The Impact of Intermediate Students’ Awareness and Use of Text Structure in Comprehending Expository Text and Its Implications

This comprehensive review of literature includes research studies dating from 1980 to 2005 on expository text structure awareness as well as studies on the effectiveness of expository text structure instruction. The purpose of this literature review is two-fold: 1) To determine the extent to which students’ expository text structure awareness facilitates reading comprehension; and 2) Whether it would be beneficial for teachers to provide explicit instruction on expository text structures. The review also includes suggestions for further research, including a detailed plan for a case study designed to fill in some of the gaps in existing research. Expository Text Structure 4 Chapter I: Introduction Interest in expository text structure awareness research emerged in 1975 with research conducted by Meyer and colleagues that established a relationship between text structure knowledge and amount of recall (as cited in Bellows, 1994). In the early and mid-1980s, several researchers studied the effects of expository text structure instruction on reading comprehension in the middle grades (Armbruster, Anderson, & Ostertag, 1987; Berkowitz, 1986; Taylor, 1982). Some researchers in the early to late 1980s also studied middle grade students’ awareness of expository text structure through a variety of methods (Englert & Hiebert, 1984; Garner & Gillingham, 1987; McGee, 1982; Richgels, et al, 1987; Smith & Hahn, 1989). In the early 1990s; however, interest in the topic diminished and little was done to apply the findings of this research in the classroom (Williams, Hall, & Lauer, 2004). In recent years, however, researchers’ interest in the explicit instruction of expository text structures has re-emerged. In 2004, researchers Hall and Williams began conducting studies about teaching expository text structure to second grade students. This research was prompted by the fact that students are exposed to very little expository text in the primary grades, and therefore students are not prepared to comprehend expository text in the middle grades. These researchers want to determine if it would be beneficial to teach students about text structure in early elementary school so that students will be better prepared to tackle expository text in middle school. A number of previous research studies have shown that explicit instruction of expository text structures improves middle school students’ reading comprehension (Armbruster, Anderson & Ostertag, 1987; Berkowitz, 1986; Englert & Hiebert, 1984; Taylor, 1982; Williams, et al, 2005). Pearson & Fielding (1991) noted that nearly any form of instruction that teaches students to Expository Text Structure 5 recognize text structure enhances comprehension (as cited in Almasi, 2003). Although the instruction of reading comprehension strategies (including instruction on expository text structures) at the middle school level has been studied a great deal by researchers, little has been done to determine whether the knowledge from this research has been applied in the classroom (Block & Pressley, 2002). Several researchers have found that awareness of text structures is highly related to reading comprehension (Dickson, Simmons, & Kameenui, 2001). According to Hall et al (2005), readers who understand how a text is structured “find greater success in identifying important information and relationships between ideas” (215). There is disagreement among theorists and researchers regarding the instruction of reading comprehension strategies, such as the explicit instruction of expository text structures. On one side of the argument are those who believe that the ability to comprehend cannot be directly taught. Some researchers suggest that expository text structure awareness is developmental, which means that as students become older and gain experience reading expository text, their awareness of expository text structures will increase (Englert & Thomas, 1987, as cited in Dickson, Simmons & Kameenui, 2001; McGee, 1982; Taylor, 1980, as cited in Bellows, 1994). Other researchers have suggested that text comprehension is dependent upon the reader’s prior knowledge, as research conducted in the late 1970s and early 1980s revealed a strong relationship between prior knowledge and reading comprehension (Beck et al. 1982; Hanson & Pearson, 1983, as cited in Fielding & Pearson, 1994). According to Willingham (2010), “Prior knowledge is vital to comprehension because writers omit information...A writer who doesn’t assume some prior knowledge on the part of her readers will write very boring prose” (para 5 and 6). Expository Text Structure 6 While some researchers believe it is important to provide students time and experience reading a variety of expository texts or to help build background knowledge, other researchers believe that explicit strategy instruction improves comprehension (Allington, 2006). For example, Janice Almasi, one who advocates providing explicit strategy instruction, suggested it is difficult for students to “discover” comprehension strategies on their own (2003). Pearson and Gallagher (1983) noted that while students may improve in their reading comprehension ability over time in the absence of any strategy training, instruction in comprehension strategies may accelerate students’ growth in reading comprehension. Statement of Problem Many middle school students struggle with reading comprehension of expository text. In my own teaching and tutoring experiences with students in middle school, I noticed that many students have difficulty comprehending expository text across content areas. At this level, the ability to comprehend expository text is imperative because content-area reading textbooks comprise a majority of the assigned reading. At the early elementary level, the emphasis of reading instruction is on narrative text, or stories. When students enter the upper elementary grades, however, reading instruction begins to shift to expository (informational) texts. By the time students reach middle school, the majority of their reading is comprised of expository text. Middle school students are surprised by the reading expectations set for them in content-area classes because they have not been taught how to read expository texts (Street, 2002). When students are asked about the strategies they use to read textbooks, they “look at us blankly” (Street, 2002, p. 35). This major shift in reading instruction causes reading comprehension difficulty for some students due to the fact that many underlying structural differences exist between narrative and expository text. Expository Text Structure 7 Research Questions Through a review of research literature, I initially sought to answer the following question: Does expository text structure awareness among middle school students facilitate reading comprehension? Once it became clear that much of the existing research suggests that expository text structure awareness facilitates comprehension, I sought to answer the following: To what extent (if at all) should middle school teachers provide explicit instruction on expository text structures to students? Definition of Terms Jean Piaget’s constructivist learning theory (1969) is one theoretical construct for this review because the awareness and use of text structure awareness involves using prior knowledge (schemata) of text structures when interacting with an expository text in order to improve reading comprehension. If what the student reads does not fit into an existing schema, the student must alter their schema or create a new schema to construct meaning (Tompkins, 2002). Using text structure awareness and prior knowledge help students construct new meaning. Interactive learning theories also guide my study of expository text structure awareness because these theories describe what readers do as they read and emphasize that readers use strategies to understand what they read (Tompkins, 2002). The metacognitive approach is an additional construct guiding my study. I define metacognition as “thinking about your thinking.” A metacognitive approach emphasizes awareness of text and provides students with a “game plan” for constructing meaning (Mueller, 1996). This approach allows students to become more independent readers. A metacognitive approach to reading comprehension instruction is one which “trains students to apply strategy to expository text as a means of determining order within text and hence, meaning” (Mueller, 1996, Expository Text Structure 8 p. 44). Awareness and use of expository text structure to comprehend expository text is a metacognitive strategy. Students’ difficulty with expository text may be attributed, in part, to students’ lack of knowledge about expository text structures. For the purposes of this study, it is important to define text structure as an “organizational pattern of text written to convey a purpose” (Dickson, Simmons, & Kameenui, 2001, p. 4). Texts can be structured in order to persuade, compare/contrast, describe, entertain, etc. The focus of my study is on expository text structures, which are text structures typically found in informational texts. The main purpose of expository text is to inform and the main purpose of narrative text is to entertain. Because of these differing purposes, narrative and expository texts are written differently and contain different underlying organizational structures. Narrative text contains one main structural pattern, sometimes referred to as a story grammar, which involves an understanding of setting, characters, problem, solution, and resolution in a story (Harp & Brewer, 2005). The story grammar provides a framework of understanding for narrative text. Unlike narrative texts, expository texts often contain multiple text structures. The following five structures are commonly seen in expository texts: description, sequence, cause/effect, compare/contrast, and problem/solution (Hall, et al, 2005). It is important to note that authors of expository text material often use several different text structures within the same text (Harp & Brewer, 2005). While some students adjust to the changes in text structures, others do not. It is not just struggling readers who experience difficulty understanding expository text. Expository Text Structure 9 Chapter II: Review of Literature My literature review is divided into two sections: Part A and Part B. Part A of my literature review will attempt to answer my first research question: Does expository text structure awareness among middle school students facilitate reading comprehension? Part A of my review focuses on expository text structure awareness research studies conducted by various researchers from the 1980s to the present day. The studies