Forty-three sixth graders, all considered competent readers, participated in two sets of tasks. First, they used a familiar textbook to locate the answers to six questions, all of which contained terms that could be found in the book's index. Second, they participated in a report-writing project in which they were asked to generate research questions, locate information to answer those questions using multiple sources, take notes, and write a research report conforming to their teacher's expectations. Despite being competent readers, who had had instruction relevant to finding information and who had completed several reports during the year, these students exhibited a wide range of performance and many difficulties. The results are discussed in relation to their implications for helping children develop independent, flexible strategies for finding and using information.
Reading to locate information is a common school and workplace expectation (Dreher, 1993). In the workplace, adults spend more time reading to locate information than reading for any other purpose (e.g., Guthrie, Seifert, & Kirsch, l986; Kirsh & Guthrie, l984; Mikulecky, l982). In school, students are expected to engage in tasks such as skimming stories to locate facts, finding definitions for new words in glossaries, and searching for information for report writing. Research on adults indicates, however, that readingtolocate performance leaves much to be desired once tasks move beyond something simple such as looking up a fact that can be accessed through an index (e.g., Dreher, l992; Dreher & Brown, l993; Dreher & Guthrie, 1990; Guthrie & Dreher, 1990; Symons & Pressley, 1993).
Similarly, research on school children indicates that many have great difficulty with seemingly easy reading-to-locate-information tasks. A number of studies (Cole & Gardner, 1979; Kobasigawa, 1983; Neville & Pugh, 1975; Wray & Lewis, 1992) indicate that children often fail to think of and apply what they have been taught about locating information. Dreher and Sammons (l994), for example, found that success rate was just over 30% when fifthgraders (all reading at least at grade level) were asked to use a textbook to locate the answers to questions, all of which contained terms that could be looked up in the index. Many did not think to use the index. Even those who did often looked up inappropriate terms or had trouble with alphabetical order. Unsuccessful children typically tried to locate the answers to these very specific questions by using the table of contents or paging through the text. Yet these students had been taught about indexes and other information access features, and they could generally find the index and other features and explain their use when asked to do so.
Despite such difficulties, children are not only expected to find information, but also to write reports. To extend our knowledge about the problems children have in locating information and to identify potential solutions to the problems, the current research first examined sixth graders' performance on tasks similar to those used in earlier work with fifth graders (Dreher & Sammons, l994). The fifth-grade data were collected near the beginning of the school year. In the current study, baseline data were obtained on endoftheyear sixth graders' success at finding answers in a textbook. Thus, the object was to obtain developmental data on success rate after almost two years of additional schooling. Second, the current study analyzed the performance of these sixth graders as they formulated research questions, searched for information in multiple, selfselected sources, took notes, and wrote reports. Because we already know that "poor" readers have many problems, this study focused on students whose teachers considered them able to read at least "at grade level." The purpose was to identify instructional needs related to posing appropriate questions, locating the information to answer the questions, and writing an adequate report.
Baseline data on students' search performance were obtained by having students use their world history textbooks to locate the answers to six questions. These questions-all with statedinthe-book answers-contained at least one key word that could be searched in the book's index, table of contents, and/or glossary (e.g., Who started special clubs called guilds? Which countries in Western Europe are often called the Low Countries?). Before answering each question, students rated how much they already knew about the answer on a 4point scale ("I know almost nothing about the answer," "I know a little about the answer," "I know quite a bit about the answer," "I am almost an expert on the answer"). The students had used this textbook periodically, along with other social studies materials, during the months that preceded this study; as noted above, at the beginning of the school year the students had participated in lessons designed to highlight this textbook's informationaccess features. The questions, however, were selected from chapters that had not been formally studied.
A week later, the students were asked to rate their prior knowledge about ancient Romans, the upcoming world history unit, on a 4point scale ("I know almost nothing about them," "I know a little about them," "I know quite a bit about them," "I am almost an expert on them"). They also responded to an open ended question on what they already knew about ancient Rome. The following day, they were given an overview of an inclass report writing activity which was conducted over five consecutive class meetings. This reportwriting project began with a topic/source planning activity in which students were asked to generate four or five questions they might want to research about "everyday life in ancient Rome." After considering these questions, students were asked to select two or three of their questions on which they would write reports, and to plan what sources they might use to locate the information they needed.
