Comparing and Combining Accommodation and Remediation Interventions to Improve the Written-Language Performance of Children With Asperger Syndrome

Abstract

The relative effectiveness of two writing accommodations, word processing (WP) and speech-recognition (SR) technology, was examined relative to the writing skills of four boys with Asperger syndrome. The more effective accommodation was then combined with the Self-Regulated Strategy Development (SRSD) writing intervention and compared with SRSD and handwriting (HW) combined to determine whether participants’ writing skills would be further improved. Using a multiple-phase alternating-treatments design with a final treatment phase, outcomes indicated that the SR accommodation improved writing fluency and quality better than WP and HW. However, participants produced the most fluent and highest quality writing when SRSD intervention was combined with SR as compared with SRSD paired with HW, SR alone, and HW alone. Implications and future directions are discussed.

Keywords

Asperger syndrome instruction writing SRSD speech recognition word processing

Researchers have suggested that students with Asperger syndrome (AS) struggle with writing (Church, Alisanski, & Amanullah, 2000; Myles et al., 2003), which could be characterized as handwriting (HW) difficulties and problems with written expression. HW challenges have been described as including poor penmanship (i.e., letter formation and spacing), difficulties mastering pencil grip, and poor motor planning skills (Fuentes, Mostofsky, & Bastian, 2009; Ghaziuddin, Butler, Tsai, & Ghaziuddin, 1994; Lopata, Volker, Hamm, Sowinski, & Thomeer, 2007; Rinehart et al., 2006). Anecdotal reports (Fondacaro, 2001) suggest that the physical demands of writing may be more onerous for students with AS than for typically developing students, possibly contributing to their reported dislike for the writing process.

Individuals with AS have been described as presenting a list of detailed facts rather than developing cohesive and qualitatively rich written products (Fein & Dunn, 2007). Students with AS seem to exhibit weaknesses in writing fluency, organization, sentence complexity, distinguishing relevant from irrelevant information, and making abstract connections and inferences (Church et al., 2000; Ozonoff, Dawson, & McPartland, 2002). Furthermore, difficulty recognizing and understanding characters’ emotional states, determining their intentions, and understanding metaphors, trickery, irony, and sarcasm (Westby, 2004) may interfere with complex character development (Baron-Cohen, 2001; Griswold, Barnhill, Myles, Hagiwara, & Simpson, 2002).

Given these well-documented difficulties, there is a paucity of research regarding the efficacy of specific writing interventions and accommodations for students with AS. Anecdotal reports borrowed from similar populations form the basis of best practice techniques throughout the literature (Myles & Simpson, 2001; Safran, Safran, & Ellis, 2003). Researchers have derived most of these recommendations from literature addressing students with writing difficulties and/or fine-motor deficits that interfere with the writing process. Suggested interventions for students with AS include the use of priming, the use of their own written work to learn to elaborate on their writing, peer modeling, graphic organizers, and self-modeling (Asaro-Saddler & Saddler, 2009, 2010; Delano, 2007; Myles et al., 2003; Safran et al., 2003).

In their research with children with dyslexia and dysgraphia, Berninger, Nielson, and Abbot (2008) suggested that writing accommodations are often necessary, but seldom sufficient, to handle the escalating writing demands placed on students with writing deficits. As children are expected to produce lengthier, better developed, and more cohesive written work when they enter later elementary years, explicit instruction in writing is essential to target expressive writing deficits that an accommodation alone does not address. Although it is important to ameliorate the physical stressors associated with the physical demands of writing particularly for students with AS, the addition of a writing intervention that could address the qualitative writing deficits would be particularly appropriate for students with AS (MacArthur, 2000; Quinlan, 2004).

Writing Accommodations

HW difficulties not only negatively affect writing fluency but researchers have shown that they also interfere with working memory, planning, and revising (Quinlan, 2004). Furthermore, when HW is not automatic, individuals may struggle with the process of transforming ideas into language concepts, and then written words (Berninger & Swanson, 1994). One of the most commonly recommended accommodations for students with AS is the use of a word processor (WP; Myles et al., 2003). It has been speculated that use of a WP may alleviate the physical demands of writing, potentially allowing students to focus more on planning and generating meaningful content (Church et al., 2000; De La Paz, 1999), although researchers have yet to assess the effectiveness of WP exclusively for students with AS (Manjiviona, 2003). Research results supporting the use of WP for students with learning disabilities and writing difficulties are mixed. Bangert-Drowns’ (1993) meta-analysis of 32 studies discovered that students with writing difficulties tended to make more revisions and create significantly longer and more creative stories when using WP compared with HW. When examining middle school students with writing disabilities, Hetzroni and Shrieber (2004) found that students made fewer spelling errors and created better organized written work when using WP.

Other researchers have not demonstrated support for the use of WP as a successful writing accommodation for children with writing difficulties. Hollenbeck, Tindal, Harniss, and Almond (1999) found no significant differences between the use of WP and HW on essay quality produced by seventh-grade students with disabilities on an Oregon statewide test. Similarly, MacArthur and Graham (1987) found no differences in language complexity, mechanical errors, quality and story structure, time and rate measures, and types of revisions between HW and work completed with WP for 11 fifth- and sixth-grade students with learning disabilities. Although it appears that there are benefits to using WP for some populations, it remains unclear which elements of writing WP can address, and whether students with AS actually benefit from this accommodation.

The use of speech-recognition (SR) technology may better alleviate the fine-motor demands of writing for those with AS, given that the stressors of writing and typing are fully removed, perhaps allowing students to express themselves more quickly and fluently (Bangert-Drowns, 1993; Cochran-Smith, 1991). Because speech-based writing alleviates the challenges of transcription that is required in HW and typing, it may enable students with AS to utilize their cognitive resources for planning and organization (De La Paz, 1999; Quinlan, 2004).

Two preliminary studies using SR yielded promising findings. MacArthur and Cavalier (2004) explored the effectiveness of SR technology with 31 high-school students with and without learning disabilities. Using a repeated measures design, use of a scribe and SR were significantly better than the HW condition in terms of quality ratings, but there were no significant differences in vocabulary diversity or length of written product across conditions. Quinlan (2004) combined SR technology with the writing intervention, Self-Regulated Strategy Development (SRSD; Graham & Harris, 1989), to improve writing fluency for 20 “fluent” writers and 21 “less fluent” writers, ages 11 to 14 years. Without SRSD, less fluent writers produced significantly fewer words and more errors in the HW condition when compared with SR. However, there were no significant differences in the quality ratings for either condition. When the authors introduced SRSD, less fluent writers produced significantly higher quality writing in both conditions, and overall the combination of SR and SRSD yielded the best outcomes. This result provides additional support that combining a writing accommodation with a treatment targeting written expression may produce better outcomes (Berninger et al., 2008; MacArthur, 2000).

