• E-mail
• Print
• Comment
• Font Size
• Digg
• del.icio.us
• Discuss article

Eyes on the Prize: Teaching Complex Historical Content to Middle School Students With Learning Disabilities

Posted on: Sunday, 2 April 2006, 06:00 CDT

By Gersten, Russell; Baker, Scott K; Smith-Johnson, Joyce; Dimino, Joseph; Peterson, Anne

ABSTRACT:

This study compared two approaches for teaching a history unit on the Civil Rights Movement (1954-1965) to middle school students with learning disabilities (LD) in general education settings. The curriculum was designed to make the content challenging, but accessible. The major text was the documentary, Eyes on the Prize (DeVinney, 1991). Readings were brief and included primary sources from the time period. Curriculum content was constant in both instruction conditions, but in the experimental condition teachers isolated the video selections into brief 2- to 3-min segments to facilitate learning and used peer dyad activities to foster active processing of content. Students with and without disabilities in the experimental condition scored significantly higher on 2 of 3 content measures.

A quiet revolution has taken place in special education in the past decade, a change formalized in both the 1997 and 2004 versions of IDEA (U.S. Department of Education, 1997, 2004). IDEA requires that students with disabilities receive "access to the general curriculum," meaning they must be taught grade-level concepts in disciplines such as history, science, and mathematics. Numerous factors led to this shift in emphasis in teaching in special education. One factor was the wide array of curriculum reform movements during the 1990s that stressed that challenging content, whether in history, math, or science, should be taught so that all students, including students with disabilities, could learn essential concepts and principles. These movements were influenced by the growing realization that "specialized" remedial programs, particularly at the middle and high school levels, almost invariably stressed "basic" academic skills at the expense of serious exposure to conceptually demanding material.

Another factor influential in curriculum access was the emerging body of research on strategy instruction and procedural facilitators. This body of research demonstrated that poor reading skills did not have to restrict students from accessing challenging content. When provided with intense and explicit instruction in how to strategically approach the learning of challenging material, these students could learn to express complex ideas in a coherent fashion (Baker, Gersten, & Scanlon 2002; Englert, Raphael, Anderson, Anthony, & Stevens, 1991).

RESEARCH ON PROVIDING ACCESS TO THE GENERAL CURRICULUM IN SOCIAL STUDIES

Observational research has consistently documented that, although students with learning disabilities (LD) often receive social studies instruction in general education classrooms, they typically do not learn very much in these classes (Zigmond, Wolery, Meng, Flumer, & Bean, 1994). Their minimal learning can be explained, in part, by the fact that traditional teaching of history relies heavily on students being able to independently read course textbooks. Most students with LD read grade level material with extreme difficulty, and these problems are exacerbated when the text is technical or abstract and the vocabulary is complex, as it is in content-area textbooks. When Zigmond et al. interviewed students with LD after social studies lessons that relied heavily on textbook reading, traditional lecture, and whole class discussion methods, they found students with LD demonstrated, at best, superficial knowledge of the key concepts that were demonstrated, explained, and reviewed by the teacher. In examining instruction in these classrooms, Zigmond et al. observed that, "students with learning disabilities. . . were selectively attentive, often focusing on an extraneous part of a lesson or explanation. [They were] easily confused and [overly] concrete in their understandings" (p. 14). Teachers also did little to ensure that students focused on core concepts, and the lesson structure itself did little to engage the attention of these students.

Passe and Beattie (1994) found that when social studies teachers tried to make "adaptations" for students with LD, they were often of little help to the students. For example, the major "adaptation" was to provide peer tutors to help students with LD do better on tests. Although this adaptation resulted in slight increases in test scores, the focus of the tutoring seemed to be limited to the memorization of factual content. It did not address higher-level concepts, such as the meaning of events or key relationships.

Passe and Beattie (1994) also found that, when teachers "individualized" academic demands for students with LD, they usually did so by lowering standards. The new standards invariably stressed the memorization of names and dates. The authors observed, for example, "when classroom peers were expected to analyze the characteristics of various nations, special education students merely had to learn to locate the nations on a map" (p. 230). In summary, special education students have received superficial access to topics in the general curriculum, but this access is rarely meaningful and it rarely covers the key concepts that students without disabilities are expected to learn.

NEW DIRECTIONS IN CURRICULUM ACCESS

To foster meaningful access to challenging content for students with LD, we focused on developing curriculum content that would not require strong or even average reading skills for accessibility. This entailed the serious use of material other than grade-level textbooks, which almost always are too difficult for special education students to independently read and understand. A large body of research suggests that various strategies and procedures based on text structures would be useful to these students in terms of comprehension of text and understanding of course content (Baker et al., 2002; Ferretti, MacArthur, & Okolo, 2001; Gersten, Fuchs, Williams, & Baker, 2001; Graham, Harris, MacArthur, & Schwartz, 1991).

To increase accessibility, we hypothesized that instruction should be highly interactive and socially mediated. For example, student achievement in reading and mathematics is enhanced when they have opportunities to work in heterogeneous dyads during parts of a lesson and are provided with clear structures to guide peer interactions (Elbaum, Vaughn, Hughes, & Moody, 1999; Graham et. al., 1991). Taken together, these studies suggest that some type of peer- mediated instruction might significantly enhance the learning of more complex content for students with LD. It might also be beneficial for average-ability students. We also reasoned that students would learn more if the teacher-directed portions of the lesson were highly interactive, providing the teacher with many opportunities to informally assess and monitor student understanding on an ongoing basis, and providing students with opportunities to articulate their understanding of the content.

PURPOSE OF THE STUDY

We believed that students with LD could learn history if (a) instruction included comprehensible and accessible materials (rather than sole reliance on traditional textbooks), and (b) incorporated instructional delivery strategies that provided numerous opportunities for students to interact with peers and the teacher during the lesson (rather than heavy reliance on lectures and whole class discussions). We targeted history because the principles and relationships explored in history are quintessentially human and of potential interest to virtually all students, regardless of reading ability.

