Printer Friendly

The effectiveness of modified cognitive strategy instruction in writing with mildly mentally retarded Turkish students.

Expository writing involves highly complex metacognitive skills and cognitive process (Englert & Mariage, 2003). Two important types of knowledge influence expository writing performance: metacognitive knowledge that guides self-regulation and text structure knowledge that guides writers in generating and organizing ideas to form well-organized texts (Englert & Mariage, 1991).

Expository writing is challenging for many students because it requires an explicit awareness of how ideas can be produced in a conventional form that communicates meanings to a distant audience (Kozulin, 2003). It is especially difficult for students with cognitive or learning disabilities (LD); (Englert, 1990). They typically have difficulty with the metacognitive aspects of regulating the writing subprocesses used to plan, draft, monitor, and revise expository texts (Englert, Raphael, Fear,& Anderson, 1988; Graham, Harris, MacArthur, & Schwartz, 1991; Wong, Wong, & Blenkisop, 1989). They also lack knowledge of the basic text structures needed "to produce well-organized texts (Englert, Raphael, Anderson, Gregg, & Anthony, 1989; Englert & Thomas, 1987; Thomas, Englert, & Gregg, 1987).

Students with mental retardation have a number of cognitive difficulties (Arabsolghar & Elkins, 2000) that further impede their performance in planning, organizing, and producing expository texts. They often fail to use effective memory and rehearsal strategies (Bray & Turner, 1986; Ellis, 1970; Engle & Nagle, 1979; Spitz, 1966; Turner, Dofny, & Dutka, 1994), and they do not spontaneously organize, chunk, rehearse, or elaborate on information in ways that facilitate learning (Belmont & Butterfield, 1971; Borkowski & Wanschura, 1974; Campione & Brown, 1977; Ellis; Kellas, Ashcraft, & Johnson, 1973; Spitz). In addition, these students typically process information more slowly (Banikowski & Mehring, 1999) and fail to establish meaningful relationships among sets of ideas.

Literacy instruction for students with mental retardation usually focuses on the teaching and mastery of isolated and linear sets of subskills (Katims, 2001). However, skilled writing requires the use of a wide array of cognitive processes in a more holistic manner (Englert, 1992). In the present study, Turkish students with mild mental retardation were taught cognitive strategies for planning, organizing, drafting, and monitoring their expository texts.

During the past 15 years, two writing intervention programs have examined the effectiveness of writing strategies at both the elementary and secondary level with children with disabilities (mainly students with LD). One program was developed by Englert and her colleagues (Englert, Raphael, Anderson, Anthony, & Steven, 1991). The Cognitive Strategy Instruction in Writing Program (CSIW) is based on four central principles drawn from a sociocultural theory of instruction (Englert & Mariage, 2003). First, this framework emphasizes the importance of immersing writers in a holistic and cognitive process in which they apply strategies related to planning, organizing, writing, editing, and revising their expository texts (Englert, 1992). Second, the CSIW framework emphasizes the importance of teachers modeling aloud strategies for these cognitive processes as they compose specific types of text (Englert, 1990; Englert & Raphael, 1988). Third, teachers apprentice students in the strategies through the use of interactive dialogues in which they prompt, scaffold, and guide students through the application of the strategies (Englert, 1990; Englert & Mariage, 1991). Fourth, teachers make text structures, writing process, and strategies visible through a series of think-sheets that provide students with structural or procedural support at each stage of the writing process by using graphic organizers, prompts, and questions that cue strategy application and self-regulation (Englert, 1990; Englert & Raphael, 1989; Graham, MacArthur, & Schwartz, 1995).

In a series of studies evaluating the effectiveness of CSIW, the expository writing skills of elementary and secondary students with LD showed significant improvement (Englert, Raphael, & Anderson, 1992; Englert et al., 1991; Hallenbeck, 1996, 1997, 2002). Compared to control students, CSIW students demonstrated better writing performance on text structures they were directly taught, transferred their knowledge to untaught text structures, and displayed greater metacognitive knowledge about writing (Englert, et al., 1992). In one study of secondary students, several students adapted the strategies they were taught to improve their performance in other high school classes (Hallenbeck, 1996).

The second instructional approach, Self-Regulated Strategy Development (SRSD), was designed by Graham and Harris and colleagues (Graham et al., 1991) to focus on the development of composition and self-regulation strategies (Harris, Graham, Mason, & Saddler, 2002). Research studies evaluating this approach demonstrated that when students with LD are taught specific strategies for planning and revising texts in combination with procedures for regulating these strategies and the writing process, there is an improvement in the quantity and quality of their writing (Danoff, Harris, & Graham, 1993; De La Paz, 1999, 2001; De La Paz & Graham, 1997; Graham & Harris, 1989; Graham & MacArthur, 1988; MacArthur, Schwartz, & Graham, 1991; Sexton, Harris, & Graham, 1998; Stoddard & MacArthur, 1993; Troia, Graham, & Harris, 1999). In a meta-analysis of SRSD writing studies, the average effect size for students with LD was 1.14 for quality, 1.86 for length, and above 2.0 for structural elements (Graham & Harris, 2003).

SRSD instruction is designed to promote students' independent use of the target writing strategies and accompanying self-regulation procedures (i.e., goal setting, self-monitoring, self-reinforcement, and self-instruction). Instruction is scaffolded so that responsibility for recruiting and using these strategies and self-regulation procedures gradually shifts from instructor to students. Students are treated as active collaborators in the learning process, and the role of student effort in learning the strategies is emphasized and rewarded. Feedback and instructional support are individualized by the instructor so that they are responsive to students' needs. Furthermore, instruction is criterion based rather than time-based (Graham & Harris, 2003; Harris & Graham, 1996).

In the current study, I modified the CSIW model (Englert et al., 1991) and applied it with adolescents with mild mental retardation attending self-contained classes in Turkey. The effectiveness of writing strategy instruction has not been evaluated with this group of students (the only exception involved a single student in De La Paz & Graham, 1997). As in other CSIW strategies, a set of think-sheets was used to help students plan, organize, edit, and revise their expository writing (Englert et al., 1992). The instruction consisted of four phases: (a) text analysis focusing on the text structure targeted for instruction (i.e., problem/solution text), (b) modeling of the writing process, (c) guided practice, and (d) independent writing. This included modeling aloud how to use the think-sheets and accompanying strategies, such as brainstorming. Students were also provided with assistance until they could apply them independently. During modeling and guided practice, dialogue about text structures and strategy use were encouraged and supported.

