Searching Informational Texts: Text and Task Characteristics That Affect Performance

Gavin T.L. Brown

For a printer-ready version of this article, click here. Use your browser’s “back” button to return to the Reading Online site.

Abstract Searching informational text involves use of text characteristics and task factors. Text characteristics include signaling devices (e.g., headings, subheadings, titles), typography (e.g., fonts, boldface or italic text), and structural features (e.g., organization, graphics, paragraphing). Task conditions depend on how explicitly or implicitly search requirements have been stated and on whether the search problem involves a single, multiple, or complex formulation. Research into young adolescent students’ use of text characteristics to locate main ideas and details in a variety of task conditions led to the development of standardized paper-and-pencil assessments. Students easily located answers using a single, declarative, verbatim search term cued through typographic and signaling features. In contrast, students found most difficult locating answers using complex, multi-word search terms and extraction into graphic organizers or structural outline note-taking frames.

Related Postings from the Archives Mediating Expository Text by Cynthia Bertelsen and John Fischer Learning From Text by Tamara Jetton and Patricia Alexander An Update on Reading in the Content Areas by Thomas Bean Policy and Practice Implications of the Programme for International Student Assessment (PISA) 2000

“Information literacy” involves the “ability to access, evaluate, and use information from a variety of sources” (Doyle, 1993, p. 138) and requires cognitive problem-solving approaches to meet a wide variety of information needs (Moore, 1995a [online abstract], 1995b, 1995c; Moore & St. George, 1989, 1991). Interest in students’ information literacy has been increasing since the 1970s (see reviews in Brown, 1997 [online abstract], 1999).

One foundational application of information literacy in reading is searching and extracting details and main ideas from informational texts. These nonfiction or expository texts are ubiquitous and important in the lives of students and adults. However, much less is known about students’ ability to locate content in informational texts than about their ability to comprehend narrative text or fiction (Dreher, 2001).

This article reviews the literature on searching and extracting details and main ideas in paper-based informational texts. (The large body of research on text searching in digital environments is outside the domain of this study; however, the research reported here may well offer valuable clues about how students search digital texts.) It describes a standardized assessment tool that was developed to measure students’ searching of informational texts in paper-based formats. An overview of results from the standardization highlights the nature of young adolescent students’ searching abilities. The article concludes with some suggestions for teacher professional development, some potential connections to searching digital texts, and some possible directions for raising overall reading comprehension.

Back to menu

Reading Informational Texts

Informational texts are reliable sources of authoritative information, presented as prose or in document format. Informational texts are most commonly thought of as prose texts — that is, textbooks, encyclopedia entries, journal articles, manuals, and so on. Document formats, such as tables, charts, graphs, maps, forms to be filled in, labels, and diagrams, are significant in communicating information efficiently and concisely (see Kirsch and Mosenthal’s 1989-93 series in the Journal of Reading, called, variously, “Understanding Documents,” “Learning From Exposition,” and “Document Strategies”). The importance of searching informational texts for main ideas and details can be seen in the prominent place given to these abilities in two recent international assessment projects: PISA, the Programme for International Student Assessment of reading literacy of 15-year-olds, and PIRLS, the Progress in International Reading Literacy Study of reading literacy of 9-year-olds.

Duke (1999, online document) has argued that young children get very little exposure to or training in reading informational texts. However, informational texts often predominate in standardized reading comprehension assessments used with older children (Daniels, 2002; Newman, 2002). The use of informational text in content area reading is commonplace in schooling once children have passed the learning-to-read stage and are fully reading to learn (normally considered at and beyond approximately age 9). One reason that informational texts may be important to older students is that they motivate them to widen their knowledge of content areas and interests (Newman). When I worked as a secondary school teacher, I noted that the changing nature and increasing presence of informational texts in the later years of schooling was such that, even in the mid-1990s, many student projects were derived, not from paper-based prose texts, but from hypertext informational texts (e.g., CD-ROMs and webpages) that could readily be incorporated into students’ own work.

Informational texts deal with relatively less familiar content and use text organizational patterns (e.g., compare and contrast, cause and effect) different from the traditional narrative structure. Understanding informational texts, then, is more difficult than understanding narrative for nearly all students (Williams, 2000). Students may need explicit scaffolding and support to be able to read expository texts effectively since they are structured so differently from narrative texts or fiction (Street, 2002).

Reading informational texts often involves reading to locate particular information (Dreher, 1993, online abstract; Guthrie & Kirsch, 1987). This is not reading to learn but looking up or consulting information in a source and then, frequently, recording that information. This is exemplified by such tasks as locating someone’s number in a telephone directory or using an encyclopedia or website to discover the mean annual hours of sunshine in New Zealand’s capital city. Locating information is strategic, selective reading to locate specific, goal-related information and often involves skimming or scanning. To skim or scan, the reader rapidly locates information that matches search requirements by seeking key elements. Characteristics of the text being read, the search task to be accomplished, and the reader all contribute to accomplishing this type of reading.