Writing Interventions

Strategy instruction is one of the most empirically supported writing interventions to date (Eckert, Codding, Truckenmiller, & Rheinheimer, 2009). SRSD, the most well-developed form of strategy instruction, is a multifaceted intervention with considerable empirical support with multiple populations (De La Paz, 2001;Graham & Harris, 2003; Harris, Graham, & Mason, 2003; Lane et al., 2008; Lienemann, Graham, Leader-Janssen, & Reid, 2006; Reid & Lienemann, 2006). Researchers have consistently demonstrated that SRSD can improve writing quality, fluency, self-efficacy, and motivational dispositions (Saddler, 2006; Saddler, Moran, Graham, & Harris, 2004), all of which represent deficits manifested by students with AS.

SRSD instructs students to improve their writing through planning, revision, and editing and attempts to increase positive student attitudes toward the writing process (Graham & Harris, 2005). Modeling and scaffolding self-instruction techniques teach students the elements of a good story and help them organize their work with the use of a graphic organizer, components that target many of the writing deficits of students with AS (Asaro-Saddler & Saddler, 2009, 2010). The graphic organizer (i.e., WWW chart) contains seven questions designed to help students formulate characters’ thoughts and feelings, which may be more difficult for those with AS than for other students. The students answer the seven questions prior to writing their story such as “Who is the main character?” and “How does the main character feel?” and then use the organizer as a guide.

SRSD has been found to improve writing fluency and story quality in students with AS. Delano (2007) combined the SRSD intervention with video self-modeling for three adolescents with AS using a multiple baseline design across responses. The SRSD package increased the number of words written (i.e., all written words regardless of spelling) and story elements, which was calculated by counting the number of premises, reasons, conclusions, and elaborations as initially defined by Graham and Harris (1989). Asaro-Saddler and Saddler (2009) also found positive results in their study examining the effectiveness of the SRSD intervention on the overall quality, measured by the 8-point holistic rating scale developed by Graham and Harris (1989) and number of story elements, for a 10-year-old male with AS using an ABA design. These findings were replicated across three younger students (two second graders and one fourth grader) with autism spectrum disorder (ASD; Asaro-Saddler & Saddler, 2010) using a multiple baseline design where number of story elements, overall holistic quality, number of words, and planning time served as the dependent variables. As there have been no studies examining generalization of these acquired skills, additional research that assesses generalization is critical to evaluate the success of selected treatment.

Purpose

We designed this study to answer three questions. First, we were interested in determining which of two writing accommodations (i.e., WP or SR) would be more successful for improving writing across measures of written fluency, accuracy, and story components for students with AS. Second, we wanted to know whether SRSD combined with the most effective accommodation would enhance written-language performance, particularly for story completeness, as measured by number of story parts. Finally, we wanted to know whether writing improvements generalized to participants’ creative writing homework assignments.

Method

Participants and Settings

Four students with AS, from Grades 4, 5, and 6, were recruited from a large urban northeast public school system in the United States. Selection was based on the following criteria: (a) formal diagnosis of AS provided by a licensed psychologist, pediatric neurologist, developmental pediatrician, or psychiatrist; (b) parent report of consistent underachievement in the area of writing; and (c) basic skill and knowledge in using WP, determined by a preliminary typing proficiency test.

Abe and Cal were in the fourth grade, Bob was in fifth grade, and Dan was in sixth. All participants received special education services along with occupational and speech therapy. Abe, Bob, and Cal were in integrated classes co-taught by special and general education teachers. Abe also was assigned a 1:1 paraprofessional and had weekly group counseling sessions. Dan was a student in a specialized program for students with AS.

We conducted the study in the students’ homes twice weekly after school hours. Sessions were 30-min for the training, baseline, and intervention phases (these details will be provided in the “Procedures” section). Phases were implemented in an area of the home where the participants completed their school assignments (e.g., dining room table, desk) with space to work free from distraction.

Materials

Basic materials required for the study included pencils, lined wide-ruled writing (21.6 × 27.9 cm) and printer paper, and a stopwatch. Participants used the WordPad program on a SONY VAIO brand laptop computer (33 × 23 × 4 cm), which does not make editing corrections, to type their stories. Dragon Naturally Speaking Preferred (2008) was the SR software used, which enabled the participants to dictate stories into the computer’s microphone. SRSD materials included a list of participants’ self-statements, a WWW graphic organizer, and three sample stories for training purposes.

Preliminary Typing Test

Prior to inclusion, all participants completed a typing test to ensure typing proficiency. Participants typed a passage from Dolphins and Sharks from The Magic Tree House series (Osborne & Boyce, 2003), with a Spache Readability Index of 3.48 (OKAPI! Online Manual, n.d.). Prior to this, all participants had been given a reading fluency assessment in which they read the Dolphins and Sharks reading passage aloud for 1 min and the number of words read correctly (WRCM) were recorded. All read more than 137 WRCM, which represented third-grade mastery level, thus their ability to transcribe the passage via typing would not be hindered by reading level (Hasbrouck & Tindal, 2006). As there is no general consensus on what constitutes a “proficient” typing speed for children, we considered participants proficient for this study if they were able to type at least 10 words per minute (Langone, Levine, Clees, Malone, & Koorland, 1996). In this study, Abe and Cal typed 30 words in 3 min (10 words per minute), Bob typed 31 words (10.3 words per minute), and Dan typed 33 words (11 words per minute).

Dependent Variables

CBM

CBM is a brief, direct, and objective measure of written expression that utilizes material from the student’s curriculum. CBM is a reliable and valid measure for written expression (Gansle, Noell, VanDerHeyden, Naquin, & Slider, 2002; Gansle, VanDerHeyden, Noell, Resetar, & Williams, 2006). CBM for written expression is divided into two types of measures, those that measure writing fluency and general production (e.g., production dependent) and those that measure writing accuracy (e.g., production independent). Production-dependent measures such as total words written (TWW) correspond with general written fluency (Gansle et al., 2006). Production-independent measures, such as percentage of correct word sequences (%CWS), yield stronger correlations with teachers’ holistic ratings (Tindal & Parker, 1989). Story starters, which researchers commonly use to prompt a student to produce a written story based on a topic sentence provided, were randomly chosen from a list of 120 compiled from AIMSweb and Shapiro (2004). A different story starter was used for each writing probe.