A primary way in which the CSIW model was modified was by incorporating specific features from the SRSD approach (Harris & Graham, 1996) into the instructional regime. These included promoting of student self-regulation (e.g., memorizing of strategy steps, self-organizing, self-editing, self-revising), criterion-based instruction, and individual feedback according to students' needs. In addition, the instructional sequence was consistent with the SRSD instructional format including emphases on modeling, guided practice, and independent practice with progress from one phase to the next based on students' attainment of specified criteria.

In summary, this study expands previous research in three ways. One, it examines the effectiveness of strategy writing instruction with students with mental retardation. Two, it is the first study to examine the effectiveness of a writing intervention that employs elements from both CSIW and SRSD. Three, it extends research on writing strategy instruction to a different culture and language. This provides an important test of the broad generalizability of such instruction. Finally, in addition to examining the immediate impact of writing strategies instruction, maintenance over time was assessed as well.



The participants of the research were chosen from two self-contained and multiage classrooms for students with mild mental retardation in Ankara, Turkey. In Turkey, education for mildly mentally retarded students occurs in self-contained classes.

Each participant met the following criteria: (a) read 70 or more words correctly per minute, (b) read and understand a problem/solution text (students read and answered 7 items from a text written at the fourth-grade level), and (c) write a personal narrative about what they did during the weekend using five or more sentences. This ensured that reading would not be a major impediment to learning the target strategies, and that students already were able to produce connected text (making it more likely strategy instruction would be appropriate).

Eight students from two self-contained classes were initially tested, with four students meeting all three criteria. An initial examination of these students' pretest problem/solution texts revealed that their written ideas were poorly organized and lacked coherence. They did not identify the writing topic and their texts did not contain sufficient information to develop their topic. With the exception of one student, their texts averaged less than 40 words. In addition, the students did not plan their text, and when I asked them to revise their papers, they made only minor surface level corrections (e.g., capitalization, punctuation, spelling).

The four participants were all males, ranging in age from 13 to 17. Turkish was the native language of all students. The first student was 17 years, 6 months old at the beginning of the study and in the eighth grade. His full-scale score on the Turkish version of the Wechsler Intelligence Scale for Children (WISC-R; Savasir & Sahin, 1994) was 67. The second student was 15 years, 6 months old at the beginning of the study and in the eighth grade. His full-scale score on the Turkish version of the WISC-R was 71. The third youngster was 15 years, 9 months old at the beginning of the study and in the seventh grade. His full-scale score on the Turkish WISC-R was 65. The fourth student was 13 years, 4 months old at the beginning of the study and in the fifth grade. His full-scale score on the Turkish WISC-R was 65.


Establishing the Structure of the Problem/ Solution Text. A commonly occurring type of text structure that underlies most social studies and science textbooks is problem/solution. This text structure was the focus of the current study and consists of three parts (Armbruster, Anderson, & Ostertag, 1989; Vacca & Vacca, 2001). The first part (problem description) provides information about whom/what the problem affects, how the problem affects those mentioned and the reasons/causes behind the problem. The second part (problem solution) explains the ways of solving the problem in connection with these reasons/causes. The third part (outcome) explains whom/what is affected in the positive sense once the problem is solved, as well as the responsibilities of people and institutions to avoid the problem in the future. Transitional words and transitional sentences are used to convey the elements of text structure related to causes/reasons and solution proposals. Examples of transitional words and transitional sentences include the following: one of the reasons is, there are many reasons, there are many possible solutions, also, another way of solving the problem is, one of the most significant ways of solving the problem is, and besides. Conjunctions (e.g., therefore, so, and, because) are also employed in problem/solution texts.

Determining the Topics for Problem/Solution Texts. For the assessment and instructional phases of the research, 30 problem/solution writing topics were developed based on actual real-world problems. These problems were described in fourth-grade science and social studies textbooks (Gurtan, 1994; Senunver et al., 1999), and they were at a similar level to those used in the students' classroom. Passage content topics that were selected included the following problems: traffic jams, erosion, air pollution, sound pollution, and earthquakes.

For the preinstructional part of the study, material that provided additional background information about the selected writing topics was developed. This was done to ensure students had knowledge about the topics they were asked to write about. This written material was developed so that it conformed to the problem/solution text structure described earlier. I rewrote the science and social studies texts (that served as the source of the material) as problem/solution passages. The passages were texts with four paragraphs. The average passage length was 194 words (range, 173-222 words per passage); the average sentence length was 7.94 words (range 6.96-8.22 words per sentence). The quality and structure of the problem/solution passages were evaluated with the help of an associate professor in English who conducted research on text structure, an assistant professor in the Department of Turkish Language who evaluated text structure of passage and linguistic compliance of the passages in Turkish, and an assistant professor in the Department of Special Education who had written life science textbooks. Based on their suggestions and feedback, the passages were revised and modified.


In the current study, the effects of the Modified CSIW on problem/solution text writing skills were assessed through the use of a multiple probe design across subjects (Tawney & Gast, 1984). The following conditions were in effect during the experiment:

Baseline Probes. During baseline, the participating students' preinstruction response rates on writing problem/solution texts were established. The primary variables of interest were the number of problem/solution elements included in each text, time spent planning, as well as total writing time, text length, coherence, and quality. At the beginning of the experimental process, one baseline probe was administered to all students. Afterward, the collection of probes continued at least three times until the baseline data stabilized in consecutive sessions. An additional baseline probe was also required for each successive student. During baseline, students were told that their texts would be typed and bound so that they could share them with their family and friends. No feedback was provided about content or quality of their texts.

Modified CSIW Instruction. Instruction started for the first student after a stable baseline was established in terms of the total number of problem/solution text structure elements that were included in the composition. Instruction continued until the first student demonstrated independent mastery of the writing strategy and was able to write texts containing each of the basic problem/solution text elements. Instruction was not started for the second student until the first student's problem/solution element score reached the criterion level (a total score of 15 points on a scale measuring text structure elements) or postinstructions' data were at least two and half times the mean number of problem/solution text elements produced during baseline. Identical procedures and criterion levels were applied to the second, third, and fourth students.