Text characteristics. When given a whole textbook in which to locate information to solve problems, adept readers use the navigational tools (or “information access tools,” as Dreher, 2001, puts it) provided — that is, the glossary, table of contents, index, or review questions — to locate information. Detecting and using the order, organization, or outline of prose text, which is often signaled by the author, enhances the location and extraction of information. Chambliss (1995) reported that the structure of informational texts consists of a generalization and one or more subtopics. The goal in reading such texts, then, is to obtain the main idea or gist. Information problems may also require students to locate details that support or provide evidence for the main ideas. León and Carretero (1995) pointed out the important role that titles of texts play in identifying the principal, logical relationship within a text. Lorch and Lorch (1995) showed how headings signal and aid readers in recognizing the organization of a text.

In addition to titles and headings that show the structural organization of information, there is a range of typographic techniques that draw attention to critical information (Harley, 1996; Waller, 1991). (As an example of advice on typographic techniques for use on websites, see Patrick Lynch and Sarah Horton’s Web Style Manual chapter on Typography.) These include the use of such features as underlining, CAPITALIZATION, change of font face, bolding, italicization, use of color, or the use of punctuation marks such as ‘single’ and “double” quotations. Judicious graphic design of pages, including the positioning of text, text boxes, white space, illustrations, or call-outs, contributes to the ease or difficulty of locating information.

Task characteristics. In addition to text characteristics, there are a number of other factors that affect whether reading to locate information can be easily accomplished. O’Donnell (1993) pointed out that the search task itself varies in difficulty. Tasks that require location of verbatim declarations, such as finding out the mean annual sunshine hours of Wellington, New Zealand, are easier than those that require incorporating several elements that are not physically contiguous or explicitly stated in the text. Dreher (1992) found that information is easier to find if readers are given the search terms, rather than needing to develop them. Brown (2001, online document) reported that single-word search terms are easier than complex or multi-word search terms to use in locating information.

Inefficient readers resort to random page selection or skimming and scanning strategies. In line with this research, Armbruster and Armstrong (1993, online abstract) reported that answers to text-explicit questions are easier to locate than those for text-implicit questions, which require some inference.

Reader characteristics. In addition to task and text factors in locating information, reader characteristics are important in determining ability to locate information (Armbruster & Armstrong, 1993; O’Donnell, 1993). The more one knows of a topic, has of the content vocabulary, or knows of complex syntactic structures used to express complex logical relationships, the easier it is to locate required information. The strategic capability of adjusting speed so as to skim rapidly for a key detail or to slow down and read carefully an appropriate text passage is critical. In addition, readers who recognize the structure — or pay attention to the structural signals — of a text or analyze the task requirements correctly will locate information more efficiently and effectively. Of course, rapid and automatic word-recognition skills are critical in freeing capacity to understand.

Note that in the survey description approach used in the research reported below it was not possible to collect data on reader characteristics prior to the study. Thus, this article reports only on the task and text factors involved in searching for information. Future research and intervention work should identify meaningful reader characteristics and use those to analyze further the impact of task and text characteristics on searching for information.

Back to menu

Information Search Model

A model based on this review of the literature on locating information in prose text is shown in Figure 1. In this model, location begins with an understanding of the search task based on various task characteristics, ranging between the most obvious and direct search task to the most complex. (For example, does the reader need to search for a single, explicitly stated or verbatim term, or is the informational problem multifaceted with implicit requirements necessitating incorporation of prior knowledge?) The reader then creates appropriate search terms, which may be taken from those supplied in the explicit statement of the search task or generated by the reader based on a subjective conceptualization of the task requirements. The ability of the reader to identify or generate explicitly usable search terms is affected by his or her characteristics as a reader, such as prior knowledge of the content, vocabulary level, or metacognitive ability.

Ideally, the reader then systematically uses the various navigational tools available (e.g., index, table of contents, or glossary) to locate a chapter or a page, within which the required information might be found. Less able searchers may ineffectively use random strategies of flipping through a potential source, while systematic use of navigation tools will require skimming and scanning strategies to locate information related to the search terms identified earlier. Once a potential chapter or page is found, the effective reader makes use of the navigational tools (e.g., headings, titles, organizational structure, typographic features, page layout) to skim or scan toward material relevant to the search terms. Once relevant material is found, the reader extracts a solution to the information problem, which is then frequently recorded or applied.