We used TWW to assess difficulties with writing fluency and production and %CWS to assess writing accuracy. TWW served as the primary dependent variable as researchers have found that for third through sixth graders, TWW has medium to large criterion validity with the Test of Written Language (TOWL), Woodcock–Johnson–Revised Writing Samples Subtest, Developmental Scoring System, Stanford-9, and the Wisconsin Knowledge and Concepts Examination (Deno, Mirkin, & Marston, 1980; Weissenburger & Epsin, 2005). TWW also has demonstrated good test–retest and interscorer reliability with 97.7% agreement (Gansle et al., 2006). TWW was computed by counting the TWW during the 3-min writing period, including titles and words spelled incorrectly. Numbers that were not spelled were not counted as words (Shinn, 1989). To compute %CWS, we counted the number of two consecutive words joined together that were spelled correctly and were grammatically correct. We then divided that score by the sum of the correct and incorrect word sequences to obtain a percentage (Shinn, 1989).

Number of Story Parts

Researchers examining the effectiveness of the SRSD intervention for students with AS often use number of story parts as a variable, as it is one of the important aspects of writing taught to students in this intervention package (Asaro-Saddler & Saddler, 2009, 2010; Reid & Lienemann, 2006). The number of story parts provides a measure of story completeness and overall story quality that corresponds with the graphic organizer that is used within the SRSD intervention (Graham & Harris, 1989). Seven key story elements were identified within each participant’s stories and included: main characters, locale, time, what the main characters wanted to do, what they did, how they felt, and how the story ends (Graham & Harris, 1989). The number of story parts identified and counted was computed in the manner consistent with previous research using this measurement (Asaro-Saddler & Saddler, 2010; Graham & Harris, 2005). We computed total number of story parts that the participant included in each of their stories by counting the number of key elements.

Accommodations and Interventions

WP

Participants used the WordPad program on a SONY VAIO brand laptop computer to write their stories. This program does not make editing corrections, thus the participants were not prompted to make changes as they typed.

SR

This study used the Dragon Naturally Speaking Preferred (2008) software. The SR software enables the participants to tell their stories orally into the computer’s microphone, rather than writing them by hand or using a keyboard.

SRSD

We introduced the SRSD intervention (Graham & Harris, 2005) in two training sessions. The first session taught the SRSD strategies and the second provided practice using the strategies. The first session began with teaching the participants to generate and utilize self-statements. These self-statements were intended to help the participants monitor and manage their writing if they got “stuck.” An example of a self-statement is “If I take my time, a good idea will come to me.” The participants then learned the POW and WWW strategies. The POW acronym stands for “pick my idea, organize my work, and write and say more.” This reminder was placed on the participants’ desk as a reminder while producing their written work. The WWW strategy is the basis of the graphic organizer and contains seven critical questions that the participants must answer while planning their stories. These questions are: “Who is the main character?” “When does the story take place?” “Where does the story take place?” “What does the main character do or want to do?” “What do the other characters do?” “What happens then?” “What happens with the other characters?” “How does the story end?” “How does the main character feel?” and “How do the other characters feel?” Once completed, the participants could refer to their WWW charts as they developed their stories.

The second training session began with modeling the use of each strategy to write a story. Next, participants had an opportunity to practice these strategies using three different story starters.

Procedures

We used a multiple-phase alternating-treatments design with a final treatment phase (Rosenberg, 1986; Scott, Glynn, & Ballard, 1988) to answer our research questions. There were five experimental phases and two training phases (SR training and SRSD training), with generalization data collected throughout the phases. The phases were initial baseline (HW), alternating treatments (HW, WP, and SR), replication of the most effective accommodation, a second alternating-treatments phase (SRSD + HW and SRSD + SR), and the replication of the most effective treatment package. Data collection was completed by the first author of this study. Each treatment session was 30 min in length and consisted of three writing probes separated by two 5-min breaks. A writing probe was initiated when the researcher presented a story starter to the participants, provided 1 min to think, and then 3 min to write their stories.

Phase 1—Initial Baseline

Participants were provided with at least five different CBM writing story starters to handwrite their stories (Shapiro, 2004). Baseline data were collected for at least five stories or until baseline data were stable.

SR Training

Prior to data collection in the first alternating-treatment phase, we trained the participants with the SR software for two 30-min sessions. The participants then practiced narrating and editing their own stories for 10 min.

Phase 2—First Alternating-Treatments Phase

Treatment conditions were counterbalanced by placing three slips of paper into a hat containing the words “handwriting,” “word processor,” and “speech recognition” and participants drew them before each session. The order in which participants drew the slips of paper indicated the order of the treatments during the session, and we ensured that there was no administration of any one treatment for more than three consecutive stories (Barlow & Hersen, 1984). Participants wrote a maximum of three stories in one 30-min session, including a 5-min break between treatment conditions during which they engaged in a puzzle activity to prevent fatigue.

Phase 3—Replication of the Most Effective Treatment

We continued the most effective accommodation, selected according to the TWW data, to better demonstrate a functional relation between the accommodation and the dependent measures, using the same procedures listed in Phases 1 and 2.

SRSD Training

The first author conducted the SRSD training across two 30-min sessions.

Phase 4—Second Alternating-Treatments Phase

Following SRSD training, we alternated SRSD + HW and SRSD + SR. Treatment conditions were counterbalanced in the same way as the previous alternating-treatment phase with no more than three stories produced per 30-min session. During this phase, participants utilized the SRSD materials independently to either handwrite their stories or dictate them using SR.

Phase 5—Replication of the More Effective Treatment

The more effective treatment continued alone for two sessions, using the procedures listed in previous phases.

Generalization

At the end of the first alternating-treatment phase, the participants completed three creative writing homework assignments that had been given to them by their respective teachers, one per condition examined (i.e., HW, SR, and WP) to determine whether gains made throughout the study generalized to homework. In keeping with the CBM procedures, we only scored the first 3 min of their writing. At the end of the second alternating-treatment phase, two additional creative writing assignments using the two conditions (i.e., SRSD + HW and SRSD + SR) were collected and scored.