Postinstruction Probes. After a student completed the instruction, three to four postinstruction probes were administered following the procedures established during baseline.

Maintenance Probes. For the first student, maintenance probes were administered 3 and 12 weeks following instruction. For the second student, maintenance probes were administered 3, 8, and 12 weeks following instruction. For the third student, maintenance probes were administered 3, 4, and 7 weeks following instruction. For the fourth student, maintenance probes were administered 3 and 8 weeks following instruction. Maintenance probes were administered following procedures established during baseline.


The order of participants was determined via random assignment. The instruction was provided by the researcher, who also administered all writing probes. Throughout the study, each student worked individually with the researcher, who maintained a log, recording the student's writing behaviors. To guide the instruction process, lesson plans were developed for each phase of instruction.

Modified CSIW instruction was applied in four phases. The first, the text instruction phase, provided the tools and the text analysis skills for understanding the problem/solution text structure. In the second, modeling phase, the researcher modeled the writing strategies for the students. The third phase, guided practice, was an interactive phase in which the researcher guided strategy use and offered feedback. In the fourth phase, students worked independently. The researcher also provided preinstruction on the problem topics in order to address gaps in students' background knowledge. Criteria were determined for transition from the modeling phase to guided practice, and from guided practice to independent practice, to ensure that students met the mastery criterion level.

For each student, modified CSIW instructional procedures were applied 3 days per week. One session was conducted per day, and each session continued until instruction was complete. For sessions longer than 45 min, a 10-min break was provided and then the session continued. For each student text structure instruction sessions lasted 30 to 60 min, modeling sessions lasted 50 to 92 min, guided practice sessions lasted 80 to 135 min, independent practice lasted 65 to 110 min. The total instructional time for the first, second, third, and fourth student was 19 hours 31 min, 9 hr 5 min, 19 hr, 12 hr 48 min, respectively. The total number of instructional sessions required by the first, second, third, and fourth student to master the strategy were 14, 10, 14, and 12, respectively. The experimental process was completed in a total of 6 months.

During the study, no information about the instructional intervention was provided to classroom teachers. They were asked to proceed with their programs in the usual way. Teachers carried out reading comprehension exercises in the Turkish language lessons during the experiment.

Preinstruction Development of Real-World Problem/Solution Topics. The problem/solution topics involved actual real-world problems (e.g., pollution, traffic, earthquakes). Because the students had limited knowledge about the problem/solution topics, a preinstruction phase was instituted in order to provide students with basic information about the topics (e.g., causes of the specific problem, effects of the problem, and its solutions). For each experimental session (baseline, text instruction, modeling, etc.), the preinstruction sessions were conducted 3 days before students were asked to write their texts. Topics were presented using pictures and direct instruction. To measure student's knowledge of the topical information following preinstruction sessions, students were asked to answer questions about the taught content (e.g., What is the problem? What are the causes of the problem?). In cases where the student could not respond or responded incompletely, the instruction proceeded each day until the students could answer the questions. Students did not have access to the pictures during the experimental phases of instruction. Daily instruction lasted for 15 min on average.


Phase 1: Text Structure Instruction. In the first instructional phase, the researcher taught the structural elements of the text genre. In the first session, the researcher presented a graphic organizer and highlighted the text features of problem/solution texts. The researcher told the students, "If you want to write a text that informs the reader about problems such as unhealthy nutrition, tooth decay, etc., you should arrange the text according to this structure." Then an overhead transparency with a problem/solution graphic organizer was presented that contained the text structure features (e.g., introduction, causes/reasons of the problem, solutions, outcomes, transitional and introductory sentences, conjunctions, and transitional words). Then using an exemplary text, the researcher identified the text features and elements, providing students with a rationale and purpose statement for each.

In the second through fourth sessions of Phase 1, the researcher presented one example and three nonexamples of problem/solution texts, telling the student that the more examples observed, the better one could write problem/solution texts. The nonexamples were missing particular elements that had proven to be problematic to students based on their performance on the baseline probes (e.g., missing transitional words, problems, causes, introductory sentences, or conjunctions; inadequate solutions; or incorrect time expressions, subject-verb disagreement, and so forth). Over the course of three sessions, the same nine nonexamples were presented to each student. At least one or two of the specific problems mentioned were included in each of the nonexamples.

For each passage, the researcher asked the student to identify the text structure elements that were present or absent based on the graphic organizer and then to evaluate whether the text should be considered well-organized or not. The researcher also highlighted transitions, conjunctions, and introductory sentences. The researcher then encouraged the students to generalize their knowledge of the text structure to other written texts by saying, "Now, you too can decide which text was written well and which text was not, and this will help you write well-organized texts."

Phase 2: Modeling. In the second phase, the researcher modeled and thought aloud about how to write a problem/solution text based on CSIW instructional techniques. The researcher implemented a multicomponent approach to teaching and writing problem/solution texts. First, using a "plan think-sheet," the researcher modeled self-talk and actions related to thinking about audience ("Who am I writing for?"), purpose ("Why am I writing this?"), and activating background knowledge ("What do I know?").

Next, the researcher modeled for the students how to organize the problem/solution text by categorizing the brainstormed information according to the text structure features shown in the graphic organizer. The researcher constructed an organizer on the board and recorded the relevant information from the brainstormed ideas into categories associated with the following questions: "What is the problem?" "Whom/what does it affect?" "What are the effects?" "What are the causes/reasons behind the problem?" "What are the ways of solving the problem?" and "What is the outcome or positive effects of the solution?" The researcher transferred each brainstormed idea into boxes on a graphic organizer, checked off the corresponding ideas on the plan think-sheet, and provided a rationale for its classification. In this way, the researcher modeled how to identify the problem and distinguish between the reasons/causes and solutions, as well as demonstrated the positive effects of the solution and the persons or institutions that might assume responsibilities for preventing the problem in the future. If information was missing from one or more of the text structure categories, the researcher modeled how to self-monitor and use the associated text structure questions to reinitiate the brainstorming process until the organizer was complete.