involve students’ text structure awareness from the middle school to the high school level and employ a variety of measures. In addition to these studies, several other researchers in the 1980s and early 1990s studied text structure awareness. The three major classifications that I have selected to organize the expository text awareness research studies are as follows: Good and Poor Readers’ Awareness of Expository Text Structures, A Comparison of Students’ Awareness of Expository Text Structures across Grade Levels, and Expository Text Structure Awareness of Students within One Grade Level or Range. After examining text structure awareness and determining whether it facilitates reading comprehension, I explore my second research question: To what extent should middle teachers provide explicit instruction on expository text structures to students? In Part B, I examine research studies that involve explicit text structure instruction and whether it improves students’ reading comprehension. The various studies involved student subjects ranging in age from grade two through twelve, with a majority of the studies including subjects in the middle grades. Part B of my literature review would be of special interest to educators because the research involves employing text structure instructional practices to improve reading comprehension. Teachers may wish to consider employing some of the strategies used in the studies. Expository Text Structure 10 Literature Review: PART A Good and Poor Readers’ Awareness of Expository Text Structures Meyer, Brandt, and Bluth (1980) conducted a research study involving 102 ninth grade students’ use of top-level structure in recall of expository text. The text passages used contained problem/solution and comparison/contrast structures. These researchers compared the recall performance of good, average, and poor readers. The data in this study showed a high correlation (p < .0001) between using top-level text structure in organizing text recall and comprehension skills in terms of total recall, central idea score, and weighted recall for the two passages used in the study. The results of the study showed that good readers are more proficient in using the author’s text structure in organizing free recalls than were poor readers (Meyer, Brandt, & Bluth, 1980). A study conducted by McGee (1982) involved a total of sixty good and poor readers from fifth grade and good readers from third grade. The purpose of this study was to determine whether good and poor readers are aware of text structure and whether text structure awareness influences oral recall. Two 125-word expository passages containing description structures were used in this study and students’ recall was tape-recorded. The students’ recorded retellings of passages were analyzed to determine whether the student used the author’s text structure and whether the retelling contained a proportionate number of superordinate and subordinate ideas. Recalls containing three idea units from the top level structure of the text and two clusters of recall on the second and third levels of the text structure hierarchy were considered to have full text structure. The results of the study showed that fifth-grade good readers are more aware of text structure than fifth-grade poor readers or third-grade good readers. After a 3 (reading ability/grade) X 2 (level of importance of idea) mixed analysis was conducted, the results Expository Text Structure 11 indicate that the retellings of the fifth-grade good readers contained proportionately more superordinate ideas than the retellings of the fifth-grade poor readers or third-grade good readers (p < .05). However, the fifth-grade poor readers displayed some awareness of text structure and their retellings contained proportionately more superordinate ideas than the third-grade good readers (p < .05). A chi-square analysis of the data was also completed and the results suggest that most third-grade good readers lack awareness of text structure, most fifth-grade poor readers had partial text structure in their recall, and most fifth-grade good readers had full text structure in their recall (p <.001). Kletzion’s 1992 study, involving tenth and eleventh-grade students, contrasts with the results obtained in the studies conducted by Meyer, Brandt, and Bluth (1980) and McGee (1982). Kletzion’s research design included conducting interviews of students as they completed cloze passages (as cited in Bellows, 1994). Kletzion (1992) found that whether readers were proficient or less proficient did not appear to affect their use of text structures (as cited in Bellows, 1994). In addition, the results of a text structure awareness study conducted by Ghaith and Harkouss (2003) indicated no significant differences between proficient and less proficient readers in terms of text structure awareness. (This study is discussed further in the “Expository Text Structure Awareness of Students within One Grade Level” category of the literature review because reading ability was not the main focus of this study.) Comparison of Students’ Awareness of Expository Text Structures Between Grade Levels In McGee’s 1982 study on awareness of text structure (discussed in the previous section), she compared a total of sixty good and poor readers in both third and fifth grade. McGee found that fifth-grade good readers displayed a better awareness of text structure than fifth-grade poor readers and third-grade good readers. However, in terms of grade level differences, fifth-grade Expository Text Structure 12 poor readers displayed some awareness of text structure, while third-grade good readers did not display an awareness of text structure in their retellings and recalled more subordinate ideas than superordinate ideas (indicated lack of text structure awareness). After a chi-square analysis of the data, McGee hypothesized that the reason most of the third-grade good readers did not use text structure in their recalls was because they had less experience reading expository text than the fifth-grade good and poor readers. McGee also noted that the text difficulty level may have been a factor influencing the results, and the text may have been easy enough for the fifth-grade good readers that they were able to attend to the text structure more than the third-grade good readers. Englert and Hiebert (1984) examined the effects of four types of expository text (sequence, comparison/contrast, description, and enumeration) on the comprehension performance of third and sixth-grade students of three ability levels: high, medium, and low reading ability. The study employed 76 third-grade students and 70 sixth-grade students. The students were provided with two topic sentences for each text structure and then were given several more sentences in which they were asked to rate the degree to which those sentences belong in the paragraph using the following scale: “(a) YES! The sentence definitely belonged in the paragraph, (b) yes?the sentence ‘sort of’ belonged in the paragraph, (c) no?—the sentence ‘sort of’ did not belong in the paragraph, and (d), NO!—the sentence definitely did not belong in that paragraph (Englert & Hiebert, 1984, p. 68). Some of the sentences given for each test item were “distractors”, meaning they did not belong in the paragraph because they did not follow the text structure of the rest of the paragraph. Other sentences were labeled as “targets,” meaning that those sentences did follow the text structure of the paragraph. To score the tests, each response was given point value from 1 to 4 depending on the accuracy of the student’s rating (Englert & Expository Text Structure 13 Hiebert, 1984). A multivariate analysis of the data was conducted using a 2 (grade) X 3 (ability: high, medium, low) X 4 (text: sequence, enumeration, comparison/contrast, description) design. The results of this study indicate that grade level was highly related to the performance on the text structure measure. Sixth-grade students of all ability levels were significantly better at recognizing “distractors” than third-grade students (p < .0001). In terms of reading ability, the results were statistically significant for both target sentences (p < .001) and distractor sentences (p < .0001). The findings of this study suggest that good readers are more sensitive to text structure than poor readers and older readers are more aware of text structure than younger readers. Smith and Hahn (1989) conducted a similar study, examining students’ sensitivity to four text structures: enumeration, comparison/contrast, description, and sequence. Two measures were used in the study to assess students’ ability to detect inconsistent text structure information: oral retell of the paragraph and response to the test administrator whether the paragraph was O.K. or if there was a problem with it (i.e. it contained intrusive information). The study involved a total of 48 students in grades 4, 6, and 8 (16 student per grade level) of average reading ability. Qualitative and quantitative data analyses were performed. In terms of qualitative analysis, oral recalls were scored on the degree to which the recall followed the text’s structure and whether the recall contained the intrusive information. Results from the qualitative analysis revealed that more eighth-grade students followed the sequence and compare/contrast text structures in their oral recalls than the fourth and sixth-grade students. All students in grades four, six, and eight followed the enumeration and description text structures in their oral recalls. While all students followed the description text structure in their oral recalls, nearly equal numbers of students included the intrusive information in their recalls, which indicates a lower awareness of the Expository Text Structure 14 description text structure. A quantitative analysis using univariate F ratios was used to assess the students’ recognition of intrusive text information for the three text types across and within grade levels. The analysis across grade levels only showed a significant difference for the description text structure (p < .05). Post hoc comparisons across grade levels showed that sixth-graders performed better than fourth-graders, but not eighth graders at identifying the intrusive information contained in a paragraph. The findings in the study conducted by Garner and Gillingham (1987) contrasts with the findings of the other studies in this category of expository text structure research. This study involved 15 students in fifth-grade and 15 students in seventh-grade of below average academic ability. The students were asked to read seven sentences on a computer screen, and then use those seven sentences to construct a good paragraph and a bad paragraph and then report what makes a good paragraph. The students’ constructed