Efficacy of a Teacher Training a Paraprofessional to Promote Communication for a Student With Autism and Complex Communication Needs

Abstract

Students with complex communication needs require well-designed supports to communicate effectively in school settings. Although paraprofessionals are often charged with supporting these students, they are rarely trained in how to promote communication, and there is limited research on how they might be trained. In this study, we used a multiple baseline across behaviors design to test the efficacy of a teacher-implemented training package to train a paraprofessional to provide opportunities for an elementary student with autism to respond, set up opportunities for her to initiate, and implement a systematic prompting hierarchy. We demonstrated a functional relation between the teacher-implemented training and paraprofessional acquisition of the strategies, and paraprofessional implementation coincided with increased student communication. This study replicates previous findings that paraprofessionals can be trained to implement evidence-based practices to promote use of alternative/augmentative communication, and extends this literature by demonstrating that a classroom teacher can deliver paraprofessional training effectively.

Keywords

staff training augmentative/alternative communication paraprofessionals autism performance feedback

All children with autism spectrum disorder (ASD) have some level of social-communication impairment (American Psychiatric Association, 2013), but a subset of these children has particularly limited verbal repertoires (Tager-Flusberg & Kasari, 2013). Augmentative and/or alternative communication (AAC) strategies can empower these children to communicate effectively and appropriately. AAC encompasses any form of nonspeech communication, including sign language, picture symbols, gestures, and speech-generating devices (Beukelman, Mirenda, & Beukelman, 2013). Researchers have demonstrated that children with ASD with limited verbal skills can utilize AAC to communicate for a variety of purposes including requesting basic wants and needs (Ganz & Simpson, 2004), socializing with peers (Chung & Carter, 2013), and sharing experiences with family members (Thunberg, Sandberg, & Ahlsén, 2009).

Although AAC has the potential to dramatically improve functional communication outcomes, its efficacy is tied to how well it is supported and promoted. This is particularly evident in school settings, where researchers have documented that AAC-users rarely communicate with adults or peers without strategic intervention (Chung, Carter, & Sisco, 2012). There is evidence that school staff can utilize a number of simple strategies to effectively teach communication to students who use AAC. Examples include providing frequent opportunities for students to respond (OTR; Bingham, Spooner, & Browder, 2007), manipulating environments or routines to set up opportunities to initiate (OTI; Douglas, Light, & McNaughton, 2013), and using a least-to-most (LTM) prompting hierarchy to provide and then subsequently fade support (Bingham et al., 2007). The first strategy, OTR, involves inviting a student response to a question or a statement. Providing OTR not only increases student communication, but can also increase student engagement, improve achievement, and decrease problem behaviors (Cavanaugh, 2013). Without training, adults often provide relatively few OTR to AAC-users compared with other students (Baxter, Enderby, Judge, & Evans, 2012). The second strategy, OTI, involves manipulating the environment, violating the child’s expectations, or other approaches that encourage a student to initiate communication. Students who use AAC may rarely initiate communication without these focused efforts (Douglas, Light, & McNaughton, 2013). The third strategy, LTM prompting, involves providing increasing levels of assistance to enable students to communicate successfully. Communication partners can use this strategy when a student does not initially communicate given an OTR or OTI (Bingham et al., 2007). Furthermore, LTM prompting is designed to fade adult assistance over time and build independence.

Despite the efficacy of these strategies, many school staff have not been trained to use them (Baxter et al., 2012). In particular, paraprofessionals typically have little or no formal training in any instructional or support strategies (Carter, O’Rourke, Sisco, & Pelsue, 2009). This is of great concern, especially when one considers that special education paraprofessionals outnumber all other special education personnel (U.S. Department of Education, 2013) and often deliver instruction to students with complex communication challenges (Carter et al., 2009).

Very few studies have focused on training paraprofessionals to teach communication. We only identified three studies that involve training paraprofessionals to increase use of AAC. Douglas, Light, and McNaughton (2013) taught paraprofessionals two strategies to promote communication during play. The first strategy involved identifying activities and providing means for communication, locating and providing vocabulary, and environmental arrangement. The second strategy involved modeling AAC use, offering opportunities for communication, responding to communication, and extending interactions. The training consisted of four 2-hr sessions spread across 2 weeks and was conducted by the author who delivered one-to-one training to each paraprofessional. Paraprofessionals intervened with three preschoolers with developmental disabilities. Introduction of training coincided with an increase in the frequency of paraprofessional-provided opportunities to communicate across all three paraprofessionals. However, a clear and substantial increase in frequency of student communication was observed for only one of the three students.