The first phase of the project involved development and extensive field-testing of the curricula materials and instructional delivery practices that might succeed with students with LD (Gersten, 2005; Gersten, Baker, Smith-Johnson, Peterson, & Dimino, 2003). We chose the Civil Rights Movement (CRM) as the topic; this period of American history transformed the nature of American society in profound ways. Our goal was for students to learn key events, the meaning of these events within a historical period, and to begin to understand the greater significance of these events in contemporary society. We wanted to evaluate our approach using rigorous research methodology: randomized controlled trials (Shadish, Cook, & Campbell, 2002). We decided that using a traditional control group (i.e., a group taught the same content but using traditional texts and teaching methods) would not make sense since we knew that virtually none of the LD students could read a grade-appropriate text.

We therefore set up a more stringent comparison condition. Students in the comparison group received the identical curriculum content as students in the experimental group, but the teacher did not utilize the interactive instructional delivery procedures we developed and adapted. Thus, the randomized controlled trials did not evaluate the curriculum per se, but evaluated whether delivering that content in these specific ways increased learning in inclusive settings.

DESIGN

Students with and without LD were matched on disability status and oral reading fluency, and then randomly assigned to the experimental teaching condition or the comparison condition. The teachers were counterbalanced across conditions. Eachteacher taught the experimental condition once and the comparison condition once, thus controlling for teacher effects. The teacher selected to teach the experimental condition in the fall in one school taught the comparison condition in the winter in the second school. The other teacher taught the comparison condition in the fall and the experimental condition in the winter. In all of the classrooms approximately half of the students were LD and half were students of average ability.

METHOD

PARTICIPANTS AND SETTING

Students. Seventy-six middle school students (47 seventh graders and 29 eighth graders) participated in the study. Thirty-six of these students were students with disabilities (33 students with LD and 3 classified as Other Health Impaired (OHI), hereafter referred to as the LD sample). These students are typically included in general education classes for most of their instructional day, especially for classes in subjects such as social studies or science (Wagner & Blackorby, 2002). For the purpose of this article, we prefer to use the term learning disabilities to refer to this sample, since neither LD nor attention deficit disorder are clearly defined and two characteristics common to both groups are reading difficulties and problems with organizational strategy. Both of these characteristics create potential problems in history instruction. We refer to all of these students subsequently as students with LD. On average, participating seventh- and eighth- grade students with LD and OHI read at a level typical for third and fourth graders, that is, 3 to 5 years below grade level (Hasbrouck & Tindal, 1992.)

Students were pretested on oral reading fluency and then matched on disability status (i.e., with or without a disability) and reading performance. A member of each matched pair was then randomly assigned to the experimental group or the control group and the other student in the pair was assigned to the other condition.

A number of measures were administered at pretest to describe the sample of students and serve as potential predictors of content learning (see Table 1). Rapid Automatized Naming (RAN; Denckla & Rudel, 1976) and Oral Reading Fluency (Shinn, Good, Knutson, Tilly, & Collins, 1992) were administered as a way to describe the processing speed of students and general reading skills, respectively. The Analogies subtest from the Woodcock Reading Mastery Tests-Revised (Woodcock, 1998) was administered as a measure of reading ability and general knowledge. We also administered a researcher-developed Matching Test as a pretest to gauge relevant background knowledge of content related to the CRM. This test consisted of a series of pictures depicting key events and figures pertaining to the CRM. Additionally, the measure included basic questions asking students to locate the CRM within recent history.

Sample Comparability. For students with LD, pretest differences between the experimental and comparison conditions on the Oral Reading Fluency, Analogies subtest, and Rapid Automatized Naming measures were not statistically significant. Students with LD read, on average, slightly more than 100 words per min accurately, a rate that was about 40% lower than their peers. The level of oral reading proficiency attained by the LD students is typically found with average-ability third and fourth graders (Hasbrouck & Tindal, 1992), indicating that these students would struggle substantially with independently reading typical grade level history texts. Note, too, the sizeable difference in Rapid Automatized Naming between the LD and average-ability samples, which is consistent with prior research (Schatschneider, Carlson, Francis, Foorman, & Fletcher, 2002). There was a statistically significant difference on the Matching Test measure favoring the experimental group for both the LD and the average-ability samples. Since students were randomly assigned to conditions, we are able to statistically adjust for this initial discrepancy in the posttest analyses by using Pedhazur's (1997) multiple regression procedure to control for initial pretest differences.

Teachers. Two experienced teachers taught the experimental and comparison groups. Both were certified to teach special education at elementary and middle school settings. Their backgrounds, however, were quite different. One teacher had taught special education at the elementary and middle school level for several years, working primarily with students with behavior disorders and LD. The other teacher had taught for 26 years. She began her career as a special education teacher and then transferred to teaching social studies at the middle school level. She had 10-years experience as a middle school social studies teacher. Both teachers actively collaborated in the development of the curriculum and assisted in the pilot study. Consequently, they did not receive any formal training on the instructional procedures for the experimental study.

TABLE 1

Pretest Data for the Experimental and Comparison Samples

Setting. The study was conducted in two middle schools in a moderate size city in the Northwest; Brazelton and Garfield (both are pseudonyms). Although Brazelton and Garfield had similar percentages of ethnic minority students (22.0% and 22.7%), in other aspects they were quite different. Brazelton was located in a primarily working class neighborhood. Families in the Garfield neighborhood included relatively wealthy, suburban and working class families, as well as children of graduate students at the local university. Garfield had a strong emphasis on appreciation of diversity. The school had large signs declaring its status as a racism-free zone, and large pictures of distinguished African Americans, Latinos, and Asian Americans were located throughout the hallways. At Garfield, instructional lessons occurred during a 41- min class period, for 25 consecutive school days during the fall semester. At Brazelton, instructional lessons occurred during a 52- min class period during the winter semester. Thus, the unit was completed in the same amount of time, but in fewer days at Brazelton.