Procedural Integrity

An educational psychology doctoral student assessed procedural integrity during 33%, 30%, 31%, and 33% of the sessions for Abe, Bob, Cal, and Dan, respectively. This independent observer listened to audiotapes while utilizing the written protocol containing participants’ oral instructions and a procedural integrity checklist to determine whether the same procedures were followed for all participants. The checklist included eight sections: typing test, baseline, first alternating-treatment phase, replication phase, two SRSD training sessions, second alternating-treatment phase, and replication. Each section was divided into subsections detailing essential steps such as providing the 3-min time limits per story, provision of 5-min breaks between probes, and not allowing the same treatment to be repeated more than three consecutive times. The independent observer checked off a subsection if completed correctly and then divided the number of correct steps by 30 (total number). Procedural integrity was 100% for all participants.

Interscorer Agreement

The same educational psychology doctoral student independently scored each writing probe to assess interscorer agreement for TWW, %CWS, and number of story parts. The doctoral student received the specific scoring guidelines for each dependent variable and recorded his calculations to compare with the researcher’s initial calculations. Total score comparisons between the doctoral student and the researcher were computed across 32% of writing probes for Abe, Cal, and Dan and 30% of Bob’s writing probes. Agreement was calculated by dividing the total number of agreement scores by the total number of agreement scores plus total disagreement scores and multiplying by 100%. For dependent variables, interscorer agreement was 100% for TWW, 93.0% for %CWS, and 93.0% for number of story parts. Interscorer agreements for Abe, Bob, Cal, and Dan were 97.7%, 95.0%, 95.8%, and 97.1%, respectively.

Results

TWW

Descriptive statistics across phases, dependent measures, and participants are provided in Table 1; Figure 1 illustrates TWW across phases for each participant. Abe’s TWW increased by 71% in the SR condition. Responding remained unchanged in HW and there was no differentiation between HW and WP. TWW remained elevated relative to the initial baseline when SR was continued alone (replication phase). Introduction of SRSD resulted in slight increases in TWW in either SR condition but not the HW condition for Abe.

Table 1.

Descriptive Statistics Across Dependent Measures.

	Baseline		HW		WP		SR		Replication		SRSD + HW		SRSD + SR		Replication
Participant	M	SD	M	SD	M	SD	M	SD	M	SD	M	SD	M	SD	M	SD
Total words written
Abe	24.00	2.36	22.20	1.92	19.80	1.30	41.00	1.58	44.33	4.16	22.20	2.17	54.25	3.30	60.00	4.24
Bob	22.75	1.75	24.25	1.83	19.62	2.39	47.28	2.63^a	52.50	0.71	25.50	3.00	56.25	1.50	58.50	2.12
Cal	21.89	1.97	25.50	0.58	16.75	0.96	55.50	5.07	56.50	0.71	25.75	1.26	65.50	2.38	68.50	3.54
Dan	22.80	2.17	27.25	4.03	32.75	2.18	66.50	1.73	67.00	1.41	25.75	2.22	76.75	2.50	81.00	7.07
Percentage of correct written sequences
Abe	98.40	3.58	98.40	3.58	100.00	0.00	100.00	0.00	100.00	0.00	100.00	0.00	100.00	0.00	100.00	0.00
Bob	83.75	10.38	83.50	11.76	82.00	11.39	92.43	3.82^a	90.20	2.83	75.50	9.33	91.50	3.42	95.50	3.54
Cal	82.44	14.09	81.00	8.72	92.00	9.56	93.50	3.11	97.00	1.41	81.50	10.08	94.00	2.94	95.00	1.41
Dan	100.00	0.00	95.25	5.85	89.00	7.96	95.00	1.41	95.00	2.83	100.00	0.00	95.25	0.96	95.50	0.71
Story parts
Abe	1.60	0.89	1.80	0.45	1.80	0.84	3.60	0.89	3.33	0.58	3.80	0.84	6.25	0.50	6.50	0.71
Bob	2.25	0.89	2.00	0.54	1.75	0.71	3.43	0.53^a	3.50	0.71	4.00	0.00	5.50	0.58	5.00	0.00
Cal	2.12	0.60	2.25	0.50	2.00	0.00	4.00	0.00	4.00	0.00	3.75	0.50	5.00	0.00	5.00	0.00
Dan	2.40	0.55	2.75	0.50	3.00	0.00	3.75	0.50	4.50	0.71	3.25	0.50	6.50	0.58	6.00	0.00

Note. HW = handwriting; WP = word processing; SR = speech-recognition technology; SRSD = self-regulated strategy development.

M and SD without outlier.

Figure 1.

Total words written.

Comparison of HW, SR, and WP demonstrated that Bob’s TWW increased by 100% in the SR condition. Consistent with Abe, responding remained unchanged in HW and there was no differentiation between HW and WP. TWW remained elevated relative to the initial baseline when SR was continued alone and introduction of SRSD did not result in increases in TWW in either condition.

Comparison of HW, SR, and WP demonstrated that Cal’s TWW increased by 150% in the SR condition. Responding remained unchanged in HW and there was differentiation between HW and WP that favored HW. TWW remained elevated relative to the initial baseline when SR was continued alone. Introduction of SRSD resulted in slight increases in TWW in the SR condition and TWW continued to be elevated in the replication phase.

Comparison of HW, SR, and WP demonstrated that Dan’s TWW increased by nearly 200% in the SR condition. Responding remained unchanged in HW and there was no differentiation between HW and WP. TWW remained elevated relative to the initial baseline when this intervention was continued alone. Introduction of SRSD resulted in increases in TWW in the SR condition and TWW continued to be elevated in the replication phase.

Percentage of Correctly Written Sequences

For Abe, percentage of correctly written sequences was undifferentiated among HW, SR, and WP (see Figure 2). Responding remained unchanged in HW when SR was continued alone. Introduction of SRSD resulted in unchanged responding.

Figure 2.

Percentage of correct word sequences.

Responding was undifferentiated across HW, SR, and WP for Bob, and continued to be similar when SR was continued in the replication phase. Some differentiation was observed when SRSD was introduced between the HW and SR conditions, favoring the combination with SR. Performance continued to be elevated when SRSD and SR were presented in the replication phase.

Undifferentiated responding was observed for Cal among HW, SR, and WP. When SR was presented alone, the percentage of correctly written sequences was elevated. Undifferentiated responding was observed between HW and SR when SRSD was introduced and performance remained high when SRSD and SR were presented in the replication phase. For Dan, responding remained undifferentiated across conditions and phases, and was consistently high.

Number of Story Parts

Abe’s number of story parts increased by more than 100% in the SR condition (see Figure 3). Responding remained unchanged in the replication phase. Introduction of SRSD resulted in increases in story parts in the SR condition and initially in the HW condition for Abe. Number of story parts continued to be elevated in the replication phase when SRSD and SR were presented in isolation.