Douglas, McNaughton, and Light (2013) conducted a similar study to train three paraprofessionals to teach preschool children with complex communication needs. One student was diagnosed with ASD, and the other two with developmental disabilities and hearing impairments. Training consisted of five online training modules that taught paraprofessionals to provide opportunities for communication, wait for the child’s communication, and then respond. Introduction of training coincided with increases in the frequency of paraprofessional-provided communication opportunities and responses across all three paraprofessionals. However, student effects were detected for only one of the three students.

In a third study conducted by a different research group, Bingham et al. (2007) trained three paraprofessionals to promote use of AAC with three students ages 8 to 16 with severe/profound intellectual disability or multiple disabilities who exhibited challenging behavior. In an initial 2-hr one-to-one training session, researchers trained each paraprofessional to provide opportunities for students to communicate, prompt communication using LTM prompting, and respond to student communication. In addition, they trained paraprofessionals to track and self-evaluate their own behavior. Introduction of training coincided with increases in paraprofessional-provided prompts and responses across all three paraprofessionals. Paraprofessional implementation coincided with clear and substantial increases in AAC use for two of the three students, and decreases in challenging behavior across all three students.

Collectively, these three studies provide important initial evidence about how to train paraprofessionals to increase use of AAC. However, these studies also have a number of limitations that leave important questions unanswered. First, paraprofessional implementation did not consistently lead to improvement of child communication. Across the three studies, clear effects on communication outcomes were only demonstrated for four of the nine students. Given these mixed outcomes, it is difficult to make clear recommendations for practice. Second, all three studies involved training paraprofessionals to simultaneously implement multiple strategies, so it is not possible to disaggregate the relative effects of each individual strategy. Therefore, it is unclear if one particular strategy or combination of strategies should be prioritized to improve student outcomes. Third, across all three studies, only one student diagnosed with ASD experienced an improved communication outcome. Further replication is needed to verify that paraprofessional training can result in improved outcomes for this population. Fourth, and perhaps most importantly, all three of the aforementioned studies involved training provided by external research teams. Therefore, it is unclear if special education teachers can effectively train paraprofessionals to teach communication.

We designed a study to address these limitations in the research literature. First, we sought to maximize the efficacy of the training by carefully selecting adult training strategies and facilitative communication strategies with a strong evidence base. Specifically, we designed a training that emphasized modeling, role-play, and performance feedback—three strategies supported by evidence in both the paraprofessional (Brock & Carter, 2013) and broader training literature (Snyder et al., 2012). We focused our training on paraprofessional-implemented OTR, OTI, and LTM—each validated in prior studies (i.e., Bingham et al., 2007; Douglas, Light, & McNaughton, 2013; Douglas, McNaughton, & Light, 2013; see review above). Second, we designed staggered training on three different strategies so that we could isolate the relative impact of each strategy. Third, we measured the impact of paraprofessional implementation on the communication of a student with ASD to further expand the research base for this specific population. Fourth, the first author—a classroom special education teacher—trained a paraprofessional in her own classroom to test the efficacy of teacher-delivered training.

Our study addressed two research questions:

Research Question 1: What is the efficacy of a training package (featuring modeling, role-play, and performance feedback) on the degree to which a paraprofessional implements OTR, OTI, and LTM prompting to promote communication for an elementary student with ASD?

Research Question 2: How does the frequency of student communication change when the paraprofessional implements each of these strategies?

Method

Participants

We recruited a paraprofessional from the first author’s classroom who worked with a student diagnosed with ASD who used AAC as a primary means of communication. The paraprofessional provided consent, the parent of the student provided permission, and the student provided assent to collect data for research purposes. The first author, a classroom teacher, implemented all paraprofessional training procedures.