Subsequently, the researcher modeled how to transform the organized ideas into well-formed problem/solution texts. The researcher introduced the writing process by saying "I am going to write a text about the problem by looking at the information in these boxes. These organized ideas will help me write a well-organized problem/solution paper. Watch me." On a whiteboard, the researcher converted the organized ideas into text, crossing each element off of the organizer as it was incorporated. Throughout the process, the researcher thought out loud about how to transform notes and phrases from the organizer into well-formed sentences. The researcher also referenced the graphic organizer of the text structure and referred students to prompt cards that contained conjunctions, transitions, and introductory sentences for the problem/solution text. At the end of the process, the researcher highlighted the transitional words and introductory sentences in the text.

After drafting the problem/solution text, the researcher modeled self-editing using a checklist by starring the parts of the paper that were well-written and putting question marks by the points that were unclear. A self-editing checklist based on the text structure features was used as a tool to guide the self-editing process. The checklist included 15 questions, nine of which were related to the presence or absence of text elements, four were related to the presence of transitional words and sentences, one related to subject-verb agreement, and one related to the language of the text. For each response to a checklist question that indicated that a particular trait was absent, the researcher modeled how to monitor and revise the text.

Students were required to rehearse the steps of the instruction until they had memorized the modeled stages; they had to name the five steps and related strategies pertaining to planning, organizing, writing, editing, and revising their texts. When they could name the five steps of the writing process and the associated strategies, the students progressed to the guided practice phase. Typically, students required from two to three sessions to attain this criterion.

Phase 3: Guided Practice. During guided practice, the researcher directed and monitored the process while students wrote their own texts, offering feedback according to students' needs. In the initial sessions, the students were allowed to use the graphic organizer representing the text structure, the cards with transitions and conjunctions, and self-editing checklists as procedural facilitators and prompts. By the last session, students were expected to construct their own think-sheets and to write their papers independently without looking at the aforementioned procedural facilitators.

Guided practice sessions were implemented as follows. The students were provided with three blank pieces of paper. First, they were asked to generate a plan think-sheet in order to write their ideas on the first paper. The researcher prompted the students to consider audience and purpose, and then they were asked to brainstorm ideas for their text. Next, students organized their ideas on the second paper. The researcher guided the students in drawing and labeling boxes to create their own graphic organizers and encouraged students to place their brainstormed ideas into the respective organizer boxes. This early emphasis on self-generated organizers was intended to foster internalization of writing strategies. Third, students generated first drafts by looking at the text structure boxes and by consulting the prompt cards associated with the use of transitional words and sentences, introductory sentences, and conjunctions. They crossed off the boxes of the organizer as the text structure elements were incorporated into their texts. Students also used a self-editing checklist to edit and revise their texts.

When students could implement the strategies and processes related to planning, organizing, drafting and editing their papers without assistance, they passed to the independent practice phase. Typically, students required from two to five sessions to attain this criterion.

Phase 4: Independent Practice. During the independent practice phase, the researcher asked the students to perform all aspects of the writing process independently. Positive and corrective feedback was provided when needed, but the researcher encouraged the students to apply the strategies on their own. After completing the text, the student read it aloud, and the researcher provided feedback about the presence or absence of the targeted text structure features (e.g., problem/solution elements, transitional words, transitional sentences and conjunctions). Independent practice was continued until the student was able to write texts that contained all basic elements and demonstrated independent mastery of strategy. Typically, students required two sessions to attain this criterion.


To ensure that instruction was delivered as planned, the following steps were taken. First, the researcher checked off steps from the daily lesson plans as they were completed. Second, a university special education intern familiar with the instructional procedures observed one session of each phase for each student throughout the intervention. During these observations, the intern was seated in the classroom in a position not to draw students' attention and to be able to observe the instructional steps. For days when the observer was present, the lesson plans in the instructional manual were used as a checklist to note completion of strategy steps. In all instructional phases, 100% reliability in treatment procedures was maintained.


The problem/solution topics were randomized across each experimental phase and session. The same topics were used for all four students in the instructional phases and writing probes. A total of seven topics were used in baseline; four topics were used in the text structure phase; ten in the modeling, guided practice, and independent practice phase; four in the postinstruction phase; and three in the maintenance phase. The direction stating the problem topic was written at the top of students' papers in all phases of assessment (e.g., write an informative essay on the problem of air pollution), and this instruction was read aloud by the researcher during the writing probes and instructional phases. During all writing probes, the students were told that they could use as much time and paper as they wanted. Students' behaviors during this process were observed, as well as the total time that students spent during planning and writing their documents.


Prior to scoring, all texts were typed and corrected for spelling, punctuation, and capitalization. All texts written before and after instruction were scored for planning and total writing time, text length, text structure elements, coherence, and quality.

Time. Planning and total writing time were measured using stopwatches. For all texts, the duration of time between the end of the researcher's direction to write and the actual start of students' writing was measured as planning time. The time between the start of the directions and the paper's completion was counted as total writing time.

Text Length. Text length was measured by counting the number of words in students' text. Prior to scoring, all texts were typed and corrected for spelling errors. The length of papers, including the title, was computed via computer.

Text Structure Elements. A scale was developed to score text structure elements. Following procedures recommended by Graham (1990), the text was divided into basic elements. Analytic scoring procedures were used to examine both the element type and completeness (De La Paz, 1999; Troia et al., 1999). For each text structure element, a score of 0 was awarded if the element was not present, a score of 1 if the element was present but only limited in its development, and a score of 2 if the element was fully developed. For example a score of 1 was awarded if only a few causes behind the problem were present, a score of 2 was awarded if most or all of the causes were present. Similarly, a score of 1 was awarded if an introductory sentence was present, and a score of 2 was awarded if the type of problem was explained in the introduction. Only the proposals for the solution to the problem were scored from 0 to 3: A score of 1 was awarded if the student wrote few plausible solution proposals, a score of 2 was awarded if some plausible solution proposals were provided, and a score of 3 was awarded if most or all possible solution proposals were provided. The total possible score on the problem/solution text scale was 15. Following the preparation of the scale, the scale items and scoring were examined by five experts in the field of Turkish language. A Likert-type scale was prepared for expert review to determine the compliance of the element scale items and related scoring criteria to the problem solution text. The items ranged from strongly agree (5) to strongly disagree (1). The experts were given a rewritten problem/solution text designed in accordance with the structure used in this study, the text element scale, and the Likert-type scale. Five experts were asked to rate the representativeness of each element scale item to the text. Experts' rates were greater than or equal to 4.0.