Figure 3.

Number of story parts.

Bob’s number of story parts increased by more than 50% in the SR condition. Responding increased in the replication phase. Introduction of SRSD resulted in increases in story parts in the SR condition and in the HW condition number of story parts was equivalent with SR alone. Number of story parts continued to be elevated in the replication phase when SRSD and SR were presented in isolation.

Cal’s number of story parts increased by 89% in the SR condition. Responding remained unchanged in the replication phase. Introduction of SRSD resulted in increases in story parts in the SR condition and the HW condition was equivalent to the SR alone condition. Number of story parts continued to be elevated in the replication phase when SRSD and SR were presented in isolation.

Dan’s number of story parts increased by more than 56% in the SR condition. Responding was unchanged during the replication phase. Introduction of SRSD resulted in increases in story parts in the SR condition, but not in the HW condition. Number of story parts continued to be elevated in the replication phase when SRSD and SR were presented in isolation.

Generalization

For TWW, all participants displayed comparable results with their respective conditions, indicating that skills generalized with respect to story length when assessing participants’ creative writing homework assignments. The same was true with respect to %CWS and the number of story parts, indicating that skills generalized with all accommodations and interventions across variables and participants.

Social Validity

As parents of all four participants witnessed all conditions, we modified the Intervention Rating Profile–15 (IRP-15; Witt & Elliot, 1985) to assess parent perceptions at the end of treatment about writing accommodations (WP and SR) and the SRSD intervention. The IRP-15 is a 15-item questionnaire requiring respondents to rate items on a 6-point Likert-type scale, ranging from 1 (strongly disagree) to 6 (strongly agree). Overall scores range from 15 to 90, with higher scores reflecting greater acceptability.

Participants read and completed the Children’s Intervention Rating Profile (CIRP; Witt & Elliot, 1985) at the end of the study and submitted it to us directly following the final replication phase. The CIRP consists of seven statements that the participant rates to specify the fairness, expected effectiveness, and potential negative outcomes of each of the accommodations and interventions used in the study (Finn & Sladeczek, 2001). The participants rated items on a 6-point Likert-type scale, ranging from 1 (disagree) to 6 (agree), with higher scores corresponding to higher treatment acceptability.

Table 2 illustrates acceptability ratings for parents and participants. Findings from the IRP-15 illustrate that parents preferred the SR intervention and the SRSD + SR treatment package. Results of the CIRP illustrate that participants preferred the SRSD + SR intervention the most and the baseline HW condition the least. All participants preferred both accommodations (i.e., WP and SR) to HW and three of the four participants had a slight preference for SR. All participants preferred SRSD + SR compared with SRSD + HW.

Table 2.

Mean Acceptability Scores on the CIRP and IRP-15.

	HW		WP		SR		SRSD + HW		SRSD + SR
Participant	M	SD	M	SD	M	SD	M	SD	M	SD
Abe	3.29	2.29	4.29	1.89	4.14	2.41	4.29	2.21	4.43	2.07
Bob	3.14	2.67	4.29	1.98	4.57	2.44	4.14	2.41	5.57	1.13
Cal	3.14	2.67	4.00	1.53	4.14	0.27	4.00	1.73	4.29	2.21
Dan	2.71	2.36	4.00	2.08	4.29	2.36	3.71	2.29	4.43	2.07
Total	21.50	1.73	29.00	1.15	29.75	1.50	28.25	2.92	32.75	4.19
Abe	1.00	0.00	2.73	0.59	4.00	0.00	4.87	0.52	5.20	1.57
Bob	1.00	0.00	5.13	0.99	5.73	0.46	5.40	0.91	5.40	1.45
Cal	1.14	1.30	2.53	0.52	5.40	1.06	5.00	0.00	5.33	1.76
Dan	1.53	0.52	4.80	0.56	5.47	1.19	5.13	0.83	5.53	1.13
Total	18.75	4.35	57.00	20.35	77.25	11.70	76.50	3.42	80.50	2.08

Note. CIRP = children’s intervention rating profile (Witt & Elliot, 1985); IRP = intervention rating profile-15 (Witt & Elliot, 1985); HW = handwriting; WP = word processing; SR = speech-recognition technology; SRSD = self-regulated strategy development.

Discussion

We had a threefold purpose for this study. First, we compared two individual writing accommodations (WP and SR) with HW to determine which would result in better written-language production, accuracy, and story quality for students with AS. Second, we examined whether the addition of the SRSD intervention to the accommodation that resulted in the better performance would have an additive effect on improvements in written language. Third, we evaluated generalization. We interpret our results to conclude that SR was superior to WP and HW and that adding SRSD to SR resulted in the highest level of performance across all participants for story quality and three of four participants for writing production. All participants exhibited high accuracy that continued throughout the study. We also observed generalization on school-based creative writing assignments that corresponded with each treatment condition.

The SR accommodation enabled our participants to produce more fluently written work than HW and WP. Thus, participants “wrote” longer stories when provided the SR accommodation. SR also was the accommodation that parents and participants preferred over WP. The effect of SR on accuracy was mixed, however. For Abe and Dan, spelling and grammar usage was near perfect and the use of either accommodation did not change these results. Bob and Cal struggled slightly more with grammar and spelling during baseline and consequently yielded stable and higher performance when using the SR accommodation. This was likely due to the fact that the SR software automatically spells the words narrated into the computer correctly and often provides appropriate punctuation without the participant instructing it to do so, although the participants sometimes had to clarify grammatical or spelling errors when the SR software did not accurately record their speech.

These improvements in writing accuracy support Quinlan’s (2004) findings that less fluent writers produced fewer spelling errors when using SR. She added that the gains made in fluency were likely due to reducing transcription interference, allowing students to use their working memory for more accurate and fluent text production. In this study, participants seemed less concerned about spelling during the SR condition (e.g., they did not ask the researcher whether words were spelled correctly as often as when using WP or within the HW condition) and did not appear to get “stuck” while writing. Thus, participants may have been able to think less about accuracy in the SR condition (De La Paz, 1999).