Teacher

The teacher was a White 25-year-old female who had been a special education teacher for 2 years. She held a bachelor’s degree in speech pathology and audiology and was in her second year of master’s program in special education. The teacher had taken a number of undergraduate and graduate courses focused on AAC.

Paraprofessional

Lindsey was a White 26-year-old female who had been a paraprofessional for 2.5 years. She held a bachelor’s degree and did not have a teaching license or certificate. She had been working with the target student for 4.5 months. In the past year, Lindsey received a total of 4 hr of formal training. None of this training focused on promoting AAC use, and Lindsey reported that she had not received any training focused on promoting AAC use prior to this study.

Student

Angela was a 7-year-old girl diagnosed with ASD and intellectual disability who qualified for her state’s alternate assessment. Her primary means of communication was a high-tech AAC device, although she occasionally used one to two word utterances to request highly preferred items (e.g., chips, cookie). Her AAC system involved a specialized application (i.e., Language Acquisition Through Motor Planning [LAMP] Words for Life) that was programmed on an iPad. LAMP is a high-tech dynamic symbol-based communication system that is paired with text and audio speech output. Angela directly selected icons by touching them with her index finger. Angela’s Individualized Education Program (IEP) goals for expressive communication included requesting help, obtaining teacher attention, initiating with a peer, and producing novel two to five word utterances. Angela rarely communicated at school. Most communication on her AAC device was limited to one to two word messages that were often off-topic and did not appear to serve any clear function. For example, during a literacy lesson about digraphs and blends, Angela would push “credit card” or her mother’s name. On rare occasions, she would request highly preferred items (e.g., favorite songs, lemonade). However, even these rare requests might be difficult for a novel communication partner to interpret. For example, a novel communication partner might not understand that when Angela pushes the icon for “juice,” she is actually requesting assistance with inserting the straw into her juice box. The Vineland Adaptive Behavior II Checklist was completed and specific weaknesses were noted for communication and socialization domains. Her cognitive score on the Developmental Profile III was the age equivalent of 7 months. On the Receptive-Expressive Emergent Language Test–Third Edition (REEL-3; she had an age-equivalent score of 3 months for receptive language and 7 months for expressive language; <1st percentile). Angela received occupational, physical, music, and speech therapy. Each week, 90 min of speech therapy were delivered in a one-on-one setting outside of the classroom.

Setting and Materials

All training and intervention sessions took place in a private school and therapy center for students 3 to 16 years old with ASD, other developmental disabilities, and typically developing peers. There were approximately 150 students at the school, with just over 90% diagnosed with a developmental disability. Approximately half of the students at the school were White/non-Hispanic and about 75% were male. Throughout the center, adult-to-student ratios ranged from 1:1 to 1:3. In this particular classroom, there was one special education teacher, three paraprofessionals, seven students with ASD, and no typically developing peers. All students were 6 to 7 years old.

All data were collected during snack time. Snack time occurred midafternoon each day with all students and teachers sitting at one large table together. Duration of snack time ranged from 10 to 25 min. Angela had a lemonade drink specifically for snack and then items such as apples, pretzels, nuts, popcorn, and raisins leftover that she did not eat at lunch. Angela’s AAC device was available and within close proximity to her across all observation sessions.

Dependent Variables

We measured four dependent variables, which included paraprofessional implementation fidelity of three different evidence-based practices designed to increase the frequency of student communication. We video-recorded all sessions and coded all dependent variables from the video after each session. Data were collected using a frequency count for all three paraprofessional behaviors and the student behavior, and then converted into rate per minute by dividing the number of occurrences by the duration of the observation (in minutes).

Paraprofessional implementation fidelity

We measured implementation of three different evidence-based practices: creating OTR, OTI, and a LTM-prompting hierarchy. OTR were defined as asking a question, giving a direction, or otherwise communicating with the student in a way that necessitates a communicative response. Examples of OTR included, “What would you like for snack?” “What is your favorite food?” “Do you want goldfish or cookies?” Nonexamples included “I see you have goldfish today” and “push in your chair.” OTI were defined as violating the child’s expectations to provoke the need for communication. This included any behavior that clearly deviated from what would be anticipated in an activity or routine, such as withholding materials (e.g., withholding a desired snack item), providing incorrect or partial materials (e.g., giving her a drink without the straw), denying access to materials, failing to initiate an expected routine, or deviating from an established schedule (e.g., skipping a scheduled snack time).