Interrater reliability was calculated for all texts. To determine interrater reliability, Pearson product-moment reliability coefficients were calculated between the scores assigned by two doctoral students in special education. The independent rater was blind to the experimental conditions. The researcher introduced the text element scale to the scorers and provided information about the scoring criteria. The scorers were also provided with problem topic analyses about the problem's effects, its reasons, and solutions for their use in deciding whether elements were appropriately developed. Interrater reliabilities between the scorers were as follows: total number of elements (.98), identifying the problem (.92), information about whom/what the problem affected (.95), information about how the problem affected those mentioned (.97), reasons/causes behind the problem (.87), ways of solving the problem (.92), explaining ways of solving the problem (.96), and outcome (.97).

Coherence. In this study, text coherence was calculated by counting the number of linguistic markers (conjunctions, transitional words, and sentences) that offered coherent relations between text segments (Graham, 1990; Sanders & Noordman, 2000). One point was awarded when these elements were used coherently within the texts. Pearson product-moment reliability coefficients were calculated between the scores assigned by two independent research assistants trained in the scoring procedures. Interrater reliability between the scorers was calculated as .99 for all texts.

Quality. A scale was developed for scoring overall quality. The scale assessed writing quality broadly, as scorers were directed to base their judgments on text structure, organization, text content, cohesion between sentences, sentence structures and variety, word diversity, and word choice. The scale scores ranged from 1 (low) to 7 (high). Two research assistants participated in a 1-hr training session where they practiced applying the scale. When scoring students' writing samples, scorers were asked to read the texts one by one, score them from 1 to 7 according to the scale, and explain which items in the texts affected the scoring. Interrater reliability was calculated for all texts produced by students during testing probes. Pearson product-moment correlation coefficients between the quality scores were .91.


Students' average scores on the writing probes are presented in Table 1. In addition, Figure 1 presents students' scores on the text element scale for each problem/solution text written during baseline, postinstruction, and maintenance.


During baseline, none of the students spent time planning their problem/solution texts in advance of writing (see Table 1). After hearing the instructions, all four students immediately began writing. None of the students engaged in any observable planning activities.

During baseline, the average writing time of students ranged from 14 to 18 min. After writing the problem/solution texts, only one of the four students (the second student) reread his text and made minor edits for spelling mistakes. In addition, the fourth student neatly rewrote his paper without making any changes. The word count of the texts written in baseline phase ranged from 29 to 100 words (see Table 1). The third student's text was over three times as long as the first student's text and twice as long as the other two students' text. This student, however, tended to include irrelevant ideas and repeated his ideas in the same or slightly different sentences.

All students received low scores on their ability to incorporate the problem/solution text structure elements during the baseline probes (see Figure 1 and Table 1). Their average scores on the element scale ranged from 1 to 4.3 (the possible maximum score was 15 points). This indicated an overall poor mastery of the problem/solution text structure elements during baseline. Only 1 out of 21 texts written by students during baseline included 5 of the 7 text structure elements identified. This text was written by the fourth student, who included 5 elements, but all of these elements were rated as undeveloped or incomplete. Except for this text, there was no successful incorporation of the problem/solution elements in the texts written during baseline. None of the texts contained the outcome or an explanation of the ways to solve the problem.


Analysis of the coherence scores of students' texts during baseline revealed low performance levels (see Table 1). Coherence scores ranged from 1.4 to 3.1. None of the students used transitional words and sentences during baseline, and their use of conjunctions was extremely limited (e.g., "and" or "because"). Finally, the quality of texts generated during baseline was low. Quality scores ranged from 1 to 1.6 (see Table 1). A text written by the fourth student in the baseline condition is presented in Appendix A.


Following instruction, there was a dramatic improvement in students' planning behavior (brainstorming and organizing). All of the students employed the planning strategies taught during instruction. The time students spent planning during the postinstruction phase ranged from 16.3 to 42.3 min (see Table 1), whereas planning time during baseline was 0 min.

Furthermore, all students made large gains in the total number of words written. The average increase in the number of words between baseline and postinstruction was 97.8 for the first student, 194.5 for the second student, 125.4 for the third student, and 135.1 for the fourth student. Likewise, the total time spent writing showed substantial improvements. The total average time increased by 75.3 min for the first student, 78.3 min for the second student, 122.5 min for the third student, and 85.1 min for the fourth student.

In addition, students' text structure element scores also improved dramatically (see Table 1 and Figure 1). With a maximum possible score of 15 points, the first student improved his average text structure elements score by 9.6 points, the second student by 13 points, the third student by 8.7 points, and the fourth student by 11.1 points. Whereas none of the students' papers included all the elements during baseline, 10 of 13 texts (76.92%) produced after instruction included all elements. Although the majority of the postinstruction texts included all seven of the problem/solution text elements, the first student showed less substantial improvements in his performance. He failed to include information about the problem's effects in two of three postinstruction texts.

The students' scores for coherence also increased following instruction (see Table 1). All the students employed conjunctions, transitional sentences, and transitional words to produce a coherently developed text. All four students employed conjunctions, transitional words and transitional sentences to signal the location of information and to offer relational statements between the propositions of their papers.

Finally, ratings of the texts' quality improved from baseline to postinstruction. With a possible high score of 7, quality scores increased by 3.3 points for the first student, 5.0 points for the second student, 3.4 points for the third student, and 4.5 points for the fourth student. A text written by the fourth student during postinstruction is presented in Appendix A.


Students completed the maintenance probes at approximately 3, 8, and 12 weeks following instruction. On the maintenance probes, planning and writing time decreased in comparison to postinstruction probes. Nevertheless, because students spent no time planning during baseline, their performance on the maintenance probes revealed a significant improvement in the application of planning strategies over what was exhibited during baseline.

Word counts revealed less consistent results. Except for the fourth student (whose maintenance length equaled or exceeded postinstruction scores), the number of words written during maintenance decreased in comparison to postinstruction, but scores were still much higher than those obtained during baseline. The only exception for these three students was the third adolescent whose first and third maintenance probes were 45 and 28 words longer than an average baseline paper. The second maintenance probe, however, was no longer than an average baseline paper.

For all students, inclusion of the text structure elements on the maintenance probes remained nearly identical to the levels obtained following instruction. None of the students' element scores decreased by more than 1 point on maintenance probes, showing that students maintained an important part of the strategy as a strong basis for constructing well-organized texts that satisfied the problem/solution text criteria.