INSTRUCTIONAL MATERIALS AND PROCEDURES FOR THE EXPERIMENTAL CONDITION

CREATING MEANINGFUL ACCESS TO THE CONTENT

Teaching Videos and Materials. The documentary Eyes on the Prize (DeVinney, 1991) was the primary content source for the CRM unit. The documentary is 6 hrs in length, but in the study we used an edited, 2-hr version developed for use in secondary schools. We divided the video into 18 segments that were shown over the 5 weeks of instruction. These individual segments ranged from 4 to 10 min in length.

The documentary begins dramatically with describing the brutal murder of Emmett Till in 1955. It follows with about a dozen other critical events, most of which were linked to important consequences that affected the nation as a whole. For example, students studied "The Little Rock 9" (i.e., the integration of Central High School in Arkansas) as a way to directly explore the significance and human impact of the Brown vs. the Board of Education Supreme Court decision. A number of other segments dealt with the involvement of children in the CRM. Furthermore, the role of children in various boycotts was the subject of one of the segments.

Readings included relevant selections from the text, Understanding American History (Carnine, Steely, & Silbert, 1996), as well as a series of articles from magazines of that era (e.g., Look, Life, Time), which presented material on Rosa Parks, the integration of Central High School, and Voter Registration training, and an excerpt from the book The Century (Jennings & Brewster, 1998).

Compare-Contrast Text Structure Activities to Promote Understanding. We scheduled four Compare-Contrast activities during the unit. We used these activities as a way to increase students' opportunities to reason with historical events in abstract ways and to build opportunities for discussions. The discussions also served as a way for students to review and recall key events and to build better understanding of linkages between events. Comparisons included juxtaposing the outcome of the Texas trials involving the racially motivated murder of an African American in 2000 with the trial of Emmett Till in the 1950s. Another comparison involved analyzing two types of protest activities, boycotts and sit-ins. Figure 1 demonstrates a typical compare-contrast activity. This involved comparing Minnijean Brown, one of the nine brave African American high school students who integrated at Central High School, to Rosa Parks, whose refusal to move to the back of the bus was the catalyst for the Montgomery bus boycott. Students were asked to compare these women's decisions, as well as to consider other issues surrounding their unique circumstances that students thought were interesting or important. The student with LD in Figure 1 noted that both figures "kept calm and didn't stop going when the going got tough." The student notes, however, that one key difference was that Minnijean was one of nine students who worked together and were supported by others outside of school, whereas Rosa Parks acted alone on the day she was arrested. Students worked on this activity over a 2-day period for approximately 15 min per day.

Narrative Text Structures to Generate Discussion Questions and Organize Material. In orchestrating the various reading selections, video segments, discussion questions, and mini-lectures into a coherent unit, we were heavily influenced by the research of McKeown and Beck (1994) on making history instruction more accessible. Presenting a historical topic as an engaging story helps students understand the events and details because the familiar narrative structure gives them a way to process the content, make predictions, and to draw causal inferences. We wanted students to discuss the various people in the CRM, both wellknown (e.g., Martin Luther King, John Lewis, Lyndon Johnson) and less well known (students such as Minnijean Brown or the unnamed student engaged in Voter Registration drives). Vivid narratives helped guide this objective.

To help students construct reasonable narratives about the various individuals they were learning about, we asked them many "How would you feel if . . . ?" questions. For example, just as students saw heavily armed police (accompanied by large dogs) order marchers to stop during the Selma March we asked, "How would you feel if you were a protester walking over the Edmond Pettus Bridge?" We also asked them to describe how they would feel undergoing training for the lunch counter sit-ins, and training for voter registration in the South. We asked them how they would feel if, like the students who integrated Central High School, they were accompanied by a guard at school. These types of questions have no right or wrong answer, but a well-reasoned response required specific knowledge about the event and context.

ACTIVITIES TO PROMOTE INTERACTIVE INSTRUCTION

Inserted Questions and Teacher Clarifications During Viewing of Videos. Although the daily video segments were brief (between 5 and 12 min), they contained a great deal of information. To enhance student understanding of this content and to increase their opportunities to engage in meaningful discussions, the experimental condition incorporated a series of 3 to 4 stop points during each video segment at which point the teacher would clarify lesson content. Teachers focused heavily on clarification (Palincsar & Brown, 1984) at some of the break points. During a 12-min video segment about events that led to the passage of the Voting Rights Act of 1965, the teacher stopped the video 90 s into the segment. Her guide gave her a precise place to stop and noted: Student Nonviolent Coordinating Committee (SNCC) working in Selma. At this particular break point, she reinforced what the speaker had just mentioned: "Since Mississippi had the worst record for registering African Americans in the country at that time, it was chosen as the target for extensive political action by the Movement" (observational field notes, Nov. 21, 1999).

FIGURE 1

Example of a Compare and Contrast Activity

The following example represents another type of clarification, during a break point in the video segment on the training for those who participated in the lunch counter sit-in (observational field notes, Nov. 15, 1999). At the first break point, 3 min into the segment, a student asked, "Why are the African Americans being treated this way?" The teacher explained they were being trained in how to respond to events that might actually happen during a sit- in. A second student followed up on this point by asking why "white people were also being trained [in sitin procedures] in the film?" This question gave the teacher the opportunity to explain that many whites did not believe in segregation and that they were going through the training with the African Americans.

The following example of inserted questions came from the same lesson, at a later point in the video. African Americans are being beaten and harassed during their sit-in at the lunch counter. At a break point the teacher asks, "What would you be thinking if you were actually there watching this happening?" One student said he was surprised that the white police were being violent. In making the connection to having waves of students who were prepared to join the sit-in, the teacher took this opportunity to make sure students understood what the term "second wave" meant. One student answered correctly that it meant "more than one group was ready," which prompted a second student to make the connection that "the sit-ins were well done, they were prepared." The notion that African Americans were well organized is a major theme of the CRM. So, at this break point, the students' comments allowed the teacher an opportunity to explain other examples of being well prepared. It also allowed the teacher to reinforce the point about preparation in the earlier discussion about the abuse African Americans were trained to undergo as part of their specific training for the sitins.

Concluding Clarification Activity. At the end of each segment, the teacher asked students if they had any questions about the segment they had just seen. The following student questions and teacher responses show how the teacher was able to use student questions to review important material from the segment and also was able to provide important background knowledge about social and political contexts in which the events took place.