The third dependent variable was LTM prompting. Implementation steps included (a) providing a 3 to 5 s delay with no prompting to give the student a chance to communicate independently without assistance, (b) reinforcing any communicative attempts with specific praise, (c) providing the next more intensive prompt (i.e., verbal, gestural, then physical) if needed, and (d) providing a 3 to 5 s response interval after each prompt. We counted a chain as correctly implemented only if implementation was consistent with the four criteria listed above.

Student communication

We also measured Angela’s communicative attempts. Communicative attempts were defined as any verbal (i.e., speech), gestural, or AAC device-supported utterance directed toward a person. We categorized the topography of each communicative attempt as (a) verbal speech, (b) use of high-tech AAC device, or (c) other (i.e., any communication that was neither verbal speech nor involved the high-tech AAC device).

Observer Training and Agreement

A secondary observer coded 30% of all sessions, balanced across experimental conditions. The first author trained the secondary observer, a graduate student, by reviewing a detailed coding manual, directing her to code practice videos, and providing feedback until the secondary observer achieved 100% agreement with the first author on two different practice videos. An agreement was defined as the secondary coder recording the same code within 5 s of the primary coder’s time stamp (i.e., either 5 s before or 5 s after). Point-by-point agreement was calculated by dividing total agreements by the total number of behaviors that the first author recorded. Overall, agreement was 94% (range = 86%–96%) across all variables and sessions. Agreement was 96% (range = 86%–100%) for OTR, 86% (range = 75%–100%) for OTI, 94% (range = 88%–100%) for LTM prompting, and 95% (range = 50%–100%) for the student’s communicative attempts.

Procedural integrity

The same secondary observer viewed 30% of all training and feedback sessions to measure and ensure procedural integrity of the training methods. The observer utilized 12 to 17 item implementation checklists for each specific training and feedback session. Overall, procedural fidelity was 94% (range = 89%–100%). Procedural integrity was calculated by dividing the number correct by the total number of opportunities.

Experimental Design

A multiple-baseline-across-behaviors design (Gast, Lloyd, & Ledford, 2014) involved staggered introduction of the independent variable (i.e., the training package) across the three paraprofessional-implemented strategies. Therefore, this design allowed opportunities to demonstrate and replicate effects for paraprofessional behavior. We measured student behavior as a secondary descriptive measure, but we did not test the efficacy of paraprofessional-implemented strategies on student behavior in the context of an experimental design. We introduced paraprofessional training for the first tier after 5 or more days of stable paraprofessional performance in the baseline conditions. We introduced the intervention in each subsequent tier after (a) baseline data were table for that tier, and (b) we detected an effect in the previous tier through visual analysis.

Procedures

Baseline

Before the introduction of intervention, Lindsey’s and Angela’s behaviors were observed during snack time. We did not provide any training or feedback during this condition. A typical snack time began with the students being instructed to get their lunch boxes out of the classroom lunch bin and to find a seat at the table. Students were allowed to have any food that was left in their lunchbox that they had not finished at lunchtime. Teachers helped the students open their items when asked, or sometimes before the students asked. Teachers sometimes sat at the table and had their own snack and they primarily talked among themselves. There were low levels of conversation between the teachers and the students, and even less between students.

Intervention

All training was delivered by the first author, a classroom teacher. Training consisted of an initial 10 to 20 min training for each target (i.e., 10 min for OTR, 10 min for OTI, and 20 min for LTM prompting), and then ongoing performance feedback after each intervention session. During the initial training session, the first author provided a rationale for why the given strategy was selected and how we anticipated it would affect student outcomes. For example, the first author explained that OTR are important for increasing engagement and opportunities for the student to communicate. Second, she defined the strategy and reviewed the associated implementation steps. For example, the author defined OTR as asking a question, giving a direction, or otherwise communicating with the student in a way that necessitates a communicative response. Third, she modeled examples of how to use the strategy, and then had Lindsey practice using the strategy through role-play. For example, they practiced greeting the student and asking the student questions. Fourth, she worked with Lindsey to brainstorm specific ways the strategy could be applied during classroom routines, and recorded these ideas on a planning sheet. Finally, she encouraged Lindsey to implement the strategy during the next observation, and invited Lindsey to ask any questions. One initial training session was implemented for each strategy, with the exception of LTM prompting—due to marked decrease in correct implementation, a second 20-min training session was conducted that matched the format of the initial training session. The first author instructed Lindsey to continue using each strategy even after training focused on a subsequent strategy (e.g., continue using OTR and OTI during LTM training).

