Teaching Language Skills to Preschool Students with Developmental Delays and Autism Spectrum Disorder Using Language for Learning

Teaching Language Skills to Preschool Students with Developmental Delays and Autism Spectrum Disorder Using Language for Learning Margaret M. Flores Kelly B. Schweck Vanessa Hinton Auburn University Abstract Language intervention using Direct Instruction (DI) has shown positive results. There is a growing body of investigation of Language for Learning (LL), a DI program, on the performance of students with autism spectrum disorders (ASD) and students with developmental delays (DD). There is need for replication and extension of research to include rural settings and students with ASD and DD. The purpose of this study was to investigate the effects of LL on the language of students with DD and ASD in a rural preschool classroom. Four preschool students with DD and ASD participated in 12 weeks of instruction, resulting in improved language skills. Key Words: Reading, direct instruction, autism spectrum disorder, preschool If you are enjoying reading this article, please consider subscribing to RSEQ or joining ACRES at http://acres-sped.org. At least one-third of students in the United States receive their education in rural school settings; yet, the majority of the research that influences policy and practice is conducted in non-rural locations (Parsley & Barton, 2015). The rural context is very different, and special considerations need to be taken into account investigating the implementation of instruction. One consideration is the unique demands teachers face in the implementation of instruction without the tools required to meet intensive and various educational challenges. For example, many rural schools are located in areas with high percentages of families who live in poverty, and schools in rural areas receive lower overall funding in comparison to schools in suburbs and larger towns (Butera & Dunn, 2005). This lack of funding may result in fewer instructional resources. An important concern in educating children who live in poverty is the high percentage of students who demonstrate language skills significantly below their peers (Henning, McIntosh, Arnott, & Dodd, 2010). This is critical because it is a known fact that language development impacts children s reading achievement (American Speech-Language-Hearing Association, 2008). Researchers also show that early intervention targeting phonological awareness without language interventions do not prevent reading difficulties for children who have disabilities or who come from families with low socioeconomic status (Henning et al., 2010). It is important for teachers in rural schools to meet the demand, incorporating instruction that builds young children s language skills in addition to administering instruction in reading and mathematics. Building language skills can be a difficult challenge for teachers who serve children with disabilities in rural schools because of limited resources and because the educational setting involves students with large variations of strengths and needs (Hulac, Wickerd, & Vining, 2013). For example, the special education teacher can often serve children with a multiple range of disability categories that include low and high incidence disabilities altogether; therefore, the educator must strategically implement instruction in a manner that balances a much wider variety of diverse characteristics, instructional learning styles, and achievement levels shown in the research. One way of balancing the various needs of students who need more intensive language and educational interventions is the use of explicit instructional programs. Explicit instructional programs, such as Direct Instruction (DI), have been shown to be effective in increasing receptive and expressive language of children with communication deficits in preschool and kindergarten settings (Adams & Engelmann, 1996; Benner et al., 2002; Tincani, Ernsbarger, Harrison, & Heward, 2005). DI is an approach that emphasizes incremental instructional steps to teach multiple skills and concepts that increase in complexity over time. DI curricula include carefully designed instructional language in the form of teacher scripts, frequent and quick student responding, immediate feedback, and many practice opportunities (Marchand-Martella, Slocum, & Martella, 2004). DI has been shown to be effective in teaching students with diverse characteristics related to achievement, ability, and background (Kinder, Kubina, & Marchand-Martella, 2005). Author Note: For all correspondence, please contact Margaret M. Flores (mflores@auburn.edu). Rural Special Education Quarterly Volume 35, No. 1 pages 3-12 2016 American Council on Rural Special Education Reprints and Permission: Copyright Clearance Center at 978-750-8400 or www.copyright.com Rural Special Education Quarterly 2016 Volume 35, Number 1 3

DI curriculum has been used to teach language skills for young children who needed communication interventions. Researchers have shown DI improved language skills for children with developmental delays (DD) and autism spectrum disorders (ASD). Benner et al. (2002) investigated the difference between the DI program Language for Learning (LL; Engelmann & Osborn, 1999) and a traditional kindergarten curriculum. Students who received LL performed significantly better than comparison students in receptive language. Woldron-Soler, Martella, Marchand-Martella, and Tso (2002) implemented LL within an integrated preschool setting. Children with DD who received 15 weeks of instruction using LL demonstrated greater gains in receptive and expressive language skills than children who did not participate in the program. In addition, students who received LL showed increased gains in social skills. Other LL research with young