A somewhat surprising finding was that the SR accommodation alone resulted in higher mean number of story parts when compared with HW and the WP accommodation. In the two previous studies examining the effect of the SR accommodation on written-language ability, only MacArthur and Cavalier (2004) found improved essay quality. Quinlan (2004) saw improvements in writing quality only when SR was paired with SRSD. When the physical demands of writing were fully alleviated, as was possible with the use of SR, the participants may have been able to draw upon more of their cognitive resources, allowing them to not only pay less attention to accuracy but also to give more attention to generating meaningful content (De La Paz, 1999). It also may be that participants were able to write lengthier stories using the SR accommodation because it did not tap their fine-motor skills, thus providing an opportunity to increase the amount written and the subsequent number of story parts in their writing samples (De La Paz, 1999).

The use of the WP accommodation, which is commonly recommended for students with AS (Myles et al., 2003), yielded poor results in terms of fluency and story length as well as story quality when compared with SR and did not benefit three of the four participants when compared with HW. Only one participant (Dan) produced lengthier written work, although this may have been due to his having more experience with using WP, an accommodation already provided in his classroom and at home. These findings are consistent with research examining students with disabilities and writing challenges (Hollenbeck et al., 1999; MacArthur & Graham, 1987) but inconsistent with Bangert-Drowns’ (1993) meta-analysis findings that reported that students wrote lengthier texts when using WP. However, the study by Bangert-Drowns only included typically developing students with writing difficulties and not students with disabilities. Thus, WP may be more beneficial for students who do not necessarily struggle with the physical demands of writing.

We found the combined use of the SR accommodation with the SRSD intervention to produce lengthier stories with higher quality when compared with using SRSD with HW for 75% of our participants. Our finding that participants did not benefit from the SRSD treatment in the HW condition beyond SR alone is contrary to other SRSD research results (Asaro-Saddler & Saddler, 2009, 2010; Delano, 2007; De La Paz, 1999; Graham & Harris, 2003; Lane et al., 2008; Lienemann et al., 2006; Reid & Lienemann, 2006; Saddler et al., 2004). This inconsistency may be attributed to the challenges the physical act of HW present in many children with AS (Fuentes et al., 2009; Ghaziuddin et al., 1994; Gross, 1994; Lopata et al., 2007; Rinehart et al., 2006) or the participants’ subsequent preference for the SR accommodation.

For TWW and number of story parts, the SRSD + SR treatment package was more effective than SR across participants, which suggests that pairing writing interventions with accommodations may produce better outcomes (Berninger et al., 2008). The SRSD + SR condition produced the same level of accuracy as seen with SR alone, suggesting that the addition of the SRSD intervention did not improve writing accuracy. It may be that SR alone is solely responsible for the improvements in accuracy for the two participants (Bob and Cal) who struggled slightly with spelling and grammar.

To date, we are aware of only one study that sought to generalize results to classroom writing assignments. Asaro-Saddler and Saddler (2010) demonstrated that participants with AS successfully transferred gains made using SRSD to another writing genre. This study results support this finding indicating that SR and SR + SRSD resulted in higher performance with classroom writing homework.

Limitations

There are several limitations that should be addressed in future research. First, although the use of WP was not a successful accommodation for these participants, only Dan used WP regularly as an accommodation in school and at home. That said, all participants were able to type 10 words per minute, which was considered proficient for this study (Langone et al., 1996). Furthermore, despite Dan’s experience using WP, SR, and SR + SRSD still resulted in the best outcomes for productivity and quality. Second, the SRSD intervention lessons used in this study were brief (2 sessions). Although Graham and Harris (2003) indicated that the SRSD intervention does not specify number of training sessions, lessons typically involve a great deal of scaffolding and practice (Asaro-Saddler & Saddler, 2010; De la Paz, 2001). If we had provided the participants additional practice and support, gains might have been greater. Participants sometimes made physical complaints (e.g., “my hand hurts”) when HW their responses onto their respective graphic organizers during the SRSD + SR condition. This may have been alleviated by creating a graphic organizer template on the computer for the participants to use.

Third, we did not directly compare all accommodation conditions with the SRSD intervention. Future researchers should examine this comparison to provide support for the current findings. Fourth, it is difficult to assess story quality in a valid and reliable manner (Greenberg, 1992). Tindal and Parker (1989) noted that many of the subjective writing measures have poor reliability and when those measures are modified to improve reliability, they tend to become less valid. Although investigators (Asaro-Saddler & Saddler, 2009, 2010; Delano, 2007) have used the number of story parts in recent SRSD studies, this measure is specific to the graphic organizer utilized in this intervention and therefore may have inflated results.

Fifth, although the SR accommodation proved to be helpful for these participants, it is not without limitations. The current version has the best voice recognition accuracy rates for adults, but there are no data on accuracy rates for children (MacArthur & Cavalier, 2004). It also is possible that voice recognition accuracy may have been compromised by these students’ prosody (Barnhill, 2001). Regardless of their preference for using this software, participants sometimes reported frustration when the software did not correctly interpret their statements. Fifth, systematic replication is required before these data can be generalized (Horner et al., 2005).

Future Research and Implications

As we did not find WP to be an effective way to address writing deficits, future researchers should focus on exploring more effective assistive technology accommodations, such as SR. At this time, research supporting SR for students with written-language difficulties is promising but sparse (MacArthur & Cavalier, 2004; Quinlan, 2004). Although SR technology has become quite advanced with reported accuracy rates between 90% and 98% for adults (MacArthur & Cavalier, 2004), there are no accuracy rates reported for children. In the future, researchers should study child accuracy rates, including evaluation of special populations like those with AS, given their atypical speech patterns, pitch, and tones (Attwood, 2006) or with other populations, such as English language learners.

Given the anecdotal evidence as well as this study’s social validity outcomes suggesting that children with AS do not like HW, future researchers should examine whether attitudes toward writing serve as a mediator for outcomes. For example, it would be interesting to note whether the use of SR could possibly improve attitudes toward the writing process by alleviating the physical strain of HW and that those improved attitudes may further improve writing outcomes. In addition, it would be important to find supporting evidence for the notion of “freeing up cognition” (De La Paz, 1999) by accommodating students’ writing deficits with assistive technology to enable them to produce more creative work.

Another possible area to explore is the use of a stopwatch for improving writing speed for children with AS, a common difficulty reported by a number of researchers (Church et al., 2000; Griswold et al., 2002). Abe and Dan asked for the stopwatch to be placed in front of them so that they could “keep track” of time. When asked about the stopwatch, both indicated that it helped them write faster. Interventions such as sentence combining that are designed to assist students with generating novel ideas through brainstorming may be useful (Graham & Perin, 2007) to explore for this particular population given their difficulties with sentence complexity (Myles et al., 2003) and character development (Baron-Cohen, 2001). Because the students in this study evidenced idiosyncratic writing deficits, some students may require specific interventions addressing spelling or grammar (Graham & Harris, 2003).