Performance feedback sessions lasted approximately 5 min, and took place after each observation session. The first author took notes during observation sessions and watched the videos after the school day ended. Performance feedback was either delivered at the end of the school day or the next morning before school. The first author provided specific examples of positive implementation, explained any missed opportunities with corrective modeling, and answered any questions the paraprofessional had at that time.

Maintenance

After effects had been demonstrated across all three behaviors, training and feedback were terminated. Maintenance probes were collected 1, 2, 3, and 4 weeks after training was terminated to gauge the degree to which Lindsey continued to maintain implementation fidelity and Angela continued to demonstrate increased rates of communication.

Social validity

Following completion of data collection, Lindsey was given a questionnaire to determine if she believed the training and intervention were effective. Using a 5-point scale (1 = not effective at all; 2 = not very effective; 3 = somewhat effective; 4 = quite effective; 5 = very effective), she indicated her perceptions of (a) the effectiveness of the training and role-play for implementing each new strategy, (b) effectiveness of the feedback (after the sessions) on helping to implement each new strategy, and (c) the effect each strategy had on increasing the student’s use of AAC device. Her perceptions of the effectiveness of each of the three intervention strategies was assessed through separate questions using the same 5-point scale. She was asked if she would be likely to participate in a similar professional development opportunity in the future and if she would recommend this kind of opportunity to a colleague (1 = not at all likely; 2 = not very likely; 3 = somewhat likely; 4 = quite likely; 5 = very likely). Finally, she was asked open-ended questions about what she liked about the intervention strategies, what she disliked about the intervention strategies, what she liked about the training process, and what she did not like about the training process.

Results

Paraprofessional training coincided with clear and substantial changes in Lindsey’s implementation of all three strategies. Together, these three effects establish a functional relation. Furthermore, paraprofessional implementation of strategies coincided with substantial increases in student communication. Although informative, student communication was a secondary variable that was not manipulated experimentally; therefore, no functional relation for student behavior can be established. Both paraprofessional and student data are displayed in Figure 1.

Figure 1.

Rate per minute of paraprofessional implementation behaviors (top three tiers) and student communication (bottom tier).

OTR

Lindsey’s rate of OTR during baseline was flat, with only one OTR across five sessions. With the introduction of the intervention, OTR immediately increased. After the first session, data were somewhat variable (range = 0.58–1.48) but maintained at a flat level through the training condition. Lindsey’s rate of creating OTR decreased somewhat during the maintenance condition (range = 0.4–1.3); however, these lower rates during maintenance probes still markedly exceeded baseline levels.

OTI

Across the 11 days of baseline, Lindsey rate of OTI was flat (i.e., all points were zero). After introduction of the intervention, the rate of OTI immediately increased. Although there was some variability (range = 0.26–1.29), data remained relatively stable through the end of the intervention condition and maintained at this level during the maintenance condition with similar variability (range = 0.43–1.29).

Systematic prompting

Lindsey’s rate of correct LTM-prompting implementation was flat in the baseline condition (i.e., at or near zero). With the introduction of the intervention, there was an immediate increase in level for the first two sessions. There was a sharp decrease in level on Day 3 after a weekend (.08). Because of this sharp drop in performance, the first author delivered a second didactic training session (labeled “second training session” in Figure 1). Specific implementation errors that were corrected during the second training session included providing two different prompts simultaneously, or not providing specific praise following a communicative attempt. The retraining session was formatted similarly to the initial training, although the first author placed greater emphasis on the specific implementation steps with the highest rates of errors. Following the second training session, the rate of correct implementation increased and then maintained for the remainder of the intervention condition, with some variability (range = 0.47–1.42). Lindsey maintained the systematic prompting hierarchy at the highest rate of the three behaviors during maintenance probes (range = 0.57–1.2).

