled with the relatively important role of informational literacies in American society (e.g., Duke, 1999; Moss, Leone, & DiPillo, 1997).

A cluster of arguments for greater attention to informational text focuses on other types of knowledge and skills that may help to develop content-area knowledge, vocabulary, and comprehension (e.g., Dreher, 2000; Guillaume, 1998). In regard to content-area knowledge, there is some limited evidence that young children can learn about the world around them through informational texts. Certainly among older children there is a relationship between informational reading and writing abilities and content-area achievement (e.g., Bernhardt, Destino, Kamil, & Rodriguez-Munoz, 1995). Specialized vocabulary is a key feature of informational text (e.g., Duke & Kays, 1998), and there is evidence that even young children do learn vocabulary from text, including text read out loud (e.g., Elley, 1989). Studies showing that teachers and/or parents attend more to vocabulary and comprehension when interacting with children around informational texts seems to reinforce the claim that informational text has vocabulary-building potential, and they also raise the possibility that general comprehension skills may be further developed through these texts (Lennox, 1995; Pellegrini, Perlmutter, Galda, & Brody, 1990; Smolkin & Donovan, 2000).

Taking the skill-building arguments one step further are those who suggest that greater attention to informational text early on may indeed buttress overall literacy development. The datum most often cited in support of this argument comes from results of the National Assessment of Educational Progress (NAEP). Results showed that fourth-grade children who report reading storybooks, magazines, and nonfiction had, on average, higher reading achievement than children who reported reading only two of these types, who in turn had higher achievement than children who reported reading only one. The interpretation (e.g., Dreher, 1998/1999) is that reading a greater variety of texts may make one a better overall reader or writer (note, however, that it is not possible to establish causality on the basis of these data).

A number of mechanisms have been suggested by which informational text may support overall literacy development. One relates to the notion discussed previously, that informational text may build background knowledge, vocabulary, and comprehension skills, which may, in turn, support reading of all kinds. A second relates to interest. It appears that at least some children have high levels of interest in informational texts or topics addressed therein. For those children, the presence of informational text in the classroom may be motivating. That motivation, in turn, may encourage children to read more or to read more productively (e.g., Caswell & Duke, 1998). A third possible mechanism by which informational text may support overall literacy development relates to home literacies. Informational text is read widely outside of schools (Venezky, 1982). The presence of informational text in early schooling may help children make links between home and school literacies and develop a more comprehensive understanding of what counts as literacy (see Duke, in press, for further discussion). This may be particularly important for children from homes in which narrative reading or writing is not common (e.g., Caswell & Duke, 1998).

Nell K. Duke is assistant professor of teacher education and learning technology and culture at Michigan State University. V. Susan Bennett-Armistead is a doctoral candidate at Michigan State University and literacy consultant to several urban early childhood programs. Ebony M. Roberts is operations manager for Black Star Educational Management, LLC in Detroit, Michigan. This article is adapted with permission from "Bridging the Gap between Learning to Read and Reading to Learn" in Literacy and Young Children, edited by Diane M. Barone and Lesley Mandel Morrow, Guilford Press, 2003, a version of Duke, N.K., Bennett-Armistead, S., & Roberts, E.M. (2002), "Incorporating informational text in the primary grades," in C.M. Roller (eds.) Comprehensive reading instruction across grade levels: A collection of papers from the Reading Research 2001 Conference. Newark, Del.: International Reading Association.


Bernhardt, E., Destino, T., Kamil, M., & Rodriguez-Munoz, M. (1995). Assessing science knowledge in an English/Spanish bilingual elementary school. Cognosos, 4, 4–6.

Britton, J., Burgess, T., Martin, N., McLeod, A., & Rosen, H. (1975). The development of writing abilities. London: Macmillan.

Caswell, L.,J., & Duke, N.K. (1998). Non-narrative as a catalyst for literacy development. Language Arts, 75, 108–117.

Chall, J.S. (1983). Stages of reading development. New York: McGraw-Hill.

Christie, F. (1984). Young children's writing development: The relationship of written genres to curriculum genres. In B. Bartlett & J. Carr (Eds.), Language in Education conference: A report of proceedings (pp. 41–69). Brisbane, CAE, Australia: Mr. Gravatt Campus.