paragraphs and verbal reports were scored on topical relatedness (meaning that the parts belong together), superordination (relation of main idea to details), and cohesion (the extent to which the sentences are tied together). The researchers examined Univariate F ratios for each of nine dependent measures (topic relatedness, superordination, and cohesion of the good paragraph; awareness of topic relatedness, superordination, and cohesion, and verbal report of topic relatedness, superordination, and cohesion. Only three of the nine ratios yielded significant results: topic relatedness awareness (p < .05), topic relatedness verbal report (p < .01), and superordination verbal report, (p < .01). In each case, the performance of the seventh-grade students was better than that of the fifth-grade students. The researchers also calculated Pearson correlation coefficients to see if there was a relationship between student performance on the good paragraph and the verbal report. The coefficients were all low and insignificant, except for the cohesion property, which the majority Expository Text Structure 15 of students did not even demonstrate in the paragraphs or mention in the verbal reports. After analysis of the paragraphs and verbal reports, Garner and Gillingham found no significant grade level differences in this study. Based on these results, Garner and Gillingham concluded that students in upper-elementary grades and secondary grades are not able to attend to structural properties of text very well, and they recommend explicit instruction in the use of text structure at these grade levels. Expository Text Structure Awareness of Students within One Grade Level/Range Taylor and Samuels (1983) conducted a study involving 14 fifth-grade students and 36 sixthgrade students from an urban elementary school. All students in this study were either at or above grade level in reading ability. The researchers constructed three 100-word expository passages and two 400-word expository passages written at a fifth-grade level. Students were asked to read two of the organized versions 100-word passage and one of the organized 400word passages. Each student also read one scrambled version of a different 100-word passage and a 400-word passage. (In the scrambled versions, the sentences were placed out of order in the 100-word passages and the paragraphs were placed out of order in the 400-word passages.) Students were asked to write down everything they recalled from the passages. Students written recalls of the passages were scored based on the number of idea units recalled and classified according to whether or not they followed the organization of the passage (for the organized passages only). Students were deemed aware of text structure if two out of three of the written recalls of the organized passages followed the organization of the passage. Researchers classified fourteen students as aware of text structure and 36 students as lacking awareness of text structure based on the scoring criteria. A regression approach to the analysis of variance was used to analyze the data. For the short passages, there was a statistically significant main Expository Text Structure 16 effect for reader awareness or no awareness of text structure (p < .05), for normal vs. scrambled text versions (p < .01), and for reader and text type interaction (p < .05). Data analysis on the longer passages yielded similar results, with significant main effects for reader awareness or no awareness of text structure (p < .001), normal vs. scrambled text (p < .01), and for reader and text type interaction (p < .05). The findings suggest that students aware of text structure have better recall on organized texts than those who are unaware of structure. When comparing students’ written recall of the normal and scrambled passages, researchers found that the students unaware of text structure scored about the same on the organized passages and scrambled passages, and students aware of text structure scored better on organized passages than on the scrambled passages. When comparing student performance on the scrambled passages, there was no significant difference found between students aware of text structure and students unaware of text structure. However, when recalling ideas from an unorganized text, both groups performed similarly. Richgels, McGee, Lomax, and Sheard (1987) examined sixth-grade students’ awareness of the following four expository text structures: collection, comparison/contrast, causation, and problem/solution. The study also examined students’ recall of texts written in those structures. Three measures of awareness were used: Use of organization in written recall, use of organization in composition, and response to an interview. The subjects in the study included 56 sixth-grade students, of which 30 were randomly selected to participate in the interview portion of the study. A multivariate analysis of variance (MANOVA) was used on the recall data. Statistically significant results were found for the passage type (p < .001), level of main ideas and details recalled (p < .001), and for passage type x level of main ideas and details (p < .001). Chi-square tests of independence were used for the measures of organization of recall (p < .001), Expository Text Structure 17 organization of composition (p < .001), and the interview (p < .001). The researchers discovered that the interview measure gave them the most valuable information about students’ awareness of expository text structure because students needed to be metacognitively aware of text structure and provide reasons and examples to defend the text structure identified. The authors found that students generally had a high awareness of the comparison/contrast structure and a low awareness of the causation structure, as found in previous studies. In terms of text structure awareness, the authors concluded that sixth-grade students showed varying degrees of text structure awareness, but the recall of ideas was stronger in students who were aware of text structure. Ghaith and Harkouss (2003) studied the role of text structure awareness in the recall of four expository text structures: description/collection, comparison/contrast, cause/effect, and problem/solution. The participants in the study included 109 students of English as a Foreign Language that were university-bound. Data collection and analysis focused on students’ text structure awareness and recall of main ideas and details from texts organized into the four different text structures under investigation. Text structure awareness was measured on a numerical scale from 0 to 7, which indicated the degree to which the organization of a written recall resembled the original passage. A score of 0= use of another structure, 1-3 = no use of structure, 4-5 = partial use of structure, and 6-7 = full use of structure. The recalls of main ideas and details were scored by calculating percentage of ideas in the written recalls as compared to the number ideas contained in the original passage. Descriptive statistics were used to calculate frequencies for ratings in the use of full structures in written recall. The Pearson Product Moment Correlation test was conducted to determine the relationship between the recall of ideas and awareness of text structures. An Analysis of Variance (ANOVA) test was also performed to Expository Text Structure 18 determine the differences between good and poor readers’ awareness of text structure. The results of the descriptive statistical analysis indicate that the students were most aware of the comparison/contrast structure and least aware of the problem/solution structure. The results of the Pearson Product Moment Correlation test showed that text structure awareness was positively related to recall of information for all text structures. The r value for each text structure type was significant at the p < .01 level. The r-values are as follows: description/collection, .50; comparison/contrast, .43; cause/effect, .55; problem/solution, .45. The ANOVA results showed no significant differences between proficient and less proficient readers in text structure awareness (p > .05 for all text structure types). Literature Review: PART B Expository Text Structure Instruction In this part of my review, I examine several research studies involving explicit expository text structure instruction and measuring the impact of such instruction on students’ reading comprehension. In Part A of my literature review, most of the research studies suggested expository text structure awareness facilitates reading comprehension to some degree. I now find it necessary to answer my second research question, which was contingent upon answering “yes” to my first research question: To what extent (if at all) should middle school teachers provide explicit instruction on expository text structures to students? Taylor (1982) studied the effects of text structure instruction on the comprehension of fifthgrade students. This study involved 48 fifth-grade students from two classes. Students were categorized as competent readers or less competent readers. Each student was then randomly assigned to the experimental instruction group or the traditional instructional group. The Expository Text Structure 19 experimental instruction group received instruction in completing a hierarchical summary after reading, while the traditional group answered questions after reading. Students were then administered a recall protocol and scored based from 1-5, where 1 indicates the student did not follow the text structure in the recall and 5 indicates the student followed the text structure. Separate ANOVAs were calculated on the recall data, organization data and short-answer scores for the health reading passage and social studies reading passage. The results of the ANOVA of the health passage recall data show a significant effect for treatment (p < .001) and reading ability (p < .001). The ANOVA of the health passage organization data showed a significant effect for treatment (p < .05) and a significant effect for reading ability (p < .01). However, the ANOVAs of the health passage recall data and organization data showed no significant effect for treatment x reading ability. The results of the analysis of short-answer scores from the health reading passage showed a significant main effect for reading ability (p < .001) and a significant main effect for question type (p < .001), but no significant main effect for treatment. The ANOVAs conducted on the data obtained from the social studies passage yielded similar results, except there was no significant effect of treatment found after performing and ANOVA on the short-answer scores of the social studies passage. Thus, Taylor concluded that the hierarchical summarization strategy did enhance students’ recall and