Communicative attempts

Across baseline sessions, the target student had a total of two communicative attempts (range = 0–0.27). After Lindsey began to provide OTR, Angela’s rate of communication showed some variability and a very small change in level. After Lindsey began to provide OTI, Angela’s rate of independent communication increased in level but was still somewhat variable. After Lindsey began implementation of LTM prompting, Angela’s overall rate of communication increased. Most of this increase can be attributed to prompted communication. The student’s communicative attempts exceeded baseline levels for the remainder of the study.

The majority of communicative attempts (81%) were independent, and the remaining 19% were prompted. In baseline, both communicative attempts were independent. During the OTR condition, eight communicative attempts were independent, two followed a verbal prompt, three followed a gestural prompt, and one communicative attempt was physically prompted. During the OTI condition, 34 communicative attempts were independent and two were physically prompted. During LTM, 57 communicative attempts were independent, 20 followed a verbal prompt, 18 followed a gestural prompt, and 32 were physically prompted. Across all conditions, 96% of all communicative attempts involved the high-tech AAC device, 1% were verbal, and the remaining 3% were categorized as involving other methods of AAC (e.g., pointing to an object in a field of choices). The student’s verbal repertoire included “wa-wa” (water), “no,” “bye,” or “done.” AAC communication consisted of a range of words from requesting specific snack items (e.g., lemonade, apples, popcorn), answering yes or no to a question, to talking about the months in a year. Other communication most commonly involved pointing to make a selection in response to an OTR.

Social Validity

Lindsey’s written responses on a questionnaire indicated that she perceived the initial training and role-play for each strategy to be very effective in helping her implement new strategies (i.e., 5 on a 5-point scale). Similarly, she indicated that the feedback following all sessions was very effective. When asked about the intervention strategies, she indicated that OTR were somewhat effective (i.e., 3 on a 5-point scale), OTI were somewhat effective, and that LTM prompting was very effective (i.e., 5 on a 5-point scale). Lindsey stated that she was very likely (i.e., 5 on a 5-point scale) to use these same strategies with the same student or a different student in the future. Her overall feedback about the training was positive, and she indicated that she would be very likely to both participate in a similar training opportunity in the future and recommend this kind of training opportunity to a colleague (i.e., 5 on a 5-point scale).

In addition to this survey data, the first author recorded observational notes about Lindsey’s intervention behavior with other students in the classroom with complex communication challenges. According to these observational notes, Lindsey created more OTR across a number of students after receiving training. She tended to provide these opportunities during natural opportunities for social conversation (e.g., arrival, dismissal, snack, lunch). She asked questions such as “What are you all doing after school tonight?” “What are you being for Halloween?” “What do you guys have for snack today” or “Who else likes popcorn?”

Discussion

In this study, we tested the efficacy of a training package that featured promising training methods (i.e., modeling, role-play, and performance feedback) on a paraprofessional’s implementation of three evidence-based practices for promoting increased communication with a student with ASD and complex communication needs. The paraprofessional implemented all three practices accurately, and the rate of student communication improved substantially. These findings extend the research literature in a number of key ways.

First, this study demonstrated that a special education teacher can train a paraprofessional to implement strategies designed to increase communication for a student with ASD and complex communication needs. Although the law requires that paraprofessionals be appropriately trained and supervised (Individuals With Disabilities Education Improvement Act, 2004), there is a lack of evidence-based models for accomplishing this goal. Practically, the individuals best positioned to provide training to paraprofessionals are likely the special education teachers who are already responsible for their supervision (Brock & Carter, 2013). Although many teachers report that they are not well prepared to train and supervise paraprofessionals (French, 2001), this study shows for the first time that if teachers were prepared to use effective training strategies, they could train paraprofessionals to implement evidence-based practices that increase communication for students who use AAC. Furthermore, findings from this study, combined with those from previous studies (i.e., Bingham et al., 2007; Douglas, Light, & McNaughton, 2013; Douglas, McNaughton, & Light, 2013) provide initial evidence across research groups that paraprofessional-implemented strategies can improve communication outcomes for students with ASD.