In an episode on the death of Emmett Till, one student asked why the jury found the men innocent. (The video moves through the sequence of events that clearly demonstrate that two men had murdered Till. In fact, right after the verdict, one of the defendants admitted to a reporter that he had committed the crime.) Another student commented on the racial aspects of the trial and typical attitudes toward African Americans in the South. A third student asked how we knew for sure that the two men killed Till. At this point, the teacher reviewed the video material in which the men bragged about murder, and reminded the students that the two men were paid $4,000 to be interviewed about the crime. During the interview, they admitted that they killed Till. This prompted a student to ask why the men were not convicted after admitting to the crime. Another student commented that you could not be tried twice for a crime after being declared innocent by jury. The teacher took the opportunity to reinforce and discuss the concept of double jeopardy (Observational field notes, Oct. 27, 1999).

PEER DYAD ACTIVITIES

A second core instructional delivery technique used was a range of activities involving heterogeneous pairs of students. Think-Pair- Share (McTighe & Lyman, 1988) is a technique in which students first work on answering a question. They either develop a succinct statement on the importance of an incident or work on a project with a peer. Then, each dyad debriefs with the entire class. We also used Think-Pair-Share activities to allow students to discuss how they would feel if they were in the situation depicted in the video or text. Students worked with their partner on each question and then individually wrote out a response. Partner groups then shared their responses with the whole class. Both members of the dyad were expected to contribute to these discussions. Following is an example from observational field notes that provides a sense of what the Think-Pair-Share activities were like in the experimental condition.

After reading the selection, the teacher asked the students to turn to their partner and discuss the following question: In your opinion what was the worst thing that happened to Minnijean Brown? Students had an opportunity to think about the answer and discuss it with their partner. The teacher asked each pair to share the answers they discussed when they were working with their partners. The response was very strong, with over half of the students' hands going up to share what they had discussed. Many students felt "being called the 'N' word" was the worst thing that happened. Another student said he and his partner debated between two incidents, "when she got beaten up or the events that occurred on the first day when Minnijean walked into the school for the first time." They decided that the first day was worse. They justified their choice by saying that on the first day more than one person was involved in the incident. Another student commented, "She was harassed and wasn't treated the way they were supposed to be." This student also stated, "The principal and teachers were not really nice." The teacher related this statement to a prediction that was made earlier in the week about how different groups of adults and children responded during the Civil Rights Movement (Observational field notes, Nov. 3, 1999).

In addition, when students read a brief passage from a text or a magazine or other primary source, experimental students alternated reading aloud to each other one paragraphs worth of material each. Students were asked to help each other out if their partner could not read a word or did not understand what a word meant. This was a means for ensuring that students with LD could receive immediate feedback from a peer if they were having difficulties reading a section.

INSTRUCTION IN THE COMPARISON CONDITIONS

The comparison group received the identical curriculum content (in terms of videos, questions, and exercises), but the teacher did not utilize the interactive instructional delivery procedures specified earlier. Instruction differed only in the following two specific ways. In the experimental condition, students worked in pairs for Think-Pair-Share activities, compare-contrast activities, and for oral reading of passages from the text or magazines. In the comparison conditions, students answered questions on their own prior to class discussion, worked on compare-contrast activities alone, and read each passage individually.

In the comparison condition, teachers asked all questions at the end of each day's 12- to 15-min video segment, whereas in the experimental condition, the questions were interspersed every 3 to 5 min. In all other respects, instruction was identical.

MEASURES

To determine if students in the experimental group learned more than in the comparison group, our primary challenge was to develop measures that would be sensitive to potentially subtle differences in learning given that the comparison group also received what we felt was a strong curriculum. Our hope was to uncover differences in students' level of understanding, which we predicted would be higher among students in the experimental condition. We developed three measures of content acquisi\tion for this purpose. Two measures were constructed along lines of traditional assessments used in content area classes: a vocabulary-matching task and a Written Exam measure that included both short answers and paragraph essays.

Because students with LD display consistent problems in writing (Baker et al., 2002) and on multiple choice or matching tasks, we included a third measure that did not rely on students' ability to demonstrate their understanding in these traditional ways. This third measure was a Content Interview, which required students to verbally articulate their understanding of the CRM in a one-on-one setting with a project staff member. The content interview measure was the only measure of the three that gave students the opportunity to explain their answers and elaborate on responses that were ambiguous or incomplete. Examiners were trained to probe student responses in a way that gave the assessment a more interactive feel. We thought this might benefit students with LD in particular, because of the difficulties these students have matching the completeness of their responses with the demands and expectations of traditional assessment tasks. Together, these three measures differed in how students were asked to respond, and most importantly, they differed in the opportunities they presented to assess the depth of knowledge students learned over the course of the intervention. All three measures were administered at posttest to the special education students, who were our primary sample. The vocabulary and written measures were given to the average ability students.

Written Exam. This measure was typical of exams given in middle school and was administered at posttest to all students. The measure consisted of six questions answered in a brief paragraph or paragraphs. Inter-scorer reliability for the measure was .89; coefficient alpha reliability was .73. Both are more than adequate for experimental design research involving groups (Shadish et al., 2002).

Scoring the Written Exam began with project staff members writing a series of bulleted points that indicated a strong response. Then, two scorers read a set of student responses to identify anchors for each of the scores. We began with anchors for the highest and lowest score for each item and then created anchors for the middle score(s). Grammar and writing mechanics were not part of the scoring criteria. Questions 1, 2, 4, and 6 used a 4-point scale; Questions 3 and 5 used a 2-point scale.

All identifying information was redacted from the exam papers, which were assigned identification numbers so that scorers were not able to identify students, instructional condition, or special education status. Two researchers scored all papers independently. For the initial scoring, the researchers checked their scores after marking several protocols to make sure they were applying the scoring criteria consistently. Then, two examiners scored each question independently. Interrater correlations for the four questions that used a 4-point scale ranged from .84 to .93, items were rescored if there was a discrepancy of 2 points or higher; this happened six times.