organization scores. Taylor then attempted to replicate the experiment using a different group of students within the same school. However, statistical analyses of scores in the second experiment indicated that the hierarchical summarization strategy did not enhance recall and organization scores. After an in-depth investigation as to why the hierarchical summarization strategy did not enhance scores in Experiment 2, Taylor noted that the difference in the quality of instruction, student inattentiveness, or student inability to master the strategy may have affected the results in Expository Text Structure 20 Experiment 2. Due to the conflicting results obtained in Experiment 1 and 2, one can conclude that expository text structure instruction may or may not be helpful in facilitating students’ comprehension. A study conducted by Taylor and Beach (1984) involving 114 seventh-grade students is similar in design Taylor’s previous study conducted with fifth-grade students described above. This study involved assigning students to either an experimental group that received instruction in creating a hierarchical summary or a traditional group answering questions after reading social studies material. However, this study utilized a pre-test/post-test design, where students were required to organize and write an essay before receiving experimental or traditional instruction and then after receiving the instruction. A recall protocol was used to score the essays based on a 4-point rating scale. Multiple ANOVAs were performed on the data obtained from the tests, including the recall scores, short-answer scores, and writing scores. A 3 x 2 x 2 ANOVA was performed on recall scores. Results revealed significant main effects for group (p < .05), passage (p < .05), and test time (p < .001). Significant group by passage interaction (p < .001), group by test time interaction (p < .01), and passage by test time interaction (p < .05), and group by passage by test time interactions (p < .01) were found. A 3 x 2 x 2 ANOVA was also performed on students short answer test scores. The analysis revealed a significant main effect for group (p < .001), test time (p < .001), group by test time (p < .01), and passage by test time (p < .01). No other significant effects were found. A 3 x 2 repeated measures ANOVA was performed on the writing scores. The analysis revealed a significant main effect for test time (p < .001) and group by test time interaction (p < .001). There was no significant effect for group. After all the ANOVAs were performed on the data, the researchers determined that the instruction provided to the experimental group was effective in enhancing recall of unfamiliar material. The post-test Expository Text Structure 21 scores of the experimental group were higher than the traditional group. The experimental group had significantly higher recall scores and post-test writing scores than the control group. The short-answer test analysis revealed that the experimental and conventional groups scored higher than the control group. Overall, the results of this study suggest that text structure instruction is effective in enhancing recall of content area materials. Berkowitz (1986) tested two instructional strategies called map-construction and map-study procedures. These strategies involve using the organization of ideas in expository text as a method for studying. The sample consisted of 99 sixth-grade students in four classes in a suburban school. Berkowitz randomly assigned one classroom to the test map-construction strategy, one classroom to test the map-study strategy and the other two classrooms traditional study procedures (answering questions and rereading). The materials used in the study were nine 600to 1,000word passages from a sixth-grade social studies book. Recall protocols and shortanswer tests were used to measure learning. The recall protocols were separated into idea units and compared to the idea units contained in the original passage. The short answer tests contained comprehension questions related to the passages. Students were subjected to four testing sessions, two of which occurred immediately after six sessions of instruction (Immediate Test 1 and Immediate Test 2), another test which occurred two weeks after treatment (Delayed Test), and a Transfer Test. The Transfer Test was designed to get students to think about the strategies they studied. Multiple analyses of variance (MANOVAs) were performed on the test measures. Results showed that there was a significant effect for treatment (p < .01). Results also revealed significant effects for reading ability (p < .01 for Immediate Tests 1 and 2, and the Delayed Test; p < .05 for the Transfer Test). Tukey’s post hoc tests showed that the mapconstruction group scored significantly higher than the other groups on Immediate Test 2 (p < Expository Text Structure 22 .01). Tukey’s post hoc tests also showed that the question-answering group gave more correct answers on the Immediate Test 1 (p < .01) than the map-study group, and the question-answering group and map-construction groups performed better than the other groups on the Immediate Test 2 (p < .01). The results suggested that the map-study procedure did not assist students any better with recall than any of the other strategies. The map-construction group, however, performed much better than the other groups on one passage, which Berkowitz noted that some passages may be better suited for map-construction than others. The type of text structure that should be taught to students to foster recall of ideas may be dependent upon the characteristics of a reading passage. Berkowitz recommended that teachers provide direct instruction in using the author’s organization of ideas to create a map for study. Armbruster, Anderson, and Ostertag (1987) studied fifth-grade students to determine if direct instruction in a problem-solution text structure would improve students’ ability to comprehend other texts containing that structure. The sample consisted of 82 fifth-grade students from four classrooms in two schools in a small city. One class in each of the two schools was assigned either to the structure training group or the traditional group (answering questions and discussing), so that one school had one structure training classroom and one traditional instruction classroom. All students reading below grade level were eliminated from the study. The materials used included workbooks designed for the text structure groups and workbooks designed for the traditional groups. Both versions of the workbook contained problem/solution social studies passages, but the text structure workbook contained specific information on the problem/solution text structure. The traditional booklet just had comprehension questions listed at the end of each passage. The duration of instructional training was eleven days. The assessments included a test containing passage and essay to write in response to the passage, a Expository Text Structure 23 10-question short answer test containing comprehension questions for the passage, a summary writing exercise, and an additional essay test about a passage located in the regular social studies text. Several forms of mixed analyses of variance were performed on the data. An analysis of the essay test results showed that the structure training group cored higher than the traditional training group (p < .01). Data analyses showed that the structure training groups recalled more information on the essay test on the main idea of a problem/solution passage. The structure training groups also wrote summaries that included more main ideas than the traditional groups, and wrote summaries that contained superior organization. As the researchers expected, there was no difference in performance on the short answer test because the questions did not require students to use text structure awareness to answer the questions. Rather, the questions required students to use content knowledge. Finally, the data indicated that using the problem/solution text structure as a skeleton for class discussion did not facilitate recall of the content in the regular social studies text, which is contrary to the rest of the findings. In a 1994 study, Bellows investigated whether providing explicit text structure instruction to remedial adolescent readers would be beneficial. Bellows used a case study approach to understand how text structure knowledge impacts comprehension. The students, Duncan (Grade 7), Anne (Grade 12), and Marsha (Grade 9), were used as subjects in this case study. Bellows provided explicit text structure instruction (description, cause/effect, problem/solution, and compare/contrast) to each student. Based on the work completed with these three students, Bellows concluded that fluent readers with good comprehension skills do not need explicit text structure instruction and that they don’t necessarily need to label the particular text structure to comprehend the text. Bellows stated that when working with Anne, her comprehension was not affected either positively or negatively by text structure awareness. Bellows also suggested that Expository Text Structure 24 the knowledge of text structures may interfere in the reading process for students who are not fluent readers with good comprehension skills because his sessions with Martha and Anne showed that text structure knowledge did not improve either of these students’ comprehension. Bellows reminds the reader that many variables exist in comprehending text, and that prior knowledge may be a bigger factor in comprehension than text structure knowledge. Although the focus of my research is on the middle school population, text structure research in recent years focuses on early elementary students. Recent research will shed more light on the effectiveness of text structure instruction, particularly the grade levels in which text structure instruction would be most appropriate. A 2005 study conducted by Williams, et al. tested a compare/contrast text structure instructional program for second graders. The study included a sample of 128 second-grade students from three elementary schools in a metropolitan area. Ten second-grade teachers volunteered to participate in the study. A pre-test/post-test research design was used. The classrooms were randomly assigned either to one of the following conditions: Text structure instruction, content only instruction, or no instruction. The materials used in the text structure groups and content groups an animal encyclopedia, trade books, and nine compare-contrast paragraphs written for the program. The text structure groups focused on the structure of the text, while the content group focused on the facts about animals. The test measures were categorized as follows: Strategy measures, Structure Outcome measures, and