Although this model is highly linear and sequential, this is obviously not the only path that real information searchers take when locating information. Moore (1995b) has shown that information problem-solving is interactive and recursive, meaning that experience at one step of the process may cause a reader to reflect upon and revise earlier decisions. For example, a reader may not find a search term in the index or table of contents, potentially resulting in a re-examination of the search task or search terms. Likewise, failure to locate required information on a certain page may encourage the reader to return to the index or table of contents for a new search destination. Furthermore, search within a page or chapter may bring new information to light that suggests to the reader a totally new interpretation or understanding of the search task, resulting in a completely new search. It should be noted that a paper-and-pencil assessment of information searching cannot properly examine readers’ interactive or recursive search behavior; such an assessment would require reflective observation and judgment by experienced teachers.

Back to menu

Assessing Reading of Informational Texts

As part of a broader development project, it was decided to develop materials within the context of this model that assessed student use of a variety of text characteristics in a variety of task conditions and which would require use of a variety of search strategies. Several questions guided this research:

• What task conditions are more difficult for students?
• What text characteristics enable effective searching?
• What note-taking formats (linear or Web style) are more difficult to use?
• Are students able to use skim and scan strategies effectively, given significant time constraints?
• What are the average levels of performance against which future performance could be measured?

Materials. Parts of the text and task elements of the model (Figure 1) were used to develop new standardized assessment tools, Finding Information in Prose Text, for use in New Zealand schools. The Finding Information in Prose Text assessments evaluate use of a wide range of typographic and organizational features characteristic of informational texts (Table 1). Certain aspects of these features are not covered either because they are evaluated in other modules within New Zealand’s Essential Skills Assessments: Information Skills (ESA: IS) package (Croft, Dunn, & Brown, 2001), of which Finding Information in Prose Text is a part, or were considered inappropriate for an assessment that focuses on the ability to locate and extract information through the use of structural or typographic features.

Thus, in order to complete the Finding Information in Prose Text assessments, students have to answer questions that require use of one of the following features:

• Signaling devices (titles, headings, & subheadings)
• Typography (bold, italic, underlining, font use, capitalization)
• Structural elements of texts (organization, linearity, supportive graphics, sections, subsections, paragraphing, topic sentences)

Two of the structural or organizational tasks are illustrated in Figures 2 and 3. In Figure 2, junior secondary school readers have to demonstrate an advanced understanding of both the text and the note-taking frame into which the main point, a major subpoint, and two detail points have to be entered. The main point can be inferred from the title, heading, and the topic sentence of the first paragraph. The subpoint is found in the topic sentence of the first bullet-point in parallel to the given subpoint. The detail points are derived from the subsequent sentences and, based on the model, constitute the explanatory process for the subpoint.

© NZCER, 2001. Extract from Essential Skills Assessment: Information Skills. Finding Information in Prose Text - Secondary (p. 8). Reproduced by permission; not to be duplicated without explicit permission of copyright holder.

Figure 3 shows a task for intermediate school readers, in which they need to complete a “mind map” type diagram based on a page of text about whales found in New Zealand waters. The diagram provides the main points and subheadings, and makes explicit many of the relationships between points and details through the use of prompt questions. The detail point answers required by the diagram are found by scanning through the structure of the text, looking for the key terms provided by the diagram and incorporating the prompt questions with a reading of the text.

© NZCER, 2001. Extract from Essential Skills Assessment: Information Skills. Finding Information in Prose Text - Intermediate (p. 9). Reproduced by permission; not to be duplicated without explicit permission of copyright holder.

Note that the graphic organizer format was not used in the secondary test and, conversely, the linear outline organizer was not used with intermediate students. This decision was taken because it was believed that the more complex linear outline organizer would be too difficult for students in Years 7 and 8 (intermediate level in New Zealand, comparable to approximately grades 6 and 7 in the United States and Canada), and because the 30-minute time restriction for the assessment prevented more in-depth exploration of secondary student ability.

These materials, combined with the model described earlier, permit a finer grained analysis of the text and task factors contributing to students’ location of information than is commonly conducted (cf. Dreher, 2001). Unlike the information assessments in New Zealand’s National Education Monitoring Project (Crooks & Flockton, 1998), the Finding Information in Prose Text assessments do not require the use of navigation tools and strategies to locate a major section or chapter of a text. Yet the need to (a) decide the nature of the search task, (b) create search terms, (c) locate information on a page using typographic, structural, or layout features, and (d) extract a solution to the search task are encompassed within the assessment questions.

It is assumed that students who score highly on the assessments are those who are better able to construct search terms for each task, make better use of navigational tools, and are better able to recognize and use the organizational structure and typographic features of text. The Finding Information in Prose Text does not measure whether readers have similar levels of prior knowledge of the content or vocabulary useful for completion of the tasks contained within the assessments. It is also assumed that effective use of text organizational structure or typographic features, the ability to construct relevant search terms for a variety of task conditions, and the ability to use systematic, term-driven search strategies are learnable by the majority of readers.