Although there was a decrease in the coherence scores from postinstruction to maintenance phases for the second and third students, their scores were still substantially higher than their baseline scores. The other two students maintained postinstruction gains.

With the exception of the third student, there was no significant change in the holistic scores for quality from postinstruction to maintenance. The third student's average postinstruction quality score was 5.0, but he received an average score of 3.0 on his second and third maintenance probes. These maintenance scores, however, were still higher than baseline scores.


The present study examined the effectiveness of Modified CSIW for students with mild mental retardation. The intervention blended together elements from CSIW (Englert et al., 1991) and the SRSD model (Harris & Graham, 1996). The intervention implemented in this study was used to inform students about the writing process as well as the role of text structure knowledge in planning, drafting, and revising problem/solution texts.

Modified CSIW instruction resulted in substantial improvements in the length, structural elements, coherence, and quality of students' text. In addition, the participating students with mild mental retardation spent more time planning and writing following instruction. They were able to independently generate and write problem/ solution texts that conformed to the basic structure of this genre (i.e., contained the problem, causes/reasons for the problem, solutions, and outcomes). Furthermore, all students wrote coherently developed texts consisting of at least four paragraphs, with a macrostructure that was supported through the provision of an appropriate introduction, development and body of the paper, as well as a conclusion. These improvements were reflected in substantial improvements in the overall quality of their papers.

Although prior research has focused on students with LD, this study demonstrates that instruction in higher-order writing processes can have a positive impact on the writing of students with mild mental retardation. Following Modified CSIW instruction, I observed that the participating students employed the planning strategies taught, including considering their audience, brainstorming ideas for their papers, grouping ideas into relevant categories using text structure as a basis for monitoring the adequacy of their texts, and editing and revising their written texts accordingly. The use of these activities was not evident during baseline; thus, their approach to writing became more sophisticated following instruction, as they employed the types of strategic behaviors typically used by more skilled writers (Graham & Harris, 2003).

As in previous CSIW work (e.g., Englert et al., 1991), the observed instruction effects were maintained over time. Overall, changes in writing behavior as well as writing performance remained stable for up to 12 weeks following instruction. Although performance on maintenance probes did not always remain as high as performance immediately following the end of instruction, long-term effects almost always eclipsed baseline scores.

In summary, the current study makes an important contribution to the literature by showing that writing strategy instruction can be effective with mildly mentally retarded students. Just as important, this study demonstrates that writing strategy instruction can be effectively adapted for students from different cultural backgrounds, different languages, and different experiences.

Further research is needed to replicate the current findings and to determine the relative contribution of instructional components as well as the variables responsible for changes in students' writing behavior. Although the instructional program implemented in this study was individualized, it should be applicable to larger group of students. For example, Englert et al. (1991) effectively employed similar procedures with groups of students with LD. Additional research is also needed to determine if the instructional program used here will be effective for teaching other text structure beyond problem/ solution text.


Findings from this study provide important implications for teaching writing to students with mild mental retardation. Most notable, directly teaching text structure and writing strategies within a process approach to writing can improve the expository writing of these students. This instruction includes (a) providing preinstruction to help students develop the background knowledge they need for the topics they are asked to write about (this may be especially crucial for students who typically lack such knowledge); (b) focusing students' attention on important textual and genre features by actively presenting and teaching text structures; (c) emphasizing and demonstrating critical writing strategies through modeling and think-alouds; (d) providing additional modeling of strategy use as needed; (e) giving extended practice and guided feedback that help students use the strategic tools they are learning; (f) supporting and scaffolding students' learning through the use of procedural facilitators, prompt cards, text structure maps, and other accommodations (Baker, Gersten, & Scanlon, 2002); (g) structuring interactive dialogues with students to provide additional assistance and learning opportunities in applying the skills and strategies taught (Englert & Mariage, 1991); and (h) modifying instruction through continuous monitoring of students' performance throughout instruction (e.g., such assessments may reveal that students need to spend more time in guided practice in order to master writing strategies). Moreover, to facilitate mastery, instruction should be criterion-based rather than time-based (Harris et al., 2002). Before beginning writing instruction, teachers should establish criteria to determine when students are ready to progress from one step to next, and only when these criteria are met should they move to the next instructional step. It is also important to determine students' prerequisite skills before beginning strategy instruction. This way needed skills or knowledge can be taught before or in conjunction with the target strategies.


Sample Texts Written by the Fourth Student in the Baseline and Postinstruction Phase



Air pollution poisonous gases emerging from factory chimneys cause air pollution. For this reason we must place filters on factory chimneys. A reason is that we must fix LPGs to cars or place filters on car exhausts, We must reduce the leadfree petrol prices. We must build factories in places outside the cities. We must increase the number of public transportation vehicles.



Earthquakes are one of the most important natural disasters of our time. People die, get injured, even their houses get ruined because of earthquakes.

There are many reasons for the loss of lives and properties in earthquakes. One of the reasons is that furniture is not fixed to the walls. Another reason is houses are built on unsafe grounds. The other reason for the loss of lives and properties in the earthquakes is that the buildings are not strong.

There are many solutions to prevent the loss of lives and properties in earthquakes. One of these solutions is to fix the furniture to the walls. Because during the earthquake, if the furniture such as bookshelves, cupboards are fixed to the walls, they will not fall down. Another solution is constructing buildings on unsafe ground should not be allowed. The other solution for loss of lives and properties in earthquakes is to live in safe buildings. There should be a report stating that the building is strong against earthquakes when buying or selling houses. Municipalities should control buildings. There should be insurance against earthquakes. One of the most important solutions is to go under the tables during the earthquakes. Because to go under the tables during the earthquakes is to take measures. Also the TV and radio programs should be organized about the problem of earthquake.

Consequently, if measures are taken to prevent loss of lives and properties in earthquakes, people will not die or get injured. Houses will not get ruined. Therefore, the authorities and we should do our duties.


Arabsolghar, A., & Elkins, J. (2000). Comparative expectations of teachers and parents with regard to memory skills in children with intellectual disabilities. Journal of Intellectual & Developmental Disability, 25, 169-179.

Armbruster, B. B., Anderson, T. H., & Ostertag, J. (1989). Teaching text structure to improve reading and writing. The Reading Teacher, 43, 130-137.

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.