Content Outcome measures. Statistical analyses of variance were performed on the data using a classroom as a unit. The results of the analyses of variance on strategy measures (recall of clue words) showed an overall effect for treatment (p = .009). Comparisons using the Scheffe pairwise comparisons show that the text structure group scored significantly higher than the content only group and the no Expository Text Structure 25 instruction group on four out of five Strategy Outcome measures. The analysis of the Structure Outcome measures (written paragraph summaries and oral responses) indicates an overall effect of treatment on three out of five measures. The first structural outcome measure requiring a written response to assess the effects of explicit teaching showed no significant main effect for treatment, but the text structure group tended to score higher on the assessment (p = .094). The next three structural outcome measures required oral responses and each measured transfer of knowledge. There was a significant main effect of treatment for two of the three transfer measures, including Immediate Transfer (p = .014) and Near Transfer (p = .043). There was no significant main effect for treatment on the Far Transfer Measure, but there was a tendency for the text structure group to score higher than the other two groups (p = .068). The Structure Transfer (the final structure outcome measure that assed the ability to transfer the instructed strategies to a different text structure) data analysis showed no significant main effect for treatment. The Content Outcome measure data analysis showed significant main effects for treatment for the vocabulary concepts measure (p = .003), but no significant treatment effect for the detail questions measure. Williams, et al. concluded the following about the results obtained in the study: Text structure program students not only learned what they were taught but were also able to demonstrate transfer of what they had learned to content beyond that used in instruction....This suggests that were not merely teaching them the content of the instructional program but also how to process a particular type of expository text (p. 546). Overall, the findings in this study suggest it is beneficial to teach text structure to students as early as second grade. Expository Text Structure 26 Part B of this literature review explored whether teaching students text structure strategies improves comprehension. Some of the studies showed that text structure instruction was effective in facilitating comprehension, while other studies obtained mixed results. The next chapter analyzes the findings from these studies in more detail in an attempt to generalize the results and make recommendations to educators. Expository Text Structure 27 Chapter III: Results and Analysis Relative to the Problem The existing research on expository text structure awareness at the middle school level is inconsistent. However, the majority of the studies showed at least some degree of relationship between text structure awareness and recall of ideas. In the research studies including good and poor readers’ awareness of text structures, some researchers found that good readers displayed more awareness of expository text structures than poor readers (Meyer, Brandt & Bluth, 1980; McGee, 1982). By contrast, other researchers found no difference between good and poor readers in their awareness of expository text structure (Kletzion, 1992, as cited in Bellows, 1994; Ghaith & Harkouss, 2003). The findings from studies comparing students’ awareness of expository text structures between two or more grade levels are also contradictory. Some researchers concluded that students in older grade levels generally have a greater awareness of text structure than the students in lower grades (McGee, 1982; Englert & Hiebert, 1984; Smith & Hahn, 1989). Garner and Gillingham (1987), however, found no significant grade level differences. The research findings on text structure awareness among students in the same grade level were more consistent. Both Richgels, et al. (1987) and Ghaith and Harkouss (2003) found that students were least aware of the causation structure and the most aware of the comparison/contrast text structure. Richgels, et al. (1987) concluded that sixth graders vary in their level of text structure awareness, but a higher awareness of text structure was linked to greater recall of ideas. Ghaith and Harkouss (2003) concluded that there is a positive relationship between expository text structure awareness and recall. Expository Text Structure 28 Due to the large inconsistencies in previous research studies on good and poor readers’ awareness of expository text structures, I plan to further examine the relationship between reading ability and expository text structure awareness in a future study. Because most of the text structure awareness studies indicated a link between text structure awareness and comprehension, I found it necessary to continue my review of literature to determine whether explicit expository text structure instruction facilitates comprehension. The results of the studies on the impact of expository text structure instruction were mixed, leaving my second research question largely unanswered. The results of only two studies consistently showed that providing students explicit instruction on text structures helps students recall more information from texts (Taylor & Beach, 1984; Williams, et al., 2005). The other studies I reviewed revealed mixed results, meaning that some of the findings within one study indicated the effectiveness of text structure strategy instruction and other findings did not. For example, Taylor’s 1982 study, which involved teaching a text structure strategy to two different groups of fifth-grade students, showed an increase in student comprehension in the first experimental group, but not in the second experimental group. Armbruster, Anderson, and Ostertag’s 1987 study of four groups of fifth-grade students showed the groups of students who received text structure strategy instruction on the problem-solution text structure recalled more information on the assessment than the control groups and also wrote summaries containing more ideas and organization than the control groups. However, when problem/solution text structure was used as a skeleton for class discussion, these researchers determined that this particular strategy did not facilitate students’ recall of information. The results of Berkowitz’s 1986 study teaching assigning groups of sixth-grade students to receive instruction on expository text structure strategy of map-construction, map-study, or no Expository Text Structure 29 strategy instruction did not have a large impact on comprehension. The group receiving mapconstruction instruction only scored significantly higher than the control group on one of the three assessment passages. The group that received the map-study strategy instruction did not perform any better on the assessment than the control group. Similarly, Bellows’ 1994 case study suggested that fluent readers do not need explicit text structure instruction, and this researcher believed that knowledge of text structures may actually interfere in reading comprehension for disabled readers. When considering the varied results of these research studies, it is important to remember that many other factors impact students’ comprehension levels of expository text, including developmental levels, text difficulty, topic familiarity, and previous exposure and experiences reading expository texts. At this point, it is indeed difficult to determine whether providing text structure instruction will foster comprehension for all students. Some of the research studies indicate that expository text structure strategy instruction is very helpful for students, and other studies show that it is not helpful. The research results are not compelling enough to determine whether or not teachers should provide explicit instruction of text structure. Thus, after examination of these studies, my second research question remains largely unanswered. Further research is necessary to determine the impact of expository text structure instructional strategies on students’ comprehension. Expository Text Structure 30 Chapter IV: Conclusion Recommendation The vast majority of the studies on expository text structure awareness and the benefits of explicit expository text structure instruction involved quantitative research methods. The only study that utilized a qualitative design was Bellows (1994). It would be worthwhile to conduct more case study research to gain insights into individual students’ levels of expository text structure awareness and whether this knowledge (or lack thereof) facilitates or interferes with comprehension. As a teacher, I want to know which text structure instructional strategies would be most beneficial for students. Further research in expository text structure instruction is necessary to determine whether or not it is worthwhile to provide explicit instruction to all students or to only provide it to certain grade levels and ability levels of students. The question that remains is: Who would most benefit from explicit text structure instruction? It is indeed difficult to generalize the findings about the impact of expository text structure strategy instruction on comprehension in order make specific instructional recommendations. One conclusion I can draw from a review of the research on expository text structure instruction, however, is that many students appear to benefit from explicit expository text structure strategy instruction spanning from the middle school grades to as young as second grade. Thus, teachers at these levels should become familiar with the different text structures and find strategies to teach these structures to foster comprehension. Some of the strategies used in the research studies contained in my review of literature would be adaptable, including the use of the hierarchical summarization strategy (Taylor, 1982), map-construction strategy, or map-study strategy (Berkowitz, 1986). I recommend teachers at a minimum point out the different text structures to students as they are encountered in text and discuss text organization and how it Expository Text Structure 31 helps the reader make sense of the material. Teachers should model how to construct graphic organizers that fit the organization of a particular text as a reading strategy. When students have to pull out key information to put in the graphic organizer, they are building meaning. It is also important to have students write summaries that follow the structure of the text, and I suggest teachers encourage students to make outlines and graphic organizers and use those as skeletons for writing about their reading. Based