Christie, F. (1987). Factual writing in the first years of school. Australian Journal of Reading, 10, 207–216.

Dalton, J., & Mallett, M. (1995). Six year olds read about fire-fighters. Reading, 29, 22–25.

Dreher, M.J. (1998/1999). Motivating children to read more nonfiction. The Reading Teacher, 42, 414–417.

Dreher, M.J. (2000). Fostering reading for learning. In L. Baker, M.J. Dreher, & J.T. Guthrie (Eds.), Engaging young readers: Promoting achievement and motivation (pp. 94–118). New York: Guilford Press.

Duke, N.K. (1999). Using nonfiction to increase reading achievement and world knowledge. Occasional paper of the Scholastic Center for Literacy and Learning, N.Y.

Duke, N.K. (2000). 3.6 minutes per day: The scarcity of informational texts in first grade. Reading Research Quarterly, 35, 202–224.

Duke, N.K. (in press). Reading to learn from the very beginning: Informational literacy in early childhood. To appear in Young Children.

Duke, N.K., & Kays, J. (1998) "Can I say 'once upon a time?'": Kindergarten children developing knowledge of information book language. Early Childhood Research Quarterly, 13, 295–318.

Duke, N.K., Martineau, J.A., Frank, K.A., Bennett-Armistead, V.S. (2003). 33.6 minutes per day: The impact of including more informational text in first grade. Unpublished manuscript, Michigan State University.

Duthie, C. (1994). Nonfiction: A genre study for the primary classroom. Language Arts. 71, 588–595.

Duthie, C. (1996). True stories: Nonfiction literacy in the primary classroom. York, Maine: Stenhouse.

Egan, K. (1986). Teaching as storytelling. Chicago: University of Chicago Press.

Egan, K. (1993). Narrative and learning: A voyage of implications. Linguistics and Education, 5, 119–126.

Elley, W.B. (1989). Vocabulary acquisition from listening to stories. Reading Research Quarterly, 24, 174–187.

Fifth Annual Report of the Secretary to the Board of Education of Massachusetts, (1842). (Testimony of Horace Mann.)

Fisher, B. (1994). Writing information books in first grade. Teaching PreK–8, 25(3), 73-75.

Flood, J., & Lapp, D. (1986). Types of text: The match between what students read in basals and what they encounter in tests. Reading Research Quarterly, 21, 284–297.

Guillaume, A.M. (1998). Learning with text in the primary grades. The Reading Teacher, 51, 476–486.

Hicks, D. (1995). The social origins of essayist writing. Bulletin Suisse de Linguistique Appliqué, 61, 61–82.

Hoffman, J.V., McCarthey, S.J., Abbott, J., Christian, C., Corman, L., Curry, C., Dressman, M., Elliott, B., Matherne, D., & Stahle, D. (1994). So what's new in the new basals? A focus on first grade. Journal of Reading Behavior, 26, 47–73.

Horowitz, R., & Freeman, S.H. (1995). Robots versus spaceships: The role of discussion in kindergartners' and second graders' preferences for science text. The Reading Teacher, 49, 30–40.

Irwin, P.A., & Mitchell, J.N. (1983). A procedure for assessing the richness of retellings. Journal of Reading, 26, 391–396.

Kamberelis, G. (1998). Relations between children's literacy diets and genre development: You write what you read. Literacy Teaching and Learning, 3, 7–53.

Kamil, M.L., & Lane, D. (1997a, March). A classroom study of the efficacy of using information text for first-grade reading instruction [Online]. Paper presented at the annual meeting of the American Educational Research Association, Chicago. Available:

Kamil, M.L., & Lane, D. (1997b, December). Using information text for first grade reading instruction: Theory and practice [Online]. Paper presented at the annual meeting of the National Reading Conference, Scottsdale, Ariz. Available:

Kletzien, S.B. (1999, December). Children's reading preferences and information books. Paper presented at the National Reading Conference, Orlando, Fla.

Kletzien, S.B., & Szabo, R.J. (1998, December). Information text or narrative text? Children's preferences revisited. Paper presented at the National Reading Conference, Austin, Texas.