After this planning session, students began their information collection. They were able to use the classroom or media center as they wished. In addition to the students' world history textbook, multiple copies of three short textbooks on ancient Rome were available in the classroom. The media center was well equipped with trade books, several sets of encyclopedias, and electronic catalog terminals. As students sought information, they kept a record of their efforts in notetaking packets. These packets had space for students to record the question they were seeking to answer, their notes pertinent to the question, and the source of the information. These packets also contained an endoftheday checklist on which students listed the sources they used. When students finished collecting information on their questions, they wrote their reports. They had been told at the beginning of the project that their reports would typically be two to three pages long, with an introduction, a conclusion and a multiparagraph body addressing the questions they had researched. They were to comply with past practice in their classrooms (e.g., to use their own words). To supplement the documents produced by the students, research assistants recorded observations throughout the project.
On the prior knowledge section of the reportwriting task, five students reported that they knew "quite a bit" about the ancient Romans. The remaining students were equally divided in reporting that they knew "almost nothing" or "a little" about them. Their open- ended responses matched their ratings and ranged from children who said they knew nothing but were looking forward to learning about the Romans to children who listed a few facts they had learned from movies they had seen on television or from earlier studies (e.g., Nero burned Rome, Romans killed Christ, "they like feeding christiens [sic] to lions").
When asked to list questions they might want to research, the students had little trouble. Most listed five, with the range from four to eight. The questions included topics such as food, religion, clothing, houses, education, the military, gender roles, social classes, and entertainment. These questions were not always well worded but were generally appropriate, answerable questions (e.g., "What did the children do for entertainment?" "How did the government work?" "What do they wear?" "What do the men do that is different than the women?" "Do the kids go to school: If so what's it like?"). But 51% of the students listed at least one question that required only a yes/no or shortphrase answer (e.g., "Did they belive [sic] in gods or godesses?" "How many people could fit in the colleceaum [sic] in ancient Rome?"); when one of these questions was later addressed in a student's report, it was implicitly reworded (e.g., what was Roman religion like?).
When asked to highlight questions from their list that they would research, 70% selected three, 23% selected two. The remaining 3 students highlighted one, two, and five questions, respectively. Although the majority highlighted three questions, they did not usually take notes on all three. Fourteen percent took notes on none of their selected questions, 23% took notes on one of the questions, 42% on two, and 21% on three.
An examination of the students' original list of questions and the ones they ended up addressing in their reports indicated that most students adhered to their plans at least in part. In their reports, 16% of the students wrote on none of their highlighted questions, 23% wrote on one of their highlighted questions, 40% on two, and 21% on three. If the highlighting on the original list is ignored, then 93% of the students included at least one question in their report from their original list. However, 35% of the students had at least one question in their report that was not mentioned on their original plans whether highlighted or not. (Rephrasings and shifts to subordinate or superordinate versions of the originals were not considered new questions.) These new questions were on topics they happened upon or that a friend was working on.
Almost all the students wrote on two or three questions as they had been asked to do. But 4 students reported on only one question. These 4 appeared to have used a length criterion; when they thought they had written quite a bit, they decided one question was enough. In addition, 8 of the students who wrote on two or three questions, can be considered to have, in effect, answered only one question. They too seemed to use a length criterion by answering one question fairly well and dispensing with the other in a sentence or two. The other students distributed the amount of report space to each of their questions rather evenly.
The quality of the students' reports ranged widely. One report made little sense; it consisted of approximately two disjointed sentences on each of several topics copied seemingly at random from the source. Four reports were very short mentions of two or three facts on each of two or three questions. But most students tried to address at least one of their questions in a fairly complete manner. Nine of the reports featured at least one elaborated answer. Unfortunately, however, 51% of the reports contained obvious copying of a paragraph or more from the source text.
Students were expected to use multiple sources for their information and to cite the sources at the end of the paper. Most listed two (40%) or three (28%) sources in their reports. Their reports and observations indicated heavy reliance on the short, Ancient Rome textbooks that were available in the classroom-78% listed one or more of these texts. Sixty percent listed one or more library books and 46% cited one or more encyclopedia articles. Although many students cited more than one source as required, most of the information in the report came from the short textbooks on Rome.