children investigated variations in implementation of the program. For example, Tincani et al. (2005) compared the effects of slow-paced and fastpaced teaching on the response opportunities, participation, accuracy, and off-task behavior of 4 pre-kindergarten students who were at-risk for school failure. Between the fast-paced and slow-paced instructional methods, there were significant differences between groups. The fast-paced group demonstrated four more responses per min. The fast-paced group provided three more correct responses and exhibited less off-task behavior than the slow-paced group. The groups did not differ in the percentage of academic responses. Within the LL literature, there is some research that includes students with ASD, a disability category that has increased in its identification over the past decade (U.S. Department of Education, 2010). Children with ASD demonstrate varied deficits in language development and communication (Charman, Drew, Baird, & Baird, 2003; Kenworthy, Wallace, Powell, Anselmo, & Black, 2012). Explicit and intensive language intervention has been effective in improving outcomes for students with ASD (National Autism Center, 2009); however, these evidence-validated methods involve one-on-one instruction that requires greater instructional resources than group instruction (Ganz, 2007). Programs such as LL could increase efficiency of instructional resources for students with ASD or other developmental disabilities. The line of DI language research using LL with students with ASD began with Ganz and Flores (2009). Three elementary students between the ages of 10 and 11 successfully completed a portion of the LL program. The researchers presented the students with one strand of the program (identification of materials) in a group format using prescribed materials and instructional procedures. A functional relation was shown between identification of materials and LL instruction. The researchers demonstrated that group instruction using frequent choral and independent responding, quick pacing, and immediate feedback was feasible and effective for teaching the target skill. Even though this research was promising, this study was limited due to its narrow scope of program presentation since only portions of the program were taught. The line of research was extended by Flores et al. (2013) to address the limitations of the study conducted by Ganz and Flores (2009). Flores et al. increased the extent of program presentation and recruited younger students for participation. Within an extended school year program, elementary students with ASD and DD, between the ages of 7 and 9, received LL instruction. The researchers presented lessons in their entirety over the course of 4 weeks. Students made significant gains in their performance across three curriculum-based progress assessments given before, during, and after instruction. Similar to the study by Ganz and Flores, the students demonstrated the ability to participate in small group instruction using LL as it was designed. This study was limited in its investigation of the feasibility and effectiveness of LL for young children since the youngest participants were 7 years of age. A follow-up study included pre-school students with ASD and DD (Flores & Ganz, in press). Students ages 3 through 10 years participated in either LL or discrete trial teaching within a 4-week extended school year program. One group received instruction using LL that was presented using program procedures in group formats. The other group, matched by ages, intellectual abilities, and language abilities, received instruction in the form of discrete trial teaching in the same behaviors and skills presented in LL. Both groups made similar gains as measured by curriculum-based assessments. LL was feasible and as effective as one-on-one instruction, but young children were also successful. In order to show that DI such as LL is an effective language intervention for young children with DD and ASD, additional research is needed. This should include research with young children within natural classroom settings. In addition, research should include diverse settings that are rural as well as urban. Tincani et al. (2005) and Ganz and Flores (2009) conducted their research in settings in large urban areas. Flores and Ganz (in press), Flores et al. (2013), and Woldron-Soler et al. (2002) conducted their studies in suburban and university-supported settings. Benner et al. (2002) conducted their research within settings in a moderately sized city. Therefore, the purpose of this study was to investigate the effects of LL on the language performance of young students with DD and ASD when implemented in a preschool setting in a rural high-need school. Two questions guided this study: (a) What are the effects of LL on the language performance of young students with developmental delays in a rural inclusive preschool setting? (b) What are the effects of LL on the language performance of young students with autism spectrum disorders in a rural inclusive preschool setting? Method The design of the study was a multiple probe across language behaviors. The language behaviors or skills chosen for instruction depended on students performance on the placement tests in the LL program. These were oral language skills that the students had not mastered. Sara and Ned s language behaviors were making action statements, identifying statements, and answering yes and no questions. Earl and Alan s behaviors were using prepositions (i.e., on, over, in front of), using opposites (i.e., big, small, full, and empty), and applying the first and second behaviors. 4 Volume 35, Number 1 2016 Rural Special Education Quarterly