Participants. Students in New Zealand usually begin their eight years of primary schooling in Year 1 at age 5; so, by Year 7, they are nominally 11 years old. Students in Years 7 and 8 are often enrolled in two-year “Intermediate” schools. Students begin up to five years of secondary schooling in Year 9 at age 13. (See Crooks, 2002, [online abstract] for further details of the New Zealand school system, and the Ministry of Education’s “Schooling in New Zealand.”)

Data were collected in July 1999 from a nationally representative standardization sample of about 2000 students, involving approximately 500 students in each of Years 7 through 10 (ages 11 to 14). (Note that this was one year prior to the PISA collection and two years before PIRLS.) The sample was proportionally stratified to ensure that students were enrolled in a cross-section of low to high socioeconomic status (SES) schools, with a balance of males and females. Only student demographic data about gender, year, and school were collected. However, general reading comprehension data from a separate small sample of students in Year 8 (n = 131) were collected in early 2000 as part of the New Zealand Council for Educational Research’s validation of the complete Essential Skills Assessments: Information Skills project.

Method. The two Finding Information in Prose Text assessments were standardized as part of 14 paper-and-pencil modular tests in the New Zealand Council for Educational Research’s new series of tests entitled Essential Skills Assessments: Information Skills. The two Finding Information in Prose Text tests examined in a 30-minute time limit students’ ability to locate information using a range of organizational and typographic features. The search tasks ranged in complexity from location of simple, verbatim terms to location through incorporation of complex, implicit terms.

To ensure that students used the strategies of skimming or scanning, two reading conditions were employed in the assessments:

• Students had to locate details in a range of passages in a speeded condition (i.e., answer 16 questions in 10 minutes) to identify those who could skim or scan to locate information
• Students had to identify and use the organizational structure of a range of passages to complete note-taking frameworks, such as tables, organizational outlines (for an example, see Figure 2), or graphic organizers (e.g., Figure 3) in a less-speeded condition (i.e., answer 15 or 16 questions in 20 minutes)

Performance was also analyzed by the text and task characteristics involved in completing the assessments. The text characteristics for the intermediate test were typographic features, headings and titles, graphic text, and tabular and graphic organizers; the secondary test text characteristics were typographic features, headings and titles, and outline organizers that examined location and extraction of main points, subpoints, and detail points. The three task characteristics common to both tests related to the distance the reader must search to locate the answer required by the question search terms, and whether the answer was verbatim or inferred. Answers that could be found within three or fewer words of the search term given in the question were classified as near, while those four or more words from the search term were classified as far. Answers that were expressed in different words than that of the search term were classified as inference.

Note that the texts used in the ESA:IS are understandably easier than those used in the PISA studies, since they were aimed at children younger than the 15-year-olds assessed in PISA (e.g., “Labour” and “Scientific Police Weapons” from PISA), but the search tasks are reasonably similar. The PISA study defined five levels of retrieving information, with Level 1 tasks similar to explicitly stated, single-word matching tasks in the ESA:IS, Level 3 tasks similar to multi-word inference tasks, and Level 5 similar to multi-word, inference, far-matching tasks. For example, a Level 5 task required students to infer the correct answer across a far distance, while a Level 3 task required students to infer across a near distance.

Differences in mean achievement were analyzed with effect sizes (Wilkinson & Task Force on Statistical Inference, 1999, online document). Effect sizes were calculated by dividing the difference in mean (M) by the average of standard deviations (SDs). Effect sizes around 0.40 are the mean effect due to regular instruction (Hattie, 1993). Thus, effect sizes lower than this indicate very little difference, while those much larger indicate real difference.

Back to menu

Results

Data from this nationally representative survey of intermediate and secondary students examined the text and task characteristics that influence ability to answer questions related to searching for details and main ideas in paper-based prose texts. The two Finding Information in Prose Text tests have acceptable psychometric characteristics (for intermediate, M = 21.5 out of 32 [67%], SD = 6.8, with estimate of reliability α = .89; for secondary, M = 18.2 out of 31 [59%], SD = 5.4, with estimate of reliability α = .9). Data collected for construct validation purposes showed a strong correlation (n = 131; r = 0.79) between the ability of students in Year 8 (nominally age 12 years) to locate information in prose text and their general reading comprehension as measured by a standardized paper-and-pencil test, the Progressive Achievement Test of Reading Comprehension (Reid & Elley, 1991).