on my review of the literature, the effectiveness of the expository text structure instruction depends on the strategy selected to pair with a particular text structure. Some strategies appear to work better with a specific type of text structure than with a different text structure. Trial and error on the teacher’s part will determine which strategies are most effective with a certain type of text structure. Another factor affecting the impact of expository text structure strategy instruction the number of opportunities given for practicing the strategy. When teachers provide students with instruction in a particular text structure strategy, they must provide multiple opportunities for the student to practice the strategy so the student can use the strategy independently in the future. Areas for Further Research Because there is a gap in existing research on expository text structure awareness using a case study design, I plan to conduct a case study involving middle school students. According to the research, students at the second grade level through the ninth grade level display awareness of expository text structures (Englert & Hiebert, 1984; Richgels et al., 1987; Smith & Hahn, 1989). I will study students in sixth grade because this grade level falls in the middle of the range of students who display awareness of text structure. My main goal is to determine the extent to which middle school students are aware of expository text structures and how they use this Expository Text Structure 32 knowledge to construct meaning from expository text. I will study students at three reading comprehension ability levels (Above Average, Average, and Below Average) to determine whether text structure awareness is related to reading comprehension ability. Another goal of my study is to understand how students initially developed the skill of identifying and using expository text structures to make sense of textbooks. Did they learn it on their own through their reading experiences or did a teacher or teachers provide explicit instruction on expository text and expository text structures? I will obtain the students’ perspective on their knowledge of expository text structures and how they obtained this knowledge, as well as their ability to apply this knowledge to construct meaning from expository text. I believe that obtaining the students’ perspective on their awareness of expository text structures has been largely ignored in many previous research studies. My study will help to fill in this gap in existing research. Several researchers have found that awareness of text structures is highly related to reading comprehension (Dickson, Simmons, & Kameenui, 2001). According to Hall et al. (2005), readers who understand how a text is structured “find greater success in identifying important information and relationships between ideas” (p. 215). It is for this reason I am going to conduct an in-depth study about middle school students’ awareness of expository text structures and their use of expository text structures to facilitate their reading comprehension. My goal is to determine whether differences exist between students of above average, average, and below average reading comprehension ability in terms of their awareness of text structure and their use of text structure in recall of expository text. I believe studying this facet of reading comprehension will help practitioners gain a greater understanding of the extent to which expository text structure awareness affects reading comprehension of expository text. This study Expository Text Structure 33 will help practitioners gain a better understanding of middle school students’ awareness of expository text structures, and can use this information to decide whether they should provide their students with explicit instruction on expository text structure. My study will contribute to the limited body of research available about middle school students’ use of expository text structures to comprehend content-area reading. My study is rather unique because I am going to conduct a qualitative research study, and the majority of the studies I have reviewed on this topic have emphasized quantitative research. While my study will contain some quantitative data in order show students’ performance on the test measures, my focus is on collecting qualitative data from students through interviews. I believe getting the students’ perspective on their awareness of expository text structure has been largely ignored in the field of research. Study Design The majority of research studies I have reviewed are quantitative in nature and/or experimental. I am most interested in studying individual students’ awareness of expository text structures. Thus, I am going to conduct a qualitative study on this topic using a comparative case study research design. I believe a case study design will provide deeper insights into this research topic. I will obtain the following information from each participant in my study: 1. The individual student’s perspective on his/her awareness and use of expository text structures to aid in reading comprehension. 2. How the student became aware of expository text structures: Did the student learn about text structure through reading experiences or did they learn it in school? 3. The student’s ability to identify a text structure from an expository text passage. Expository Text Structure 34 4. The extent of the student’s use of the author’s text structure in oral retellings of expository passages. In order to obtain this information, I designed an interview protocol to use with each student in my study (See Appendix C). I will also utilize two test measures: an oral recall of an expository text passage (See Appendix A) and a written test in which students need to identify the underlying text structure of a passage (See Appendix B). Participants Before beginning my study, I will obtain permission to conduct research on human subjects through the Human Subjects Committee at Northern Michigan University by submitting the “Application for Review of Research Involving Human Subjects.” (Please see Appendix D to view the completed application form.) A total of nine students at the sixth grade level will participate in my study. Students from multiple classrooms may be used in order to properly represent the three groups of reading comprehension ability that I wish to study. I will study students classified as Above Average, Average, and Below Average in reading comprehension ability. The students selected to participate must meet the following criteria: the student volunteers to participate in the study and the student obtains permission from his/her parents to participate through the signing of a consent form (Please see Appendix E to view the consent form used for this study). The selected students will be assigned to an ability group based upon their reading comprehension ability as determined by the student’s classroom teacher(s) and standardized reading assessment scores. Each ability group will consist of three students. It is important to note, however, that the students will not be aware of this ability classification system. Knowing this information could potentially damage the students’ self-esteem and affect the outcome of the study negatively. Expository Text Structure 35 Materials My study will involve determining what middle school students know about expository text structures and whether they are able to use this knowledge to comprehend expository text material. I will collect data from the students by interviewing them on an individual basis and by administering two test measures: A written test in which the student must identify the text structures in expository passages and an oral retelling after reading an expository text passage. (See Appendix A for the Oral Retelling Measure, Appendix B for the Written Test Measure, and Appendix C for the Interview Protocol.) The text passages for the written test measure will be selected from the students’ content-area reading textbooks, and therefore, the test does not appear completed at this time. However, the general format for the written test measure appears in Appendix B. My study builds on the research on text structure awareness conducted by McGee (1982), Richgels et al. (1987), and Ghaith & Harkouss (2003). Each of these researchers studied text structure awareness from the middle school to the high school level using a variety of measures. My interview protocol and written test measure are not based on these prior studies. However, the oral retelling test measure is based on McGee’s main method of data collection in her 1982 study. For the oral retelling passage, I will select one passage that contains a compare/contrast structure. The reason behind the selection of this particular text structure for the oral retelling measure is due to the inconsistencies of students’ awareness of this text structure in prior studies. Some researchers found that students are most aware of the comparison/contrast structure (Richgels, et al, 1987; Ghaith & Harkouss, 2003), while others found that students are least aware of the comparison/contrast structure (Englert & Hiebert, 1984; Smith & Hahn, 1989). In my study, I will shed more light on students’ awareness of the compare/contrast text structure. Expository Text Structure 36 The main focus of my study will be on the interview protocol, which I believe will help practitioners gain a greater understanding of adolescents’ awareness and use of expository text structures. According to Baker & Brown (1984), the most obvious way to discover students’ knowledge about a topic is to simply ask the students what they know (as cited in Garner & Gillingham, 1987). Although interviews were utilized in some prior studies (Elliot, 1980, as cited in Richgels, et al 1987; Garner & Gillingham, 1987; Richgels, et al, 1987; Bellows, 1994), none of the studies placed a major emphasis on this measure to determine expository text structure awareness. Sampling My sampling procedure will be purposeful random in that I will be selecting middle school students at random from a classroom (or classrooms) of students volunteering to participate in the study. I categorize the sample as a maximum variation because I will be selecting students whose reading comprehension level is above average, average, and below average. I selected this sampling procedure because I want to study students at all levels of reading comprehension ability. In my data analysis, I will compare/contrast the three ability groups and determine whether a relationship exists between reading comprehension and students’ awareness of expository text structures. My sampling unit is individual students. The size of my sample is nine students. There will be three subgroups consisting of three students per subgroup. The subgroups