Setting The study took place in an inclusive pre-school classroom in a public elementary school in a rural county in the southeastern United States. The population of the county (land area of 550 square miles) was 35,000. The median family income was $32,186.00, and 21% of family incomes were below the poverty level (United States Census Bureau, 2013). The pre-school class served students with and without disabilities who attended for full and half school days, as well as students who attended 5 days per week and 3 days per week. For students with disabilities, special education services were provided within the pre-school classroom according to their individualized education programs (IEPs). The classroom teacher was certified in early childhood special education and was assisted by three paraprofessionals. The study lasted 15 weeks and included collection of data regarding the students language performance during baseline condition, prior to intervention, and during maintenance condition to assess language skills (e.g., making statements, answering questions, identifying opposites, using prepositions) mastered during intervention conditions. The data collected over the 15 weeks were checked for reliability agreement, which is discussed later in the methods section; reliability agreement was not checked prior to the study. The intervention was implemented for 12 weeks, and instructional sessions occurred 3 days per week. The researchers provided instruction within the classroom during regularly scheduled instructional rotations that included one-on-one instruction with the classroom teacher and learning centers (e.g., dramatic play, gross motor skills, art, numeracy, writing, literacy). The researchers provided instruction because the teacher and paraprofessionals did not have time to devote to the intervention; during this time, the teacher and paraprofessionals were responsible for other instructional needs. The students rotated among centers that were self-selected for an hour each morning, and the students who participated in the study worked with the researchers during one of the 10-min rotation periods. Instruction took place at a table located in the front of the room; there were no barriers that separated the table, but it was not near other activity areas within the room. Participants Of the 12 students with disabilities served in the classroom, 4 participated in the study. The criteria for participation were (a) parent permission to participate, (b) eligibility for special education services, and (c) placement in the Language for Learning program (Engelmann & Osborn, 2008). Sara, Ned, Earl, and Alan lived in homes in which English was the only language spoken. Recent evaluation information was available, but the researchers collected additional information regarding cognitive functioning. They used a nonverbal cognitive assessment since all of the students IEPs indicated deficits in communication. Background and assessment information for Sara, Ned, Earl, and Alan are located in Table 1. Sara, age 4 years, participated in early intervention services the year prior to her enrollment in preschool. Sara was eligible for services under the DD category and demonstrated delays in adaptive, communication, cognitive, and motor domains. She received speech/language and occupational therapy services on a weekly basis in addition to special education services 5 days per week. Sara communicated using three-word phrases. She followed verbal directions but not consistently without prompting. A number of deficits were noted within her IEP; Sara did not point to pictures, name objects presented in pictorial form, call others by name, use appropriate pronouns, or demonstrate understanding of positional concepts. Ned, age 4 years, participated in a Head Start early education program prior to enrollment in pre-school. He was eligible for special education services under the DD category and demonstrated delays in the communication, motor, cognitive, and personal/social domains. Ned received speech/language and occupational therapy on a weekly basis in addition to special education services 5 days per week. Ned communicated using phrases of three words or more, but his speech was difficult to understand. He repeated phrases rather than using spontaneous utterances. Ned followed one-step verbal directions, but his performance was inconsistent. Ned had tantrums several times per day, lasting from 1 to 10 min. Ned used gestures, identified objects, and indicated negatives. Ned had difficulty expressing actions, descriptions, and possession. Earl, age 4 years, participated in early intervention services prior to enrollment in preschool. He was eligible for special education services under the ASD category. The eligibility evaluation team included professionals with expertise and appropriate training in evaluation and identification of ASD. The Autism Diagnostic Observation Schedule (Lord, Rutter, DiLavore, & Risi, 2000) and Autism Diagnostic Interview-Revised (Rutter, LaCouteur, & Lord, 2007) indicated deficits in communication, deficits in social interaction, and stereotyped or restrictive behaviors. He received special education services 3 days per week. His IEP indicated that his speech and language had improved significantly, using intelligible speech and increasing the length of his sentences to five words. Earl consistently followed simple one-step directions. Earl had difficulty retelling stories or events and expressing descriptions and possession. He did not attend to learning tasks for more than 5 min. Alan, age 4 years, was eligible for special education services under the category of DD in the adaptive, personal/ social, communication, motor, and cognitive domains. Alan received speech/language and occupation therapy on a weekly basis in addition to special education services 3 days per week. He communicated using three-word sentences and followed one-step directions. Alan matched objects to pictures, named body parts, and attended to learning tasks for up to 10 min. He could not consistently demonstrate the following: (a) describe how objects or pictures were similar or different, (b) appropriately use pronouns, or (c) demonstrate understanding of positional concepts. Materials The instructional materials consisted of the Language for Learning (Engelmann & Osborn, 2008) presentation book and objects prescribed by the instructional lessons (e.g., yellow wooden pencil used within lessons that involved differentia- Rural Special Education Quarterly 2016 Volume 35, Number 1 5