Mean scores by year level, gender, and school SES from the standardization for both tests, broken out by test section, are displayed in Table 2. In both the intermediate and secondary tests, students performed substantially better on the 10-minute rapid location section than in the 20-minute organizational and note-taking section, with average scores of 74% compared to 61% for intermediate-level students and 66% compared to 50% for secondary. For most students, the time limit and number of questions did not constitute a significant barrier to the location of answers using the typographic features of the various texts. On the other hand, the organizational section represented a more challenging set of tasks, despite allowing twice the time to complete this section of the assessment.

For each test, students in the upper of the two years outperformed those in the lower year by a small but statistically significant margin (effect sizes ranged from 0.27 to 0.38). Girls outperformed boys by a small yet statistically significant margin, regardless of year (effect sizes ranged from 0.17 to 0.37). These small effect sizes suggest little meaningful growth in information location ability in the one-year gap between year groups.

Students in higher socioeconomic schools had higher mean scores than those in lower socioeconomic schools. This is significantly more obvious among secondary students, where effect sizes ranged from 1.22 to 1.62, compared to effect sizes of only 0.19 to 0.56 for intermediate students. This difference suggests that the gap between students in “have” and “have not” schools increases as students progress through compulsory schooling — an example consistent with the “Matthew effect” (Stanovich, 1986). However, as in other ESA: IS test standardization results (Brown, 2001), this difference between SES categories is tempered by the fact that the range of scores in every socioeconomic grouping was 0% (or nearly 0%) or to 100% for both parts of the test across both forms. In other words, some student or students got the highest possible mark, regardless of enrollment in a low SES school, while other students got the lowest possible marks, despite enrollment in a high SES school. It would appear then that school SES is a poor proxy for an individual student’s achievement. Note that this is a similar pattern of results to that found in PISA, in which New Zealand students showed considerable variance within schools.

The most pedagogically significant result was the difference in achievement between the two major sections of the two tests. For all students, the 10-minute speeded section that required students to skim or scan to locate information using primarily typographic features was easier than the 20-minute section on using the organization of informational texts. This difference is significant across year level, gender, and school SES, with older students, girls, and those in higher SES schools outperforming their counterparts. For Year 7 and 8 students, the average mean effect size was 0.66 (range, 0.55 to 0.89); for Year 9 and 10 students, the average mean effect size was 0.73 (range, 0.41 to 0.94). Thus, there was a difference of some two-thirds to three-quarters of a standard deviation in student ability to locate typographically signaled information compared to using the organization of informational texts to locate and extract information.

A more detailed analysis of the aspects of finding information in prose text that students found easier and harder is provided in Table 3, which shows the main text characteristics assessed in the Finding Information in Prose Text tests by test level and the three task characteristics used to classify test questions. All students found skimming and scanning for nearby answers (i.e., fewer than four words from the search term given in the task) requiring the use of typographic cues and headings and titles easiest to complete. The typographic features questions were text explicit, declarative, verbatim search tasks that required finding one nearby term. For the seven questions of this type in the intermediate test, students scored a mean of 82% correct; secondary students, on a similar six questions, scored a mean of 70% correct. Making use of headings and titles to locate information was also easy (78% mean correct for two items on the intermediate test and 81% correct for two items on the secondary test) but became somewhat more difficult as location of answers depended on inference (63% correct for four items on the intermediate test and 58% correct on three items on the secondary test).

Extracting information into tabular or graphic notes (e.g., Figure 3) for intermediate students and an outline organizer based on inference answers (e.g., Figure 2) for secondary students were the hardest location tasks, despite significantly more time allowed to complete this section of the assessment. These questions required students to incorporate information from different parts of the text or to infer from implicit statements in the text. These questions had multiple or complex search terms and the answers were expressed in language other than that of the question search terms. Intermediate students found completing the graphic organizer (i.e., Figure 3) most difficult (a total of seven items for a mean score of 51% correct), while secondary students found completing the linear outline organizer (i.e., Figure 2) most difficult (a total of six items answered correctly on average by only 43% of students). These data compare appropriately with the 2000 PISA results for New Zealand that found 23% of 15-year-olds at Level 3, 25% at Level 4, and 22% at Level 5 for retrieving information. (Download a spreadsheet of PISA results; requires Microsoft Excel or compatible software to view.)

Back to menu

Discussion

It should be noted that, due to test-construction processes wherein extremely easy or difficult items were removed prior to standardization, the full range of student ability could not be identified with these data. In other words, some students may be able to locate information in much more complex texts with much more challenging search requirements. In addition, note that although the two tests are similar in their intent and coverage, they are not parallel forms; thus, achievement levels should not be expected to increase across both forms. Future research with linked forms or items would allow analysis of growth using a consistent scale and consistent item formats. Although reader characteristics such as vocabulary size, prior knowledge of content contained in assessment tasks, prior knowledge of informational text structures, prior knowledge of text typographic features, reading speed, and ability to vary reading speed are known to influence literacy performance, the data collected here were not intended to examine the role of such features. Future research should examine the impact that individual student’s reading abilities may have on performance on these assessments.