Most students tried to comply with the structure the teacher had specified-an introduction, multiple paragraphs, and a conclusion. Seventyseven percent had all three structures and another 16% had all but the conclusion. The remaining 7% had neither an introduction nor a conclusion. In their introductions, almost all the children made an attempt to introduce the questions they planned to write about in the report. For example, one child wrote, "In this report I'm going to tell you what the Romans did for fun. I'll also tell you how the Romans dressed for everyday life. You'll also learn about what an adverag [sic] Roman house is like. I hope you enjoy my report and learn something too." Similarly, another student began by stating, "In the past week or two my classmates and I have been answering our questions about Ancient Rome. The two questions I asked myself were what type of government did Ancient Rome have and who are some famous Romans."
Just over 40% of those who wrote conclusions basically said that they hoped we learned a lot and liked their reports (e.g., "I hope you enjoyed my report. I know I learned alot [sic] about this interesting place"); the others went a bit further, making a stab at a summary by restating what their topics had been (e.g., "I learned a lot from this report. I learned how the Romans dressed and what cosmetics they used. I learned about the Romans [sic] houses. I also learned about what the ancient Romans ate. I found out things I never knew before. In my opinion, this report was very interesting.").
On the reportwriting task, these students demonstrated that they could generate generally appropriate questions and follow through to write a report on at least some of them. But there were some difficulties in question formation evidenced by the occurrence of a considerable number of questions that could be answered with one word. In addition, it was not uncommon for good questions to be dropped along the way and replaced by a topic the student happened upon or that a friend suggested. Although switching questions is not in itself a bad thing, the reasons for switching were not usually to get a better topic but because it was easier to switch than to find the information that they originally wanted. Observations indicated, for example, that one student persisted in looking for information on his topic on three consecutive workdays; but his persistence was rare. Although many factors enter into persistence, difficulty finding information was certainly one factor.
These students generally adhered to the required report structure and to the requirement of citing more than one resource, but the conclusions were typically contentfree. In addition, although more than one source was usually cited, often only one source was actually used. Typically, there was heavy reliance on the short, easytoread Ancient Rome textbooks that were available in the classroom. Part of the reason for this reliance may have been that one of these books is all students needed to get enough information to comply with the task. Furthermore, some students wasted so much time chatting during the work sessions that they only had time to grab quick answers; but other reasons may enter into this reliance. For example, although these were capable readers, and although many did look at multiple sources, observations indicated that they found it difficult to integrate information across sources. They also had trouble locating information in longer books and in encyclopedia articles. The short chapters on a single topic in the Ancient Rome textbooks meant that students could use the table of contents to access information instead of techniques that were more difficult such as using indexes in longer books and locating and skimming an encyclopedia article.
The content of the reports was often superficial, and quite often was obviously copied. Observations indicated that students knew they were not supposed to copy; they were overheard reminding each other and asking questions such as how a list could be put into one's own words. But many just did not quite know how to avoid it, and others made no attempt not to copy. Moreover, a number of students stopped with one question in their reports or wrote an elaborated response on only one question with a sentence or so for the others.
These students were at least adequate readers in a good school who had had instruction relevant to finding information, and who had completed several reports during the year. Yet the process of posing questions, locating information and writing a report was rather challenging for them. In most of their previous report writing experiences, they had had considerable help from the teacher or worked in groups in which stronger students took the lead. Indeed, previous reports were often taken home where parents helped. In the current study, they could and did ask for help but they were encouraged to do it themselves, and the entire project was done at school. Under these conditions, automatic application of skills they had been taught was not always evident.