Table 1. Student Demographic Information Student Age Cultural Background Eligibility Sara 4 Latina Developmental Delay Ned 4 African American Developmental Delay Earl 4 White Autism Spectrum Disorders Alan 4 White Developmental Delay Assessment of Developmental Domains a Adaptive: 76 Communication: 66 Motor: 79 Cognitive: 64 Personal/Social: 77 Communicaations: 58 Motor: 64 Cognitive: 55 Adaptive: 89 Personal/Social: 86 Communication: 92 Adaptive: 61 Personal/Social: 77 Communication: 81 Motor: 68 Cognitive: 78 Assessment of Cognitive Ability b 83 95 119 74 a Standard scores obtained from the Battelle Developmental Inventory II (Newborg, 2005) b Standard scores obtained from the Leiter International Performance Scale-Revised (Roid & Miller (2002) tion of whole and parts of a pencil). The first author created assessment probes to assess learning. The probes consisted of pictures shown to the student, scripts read by the researchers, and space to record the student s response. Sample probe items are included in Figure 1. The assessment tasks presented were similar to those taught within the program. For example, Language for Learning lessons related to the prepositions over, on, and in front included a picture of a car with dogs positioned around and on the car. During a lesson, the researcher pointed to a dog and asked the student, Where is this dog? An assessment probe for prepositions had a picture of a car with frogs positioned around and on the car. The researcher used the same pointing and questioning procedures to assess the students knowledge, pointing to a frog and asking students to tell where it was. In other program lessons, the pictures were a horse, cat, or table with balloons positioned around the objects. The corresponding assessment probes were pictures of animals (frog, dog, and cat) with balls or insects. Assessment probes for tasks related to action statements included a researcher script and a space for recording the student s answer. These probes were similar to the action and description lessons within the Language for Learning program. For example, action statements involved asking the student to touch his or her head and tell what he or she was doing. The script read, Touch your head; what are you doing? There was a blank line near the script; the researcher wrote the student s answer word for word. Procedures Instruction progressed according to the students learning. Prior to instruction, the researchers administered the Language for Learning (Engelmann & Osborn, 2008) placement test. The placement tests for Sara and Ned indicated that they began with Lesson 1. The placement tests for Earl and Alan indicated that they began with Lesson 31. The researchers initiated baseline assessments for each behavior and instruction in the first behavior until five data points had been collected and a stable baseline was established. They defined a stable baseline as follows: The last three data points in the baseline path would vary no more than 20% from the mean of the data points. In more practical terms, items correct could not vary by more than one across the three probes. During the learning center rotation, the students participated in instruction in groups of two with the first author, a certified special education teacher with specific professional development in DI procedures and experience in using Language for Learning. The classroom learning rotations consisted of 10-min periods that began and ended with an auditory signal of a timer. When the timer signaled the end of a rotation, the researcher went to the students and asked each to come with her to the table to play the picture/action game. At times, instructional groups were one-on-one when a student was absent. Each session began with individually administered assessment probes; one of the researchers asked each student to come one at a time. The researcher administered probes prior to instruction so that learning from the previous session could be measured. After assessment, both students sat on either side of the researcher and instruction began. The researcher provided instruction according to the procedures prescribed in the Language for Learning presentation book. These procedures include the use of scripted language, teacher behaviors, teacher gestures, pacing, and positive affect. The researcher presented lesson activities according to the script but without 6 Volume 35, Number 1 2016 Rural Special Education Quarterly