Although locating information in prose text is considered a relatively easy task (Crooks & Flockton, 1998), the results here more clearly identify the architecture of student ability to locate information. Students are able to use typographic features and headings and titles to identify information related to verbatim, single-word search terms, even under considerable time pressure. However, extraction of information based on organizational features into graphic or outline note frames is a skill that only about half of students were able to exercise in test conditions. In addition, differences between the sexes and year levels were relatively trivial, though at the secondary level, differences between students in high and low SES background schools were substantial. It is unlikely that the achievement results reported here were due only to test artifact: Dreher (2001) has suggested that students do not actually apply the knowledge they have about using information access tools without careful scaffolding or prompting.

It has been suggested that comprehension and locating information are quite distinct aspects of reading (Dreher, 2001); however, these data suggest that there is a strong connection between being able to locate information and utilize text structure and general reading comprehension. These results suggest that achievement in both fields is significantly related and that improvement in one area may have a positive affect on achievement in the other. This is consistent with the general thrust of research (e.g., Dymock, 1998), which “suggests that almost any approach to teaching the structure of informational text improves both comprehension and recall of key text information” (Duke & Pearson, 2002, p. 217). In other words, improving students’ skills in locating information within prose text through a focus on organizational structure of texts, practice at creating search terms for implicitly stated tasks, locating material through inference, and extracting information into graphical and linear note-taking frames may increase general reading comprehension.

Although the texts used in these assessments were static, paper-based materials, the results may have some relevance to the use of the digital texts whose role and presence in our society is increasing. The model involves construction of appropriate search terms, which are essential to finding required information in such data-laden resources as CD-ROM encyclopedias or websites. The model also requires strategic use of various navigation tools and the identification and use of a text’s internal organizational and structural elements. The presence of digital search tools may actually encourage more hunt-and-peck strategies if reader thinking is not guided by clear construction of the search task. The presence of other digital text characteristics — for example, animation or hypertext branches — may actually distract readers rather than enable more rapid location and extraction of sought-after information.

On the positive side, if webpages are written as chunks designed for scanning rather than reading (as recommended by such sources as the Web Style Guide and Sun Microsystems’ Writing for the Web), the texts ought to be relatively short and highly focused on the content identified by the page heading. This may result in easier location of required information within the page, especially when combined with in-page search tools. It is expected that the assessment tools described here can contribute to students’ learning to search and use digital texts.

The primary objective of this research was to identify and describe the structure of students’ ability to search informational texts purposefully. However, this research has some implications for the professional development of teachers. Potential approaches and appropriate professional development materials are listed here as a tentative beginning point. Generally, the data suggest that professional development should include making teachers aware that student performance is not solely a function of prior knowledge of topics or of vocabulary. The ability to search informational texts, to recognize the organizational structure of text, or to construct appropriate search terms based on incorporation of various implicit elements are sophisticated skills that can be taught. See, for example, the professional development resources and suggestions in the following:

• Chapter 8 of Beth M. Arthur’s online text “Making a Difference — One Student at a Time”
• “Strategic Processing of Text,” an ERIC digest by Joanna P. Williams
• “Effective Practices for Developing Reading Comprehension,” a review of research by Nell Duke and David Pearson, published in the third edition of What Research Has to Say About Reading Instruction
• Edith A. Slaton’s conference paper, “Building Awareness of Text Structure Through Technology,” describing computer software (e.g., Timeliner, KidPix, and Kidspiration) that can assist students to understand the structure of informational text
• Kane’s “The view from the discourse level: Teaching relationships and text structure” The Reading Teacher (vol. 52, no. 2, pp. 182-184), which identifies texts for children that have specific informational structures
• Swafford, Akrofi, Rogers, and Alexander’s “Primary-grade students’ interactions with one another as they read and write informational texts” from the 48th Yearbook of the National Reading Conference, which describes how primary grade students can interact with one another in the context of reading and learning from informational texts
• An expository text structure glove is described by Newman (2002) in “Comprehension Strategy Gloves,” from The Reading Teacher (vol. 55, no. 4, pp. 329-332)
• Examples from various assessments (e.g., Finding Information in Prose Text, PISA, PIRLS), which can serve as teaching materials to extend to a new topic, text type, or text difficulties

The real challenge in searching for main ideas and details, whether within paper-based documents or digital environments, is not the location of verbatim terms, but lies rather in the incorporation of implicit terms that need to be inferred across passages of text. The results of this research suggest that if teachers draw student attention to and assist students’ in learning various navigation tools, appropriate strategies for searching and constructing search terms, using typographic features, and organizational cues and patterns, then student ability to locate information should improve. Naturally, the goal in students’ learning to search for main ideas and details by utilizing a variety of text and task characteristics is not simply rapid location of materials, even across distances and requiring inference. Rather, these skills are used to solve informational problems. Unless presented, developed, and exercised in such a context, they will remain rather pointless (for expansion on this issue see McKenzie, 1998, online document, posted at the author’s From Now On website). Helping students adopt a problem-solving approach to locating information and setting them tasks that require a diversity of text and task challenges seem essential elements of reading education.