In the interview, students had to answer the following posttest question both in writing and orally: "Why do you think the Eyes on the Prize documentary began with the Emmett Till segment?" Woven implicitly throughout the Eyes on the Prize documentary was the critical idea that the popular media's depiction of specific events in the southern states had a profound influence in helping to create and sustain the movement. According to the documentary, Emmett Till was an impetus for the CRM, not because of the brutality of his murder, as horrific as it was, but because the event literally marked the first time that the popular media gave widespread coverage to specific acts of racism in the South. Their coverage served to portray southern racism in a way that average Americans across the country found abhorrent. Students needed to understand the fact that the media's focus on Emmett Till was important in launching a pattern of media attention that became an integral part of the movement.

To illustrate their understanding, students were allowed to provide their knowledge both orally and in writing. The following is the written response of Jaime, one of the students with LD, whose overall score was just slightly below the mean for the sample of students with LD:

"Because he got deformed from the bets. And he was from the south and had a open cascet (sic) ceremony." We keyboarded in all the written responses and corrected any errors due to spelling so that spelling or handwriting would not bias scoring. Thus, our entry looked like this:

"Because he got deformed from the (unclear) [Note: Emmett Till responded to a challenge from a friend to go into the store and talk to a white woman. His body was found in a river, badly deformed]. And he was from the South and had an open casket ceremony."

Content Interview. This measure adapted procedures developed by Campione and Brown (1987). The interactive nature of this assessment gave students the strongest opportunity to demonstrate the extent of their learning. The Content Interview was administered at posttest to all students with LD and a handful of students without disabilities (for exploratory purposes). The five questions were similar in scope to the Written Exam questions. One item was asked in both the written exam and the Content Interview: Why did the Eyes on the Prize documentary begin with the Emmett Till segment? The prompts interviewers used to probe responses were standardized in order to quantitatively score the interview in a reliable fashion.

Scoring. In general, raters disregarded inaccurate information if the general theme of the response was on target. As with the Written Exam, the two raters developed a series of anchors to guide their scoring. In this case, all items were scored on a 0 to 3 scale. Raters who were blind to intervention condition scored all responses. If there was a minor discrepancy in scores between the two raters, the student received the average of the two scores for that item. Interrater correlations were computed for each item. Items with a correlation below .80 were rescored. The two raters were given those items on which there was not agreement. They were not given their original scores and were asked to rescore the item. After rescoring, the correlation was again calculated. For all items, the interrater correlation exceeded the .80 criterion.

Oral Responses. Following is a sample interview response from the same student, Jaime, to demonstrate the types of prompts used, and to provide a sense of the nature of responses by many of the students.

(Interviewer) I: Why do you think the Eyes on the Prize documentary began with the Emmett Till segment?

Jaime: Maybe because that started the Civil Rights Movement. [Note how this initial response is essentially inaccurate. Note, too, the lack of confidence. Below are the series of prompts and Jaime's responses: Note how Jaime does demonstrate his knowledge when presented with a series of prompts that were nonjudgmental and did not provide evidence regarding the correct answer.]

I: How did it start it?

Jaime: Maybe because they just got mad about it and they decided to do something about it.

I: Who got mad?

Jaime: The African Americans. And so . . . because they killed him. They killed Emmett Till. And they were like tired of it so they like tried to do something.

I: O.K. What did they try to do?

Jaime: They would like try to go to marches and stuff. They were . . . so they would be able, wouldn't be like discriminated.

I: Tell me a little bit about Emmett Till. Who was he, and what happened?

Jaime: He was a boy from the North. They weren't really racist up there. So, he wasn't used to that. In the South, he went down to visit and he talked to a European American and he got killed just for doing that.

I: And then what happened to the people who killed him?

Jaime: They didn't get . . . they weren't accused of it. But later TV like stations paid them to tell the truth, their side of the story. And they said that they did kill him.

Note how much richer Jaime's response becomes when he is able to answer the identical question through an interactive interview format rather than in writing. Although the initial response is weak, because the interviewer was able to ask a very pointed question Jaime was able to provide more information about the topic. The incremental nature of the experimenter-student interaction allowed Jaime to demonstrate what he knew about the relevant content.

Jaime addresses, albeit indirectly, the issue of the media's involvement in publicizing the brutal murder of Emmett Till, which served as a critical role in the CRM. Jaime's response expresses an awareness of the media's role and includes detail of attitude differences about race between the North and South at that time. He also is clearly trying to express his outrage at the murderers being declared innocent. He says, "And they didn't get . . ." and seems unable to come up with a suitable word for the thought he is striving for (i.e., a word such as "justice" or "convicted"). He also says "They weren't accused of it" when he clearly means, "They weren't convicted for it." Finally, he never explicitly answers the question of why the documentary started with the event. Thus, his score is in the middle range. Yet, he does show understanding of key issues covered in the unit.

Matching Test. The Matching Test contained 18 terms and provided a rough snapshot of students' knowledge of definitions and critical terms and key figures associated with the CRM. Items included key vocabulary, core concepts such as integration and boycott, terms such as Jim Crow Laws and key figures such as John Lewis. The Matching Test was administered to all students at pre- and posttest. Coeffic\ient alpha internal consistency reliability was .86.

FIDELITY OF IMPLEMENTATION

Weekly observations were conducted in both instructional conditions by an observer with extensive experience observing and coaching teachers. Before each observation, the observer conferred with the teacher to ascertain the day's lesson. The daily lesson plan listed the lesson's activities. The observer wrote the titles of the activities on the observation protocol and during the lesson recorded whether these activities were covered.