reading directly from the book. Instruction used quick pacing, allowing for pauses when appropriate, in a manner that showed enthusiasm for each task. The researcher asked the students to respond chorally and in unison, as set by the program. The researcher also followed correction procedures; the researcher modeled correct responses, asked the students to respond with the researcher, and asked the students to respond independently. The researchers presented lessons in order and included the activities within the presentation book. They collected data for specific behaviors within the program, but instruction was inclusive of all behaviors within the program. Sara and Ned participated in Lessons 1 through 23, and the behaviors for which data were collected included action statements, identity statements, and yes and no questions. Earl and Alan participated in Lessons 31 through 50, and the behaviors for which data were collected included prepositions (i.e., on, over, in front of), opposites (i.e., big, small, full, and empty), and application of the first and second behaviors. Instruction progressed according to students learning. When skills were demonstrated at mastery, the researcher presented instruction on the next skill; the researchers defined mastery as three probes with 100% accuracy. For example, Sara and Ned s first behavior was action statements using a sentence. Lessons 1 through 5 included description of actions (e.g., asking the students to stand up or sit down and tell what they were doing) and identification of pictures using two words (e.g., a dog ). Sara and Ned s second behavior was describing a picture using a whole sentence (e.g., This is a dog ), and instruction in this skill did not begin until Lesson 6. Instruction did not progress past Lesson 5 until students described their actions using a complete sentence ( I am sitting down and I am standing up ). When students mastered this skill, instruction progressed to Lesson 6 that included identity statements. The researcher presented the student with a picture, and he/she responded with a sentence (e.g., This is a ball ). The third behavior for Sara and Ned was correct response to yes and no questions. This skill began with Lesson 15, but instruction in this lesson only began if students had mastered the previous skill. Accurate responses to yes and no questions involved the presentation of a picture (e.g., picture of a cat), and asking the student if the picture was a cat (the correct answer was Yes ) or asking the student if the picture was a ball (the correct answer was No ). Earl and Alan s first behavior was descriptions using the prepositions over, on, and in front when presented with a picture. Instruction related to this behavior began with Lesson 31. Earl and Alan s second behavior was describing opposites using the words big, small, full, and empty. Instruction for this behavior began with Lesson 41, presented after they had mastered the first behavior. The researcher presented Earl and Alan with pictures of objects that were full and empty or animals that were big and small and asked them to differentiate the pictures. Assessment procedures were the same as those used during instruction, but the picture stimulus included different objects or animals. The third behavior was concept application using the two previous behaviors. Instruction in this behavior began with Lesson 46 but was not presented until the students mastered the second behavior. The researcher presented Earl and Alan with a picture in which multiple people, objects, or animals were positioned differently or doing different actions; however, one person, object, or animal was over, on, or in front of something. The instructor stated a rule about the person, object, or animal (e.g., The pot on the table is full; The man on the box will laugh ). Then, the researcher asked the students about the different people, objects, or animals, and their answers required application of the rule ( Is this man on the box? So, what do you know about his man? or Is this pot on the table? So, what else do you know about this pot? ). Earl and Alan responded to the first question using yes or no and answered the second question with a complete sentence. Assessment procedures were similar to instruction but did not include the yes or no question, just the question that required a complete sentence (e.g., The question was, What do you know about this bucket? and the answer was either This bucket is full or This bucket is not full ) Reliability The researchers collected treatment integrity data for 30% of the instructional lessons for each student. An observer watched administration of assessment probes and lessons as they occurred within the classroom and completed a checklist of researcher behaviors. The treatment integrity checklist included a list of behaviors, and the observer indicated either presence or absence of those behaviors. The checklist was based on an observation scale by Marchand-Martella, Lignugaris-Kraft, Pettigrew, and Leishman (1995) and had been used in previous research (Flores et al., 2013). The observer was a certified special education teacher who had received specific professional development in DI and had experience in using Language for Learning (Engelmann & Osborn, 2008). The researchers measured reliability of data collection through examination of assessment probes; two researchers inspected all assessment probes for each student, and the researchers calculated interobserver agreement as follows: The number of item agreements divided by total number of item agreements and disagreements. Results All of the students began baseline condition across all behaviors at the same time, and the researchers collected at least five baseline data points before beginning the intervention. They inspected the data visually in order to determine the presence of a functional relation. Graphs for Ned, Sara, Earl, and Alan are shown respectively in figures two, three, four, and five. Baseline data across all students were stable. For this study, the researchers defined baseline stability was defined as follows: The last four data points within a path varied no more than 10%. Baseline data paths across students and behaviors showed a neutral trend. Intervention Data for Ned For the first behavior, action statements, the level for Ned s data path was 63%. The data ranged from 0% to 100%. The data path showed an upward trend, and there was a seven percent overlap between baseline and intervention data. For the second behavior, identity statements, the level for Ned s Rural Special Education Quarterly 2016 Volume 35, Number 1 7