are: 1) Above Average Reading Comprehension, 2) Average Reading Comprehension, and 3) Below Average Reading Comprehension. I believe this sample size is sufficient because I will study three students in each category, which will allow me to determine whether there are differences between students in the same category. Expository Text Structure 37 Data Collection Methods My data collection will focus primarily on the emic perspective because my goal is to obtain the students’ perceptions about their knowledge of expository text structure as it relates to their reading comprehension. The etic perspective will also be represented in my study, as I will be interpreting the data, imposing meaning on the data, and identifying patterns in the data. The focus of my data collection is on explanation. I will be collecting data from students in order to identify patterns across the cases (students). I am looking for a relational pattern in the data. The data collection instruments used will include an individual interview with each student and two different test measures. One test measure will assess the student’s ability to identify expository text structures and the other test measure will be used to determine whether the student uses text structure in his/her retelling of an expository passage. I will be actively involved in the data collection for the interviews. For the test measures, I will only provide directions for the students before they begin taking the test. Data Collection Procedures After identifying student participants, I will work with each student individually to collect data for my study. I will measure students’ awareness of expository text structure as it relates to reading comprehension in two ways: Through an oral and written test measure, and an interview protocol. During the first session with the student, I will conduct the oral and written test measures. These measures will be conducted with each student in one 50-minute session. I will administer the oral retelling assessment first, followed by the written test measure. The oral retelling assessment is two-fold: The student will read the expository text passage and then the student will orally retell the passage to me as if I had not read it and I wanted to know what it was about. The oral retelling will be recorded on audiotape. After the oral retelling assessment, Expository Text Structure 38 I will administer the written test measure. The written test measure consists of ten items. For each item, the student will read an expository passage and then identify the underlying text structure from a list of the five following structures: Compare/Contrast, Description, Cause/Effect, Problem/Solution, and Chronological Sequence. Students will be given a brief definition of each structure type before beginning the test, as they may not be familiar with this terminology. My second measure is an interview protocol. I will conduct the interview during the second 50 minute session with each individual student. I will elicit student responses by asking specific questions about their awareness of expository text structure and reading comprehension (See Appendix C for list of interview questions). I will record each interview on audiotape. Data Analysis Procedures My main methods of analysis for this study will be reflective analysis and structural analysis. I will utilize reflective analysis for the interview data collected. I believe using reflective analysis in my study will provide the etic perspective of my case study because in reflective analysis, the researcher relies on his/her intuition and personal judgment to analyze the data. For the interview measure, I will review the audio-taped interviews and look for patterns in the students’ responses. I analyzing the interviews using reflective analysis, I will look for patterns and themes to emerge across individual cases and ability groups. I will use structural analysis in interpreting the students’ oral retellings of an expository text passage. In analyzing students’ retellings, I will record signal words (words that signal the author’s text structure) the student uses in his/her retelling. I will also record the number of super ordinate (main ideas) and subordinate ideas (supporting details) the student includes in his/her oral retelling. Expository Text Structure 39 I will calculate some descriptive statistics for the written test measure and the oral retelling of an expository text passage. I will calculate the mean, median, and range of scores for each subgroup. The written test scores will be calculated based on the number text structures identified correctly. I will then convert the raw scores to a percentage of items answered correctly. After converting the scores into percentages, I will calculate the mean and median scores. For the test measure, I will develop a table that represents the mean and median scores for each subgroup and compare the subgroups’ scores on the test to see if differences exist between the students who have below average, average, and above average reading comprehension abilities. The range of raw scores will also be calculated for each of the three subgroups to see if any variation exists among the students contained in one subgroup. For the retelling test measure, I will audiotape the students’ oral retelling. There will be two assessment criteria for this measure. The students will be assessed on their ability to retell the passage using the author’s underlying text structure, as evidenced by their use of signal words, which are words that are commonly used in a specific text structure. The students will also be assessed on their recall of super ordinate (main ideas) and subordinate ideas (supporting details). I will determine the number of super ordinate and subordinate ideas in the passage, and students will be scored based on the number of ideas contained in their oral retelling. The students’ final score will be the percentage of ideas the students recalls in the retelling of the passage. Limitations of Study Reliability and Validity Concerns. For the written test measure, the validity concern is that the tests will indeed measure what they are supposed to measure. One concern is that the text level for written passages on the test measures will be too difficult for students in the study, or that students may not have sufficient prior knowledge of the topics of the passages. The text Expository Text Structure 40 difficulty and/or lack of students’ prior knowledge are factors that could interfere with the results of the test measures. To minimize this concern as much as possible, I will select passages for the test that are at the grade level of the students. I will choose passages from their grade level content-area reading textbooks. My goal is for the test to measure the students’ ability to identify expository text structures. Thus, I will structure the written so that each expository text passage selected to be included on the test will contain only one specific text structure, and the student will be instructed to identify the correct text structure. There are several types of validity and reliability issues for the interview protocol. Obviously, the validity of the interview measure depends on students being honest about what they know about expository text structure. The validity also depends on students’ ability to articulate their knowledge verbally in an interview setting. In interviews, students sometimes fail to report knowledge of highly automated processes because they lack the verbal ability to do so (Bellows, 1994). Furthermore, the reliability of the interview measure is dependent on developing a set of questions that will elicit all the information I need from the students. My interpersonal skills as an interviewer also impacts reliability in this study. After I complete the interviews, I will compile the interview data, look for trends in the data, describe any inconsistencies in the trends, and attempt to provide a reasonable explanation for any inconsistencies in the data. Generalization of Results The findings are limited by the case study research design, but will provide rich information on a small number of students of three ability levels. The short passages also limit the generalization of the results because they contain only one text structure. These passages are unlike passages students are likely to encounter in textbooks, as very few passages actually Expository Text Structure 41 reflect a single text structure (Richgels, et al, 1987). Moreover, many textbooks are organized and written poorly (Dickson, Simmons & Kameenui, 2001), lack clear structure, and often switch between text structures frequently (Hall, et al, 2005). Thus, poorly written textbooks may interfere with a students’ use of text structure as a strategy for reading comprehension. Each of these factors affects the generalizability of the results. Expected Results Based on prior studies, I anticipate that I will obtain different results for each ability group of students. In terms of performance on the oral and written test measures, I predict that there will be a marked difference in scores between students in the Above Average category and Below Average category. I predict the following results: Above Average students will have high scores on the oral and written test measures, Average students will obtain average scores on these measures, and Below Average students will achieve below average scores on the test measures. I also anticipate individual differences in scores within each subgroup. In terms of the interview measure, I anticipate that students in the Above Average group will be better able to articulate their responses than those of the Average and Below Average groups. I believe many individual differences will also exist within each subgroup in their performance on the test measures and in their responses to the interview questions. Expected Findings I expect that all students in all subgroups will display some awareness of text structure, based on the results of previous research studies. In terms of expected differences between the subgroups, I believe the students in the Above Average group will perform better on the test measures and interview than those in the Average and Below Average groups. I believe the Expository Text Structure 42 students in the Average group will perform slightly better on the test measures and interview than those in the Below Average group. To generalize further, the expected results indicate that students who are above average readers (in terms of reading comprehension ability) are more aware of text structure than students who are average or poor readers In addition, students in the Average group are slightly more aware of text structure than those students in the Below Average group. Thus, from these expected