Lennox, S. (1995). Sharing books with children. Australian Journal of Early Childhood, 20, 12–16.

Moffett, J. (1968). Teaching the universe of discourse. Boston: Houghton Mifflin.

Monson, D.L. & Sebesta, S. (1991). Reading preferences. In J. Flood, J.M. Jensen, D. Lapp, & J.R. Squire (Eds.), Handbook of research on teaching the English language arts (pp. 664–673). New York: Macmillan.

Moss. B. (1997) A qualitative assessment of first graders' retelling of expository text. Reading Research and Instruction, 37, 1–13.

Moss, B., Leone, S., & DiPillo, M.L. (1997). Exploring the literature of fact: Linking reading and writing through information trade books. Language Arts, 74, 418–429.

Moss, B., & Newton, E. (1998, December). An examination of the informational text genre in recent basal readers. Paper presented at the National Reading Conference, Austin, Texas.

Oyler, C., & Barry, A. (1996). Intertextual connections in read-alouds of information books. Language Arts, 73, 324–329.

Pappas, C.C. (1991a). Fostering full access to literacy by including information books. Language Arts, 68, 449–462.

Pappas, C.C. (1991b). Young children's strategies in learning the "book language" of information books. Discourse Processes, 14, 203–225.

Pappas, C.C. (1993) Is narrative "primary"? Some insights from kindergartners' pretend readings of stories and information books. Journal of Reading Behavior, 25, 97–129.

Pelligrini, A.D., Perlmutter, J.C., Galda, L., & Brody, G.H. (1990). Joint reading between black Head Start children and their mothers. Child Development, 61, 443–453.

Pressley, M., Rankin, J., & Yokoi, L. (1996). A survey of instructional practices of primary teachers nominated as effective in promoting literacy. Elementary School Journal, 96, 363–384.

Purcell-Gates, V. (1988). Lexical and syntactic knowledge of written narrative held by well-read-to kindergarten and second graders. Research in the Teaching of English, 22, 128–160.

Purcell-Gates, V., McIntyre, E., & Freppon, P. (1995). Learning written storybook language in school: A comparison of low-SES children in skills-based and whole language classrooms. American Educational Research Journal, 32, 659–-685.

Read, S. (2001). "Kid mice hunt for their selfs": First and second graders writing research. Language Arts, 78, 333–342.

Robinson, C.C., Larsen, J.M. Haupt, J.H., & Mohlman, J. (1997). Picture book selection behaviors of emergent readers: Influence of genre, familiarity, and book attributes. Reading Research and Instruction, 36, 287–304.

Sawyer, W., & Watson, K. (1987). Questions of genre. In I. Reid (Ed.), The place of genre in learning: current debates (pp. 46–57). Geelong, Victoria, Australia: Deakin University, Centre for Studies in Literary Education.

Schmidt, W.H., Caul, J., Byers, J.L., & Buchmann, M. (1984). Content of basal selections: Implications for comprehension instruction. In G.G. Duffy, L.R. Roehler, & J. Mason (Eds.), Comprehension instruction: Perspectives and suggestions (pp. 144–162). New York: Longman.

Selsam, M. (1973). How kittens grow. New York: Scholastic.

Smith, F. (1992). Reading the bear facts: Information books and learning in the primary classroom. English in Education, 26, 17–25.

Smith, N.B. (1986). American reading instruction. Newark, Del.: International Reading Association.

Smolkin, L.B., & Donovan, C.A. (2000). The contexts of comprehension: Information book read alouds and comprehensive acquisition. (Report No. 2-009). Ann Arbor, Mich. Center for the Improvement of Early Reading Achievement.

Soutter-Perrot, A. (1993) Earthworms. Mankam, Minn.: Creative Editions.

Venezky, R.L. (1982). The origins of the present day chasm between adult literacy needs and school literacy instruction. Visible Language, 16, 112-127.

Watts, B. (1988). Potato. Englewood Cliffs, N.J.: Silver Burdett Press.

Yopp, R.H., & Yopp, H.K. (2000). Sharing informational text with young children. The Reading Teacher, 53, 410-423.

American Educator, Spring 2003