data path was 63%. There was a range in data from 25% to 100%. The data path showed an upward trend, and there were no overlapping data points between baseline and intervention data. For the third behavior, answering yes and no questions, the level for Ned s data was 76% with a range from 28% to 100%. The data path showed an upward trend, and there were no overlapping data points between baseline and intervention conditions. Intervention Data for Sara For the first behavior, action statements, the level for Sara s data path was 62%. The data ranged from 25% to 100%. The data path showed an upward trend, and there was no overlap between baseline and intervention data. For the second behavior, identity statements, the level for Sara s data path was 67%. There was a range in data from 6% to 100%. The data path showed an upward trend, and there were 11% overlapping data points between baseline and intervention data. For the third behavior, answering yes and no questions, the level for Sara s data was 63% with a range from 14% to 100%. The data path showed an upward trend, and there was 20% overlap in data points between baseline and intervention conditions. Intervention Data for Earl For the first behavior, prepositions, the level for Earl s data path was 63%. The data ranged from 65% to 100%. The data path showed an upward trend, and there was no overlap between baseline and intervention data. For the second behavior, descriptions of opposites, the level for Earl s data path was 60%. There was a range in data from 16% to 100%. The data path showed an upward trend, and there was no overlap of data points between baseline and intervention conditions. For the third behavior, application of prepositions and opposites, the level for Earl s data was 72% with a range from 32% to 100%. The data path showed an upward trend, and there was no overlap in data points between baseline and intervention conditions. Intervention Data for Alan For the first behavior, prepositions, the level for Alan s data path was 54%. The data ranged from 16% to 100%. The data path showed an upward trend and there was no overlap between baseline and intervention conditions. For the second behavior, descriptions of opposites, the level for Alan s data path was 62%. There was a range in data from 0% to 100%. The data path showed an upward trend, and there was nine percent overlap of data points between baseline and intervention conditions. For the third behavior, application of positions and opposites, the level for Alan s data was 72% with a range from 32% to 100%. The data path showed an upward trend, and there was no overlap in data points between baseline and intervention conditions. Effect Size The researchers calculated effect sizes using Tau-U. This index for analysis includes calculation of trends within phases (or conditions) as well as non-overlap between phases (or conditions; Parker, Vannest, Davis, & Sauber, 2011). The researchers found an effect size for each student s behaviors, as well as each student s overall performance. They did not detect any trends in baselines conditions for any student. For Ned s action statements, the intervention had a strong effect (Tau=0.93). For his identity statements, the intervention had a strong effect (Tau=1). For answers to yes and no questions, the intervention had a strong effect (Tau=1). Overall, the intervention had a strong effect for Ned across behaviors (Tau=0.98). For Sara s action statements, the intervention had a strong effect (Tau=1). For her identity statements, the intervention has a strong effect (Tau=0.97). For answers to yes and no questions, the intervention had a strong effect (Tau=.89). Overall, the intervention had a strong effect for Sara across behaviors (Tau=0.94). For Earl s prepositions, the intervention had a strong effect (Tau=1). For his descriptions of opposites, the intervention had a strong effect (Tau=1). For application of concepts, the intervention had a strong effect (Tau=1). Overall, the intervention had a strong effect on Earl s performance across behaviors (Tau=1). For Alan s prepositions, the intervention had a strong effect (Tau=1). For his descriptions of opposites, the intervention had a strong effect (Tau=0.92). For application of concepts, the intervention had a strong effect (Tau=0.90). There was a strong overall effect for Alan across behaviors (Tau=0.92). Treatment Integrity and Inter-observer Agreement Treatment fidelity was 100% throughout the study. With regard to consistency in data collection, interobserver agreement for the probes was as follows: 94% agreement for Ned s action statement, 100% agreement for Ned s identity statement probes, 100% for Ned s yes and no probes, 92% agreement for Sara s action statement probes, 100% agreement for Sara s identity statement probes, 100% agreement for Sara s yes and no probes, 100% agreement for Earl s preposition probes, 100% agreement for Earl s opposite probes, 100% agreement for Earl s application probes, 100% agreement for Alan s preposition probes, 100% agreement for Alan s opposite probes, and 100% agreement for Alan s application probes. Social Validity The researchers examined social validity through teacher interviews before and after the study. Prior to the study, the students teacher reported the following: (a) Students demonstrated significant deficits in the behaviors chosen for the study, (b) there was a need for additional instruction, and (c) the behaviors chosen for the study were important to each student s future classroom success. For example, Ned and Sara did not consistently follow simple directions, and this would negatively impact their performance in an inclusive kindergarten classroom. After the study, the teacher reported (a) the students language skills improved, (b) there was noticeable improvement in their language in the classroom, and (c) she would recommend the program to others. 8 Volume 35, Number 1 2016 Rural Special Education Quarterly