Second, results from this study provide insight into the relative impact of different paraprofessional-implemented practices on student communication. When the paraprofessional created more OTR, the student’s rate of communicative attempts remained unchanged. When the paraprofessional set up OTI, the student’s independent communication increased markedly. We suspect that the difference in effects between the two strategies relates to the student’s motivation to communicate. When the paraprofessional provided OTRs (e.g., “How are you doing?”), the student did not have an opportunity to access a highly desired object or interactions; the only natural reinforcer was further social interaction. In contrast, when the paraprofessional provided OTI (e.g., provided a juice box with no straw), the student could only access a highly desired object (e.g., the juice box) through communication. Finally, when the paraprofessional began to use LTM prompting, the student’s prompted communication increased markedly, but her independent communication remained relatively unchanged. This change suggests that prior to LTM prompting, student communication was mostly limited to utterances that were within her repertoire and allowed her to access highly desired objects or interactions. With LTM prompting, the student began to use vocabulary not yet in her independent repertoire.

Related to this point, based on her responses to the social validity questionnaire and her maintenance of strategies, the paraprofessional also perceived that OTIs and LTM prompting (and not OTRs) had the greatest impact on student communication. In the social validity survey, she indicated that she perceived OTRs to be just somewhat effective. Furthermore, she maintained implementation of OTIs and LTM prompting and much higher rates than OTRs after training was terminated, suggesting she prioritized the strategies she perceived to be most effective.

Third, baseline data from this study suggest that paraprofessionals may not naturally perceive nonacademic times as rich opportunities to promote student communication. Lindsey rarely provided any opportunities for communication opportunities, nor did she prompt communication behaviors prior to training. When considered alongside the results from previous studies, results from this study strengthen existing evidence that many paraprofessionals are quite willing and capable of promoting student communication. However, they need training and guidance to both identify missed opportunities and implement effective strategies.

Implications for Practice

Findings from this study have important implications for practice. First, special educators can and should provide focused training to paraprofessionals that enables them to increase the communication of students who use AAC. To provide training, special educators themselves must be skilled in promoting communication. We recommend utilizing a combination of modeling and ongoing performance feedback. Although we recommend focused training for paraprofessionals, we also advise teachers to thoughtfully consider how they allocate responsibilities to paraprofessionals. Licensed teachers should be responsible for teaching new communication acts to students with complex communication, although it may be appropriate for paraprofessionals to target the improvement of these skills. It is important for teachers to think carefully when allocating classroom responsibilities, and to prioritize teacher-implemented focused interventions with students with the most complex support needs (Giangreco, Doyle, & Suter, 2012). One barrier that teachers frequently cite to providing higher quality training and supervision is time (e.g., Breton, 2010); administrators could help mitigate this barrier by providing designated time for training and planning when possible.

Second, teachers should consider individual student characteristics and communication profiles when designing trainings for paraprofessionals. Implementation of the strategies discussed in this study would differ based on student characteristics. For example, we trained a paraprofessional to use a physical prompt as the most intrusive prompt in the prompting hierarchy because the student was using direct selection on a high-tech AAC device; a model prompt might be more appropriate for a student who uses sign language. Similarly, the impact of OTI is linked to student’s motivation, so teachers should consider what items or activities are highly motivating. It may be helpful to implement a preference assessment.

Third, given how rapidly the parparofessional acquired OTI and the immediate impact on student outcomes, teachers might consider prioritizing OTI. In contrast, OTR had less of an impact on student communication and LTM prompting was more difficult for the paraprofessional to acquire. We also note that we staggered training on practices in a particular order for the purpose of maintaining experimental control and gauging the impact of individual strategies on student communication; in everyday practice, it might be more efficient for teachers to provide training on more effective strategies first (e.g., OTI) or focus on more than one strategy at the same time.