Thus, as in related research with children, the question arises of how best to help students develop independent, flexible strategies for finding and using information. Teachers' manuals and curriculum guides include research related instruction (e.g., Armbruster & Gudbrandsen, 1986). Furthermore, students such as those in the current study have been exposed to researchrelated instruction-that fact is obvious in that they and students in similar studies can answer questions about informationaccess features, can pass tests on such features, and can often use such features when prompted (even if they do not do so spontaneously) (e.g., Cole & Gardner, 1979; Dreher & Sammons, l994; Kobasigawa, 1983; Neville & Pugh, 1975; Wray & Lewis, 1992). However, if large numbers of students are to produce the kind of reports we would really like to see, we need to do more than produce what is often inert knowledge (e.g., Bereiter & Scardamalia, 1985; Brown & Campione, l981). We need to teach informationseeking strategies by providing systematic guided practice in a meaningful context so that children are more likely to transfer their verbal knowledge to usable strategies. In particular, a problem like massive copying of the original sources or giving up when information is not found on a first try is not solved by simply telling students not to do it. Thus, research is needed on improving instruction related to locating and using information. Rather than teaching researchrelated skills out of context as is typically the case, a productive approach may be to examine search strategy instruction that is integrated with the demands of inquirybased contentarea projects (e.g., Blumenfeld et al., 1991; Brown, 1992; Dreher, 1995).
Bereiter, C. , & Scardamalia, M. (1985). Cognitive coping strategies and the problem of "inert" knowledge. In S. Chipman, J. W. Segal, & R. Glaser (Eds.). Thinking and learning skills: Current research and open questions (Vol. 2, pp. 65-80). Hillsdale, NJ: Erlbaum.
Blumenfeld, P. C., Soloway, E., Marx, R. W., Krajcik, J. S., Guzdial, M., & Palincsar, A. (1991). Motivating projectbased learning: Sustaining the doing, supporting the learning. Educational Psychologist, 26, 369-398.
Brown, A. L. (1992). Design experiments: Theoretical and methodological challenges in creating complex interventions in classroom settings. Journal of the Learning Sciences, 2, 141-178.
Brown. A. L., & Campione, J. C. (l981). Inducing flexible thinking: A problem of access. In M. Friedman, J. P. Das, & N. O'Connor (Eds.), Intelligence and learning (pp. 515-529). New York: Plenum.
Cole, J., & Gardner, K. (1979). Topic work with firstyear secondary pupils. In E. Lunzer & K. Gardner (Eds.), The effective use of reading (pp. 167-192). Great Britain: Heinemann.
Dreher, M. J. (1995). How well can our students read to locate information? Reading Today, 12, 40.
Dreher, M. J. (1992). Locating information in textbooks. Journal of Reading, 35, 364-371.
Dreher, M. J. (1993). Reading to locate information: Societal and educational perspectives. Contemporary Educational Psychology, 18, 129-138.
Dreher, M. J., & Brown, R. (1993). Planning prompts and indexed terms in textbook search tasks. Journal of Educational Psychology, 85, 662-669.
Dreher, M. J., & Guthrie, J. T. (1990). Cognitive processes in textbook chapter search tasks. Reading Research Quarterly, 25, 323-339.
Dreher, M. J., & Sammons, R. B. (1994). Fifth graders' search for information in a textbook. Journal of Reading Behavior, 26, 301-314.
Guthrie, J. T., & Dreher, M. J. (1990). Literacy as search: Explorations via computer. In D. Nix and R. Spiro (Eds.), Cognition, education, and multimedia: Exploring ideas in high technology (pp. 65-114). Hillsdale, NJ: Erlbaum.
Guthrie, J. T. , Seifert, M., & Kirsch, I. S. (1986). Effects of education, occupation, and setting on reading practice. American Educational Research Journal, 23, 151-160.
Kirsch, I., & Guthrie, J. T. (l984). Adult reading practices for work and leisure. Adult Education Quarterly, 34, 213-232.
Kobasigawa, A. (l983). Children's retrieval skills for school learning. The Alberta Journal of Educational Research, 29(4), 259-271
Mikulecky, L. (l982). Job literacy: The relationship between school preparation and workplace actuality. Reading Research Quarterly, 17, 400-419.
Neville, M. H., & Pugh, A. K. (1975). Reading ability and ability to use a book: A study of middle-school children. Reading, 9, 23-31.
Symons, S., & Pressley., M. (1993). Prior knowledge affects text search success and extraction of information. Reading Research Quarterly, 28, 250261.
Wray, D., & Lewis, M. (1992). Primary children's use of information books. Reading, 19, 19-24.