Back to menu

References

Armbruster, B.B., & Armstrong, J.O. (1993). Locating information in text: A focus on children in the elementary grades. Contemporary Educational Psychology, 18, 139-161. Abstract retrieved August 2003, from http://dx.doi.org/10.1006/ceps.1993.1015
Back

Brown, G. (1997). Information skills in the New Zealand curriculum: A blueprint for education? Paper presented at the Annual Conference of the New Zealand Association for Research in Education, December 4-7, Auckland. (ERIC Document Reproduction Service No. ED 429 618) Abstract retrieved August 2003, from www.edrs.com/Webstore/Detail.asp?q1=@Meta_PubID%20445033&txtSort=Meta_DocID[D]&txtMaxdisplayed=10&txtDocType=ED
Back

Brown, G. (1999). Information literacy curriculum & assessment: Implications for schools from New Zealand. In J. Henri & K. Bonanno (Eds.), The information literate school community: Best practice (pp. 55-74). Wagga Wagga, Australia: Charles Sturt University, Centre for Information Studies.
Back

Brown, G. (2001). Locating categories & sources of information: How skilled are New Zealand children? School Library Media Research, 4.
Back

Chambliss, M.J. (1995). Text cues and strategies successful readers use to construct the gist of lengthy written arguments. Reading Research Quarterly, 30(2), 778-807.
Back

Croft, C., Dunn, K., & Brown, G. (2001). Essential skills assessments: Information skills (manual). Wellington, New Zealand: New Zealand Council for Educational Research.
Back

Crooks, T.J. (2002). Educational assessment in New Zealand schools. Assessment in Education: Principles, Policy & Practice, 9(2), 237-253. Abstract retrieved August 2003, from taylorandfrancis.metapress.com/app/home/contribution.asp?wasp=e2eygx3dwr6tuhb9kc3u&referrer=parent&backto=issue,6,9;journal,3,13;linkingpublicationresults,1,1
Back

Crooks, T.L., & Flockton, L. (1998). Information skills: Assessment results 1997 (National Education Monitoring Rep. No. 7). Dunedin, New Zealand: University of Otago, Educational Assessment Research Unit.
Back

Daniels, H. (2002). Expository text in literature circles. Voices From the Middle, 9(4), 7-14.
Back

Doyle, C.S. (1993). The Delphi method as a qualitative assessment tool for development of outcome measures for information literacy. School Library Media Annual, 11, 132-144.
Back

Dreher, M.J. (1992). Searching for information in textbooks. Journal of Reading, 35(5), 364-371.
Back

Dreher, M.J. (1993). Reading to locate information: Societal and educational perspectives. Contemporary Educational Psychology, 18, 129-138. Abstracted retrieved August 2003, from dx.doi.org/10.1006/ceps.1993.1014
Back

Dreher, M.J. (2001). Children searching and using information text: A critical part of comprehension. In C.C. Block & M. Pressley (Eds.), Comprehension instruction: Research-based best practices. New York: Guilford.
Back

Duke, N.K. (1999). The scarcity of informational texts in first grade (Report No. 1-007). Ann Arbor, MI: University of Michigan, School of Education, Center for the Improvement of Early Reading Achievement. Retrieved August 2003, from www.ciera.org/library/reports/inquiry-1/1-007/Report%201-007.html
Back

Duke, N.K., & Pearson, P.D. (2002). Effective practices for developing reading comprehension. In A.E. Farstrup & S.J. Samuels (Eds.), What research has to say about reading instruction (3rd ed., pp. 205-242). Newark, DE: International Reading Association.
Back

Dymock, S.J. (1998). A comparison study of the effects of text structure training, reading practice, and guided reading on reading comprehension. In T. Shanahan, F.V. Rodriguez-Brown, C. Worthman, J.C. Burnison, & A. Cheung (Eds.), 47th Yearbook of the National Reading Conference (pp. 90-102). Chicago, IL: National Reading Conference.
Back

Guthrie, J.T., & Kirsch, I.S. (1987). Distinctions between reading comprehension and locating information in text. Journal of Educational Psychology, 79(3) 220-227.
Back

Harley, J. (1996). Text design. In D. Jonassen (Ed.), Handbook of research for educational communications and technology (pp. 795-820). Mahwah, NJ: Erlbaum.
Back