The observer also took detailed notes throughout the lesson. In part, these notes represent documentation of how the teaching unit and instructional strategies in the experimental group were actually implemented. Observation notes described the nature and degree of students' responses, the extent of their participation, interactions between partners, and attitudes toward instruction. These notes documented any instances where the experimental and comparison teachers used any strategy reserved for the other instructional condition. We instructed the observer to discuss any deviations from the lesson plan; however neither teacher deviated from the instructional delivery procedures specified in the guide for each condition during either round of the study implementation. Teachers in the comparison condition never incorporated partnerreading, Think- Pair-Share (McTighe & Lyman, 1988), or inserted questions or clarifications into their lessons or showing of videos. In addition to the weekly fidelity observations, the two senior authors also observed each classroom more informally approximately once or twice a week for both rounds of the study. We did not detect any deviations from the prescribed instructional procedures.

DATA ANALYSIS

We conducted three hierarchical multiple regressions following procedures described in Pedhazur (1997). In each case, we first entered the Matching pretest score (X^sub 1^), next a dummy variable (X^sub 2^), which was coded 1 for experimental, 0 for comparison) and, finally a term to represent potential interactions, X^sub 3^ which is the product of X^sub 1^ and X^sub 2^. This approach allowed us to statistically control for pretest variables on the Matching Test, and to explore whether there were any interactions between scores on the Matching test and the intervention.

RESULTS

POSTTEST PERFORMANCE: STUDENTS WITH LEARNING DISABILITIES

For the Content Interview, the main effect for intervention condition was significant, t(32) = 2.244, p = .032. The interaction was nonsignificant, t(32) = 1.925, (p = .063). The vocabulary matching test was a significant predictor for the whole sample, t(32) = 3.804, p = .001. However, when we computed separate correlations for each sample, we found the pre- and postcorrelation was .028 for the intervention group, which is not significantly different from zero. In contrast, the correlation was .624 for the comparison group, which was significant, and moderately high.

For the written essay test, again, the effect for the intervention condition was significant, t(32) = 2.428, p = .021. The interaction was also nonsignificant, t(32) = 1.463, p = .153. The effect size was 1.00, indicating a large effect. Pretest scores on the matching test were again significant predictors, t(32) = 2.772, p = .009; and the analysis statistically controlled for any potential impact for pretest discrepancies on the matching measure. The correlation between the matching pretest and essay scores was .1 (not significant) for the intervention sample, but a significant .63 for the comparison sample, again suggesting a moderately strong relationship between prior background knowledge and final essay test performance for comparison students.

For the Matching Test, the same multiple regression was conducted. In this case, the interaction between pretest scores and posttest scores was significant, t(32) = 2.814, p = .008. For this measure, the pretest scores explained 53% of the variance for the comparison group, which is large and significant. In contrast, pretest scores explained 1% of the variance of the intervention sample.

POSTTEST PERFORMANCE: AVERAGE-ABILITY STUDENTS

Table 2 also presents posttest performance for the average- ability students (without LD) on (a) the written exam and (b) the matching/vocabulary test. (These students were not administered the Content Interview measure, because of its individually administered format, and because students without disabilities were not the major focus of the study.) We initially analyzed the data with the same Pedhazur (1997) procedure. However, unlike the LD sample, these data were not adequately explained by the multiple regression. On the Written Exam measure, none of the three terms explained significant proportions of the variance. For the interaction between pretest matching test score and condition, β was .41, t(36) = .802, p = .428, the β was -.09 for intervention condition, t(36) = .200, p = .842, and for the pretest, β was .328, t(36) = 1.682, p = 10. Given the fact that students were randomly assigned to conditions, we saw the t test on posttest scores as the only analysis option. Results indicated a significant effect favoring the intervention group, t(38) = 2.698, p = .01. At .76, the effect size on this measure was sizeable and favored students in the intervention condition. For both students with and without disabilities, the largest impact of the intervention was evident on the Written Exam measure, but only for students with LD was the impact statistically significant.

On the Matching test for average-ability students, neither the interaction nor the main effect were significant, t(36) = .257, p = .799, and t(36) = .129, p = .898. Nor did the pretest significantly predict posttest performance, t(36) = 1.239, p = .223. A t test conducted on the raw means was also not significant, t(38) = .99, p = .33. Note, however, that students in both conditions score close to a perfect score (18) on this measure at posttest, indicating a potential ceiling effect.

Analysis of Teacher Effects. Although counterbalancing and monitoring fidelity of implementation precluded each teacher's effectiveness from biasing intervention impact, we were curious about the relative effectiveness of the two teachers across conditions. We conducted t tests to examine whether either teacher was more effective with the students with LD, regardless of experimental condition. These analyses indicated nonsignificant differences for students with LD on each of the criterion measures. Effect sizes were .32 on the Matching test to -.18 on the Written Exam and -.14 on the Content Interview. The results suggest that one of the teachers may have been somewhat stronger in the area of understanding relationships, while the other teacher may have been somewhat stronger in explicitly conveying key facts and terminology to her students, regardless of condition. However, these effects are negligible on two measures and small on the more factual/ definitional measure. In any case, counterbalancing ensures that these minimal differences did not confound results.

DISCUSSION

This study demonstrates that students with LD can learn relatively complex grade-level material in American history when they are provided with instruction that supports active involvement in the learning process and materials that do not rely heavily on textbook reading and are adjusted and monitored for comprehensibility. Students with LD demonstrated superior levels of performance on both written and oral examinations that asked them to discuss key issues and figures in the CRM. Effect sizes on both measures were large with values of 1.0 on the Written Exam measure and .72 on the Content Interview measure. Although impact was not significant on the Matching Test measure, which involved knowledge of definitions of key terms and key figures, the moderate effect size (.56) favored students in the experimental condition.

TABLE 2

Student Performance for Both Samples on Posttest Criterion Measures

The research design did not allow us to discern the impact of the CRM curriculum content independent of instructional delivery. The descriptive data suggest that when students with LD are provided with accessible curricula materials but not highly interactive instruction, they learn some of the material, but significantly less than when they are also provided with specialized delivery techniques. To determine impact of the CRM content in isolation would have required a third group using traditional textbooks and essay exams. This did not seem feasible or justifiable, given that approximately half of the students would certainly have had major difficulties reading grade-level texts independently, the way they would be expected to in typical middle school classrooms.