Because the most positive outcomes resulted from the combination of the SR accommodation and the SRSD intervention, additional research is needed to provide support for this treatment package. It would be interesting to know whether the SR and SRSD combination could work for other students with writing problems or if this particular combination is unique to children with AS. In addition, researchers might examine SRSD treatment dosage. Studies could explore whether increasing the number and/or length of SRSD sessions would have a positive impact on writing outcomes or if fewer sessions would be sufficient when accompanied by booster sessions.

Footnotes

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the authorship and/or publication of this article.

Funding

The author(s) received no financial support for the research and/or authorship of this article.

References

Asaro-Saddler

Saddler

(2009). Effects of planning instruction on a young writer with Asperger’s syndrome. Intervention in School and Clinic, 44, 268–275.

Asaro-Saddler

Saddler

(2010). Planning instruction and self-regulation training: Effects on writers with autism spectrum disorders. Exceptional Children, 77, 107–118.

Attwood

(2006). Asperger’s syndrome: A guide for parents and professionals. London, England: Jessica Kingsley.

Bangert-Drowns

R. L.

(1993). The word processor as an instructional tool: A meta-analysis of word processing in writing instruction. Review of Educational Research, 63, 69–93.

Barlow

D. H.

Hersen

(1984). Single case experimental designs: Strategies for studying behavior change. Elmsford, NY: Pergamon Press.

Barnhill

G. P.

(2001). What’s new in AS research: A synthesis of research conducted by the Asperger syndrome project. Intervention in School and Clinic, 36, 300–305.

Baron-Cohen

(2001). Theory of mind and autism: A review. Special issue of the International Review of Mental Retardation, 23, 169–176.

Berninger

V. W.

Nielson

K. H.

Abbot

R. D.

(2008). Writing problems in developmental dyslexia: Under-recognized and under-treated. Journal of School Psychology, 46, 1–21.

Berninger

V. W.

Swanson

H. L.

(1994). Modifying Hayes and Flower’s model of skilled writing. In Butterfield

(Ed.), Children’s writing: Toward a process theory of development of skilled writing (pp. 57–81). Greenwish, CT: JAI Press.

10.

Church

Alisanski

Amanullah

(2000). The social, behavioral, and academic experiences of children with Asperger syndrome. Focus on Autism and Other Developmental Disabilities, 15, 12–20.

11.

Cochran-Smith

(1991). Word processing and writing in elementary classrooms: A critical review of related literature. Review of Educational Research, 61, 107–155.

12.

Delano

M. E.

(2007). Improving written language performance of adolescents with Asperger syndrome. Journal of Applied Behavior Analysis, 40, 345–351.

13.

De La Paz

(1999). Composing via dictation and speech recognition systems: Compensatory technology for students with learning disabilities. Learning Disability Quarterly, 22, 173–182.

14.

De La Paz

(2001). Teaching writing to students with attention deficit disorders and specific language impairment. Journal of Educational Research, 95, 37–47.

15.

Deno

S. L.

Mirkin

P. K.

Marston

(1980). Relationships among simple measures of written expression and performance on standardized achievement tests (Research Report No. 22). Minneapolis: University of Minnesota Institute for Research on Learning Disabilities.

16.

Dragon Naturally Speaking Standard (10.0th version) [Computer program]. (2008). Burlington, MA: Lernout & Hauspie.

17.

Eckert

T. L.

Codding

R. M.

Truckenmiller

A. J.

Rheinheimer

J. L.

(2009). Improving children’s fluency in reading, mathematics, spelling, and writing: A review of evidenced-based academic interventions. In Akin-Little

Little

S. G.

Bray

M. A.

Kehle

T. J.

(Eds.), Behavioral interventions in schools: Evidenced-based positive strategies (pp. 111–124). Washington, DC.: American Psychological Association.

18.

Fein

Dunn

M. A.

(2007). Autism in your classroom: A general educator’s guide to students with autism spectrum disorders. Bethesda, MD: Woodbine House.

19.

Finn

C. A.

Sladeczek

I. E.

(2001). Assessing the social validity of behavioral interventions: A review of treatment acceptability measures. School Psychology Quarterly, 16, 176–206.

20.

Fondacaro

D. M.

(2001, February). Asperger syndrome: A qualitative study of successful educational interventions. Paper presented at the Eastern Educational Research Conference, Chester, PA.

21.

Fuentes

C. T.

Mostofsky

S. H.

Bastian

A. J.

(2009). Children with autism show specific handwriting deficits. Neurology, 73, 1532–1537.

22.

Gansle

K. A.

Noell

G. H.

VanDerHeyden

A. M.

Naquin

G. M.

Slider

N. J.

(2002). Moving beyond total words written: The reliability, criterion validity, and time cost of alternate measures for curriculum-based measurement in writing. School Psychology Review, 31, 477–497.

23.

Gansle

K. A.

VanDerHeyden

A. M.

Noell

G. H.

Resetar

J. L.

Williams

K. L.

(2006). The technical adequacy of curriculum-based and rating-based measures of written expression for elementary school students. School Psychology Review, 35, 435–450.

24.

Ghaziuddin

Butler

Tsai

Ghaziuddin

(1994). Is clumsiness a marker for Asperger syndrome? Journal of Intellectual Disability Research, 38, 519–527.

25.

Graham

Harris

K. R.

(1989). Improving learning disabled students’ skills at composing essays: Self-instructional strategy training. Exceptional Children, 56, 201–214.

26.

Graham

Harris

K. R.

(2003). Students with learning disabilities and the process of writing: A meta-analysis of SRSD studies. In Swanson

L. H.

Harris

Graham

(Eds.), Handbook of learning disabilities (pp. 323–344). New York, NY: Guilford.

27.

Graham

Harris

K. R.

(2005). Improving the writing performance of young struggling writers: Theoretical and programmatic research from the center on accelerating student learning. Journal of Special Education, 39, 19–33.

28.

Graham

Perin

(2007). Writing next: Effective strategies to improve writing of adolescents in middle and high schools (A report to Carnegie Corporation of New York, Alliance for Excellent Education, pp. 1–77). Washington, DC: Alliance for Excellent Education.

29.

Greenberg

(1992). Validity and reliability: Issues in the direct assessment of writing. WPA: Writing Program Administration, 16, 7–22.