Hattie, J. (1993). Measuring the effects of schooling. set: Research Information for Teachers, 2, 4.
Back

León, J.A., & Carretero, M. (1995). Intervention in comprehension and memory strategies: Knowledge and use of text structure. Learning and Instruction, 5, 203-220.
Back

Lorch, R.F., Jr., & Lorch, E.P. (1995). Effects of organizational signals on text-processing strategies. Journal of Educational Psychology, 87(4), 537-544.
Back

McKenzie, J. (1998). Grazing the Net: Raising a generation of free range students [special online of Phi Delta Kappan article, September 1998]. Retrieved August 2003, from www.fno.org/text/grazing.html
Back

Moore, P. (1995a). Information problem solving: A wider view of library skills. Contemporary Educational Psychology, 20, 1-31. Abstract retrieved August 2003, from dx.doi.org/10.1006/ceps.1995.1001
Back

Moore, P. (1995b). The dynamics of information problem solving. set: Research Information for Teachers, 2, 3.
Back

Moore, P. (1995c). Information literacy: Past approaches and present challenges. New Zealand Annual Review of Education, 5, 137-151.
Back

Moore, P., & St. George, A. (1989). The information quest: A look at children as information seekers. set: Research Information for Teachers, 2, 9.
Back

Moore, P., & St. George, A. (1991). Children as information seekers: The cognitive demands of books and library systems. School Library Media Quarterly, 19(3), 161-168.
Back

Newman, G. (2002). Comprehension strategy gloves. The Reading Teacher, 55(4), 329-332.
Back

O’Donnell, A. (1993). Searching for information in knowledge maps and texts. Contemporary Educational Psychology, 18, 222-239. Abstract retrieved August 2003, from dx.doi.org/10.1006/ceps.1993.1018
Back

Reid, A.N., & Elley, W.B. (1991). Progressive achievement tests of reading. Wellington, New Zealand: New Zealand Council for Educational Research.
Back

Stanovich, K.E. (1986). Matthew effects in reading: Some consequences of individual differences in the acquisition of literacy. Reading Research Quarterly, 16, 360-407.
Back

Street, C. (2002). Expository text and middle school students: Some lessons learned. Voices From the Middle; 9(4), 33-38.
Back

Waller, R. (1991). Typography and discourse. In R. Barr, M.L. Kamil, P. B. Mosenthal, & P.D. Pearson. (Eds.), Handbook of reading research (Vol. 2, pp. 341-380). New York: Longman.
Back

Wilkinson, L., & Task Force on Statistical Inference. (1999). Statistical methods in psychology journals: guidelines and explanations. American Psychologist, 54(8), 594-604. Retrieved August 2003, from www.apa.org/journals/amp/amp548594.html
Back

Williams, J. (2000). Strategic processing of text: Improving reading comprehension for students with learning eisabilities (ERIC/OSEP Digest No. E599). Arlington, VA: ERIC Clearinghouse on Disabilities and Gifted Education.
Back

About the Author

Gavin Brown, senior project manager for the Assessment Tools for Teaching and Learning Project at the University of Auckland, is a former secondary school English literature and language teacher. Although he trained in Montreal, Gavin has done all his teaching in New Zealand. After leaving teaching, Gavin worked as an assessment researcher at the New Zealand Council for Educational Research, where he was involved with the Assessment Resource Bank and the Essential Skills Assessments: Information Skills project from which the research presented here is drawn. Gavin’s doctoral thesis is on primary school teachers’ conceptions of assessment and how those beliefs relate to teachers’ understanding of learning, teaching, curriculum, and efficacy. His research interests include large-scale testing, literacy, and measurement. Contact him by e-mail at gt.brown@auckland.ac.nz.

Back to top

Note: This research was carried out with Purchase Agreement funding from the New Zealand Ministry of Education while the author was at the New Zealand Council for Educational Research (NZCER). The author acknowledges the leadership of Cedric Croft and the work of Karyn Dunn, both of NZCER, without whose contribution this work would not have been completed. An earlier version of this article was presented with Karyn Dunn at the International Reading Association’s 18th World Congress on Reading, held in Auckland, July 2000. The author also acknowledges with appreciation the encouragement and insight of Professor Jean Dreher, and the comments and suggestions of the editors and anonymous reviewers, which were most helpful in revising this article.

For a printer-ready version of this article, click here.

Citation: Brown, G.T.L. (2003, September/October). Searching informational texts: Text and task characteristics that affect performance. Reading Online, 7(2). Available: http://www.readingonline.org/articles/art_index.asp?HREF=brown/index.html

Reading Online, www.readingonline.org
Posted September 2003
© 2003 International Reading Association, Inc.   ISSN 1096-1232