In terms of the CRM focus, we see enormous potential in covering critical issues and topics in 20th Century history with middle school students both with and without LD. Contemporary material is more easily accessible to them in a variety of formats, and relevance to other current topics can be more easily drawn. Regarding the CRM specifically, View (2004) draws attention to the pulse of its powerful narrative dimension noting, "The real story of ending formal racial segregation in the United States is a very human story; one that includes strategic brilliance, logistical messiness, exalted joy, heart-rending sorrow, unbelievable courage, sharp tactical conflicts, and near-religious personal transformations" (pp. 28-29).

IMPLICATIONS OF FINDINGS

On the Written Exam measure and the Matching Test measure, the performance of students with LD in the experimental condition was similar to the performance of average-ability students in the comparison condition (see Table 2). This provides evidence that curriculum access for students with LD was comparable to the access provided average achiev\ing students. Average-ability students appeared to benefit from the instruction provided in the experimental condition; they significantly outperformed students in the comparison condition on the written test, which required the ability to integrate material from the unit. The effect size was large, .70. Taken together, the findings from this study lead us to conclude that the type of instruction provided in the experimental condition is likely to be helpful to the special education students, while, in no way, creating any problems for the average-ability students in terms of content acquisition.

Our LD sample size was large enough to allow us to perform some secondary analysis. Pearson correlation coefficients indicated, with traditional instruction, students who began the study with the most background tended to benefit the most from the unit. However, this was not the case for LD students in the intervention group. The interactive instructional techniques seemed to have the effect of leveling the playing field for students in the experimental condition. Even students beginning with minimal knowledge, were able to benefit from the lessons.

IMPLICATIONS FOR TEACHING AND FURTHER RESEARCH

Because the students were randomly assigned to instructional conditions, teachers were counterbalanced across the two conditions, and implementation fidelity was carefully monitored, the study allows us to infer with a reasonable degree of confidence that the combination of instructional variables in the experimental condition led to the superior outcomes. In terms of the relevance of the findings for day-to-day instruction in middle school classrooms, our sense is that much of the approach we tested would be feasible for inclusive middle school classrooms. Although a great deal of time went into the preparation of this teaching unit, it seems reasonable to apply the instructional framework to the same and other content in the context of normal realities middle school teachers face in preparing lesson content. These types of instructional techniques seem particularly appropriate in courses such as American history for students with LD, where a wide array of curricula could be used, where there could be only minimal reliance on independent reading of grade level textbooks, and where considerable focus on techniques that foster dialogue, analysis, and verbal interaction with peers and the teacher about the content and concepts fundamental to understanding could lend instructional focus. The use of videos, DVDs, and Web site material would seem to be strong content sources for teaching contemporary history. In our project, the use of multimedia for teaching 20th Century history was possible due to the plethora of material available on the CRM.

The results support students with LD working in heterogeneous pairs on a range of activities that are carefully structured for them. Our observations also suggest that students with LD benefited from the sustained, ongoing interactions with their partners. The experience also seemed to benefit the average-ability students, as evidenced by our field notes and their higher scores on the essay test. History may be a powerful venue for peer-mediated learning, in part because the content is accessible on many levels and thus relevant to a wide range of students. Moreover, content on topics involving human rights and attempts to resolve human conflict would be of high interest to virtually all students. One implication of the study for teaching history is to include activities that involve the consistent use of heterogeneous student pairs to process lesson content.

Similarly, the intentional use of redundancy (e.g., presenting both a magazine article and video on a key figure or key event) seemed quite beneficial in helping students solidify their knowledge and subtly introduce them to alternate interpretations of events. This approach, and the "How would you feel if" questions, could be easily integrated into the teaching of many historical topics. We strongly recommend that teachers intersperse questions or brief discussions while showing videotapes or DVDs, especially when presenting rather abstract material in history or social science. Rather than disrupting the flow of the content, as one might reasonably fear, inserted questions (Andre, 1979) appear to slow things down so that students have more time and opportunity to actively process troubling, unfamiliar, and often abstract content.

Significant findings on both the Written Exam and Content Interview documented that the instructional approach led to better understanding of content associated with the CRM. We would encourage teachers and researchers to experiment with using any or several of the specific techniques we incorporated in the experimental condition, because we think that all of the specific techniques used show promise, and the combination was effective for both the special education students and the students without disabilities. Future research might help unpack the relative contributions of the individual components as well as examine the ease of implementation of the various techniques with a large sample of practicing classroom teachers. Our sense is that most of the techniques will be feasible to implement, although high-quality implementation will likely vary considerably among teachers.

[S]pecial education students have received superficial access to topics in the general curriculum, but this access is rarely meaningful and it rarely covers the key concepts that students without disabilities are expected to learn.

Presenting a historical topic as an engaging story helps students understand the events and details because the familiar narrative structure gives them a way to process the content, make predictions, and to draw causal inferences.

This study demonstrates that students with LD can learn relatively complex grade-level material in American history when they are provided with instruction that supports active involvement in the learning process and materials that do not rely heavily on textbook reading and are adjusted and monitored for comprehensibility.

REFERENCES

Andre, T. (1979). Does answering higher-level questions while reading facilitate productive learning? Review of Educational Research, 49, 280-318.

Baker, S., Gersten, R., & Scanlon, D. (2002). Procedural facilitators and cognitive strategies: Tools for unraveling the mysteries of comprehension and the writing process, and for providing meaningful access to the general curriculum. Learning Disabilities Research and Practice, 17, 65-77.

Campione, J., & Brown, A. (1987). Linking dynamic assessment with school achievement. In C. S. Lidz (Ed.), Dynamic assessment (pp. 82- 116). New York: Guilford.

Carnine, Steely, D., & Silbert, J. (1996). Understanding U.S. History-Reconstruction to world leadership 2. Eugene, OR: University of Oregon.

Denckla, M. B., & Rudel, R. (1976). Rapid automatized' naming (ran): Dyslexia differentiated from other learning disabilities. Neuropsychologia, 14, 471-479.

DeVinney, J. A. (Writer/Producer). (1991). Eyes on the prize- America's civil rights years, 1954-1965 [Documentary] . United States: Blackside Inc.