findings, one can conclude that text structure awareness is positively related to reading comprehension ability. These results would be very similar to those found in McGee (1982), Meyer, Brandt, & Bluth (1980), and Richgels, et al (1987). I believe that students in all subgroups will display some level difficulty with the compare/contrast text structure in the oral retelling. The compare/contrast structure is more difficult for young students to comprehend (compared to adults) because they lack “cognitive sophistication and memory capabilities for recalling the parallel features and attributes of two concepts being compared” (Englert & Hiebert, 1984). From these expected results, one can conclude that students have less awareness of the compare/contrast text structure than other text structures. These expected results would be comparable to the findings of Englert & Hiebert (1984) and Smith & Hahn (1989). Implications and Suggestions for Further Research Many researchers have studied the effects of teaching expository text structures on reading comprehension, and several findings from these researchers indicate that training in expository text structures appears to facilitate reading comprehension. Based on the expected findings in this study, it would make sense for teachers to take the time to teach students about expository text structures in order to potentially help students improve their reading comprehension Expository Text Structure 43 abilities. The test measures and interviews used in this study reveal what students know about expository text structures. Using these assessments in the classroom would help teachers gain a better understanding of what their students know about expository text structures. Teachers could then provide any necessary instruction on text structures to help their students improve their reading comprehension. Providing students with explicit instruction on expository text structure will help them construct new schemata for comprehending expository text. Such instruction also helps students increase their metacognitive awareness of text structures, which will ultimately allow students to become more independent, successful readers of expository text. My proposed study is similar in some respects to those studies conducted by Richgels, et al (1987), McGee (1982), and Ghaith and Harkouss (2003). However, my planned research study differs from the previous studies in two ways. First, I will place a much greater emphasis on the use of the interview protocol to discover students’ awareness of expository text structures. Second, I will design a new written test measure not used by previous researchers to measure students’ ability to identify expository text structures. Because my research methods are somewhat different from the methods used in preceding studies, I believe my findings will be more generalizable if my study is replicated using the same research methodology. Specifically, this study should be replicated with groups of students at other grade levels in the middle school (including grades 5, 7, and 8) to make the research applicable to the whole age range of middle school students. Summary and Conclusion A great deal of further research is still needed in the area of expository text structure awareness. Due to the conflicting results of previous studies, more studies are needed to determine the text structures in which students are most aware and least aware. More research is Expository Text Structure 44 also necessary to determine when it would be most developmentally appropriate to teach students about expository text structure awareness. Hall, et al (2005) recently conducted research on teaching expository text structure to second grade students. The results of the preliminary study conducted by Hall, et al (2005) suggest that it is beneficial to provide young children with instruction on expository text structures. Williams, Hall, and Lauer (2004) and Hall, et al (2005), are among the first researchers to explore the possibility of expository text structure instruction at the early elementary level. Williams, Hall, and Lauer (2004) note that very few instructional programs have been developed for teaching expository text structure, so research to develop and evaluate instructional programs is also needed. Additional research is also required to determine the most effective ways to teach expository text structures (Hall, et al, 2005). The purpose of the literature review was two-fold: 1) To determine middle school students’ awareness and use of expository text structures to facilitate their reading comprehension; and 2) To determine what extent teachers should provide explicit instruction in expository text structures. Many students struggle with reading comprehension of expository text in middle school due to limited exposure to such texts in elementary school. There are several teaching strategies for fostering students’ comprehension of expository text in addition to expository text structure awareness, such as teaching vocabulary and text signals (headings, overviews, etc.) (Hall, et al, 2005). Any comprehension strategy instruction that teachers can provide to help students improve their reading comprehension skills is worthwhile, and expository text structure awareness is one comprehension strategy that should not be overlooked due to its potential to increase students’ reading comprehension. According to Dickson, Simmons & Kameenui Expository Text Structure 45 (2001), “students who are aware of or have had instruction in text structures demonstrate better global comprehension (an understanding of the main ideas) than those who lack awareness or have not had instruction” (p. 23). Students’ ability to comprehend expository text is imperative because expository texts make up the bulk of the reading students will face in middle school, high school and beyond. Therefore, I believe all educators should make an effort to learn more about how to increase students’ reading comprehension of expository texts. Implementing instructional strategies to facilitate expository text comprehension will help prepare students for the many challenging reading tasks that lie ahead. Expository Text Structure 46 REFERENCESAllington, R.L. (2006). What really matters for struggling readers: Designingresearch-based programs. (2 Ed.). Upper Saddle River, NJ: PearsonEducation, Inc.Almasi, J.F. (2003). Teaching strategic processes in reading. New York:The Guilford Press.Armbruster, B.B., Anderson, T.H. & Ostertag, J. (1987). Does textstructure/summarization instruction facilitate learning from expository text?Reading Research Quarterly, 22(3), 331-346.Bellows, B.P. (1994). Does knowing about text structures help disabled, adolescentreaders? An exploratory study of adolescents’ awareness and use of globalcoherence. Paper presented at the Annual Meeting of the National ReadingConference. San Diego, CA, November 30-December 3, 1994.Berkowitz, S.J. (1986). Effects of instruction in text organization on sixth-gradestudents’ memory for expository reading. Reading Research Quarterly, 21(2), 161-178.Block, C., & Pressley M. (Eds.) (2002). Comprehension instruction: Research-basedbest practices. New York: The Guilford Press.Brewer, J.H., & Harp, B. (2005). The informed reading teacher: Research-basedpractice. Upper Saddle River, NJ: Pearson Education, Inc. Dickson, S.V., Simmons, D.C., & Kameenui, E.J. (2001). Text organization and itsrelation to reading comprehension: A synthesis of the research. Washington, D.C.: Expository Text Structure 47 National Center to Improve the Tools of Educators. Retrieved November 3, 2005,from http://idea.uoregon.edu/~ncite/documents/techrep/tech17.htmlEnglert, C., & Hiebert, E.F. (1984). Children’s developing awareness of textstructure in expository materials. Journal of Educational Psychology, 76,65-74.Garner, R., & Gillingham, M.G. (1987). Students’ knowledge of text structure.Journal of Reading Behavior, 19(3), 247-259.Ghaith, G.M., & Harkouss, S.A. (2003). Role of text structure awareness in theRecall of expository discourse. Foreign Language Annals, 36(1), 86-96.Hall, K.M., Sabey, B.L., & McClellan, M. (2005). Expository textcomprehension: Helping primary-grade teachers use expository texts to fulladvantage. Reading Psychology, 26, 211-234.McGee, L.M. (1982). Awareness of text structure: Effects on children’s recall ofexpository text. Reading Research Quarterly, 17(4), 581-590.Meyer, B.J., Brandt, D.M., & Bluth, G.J. (1980). Use of Top-Level Structure in Text: KeyFor Reading Comprehension of Ninth-Grade Students. Reading Research Quarterly,16(1), 72-103. Mueller, M.E. (1996). Using metacognitive strategies to facilitate expository textmastery. Educational Research Quarterly, 20(3), 41-65. Pearson, P.D., & Gallagher, M.C. (1983). The instruction of reading comprehension.Contemporary Educational Psychology, 8, 317-344. Richgels, D.J., McGee, L.M., Lomax, R.G., & Sheard, C. (1987). Awareness offour text structures: Effects on recall of expository text. Reading Research Expository Text Structure 48 Quarterly, 22(2), 177-196.Smith, T.F., & Hahn, A.L. (1989). Intermediate-grade student’s sensitivity tomacrostructure intrusions. Journal of Reading Behavior, 21(2), 167-180.Street, C. (2002). Expository text and middle school students: Some lessons learned.Voices from the Middle, 9(4), 33-38.Taylor, B.M. (1982). Text structure and children’s comprehension and memory forexpository material. Journal of Educational Psychology, 74(3), 323-340.Taylor, B.M., & Beach, R.W. (1984). The effects of text structure instruction on middle-grade students’ comprehension and production of expository text. Reading ResearchQuarterly, 19(2), 134-146.Taylor, B.M., & Samuels, S. J. (1983). Children’s use of text structure in the recall ofexpository material. American Educational Research Journal, 20(4), 527-528.Tompkins, G. (2002). Literacy for the 21 Century. (3 Ed.). Upper Saddle River,NJ: Pearson Education, Inc.Williams, J.P., Hall, K.M., & Lauer, K.D. (2004). Teaching expository textstructure to young at-risk learners: Building the basics of comprehensioninstruction. Exceptionality, 12(3), 129-144.Williams, J.P., Hall, K.M., Lauer, K.D., Stafford, B., DeSisto, L.A., & DeCani, J. S. (2005).Expository text comprehension in the primary grade classroom. Journal of EducationalPsychology, 97(4), 538-550. Willingham, D. (2010, n.d.). Reading is not a skill—And why this is a problem for thedraft national standards. The Washington Post. Retrieved fromhttp://voices.washingtonpost.com/answer-sheet/daniel-willingham/willingham-reading-is-not-a-sk.html Expository Text Structure 49 Words that signal a text’s organizational structure (n.d.). Somers Central School DistrictWebpage. Retrieved November 3, 2005, fromhttp://www.somers.k12.ny.us/intranet/reading/signalwords.html Expository Text Structure 50