Discussion The purpose of this study was to investigate the effects of DI, such as LL, on the language performance of young students with DD and ASD, when implemented in a rural preschool setting. The researchers incorporated instruction using LL into the preschool classroom routine without difficulty. A multiple probe across behaviors design demonstrated functional relations across three behaviors for each of the 4 students who participated in the study. In addition, the overall effect sizes for the LL intervention were strong for each of the students. The current study extends the literature (e.g., Benner et al., 2002; Woldron-Soler et al., 2002; Flores et al., 2013) regarding DI for students with DD and ASD by investigating the effects of implementation of a Language for Learning program within an inclusive rural classroom setting in a manner that balanced the wide variety of diverse student characteristics, instructional learning styles, and achievement levels of students in the classroom. The researchers administered instruction in rotating stations as a way to individualize instruction based on unique student strengths and weaknesses. All students in the classroom received academic instruction, but the researchers also pulled students to receive intensive instruction based on learning needs. Students who required more intensive language interventions received DI. All students in this study made progress across all behaviors and also maintained progress after instruction ended. Teachers who provide instruction for students with significant language needs can implement instruction using DI procedures and focusing on language behaviors. Language behaviors involve teaching students to make action statements, make identity statements, answer yes and no questions, make descriptions using prepositions, make descriptions using opposites, and apply prepositions and opposites in discussion prompted by teacher questions. Interventions can occur during station teaching as a way to differentiate instruction. It is noted that Earl had difficulty attending to instructional tasks for more than 5 min prior to the study. Earl s attention during LL instruction was not consistent at the beginning of the study; however, he attended throughout the 10-min sessions by the end of the study. It is not known what brought about this change in behavior; however, the structure and pacing of LL may have contributed to this, as demonstrated by Tincani et al. (2005). The change in student attention is an important finding because teachers may be hesitant to include students in instruction similar to DI due to problems with attention; however, this study demonstrated it is possible for students to learn appropriate instructional behaviors while participating in the program because students demonstrated generalization of language skills to other instructional activities. The students language skills improved as measured by the assessment probes but also outside of LL instruction. The classroom teacher reported that Ned s and Sara s compliance improved and that they answered yes and no questions accurately during other instructional activities. Earl s and Alan s language skills outside of LL instruction improved as well; their descriptions of objects included correct prepositions and were given with greater detail than previously. Over the course of the 12-week intervention, instruction was implemented according to LL program procedures that included repetition of lessons until mastery was demonstrated. The researcher and students repeated LL lessons; the researchers completed approximately 20 program lessons for each student over the course of 30 instructional days. The discrepancy between the length of the intervention and the number of instructional sessions was due to school holidays and field trips. They repeated approximately 30% of program lessons, but they did not distribute these lessons evenly across the study. Repetition of lessons occurred at the beginning rather than the end of the study. This may have been due to lack of familiarity with the program methods. Choral responding was difficult for all of the students, and corrections occurred throughout the study but decreased as the study progressed. Limitations and Future Research This study was limited with regard to generalization of findings; single-case research requires replication and further investigation to accomplish this. In addition, the study lacked measures of generalization for each student behavior. The researchers noted teacher reports, but did not develop any formal procedures or measures. Future research should include data collection regarding students use of intervention behaviors in other natural contexts. Another limitation of this study was implementation of LL by a researcher rather than the classroom teacher. This allowed for efficient execution and may have increased fidelity; however, implementation by a classroom teacher would provide more realistic results and information. In addition, teacher implementation may allow for further investigation regarding the potential differences between research and practice in rural settings. Finally, future research should continue investigations of LL in comparison to other curricula and teaching methods in rural locations. Although research has shown that LL is an effective method, it is important to investigate its effects compared to other DI programs and practices. Furthermore, LL research should include instructional packages that incorporate other activities. For example, there may be combinations of activities that can be added to LL to increase its effects. Within preschool settings, incidental teaching may be one such practice that occurs regularly in settings that employ evidence-based strategies. Language behaviors included in LL could be easily incorporated in play and other engaging activities that occur in early education settings. In conclusion, this study showed LL can be effectively implemented in rural settings. Research should continue to include implementation within high need settings to inform the field of efficient and effective practices. Rural Special Education Quarterly 2016 Volume 35, Number 1 9