Elbaum, B., Vaughn, S., Hughes, M., & Moody, S. W. (1999). Grouping practice and reading outcomes for students with disabilities. Exceptional Children, 65, 399-415.

Englert, C. S., Raphael, T. E., Anderson, L. M., Anthony, H. M., & Stevens, D. D. (1991). Making writing strategies and self-talk visible: Cognitive strategy instruction in regular and special education classrooms. American Educational Research Journal, 28, 337- 372.

Ferretti, R. P., MacArthur, C. D., & Okolo, C. M. (2001). Teaching for historical understanding in inclusive classrooms. Learning Disability Quarterly, 24, 59.

Gersten, R. (2005). Behind the scenes of a research study. Learning Disabilities Research & Practice, 20, 200-212.

Gersten, R., Baker, S., Smith-Johnson, J., Peterson, A., & Dimino, J. (2003). Providing meaningful access to the general curriculum through use of design experiments (Technical Report 03- 02). Long Beach, CA: Instructional Research Group.

Gersten, R., Fuchs, D., Williams, J., & Baker, S. (2001). Teaching reading comprehension strategies to students with learning disabilities. Review of Educational Research, 71, 279-320.

Graham, S., Harris, K. R., MacArthur, C. A., & Schwartz, S. (1991). Writing and writing instruction for students with learning disabilities: Review of a research program. Learning Disability Quarterly, 14, 89-114.

Hasbrouck, J. E., & Tindal, G. (1992). Curriculum-based oral reading fluency norms for students in grades two through five. TEACHING Exceptional Children, 24, 41-44.

Jennings, P., & Brewster, T. (1998). The century. New York: Doubleday.

McKeown, M. G., & Beck, I. L. (1994). Making sense of accounts of history: Why young students don't and how they might. In G. Leinhardt, I. Beck & C. Stainton (Eds.), Teaching and learning in history (pp. 1-26). Hillside, NJ: Lawrence Erlbaum.

McTighe, J., & Lyman, F. T., Jr. (1988). Cueing thinking in the classroom: The promise of theory-based tools. Educational Leadership, 45, 18-24.

Palincsar, A. S., & Brown, A. L. (1984). Reciprocal teaching of comprehension-fostering and comprehension-monitoring activities. Cognition and Instruction, 1, 117-175.

Passe, J., & Beattie, J. (1994). Social studies for students with mild disabilities: A progress report. Remedial and Special Education, 15, 227-233.

Pedhazur, E. J. (1997). Multiple regression in behavioral research (3rd Ed.). Fort Worth, TX: Harcourt Brace.

Schatschneider, C., Carlson, C. D., Francis, D. J., Foorman, B., & Fletcher, J. (2002). Relationships of rapid automatized naming and phonological awareness in early reading development: Implications for the double deficit hypothesis. Journal of Learning Disabilities, 35, 245-256.

Shadish, W. R., Cook, T. D., & Campbell, D. T. (2002). \Experimental and quasi-experimental designs for general causal inference. Boston: Houghton Mifflin.

Shinn, M. R., Good, R. H., Knutson, N., Tilly, W. D., & Collins, V. (1992). Curriculum-based measurement of oral reading fluency: A confirmatory analysis of its relation to reading. School Psychology Review, 21, 459-479.

U.S. Department of Education. (1997). IDEA proposed regulations, 42 fed. Reg. 55026, 55115, to be codified at 34 c.F.R. 300.660- 662.

U.S. Department of Education. (2004). IDEA proposed regulations, 42 fed. Reg. 55026, 55115, to be codified at 34 c.F.R. 300.660- 662.

View, J. (2004). Teaching civil rights: Putting the movement back in the story. Education Week, pp. 28-29.

Wagner, M., & Blackorby, J. (2002, July). Measuring student outcomes. Paper presented at the 2002 OSEP Research Project Director's Conference, Washington, DC.

Woodcock, R. W. (1998). Woodcock reading mastery tests-revised. Circle Pines, MN: American Guidance Service.

Zigmond, N., Wolery, R., Meng, Y., Flumer, D., & Bean, R. (1994, April). Students with learning disabilities in elementary-level mainstream history classes: What gets taught? What gets learned? Paper presented at the annual meeting of the American Educational Research Association, New Orleans, LA.

RUSSELL GERSTEN

Instructional Research Group

SCOTT K. BAKER

Pacific Institutes for Research

JOYCE SMITH-JOHNSON

Eugene 4J School District, Oregon

JOSEPH DIMINO

Instructional Research Group

ANNE PETERSON

Eugene 4J School District, Oregon

ABOUT THE AUTHORS

RUSSELL GERSTEN (CEC CA Federation), Director, Instructional Research Group, Signal Hill, California. SCOTT K. BAKER (CEC OR Federation), Director, Pacific Institutes for Research, Eugene, Oregon. JOYCE SMITH-JOHNSON, Teacher, Eugene 4J School District, Eugene, Oregon. JOSEPH DIMINO (CEC CA Federation), Research Associate, Instructional Research Group, Signal Hill, California. ANNE PETERSON (CEC OR Federation), Teacher, Eugene 4J School District, Eugene, Oregon.

Address all correspondence to Russell Gersten, Instructional Research Group, 2525 Cherry Ave., Suite #300, Signal Hill, CA 90755 (e-mail: rgersten@inresg.org).

This research was supported in part by a Grant H324C9900063 from the Research to Practice Division of the Office of Special Education Programs of the U.S. Department of Education.

The authors woiild like to acknowledge the extremely useful feedback we received from Margaret McKeown, Susan Unok Marks, and Steve Graham. We would also like to thank Madhavi Jayanthi, Lana Edwards, Jonathan R. Flojo, and Michelle Spearman of Instructional Research Group for their editorial assistance and Susan Unok Marks for her contribution to this work. We also would like to acknowledge the vibrant support for the project provided by Paul Jorgensen, building principal at Eugene 4-J School District.

Manuscript received October 2004; accepted May 2005.

Copyright Council for Exceptional Children Spring 2006

Source: Exceptional Children

More News in this Category