Banikowski, A. K., & Mehring, T. A. (1999). Strategies to enhance memory based on brain research. Focus on Exceptional Children, 32, 1-16.

Belmont, J. M., & Butterfield, C. E. (1971). Learning strategies as determinants of mental deficiencies. Cognitive Psychology, 2, 411-420.

Bray, N. W., & Turner, L. A. (1986). The rehearsal deficit hypothesis. In N. R. Ellis & N. W. Bray (Eds.), International review of research in mental retardation (Vol. 14, pp. 47-71). New York: Academic Press.

Borkowski, J. G., & Wanschura, P. B. (1974). Mediational processes in the retarded. In N. R. Ellis/Ed.), International review of research in mental retardation (Vol. 7, pp. 1-50). New York: Academic Press.

Campione, J. S., & Brown, A. L. (1977). Memory and meta memory development in educable mentally retarded children. In R. V. Kail, Jr., & J. W. Hagen (Eds.), Perspectives on the development of memory and cognition (pp. 367-406). Hillsdale, NJ: Lawrence Erlbaum Associates.

Danoff, B., Harris, K. R., & Graham, S. (1993). Incorporating strategy instruction within the writing process in the regular classroom: Effects on normally achieving and learning disabled students' writing. Journal of Reading Behavior, 25, 295-322.

De La Paz, S. (1999). Self-regulated strategy instruction in regular education settings: Improving outcomes for students with and without learning disabilities. Learning Disabilities Research & Practice, 14, 92-106.

De La Paz, S. (2001). Teaching writing to students with attention deficit disorders and specific language impairments. Journal of Educational Research, 95, 37-47.

De La Paz, S., & Graham, S. (1997). Strategy instruction in planning: Effects on the writing performance and behavior of students with learning difficulties. Exceptional Children, 63, 167-181.

Ellis, N. R. (1970). Memory processes in retardates and normals. In N. IL Ellis (Ed.), International review of research in mental retardation (Vol. 4, pp. 1-33). New York: Academic Press.

Engle, R. W., & Nagle, R. J. (1979). Strategy training and semantic encoding in mildly retarded children. Intelligence, 3, 17-30.

Englert, C. S. (1990). Unraveling the mysteries of writing through strategy instruction. In T. E. Scruggs & B. Y. L. Wong (Eds.), Intervention research in learning disabilities (pp. 186-223). New York: Springer-Verlag.

Englert, C. S. (1992). Writing instruction from a socio-cultural perspective: The holistic, dialogic, and social enterprise of writing. Journal of Learning Disabilities, 25, 153-172.

Englert, C. S., & Mariage, T. (1991). Shared understandings: Structuring the writing experience through dialogue. Journal of Learning Disabilities, 24, 330-342.

Englert, C. S., & Mariage, T. (2003). The socio-cultural model in special education interventions: Apprenticing students in higher-order thinking. In L. H. Swanson, K. Harris, & S. Graham (Eds.), Handbook of learning disabilities (pp. 450-467). New York: Guilford.

Englert, C. S., & Raphael, T. E. (1988). Constructing well-formed prose: Process, structure, and metacognitive knowledge. Exceptional Children, 54, 513-520.

Englert, C. S., & Raphael, T. E. (1989). Developing successful writers through cognitive strategy instruction. In J. E. Brophy (Ed.), Advances in research on teaching (Vol. 1, pp. 105-15!). Greenwich, CT: JAI Press.

Englert, C. S., Raphael, T. E., & Anderson, L. M. (1992). Socially mediated instruction: Improving students' knowledge and talk about writing. The Elementary School Journal, 92, 411-449.

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

Englert, C. S., Raphael, T. E., Anderson, L. M., Gregg, S. L., & Anthony, H. M. (1989). Exposition: Reading, writing, and the metacognitive of learning disabled students. Learning Disabilities Research, 5(1), 5-24.

Englert, C. S., Raphael, T. E., Fear, K. L., & Anderson, L. M. (1988). Students' metacognitive knowledge about how to write informational texts. Learning Disability Quarterly, 11, 18-46.

Englert, C. S., & Thomas, C. C. (1987). Sensitivity to text structure in reading and writing: A comparison of learning disabled and non-learning disabled students. Learning Disability Quarterly, 10, 93-105.

Graham, S. (1990). The role of production factors in learning disabled students' compositions. Journal of Educational Psychology 82, 781-791.

Graham, S., & Harris, K. R. (1989). A components analysis of cognitive strategy instruction: Effects on learning disabled students' composition and self-efficacy. Journal of Educational Psychology, 81, 353-361.

Graham, S., & Harris, K. R. (2003). Students with learning disabilities and the process of writing: A meta-analysis of SRSD studies. In L. H. Swanson, IC Harris, & S. Graham (Eds.), Handbook of learning disabilities (pp. 323-344). New York: Guilford.

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.

Graham, S., & MacArthur, C. (1988). Improving learning disabled students' skills at revising essays produced on a word processor: Self-instructional strategy training. Journal of Special Education, 22, 133-152.

Graham, S., MacArthur, C., & Schwartz, S. (1995). The effects of goal setting and procedural facilitation on the revising behavior and writing performance of students with writing and learning problems. Journal of Educational Psychology 87, 230-240,

Gurtan, S. (1994). Environment, health, traffic textbook 4th grade. Ankara, Turkey: Tekisik Publishing.

Hallenbeck, M. J. (1996). The cognitive strategy in writing: Welcome relief for adolescents with learning disabilities. Learning Disabilities Research & Practice, 11, 107-119.

Hallenbeck, M. J. (1997). From the inside out: Adolescent; with learning disabilities think and talk about writing. Unpublished doctoral dissertation, Michigan State University, East Lansing.

Hallenbeck, M. J. (2002). Taking charge: Adolescents with learning disabilities assume responsibility for their own writing. Learning Disability Quarterly, 25, 227-246.

Harris, K. R., & Graham, S. (1996). Making the writing process work: Strategies for composition and self-regulation. Cambridge, MA: Brookline.

Harris, K. R., Graham, S., Mason, L. H., & Saddler, B. (2002). Developing self-regulated writers. Theory Into Practice, 41, 110-116

Katims, D. S. (2001). Literacy assessment of students with mental retardation: An exploratory investigation. Education and Training in Mental Retardation and Developmental Disabilities, 36, 363-372.