Figure 1. Examples of Probe Items Look at these dogs. Here is the rule. The dog under the table will eat. 10 Volume 35, Number 1 2016 Rural Special Education Quarterly

Figure 2. Results for Ned Figure 4. Results for Earl Figure 3. Results for Sara Figure 5. Results for Alan Rural Special Education Quarterly 2016 Volume 35, Number 1 11

References Adams, G. L., & Engelmann, S. (1996). Research on Direct Instruction: 25 years beyond DISTAR. Seattle, WA: Educational Achievement Systems. American Speech-Language-Hearing Association (2008). Roles and responsibilities of speech-language pathologists in early intervention: Technical report. Retrieved from www.asha.org/policy. Benner, G. J., Trout, A., Nordness, P. D., Nelson, J. R., Epstein, M. H., Knobel, M., Epstein, A., Maguire, K., & Birdsell, R. (2002). The effects of the Language for Learning program on the receptive language skills of kindergarten children. Journal of Direct Instruction, 2(2), 67-74. Butera, G. & Dunn, M (2005). The case for cases in preparing special educators for rural schools. Rural Special Education Quarterly, 24, 22-27. Charman, T., Drew, A., Baird, C, & Baird, G. (2003). Measuring early language development in pre-school children with autism spectrum disorders using the Mac Arthur Communicative Development Inventory (Infant Form). Journal of Child Language, 30, 213-236. Engelmann, S., & Osborn, J. (2008). Language for learning. Columbus, OH: SRA/McGraw-Hill. Flores, M. M., & Ganz, J. B. (in press). Comparison of direct instruction and discrete trial teaching on the language development of students with autism and developmental disabilities. Exceptionality. Flores, M. M., Nelson, C., Hinton, V., Franklin, T. M., Strozier, S. D., Terry, S. L., & Franklin, S. (2013). Teaching reading comprehension and language skills to students with autism spectrum disorders and developmental disabilities using Direct Instruction. Education and Training in Autism and Developmental Disabilities, 48, 41-48. Ganz, M. L. (2007). The lifetime distribution of the incremental societal costs of autism. Archives of Pediatrics & Adolescent Medicine, 161, 343-349. Ganz, J. B., & Flores, M. M. (2009). The effectiveness of direct instruction for teaching language to children with autism spectrum disorders: Identifying materials. Journal of Autism & Developmental Disorders, 39, 75-83. doi:10.1007/s10803-008-0602-6 Henning, C., McIntosh, B., Arnott, W., & Dodd, B. (2010). Long-term outcome of oral language and phonological awareness intervention with socially disadvantaged preschoolers: the impact on language and literacy. Journal of Research in Reading, 33, 231-246. Hulac, D. M., Wickerd, G., & Vining, O. (2013). Allowing students to administer their own interventions: An application of the self-adminstered folding-in technique. Rural Special Education Quarterly, 32, 31-36. Kenworthy, L., Wallace, G. L., Powell, K., Anselmo, A. M., & Black, D. O. (2012). Early milestones predict later language, but not autism symptoms in higher functioning children with autism spectrum disorders. Research in Autism Spectrum Disorders, 6, 1194-1202. Kinder, D., Kubina, R., & Marchand-Martella, N. E. (2005). Special education and direct instruction: An effective combination. Journal of Direct Instruction, 5(1), 1-36. Lord, C., Rutter, M., DiLavore, P. C., & Risi, S. (2000). Autism diagnostic observation schedule. Los Angeles, CA: Western Psychological Services. Marchand-Martella, N., Lignugaris-Kraft, B., Pettigrew, T., & Leishman, R. (1995). Direct instruction observation system. Logan, UT: Utah State University. Marchand-Martella, N. E., Slocum, T. A., & Martella, R.C. (2004). Introduction to Direct Instruction. Boston, MA: Ally and Bacon. National Autism Center (2009). Evidence-based practice and autism in the schools: A guide to providing appropriate interventions to students with autism spectrum disorder. Randolph, MA: National Autism Center. Newborg, J. (2005). Battelle Developmental Inventory, 2 nd Edition. Itasca, IL: Riverside Publishing. Parker, R.,I., Vannest, K. J., Davis, J. L., & Sauber, S. B. (2011). Combining non-overlap and trend for single-case research: Tau-U. Behavior Therapy, 42, 284-299. Parsley D. & Barton, R. (2015). The myth of the little red schoolhouse: Challenges and opportunities for rural school improvement. Peabody Journal of Education, 90, 191-193. Roid, G. H., & Miller, L. J. (2002). Leiter International Performance Scale- Revised. Wood-Dale, IL: Stoelting Co. Rutter, M., LeCouteur, A., & Lord, C. (2007). Autism diagnostic interviewrevised. Los Angeles, CA:Western Psychological Services. Tincani, M., Ernsbarger, S., Harrison, T. J., & Heward, W. L. (2005). Effects of instructional paces on pre-k children s rate, accuracy, and off-task behavior in the Language for Learning program. Journal of Direct Instruction, 5(1), 97-109. U.S. Department of Education (2010). Children with Disabilities Under Part B of the Individuals with Disabilities Education Improvement Act. Data Analysis System (DANS). OMB # 1820-0043. Woldron-Soler, K. M., Martella, R. C., Marchand-Martella, N. E., & Tso, M. E. (2002). Effects of a 15-week Language for Learning implementation with children in an integrated pre-school. Journal of Direct Instruction, 2(2), 75-86. 12 Volume 35, Number 1 2016 Rural Special Education Quarterly