30.

Griswold

D. E.

Barnhill

G. P.

Myles

B. S.

Hagiwara

Simpson

R. L.

(2002). Asperger syndrome and academic achievement. Focus on Autism and Other Developmental Disabilities, 17, 94–102.

31.

Gross

(1994). Asperger syndrome: A label worth having? Educational Psychology in Practice, 10, 104–110.

32.

Harris

K. R.

Graham

Mason

L. H.

(2003). Self-regulated strategy development in the classroom: Part of a balanced approach to writing instruction for students with disabilities. Focus on Exceptional Children, 35, 1–16.

33.

Hasbrouck

Tindal

G. A.

(2006). Oral reading fluency norms: A valuable assessment tool for teachers. Reading Teacher, 59, 636–644.

34.

Hetzroni

O. E.

Shrieber

(2004). Word processing as an assistive technology tool for enhancing academic outcomes of students with writing disabilities in the general classroom. Journal of Learning Disabilities, 37, 143–154.

35.

Hollenbeck

Tindal

Harniss

Almond

(1999). The effect of using computers as an accommodation in a statewide writing test. Eugene, OR: University of Oregon, BRT.

36.

Horner

R. H.

Carr

E. G.

Halle

McGee

Odom

Wolery

(2005). The use of single-subject research to identify evidence-based practice in special education. Exceptional Children, 71, 165–179.

37.

Lane

K. L.

Harris

K. R.

Graham

Weisenbach

J. L.

Brindle

Morphy

(2008). The effects of self-regulated strategy development on the writing performance of second-grade students with behavioral and writing difficulties. Journal of Special Education, 41, 234–253.

38.

Langone

Levine

Clees

T. J.

Malone

Koorland

(1996). The differential effects of a typing tutor and microcomputer-based word processing on the writing samples of elementary students with behavioral disorders. Journal of Research on Technology in Education, 29, 141–159.

39.

Lienemann

T. O.

Graham

Leader-Janssen

Reid

(2006). Improving the writing performance of struggling writers in the second grade. Journal of Special Education, 40, 66–78.

40.

Lopata

Volker

M. A.

Hamm

E. M.

Sowinski

J. E.

Thomeer

M. L.

(2007). Motor and visuomotor skills of children with Asperger disorder: Preliminary findings. Perceptual & Motor Skills, 104, 1183–1192.

41.

MacArthur

C. A.

(2000). New tools for writing: Assistive technology for students with writing difficulties. Topics in Language Disorders, 20, 85–100.

42.

MacArthur

C. A.

Cavalier

A. R.

(2004). Dictation and speech recognition technology as test accommodations. Exceptional Children, 71, 43–58.

43.

MacArthur

C. A.

Graham

(1987). Learning disabled students’ composing under three methods of text production: Handwriting, word processing, and dictation. Journal of Special Education, 21, 22–42.

44.

Manjiviona

(2003). Assessment of specific learning difficulties. In Prior

(Ed.), Learning and behavior problems in Asperger syndrome (pp. 55–84). New York, NY: Guilford.

45.

Myles

B. S.

Rome-Lake

Barnhill

G. P.

Huggins

Hagiwara

Griswold

D. E.

(2003). Written language profile of children and youth with Asperger syndrome: From research to practice. Education and Training in Developmental Disabilities, 38, 362–369.

46.

Myles

B. S.

Simpson

R. L.

(2001). Effective practices for students with Asperger syndrome. Focus on Exceptional Children, 34, 1–14.

47.

OKAPI! Online Manual. (n.d.). Retrieved from www.jimwrightonline.com/htmdocs/tools/okapi/okapimanual/okapimanual.php

48.

Osborne

M. P.

Boyce

N. P.

(2003). Dolphins and sharks. New York, NY: Random House Books for Young Readers.

49.

Ozonoff

Dawson

McPartland

(2002). A parent’s guide to Asperger syndrome & high functioning autism. New York, NY: Guilford.

50.

Quinlan

(2004). Speech recognition technology and students with writing difficulties: Improving fluency. Journal of Educational Psychology, 96, 337–346.

51.

Reid

Lienemann

T. O.

(2006). Self-regulated strategy development for written expression with students with attention deficit/hyperactivity disorder. Exceptional Children, 73, 53–68.

52.

Rinehart

N. J.

Bellgrove

M. A.

Tonge

B. J.

Brereton

A. V.

Howells-Rankin

Bradshaw

J. L.

(2006). An examination of movement kinematics in young people with high-functioning autism and Asperger’s disorder: Further evidence for a motor planning deficit. Journal of Autism and Developmental Disorders, 36, 757–767.

53.

Rosenberg

M. S.

(1986). Error-correction during oral reading: A comparison of three techniques. Learning Disability Quarterly, 9, 182–192.

54.

Saddler

(2006). Increasing story-writing ability through self-regulated strategy development: Effects on young writers with learning disabilities. Learning Disability Quarterly, 29, 291–305.

55.

Saddler

Moran

Graham

Harris

K. R.

(2004). Preventing writing difficulties: The effects of planning strategy instruction on the writing performance of struggling writers. Exceptionality, 12, 3–17.

56.

Safran

S. P.

Safran

J. S.

Ellis

(2003). Intervention ABCs for children with Asperger syndrome. Topics in Language Disorders, 23, 154–165.

57.

Scott

Glynn

Ballard

(1988). Reducing self-injury in a boy with intellectual disability using visual screening, toy play training, and contingent social interaction. Behaviour Change, 5, 117–127.

58.

Shapiro

E. S.

(2004). Academic skills problems: Direct assessment and intervention (3rd ed.). New York, NY: Guilford.

59.

Shinn

M. R.

(1989). Curriculum-based measurement: Assessing special children. New York, NY: Guilford.

60.

Tindal

Parker

(1989). Assessment of written expression for students in compensatory and special education programs. Journal of Special Education, 23, 169–183.

61.

Weissenburger

J. W.

Epsin

C. A.

(2005). Curriculum-based measures of writing across grade levels. Journal of School Psychology, 43, 153–169.

62.

Westby

(2004, April). Reading between the lines for social and academic success. Paper presented at 9th Annual Autism Spectrum Disorders Symposium, Social and Academic Success for Students with ASD, Providence, RI.

63.

Witt

J. C.

Elliot

S. N.

(1985). Acceptability of classroom strategies. In Kratochwill

T. R.

(Ed.), Advances in school psychology (pp. 251–288). Mahwah, NJ: Lawrence Erlbaum.