Kellas, T. J., Ashcraft, M. H., & Johnson, N. S. (1973). Rehearsal processes in the short-term memory performance of mildly retarded adolescents. American Journal on Mental Retardation, 77, 670-679.

Kozulin, A. (2003). Psychological tools and mediated learning. In A. Kozulin, B. Gindis, V. S. Ageyev, & S. M. Miller (Eds.), Vygotsky's educational theory in cultural context (pp. 15-38). Cambridge, UK: Cambridge University Press.

MacArthur, C., Schwartz, S., & Graham, S. (1991). Effects of a reciprocal peer revision strategy in special education classrooms. Learning Disabilities Research and Practice, 6, 201-210.

Sanders, T. J. M., & Noordman, L. G. M. (2000). The role of coherence relations and their linguistic markers in text processing. Discourse Processes, 29, 37-60.

Savasir, I., & Sahin, N. (1994), Wechsler Intelligence Scale for Children (WISC-R). Ankara, Turkey: Turkish Psychologists Association.

Senunver, G., Karabulut, E., Kesim, H. M., Turgut, R., Ercan, N., Kucukbaycan, M. et al. (1999). Social studies textbook 4th grade. Istanbul, Turkey: National Education Publishing.

Sexton, M., Harris, K. R., & Graham, S. (1998). Self-regulated strategy development and the writing process: Effects on essay writing and attributions. Exceptional Children, 64, 295-311.

Spitz, H. H. (1966). The role of input organization in the learning and memory of mental retardates. In N. R. Ellis (Ed.), International review of research in mental retardation (Vol. 2, pp. 29-54). New York: Academic Press.

Stoddard, B., & MacArthur, C. (1993). A peer editor strategy: Guiding learning disabled students in response and revision. Research in the Teaching of English, 27, 76-103.

Tawney, J. W., & Gast, D. L. (1984). Single subject research in special education. Columbus, OH: Merrill.

Thomas, C. C., Englert, C. S., & Gregg, S. (1987). An analysis of errors and strategies in the expository writing of learning disabled students. Remedial and Special Education, 8(1), 21-30.

Troia, G. A., Graham, S., & Harris, K. R. (1999). Teaching students with learning disabilities to mindfully plan when writing. Exceptional Children, 65, 235-252.

Turner, L. A., Dofny, E. M., & Dutka, S. (1994). Effects of strategy and attribution training on strategy maintenance and transfer. American Journal on Mental Retardation, 98, 445-454.

Vacca, R. T., & Vacca, J. L. (2001), Content area reading: Literacy and learning across the curriculum. Needham Heights, MA: Allyn & Bacon.

Wong, B. Y. L., Wong, R., & Blenkisop, J. (1989). Cognitive and metacognitive aspects of learning-disabled adolescents' composing problems. Learning Disability Quarterly 12, 300-322.

RUYA GUZEL-OZMEN, Faculty of Education, Department of Special Education, Gazi University, Besevler-Ankara, Turkey.

I would like to thank Professor Carol Sue Englert for her continued support and willingness to share her expertise with me in the creation of this

Address all correspondence to Ruya Guzel-Ozmen, Faculty of Education, Department of Special Education, Gazi University, 06500 Besevler-Ankara, Turkey (e-mail:

Manuscript received October 2004; accepted May 2005.


Gazi University

Student Average Scores During Each Experimental Phase

 Student 1

Writing Measures B PI M1 M2 M3

Planning Time 0.0 37.6 48.0 28.0 --
Writing Time 18.0 93.3 117.0 69.0 --
Length 29.2 127.0 125.0 99.0 --
Elements 1.0 10.6 10.0 10.0 --
Overall Quality 1.0 4.3 4.0 4.0 --
Coherence 2.0 12.3 15.0 11.0 --

 Student 2

Writing Measures B PI M1 M2 M3

Planning Time 0.0 16.3 14.0 9.0 6.0
Writing Time 14.0 92.3 83.0 65.0 52.0
Length 45.5 240.0 205.0 154.0 151.0
Elements 2.0 15.0 14.0 15.0 14.0
Overall Quality 1.0 6.0 6.0 6.0 5.0
Coherence 1.4 22.6 18.0 18.0 15.0

 Student 3

Writing Measures B PI M1 M2 M3

Planning Time 0.0 42.3 38.0 24.0 18.0
Writing Time 16.1 138.6 115.0 84.0 63.0
Length 100.6 226.0 145.0 100.0 128.0
Elements 4.3 13.0 11.0 12.0 13.0
Overall Quality 1.6 5.0 4.0 3.0 3.0
Coherence 3.1 19.3 18.0 11.0 9.0

 Student 4

Writing Measures B PI M1 M2 M3

Planning Time 0.0 20.2 20.0 15.0 --
Writing Time 16.1 101.2 100.0 49.0 --
Length 46.4 181.5 245.0 184.0 --
Elements 3.4 14.5 15.0 15.0 --
Overall Quality 1.5 6.0 7.0 6.0 --
Coherence 1.7 17.2 20.0 21.0 --

Note: B = Baseline, PI = Post-Instruction, M1 = First Maintenance,
M2 = Second Maintenance, M3 = Third Maintenance
COPYRIGHT 2006 Council for Exceptional Children
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2006 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Guzel-Ozmen, Ruya
Publication:Exceptional Children
Geographic Code:7TURK
Date:Mar 22, 2006
Previous Article:Eyes on the prize: teaching complex historical content to middle school students with learning disabilities.
Next Article:Increasing student participation in IEP meetings: establishing the self-directed IEP as an evidenced-based practice.

Related Articles
Impact of Diana, Larry P., and P.L. 94-142 on minority students.
The Task Demonstration Model: a concurrent model for teaching groups of students with severe disabilities.
Differences in mathematics performance between students with learning disabilities and students with mild retardation.
Retention or promotion? Wrong question.
The effects of cognitive organizers to facilitate content-area learning for students with mild disabilities: a pilot study.
TWA + PLANS strategies for expository reading and writing: effects for nine fourth-grade students.
A meta-analysis on teaching mathematics to students with significant cognitive disabilities.
Defending and Parenting Children Who Learn Differently: Lessons from Edison's Mother.

Terms of use | Privacy policy | Copyright © 2018 Farlex, Inc. | Feedback | For webmasters