Limitations and Future Research

There are several limitations to this study. First, this paraprofessional was selected from the first author’s classroom, and is likely not representative of all paraprofessionals. Researchers might implement similar studies with other paraprofessionals with different characteristics (e.g., different levels of experience, age) to determine if our effects replicate. Second, we did not collect generalization data, so it is unclear if the paraprofessional would be able to generalize these strategies to new students and situations, or if the student’s rate of communication increased in other situations. In future studies, researchers could collect these data and program for generalization as needed. They should focus on a range of contexts other than food-based activities, and they might also focus on training peers to implement these strategies to promote communication. Third, two of the three paraprofessional-implemented strategies focused on changing opportunities for communication. Although beneficial, these strategies did not actually teach or increase communication skills not already within the student’s repertoire. In future studies, researchers might focus more heavily on strategies that teach new communication acts.

Conclusion

During the school day, there are often many missed natural opportunities to promote and expand communication for students with ASD and complex communication needs. Given their lack of formal training, it is unsurprising that paraprofessionals often struggle to identify and capitalize on these opportunities. This study demonstrates that teacher-delivered training can enable a paraprofessional to increase student communication. Although this study focused only on one student during one part of the school day, this simple training package could likely be used to target any student who has complex communication challenges during any part of the school day. If all special education staff were trained to provide and capitalize on communication opportunities across the school day, they could have a profoundly positive impact on the lives of students with complex communication challenges.

Footnotes

Acknowledgements

We would like to thank Dr. Diane Sainato for her advice and support. We would also like to thank Alana Oif for her assistance with data collection.

Rachel Seaman is now a Postdoctoral Fellow at Emory University.

Declaration of Conflicting Interests

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

Funding

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

References

American Psychiatric Association. (2013). Diagnostic and statistical manual of mental disorders (5th ed.). Arlington, VA: American Psychiatric Publishing.

Baxter

Enderby

Judge

Evans

(2012). Barriers and facilitators to use of high technology augmentative and alternative communication devices: A systematic review and qualitative synthesis. International Journal of Language & Communication Disorders, 47, 115–129. doi:10.1111/j.1460-6984.2011.00090.x

Beukelman

D. R.

Mirenda

Beukelman

D. R.

(2013). Augmentative and alternative communication: Supporting children and adults with complex communication needs (4th ed.). Baltimore, MD: Paul H. Brookes.

Bingham

M. A.

Spooner

Browder

(2007). Training paraeducators to promote the use of augmentative and alternative communication by students with significant disabilities. Education & Training in Developmental Disabilities, 42, 339–352.

Breton

(2010). Special education paraprofessionals: Perceptions of preservice preparation, supervision, and ongoing developmental training. International Journal of Special Education, 25, 34–45.

Brock

M. E.

Carter

E. W.

(2013). A systematic review of paraprofessional-delivered instruction to improve outcomes for students with intellectual and developmental disabilities. Research & Practice for Persons With Severe Disabilities, 38, 211–221. doi:10.1177/154079691303800401

Carter

E. W.

O’Rourke

Sisco

L. G.

Pelsue

(2009). Knowledge, responsibilities, and training needs of paraprofessionals in elementary and secondary schools. Remedial and Special Education, 30, 344–359. doi:10.1177/0741932508324399

Carter

E. W.

Sisco

L. G.

Brown

Brickham

Al-Khabbaz

Z. A.

(2008). Peer interactions and academic engagement of youth with developmental disabilities in inclusive middle and high school classrooms. American Journal on Mental Retardation, 113, 479–494.

Cavanaugh

(2013). Performance feedback and teachers’ use of praise and opportunities to respond: A review of the literature. Education & Treatment of Children, 36, 111–137.

10.

Chung

Y. C.

Carter

E. W.

(2013). Promoting peer interactions in inclusive classrooms for students who use speech-generating devices. Research & Practice for Persons With Severe Disabilities, 38, 94–109. doi:10.2511/027494813807714492

11.

Chung

Y. C.

Carter

E. W.

Sisco

L. G.

(2012). A systematic review of interventions to increase peer interactions for students with complex communication challenges. Research & Practice for Persons With Severe Disabilities, 37, 271–287. doi:10.2511/027494813805327304

12.

Douglas

S. N.

Light

J. C.

McNaughton

D. B.

(2013). Teaching paraeducators to support the communication of young children with complex communication needs. Topics in Early Childhood Special Education, 33, 91–101. doi:10.1177/0271121412467074

13.

Douglas

S. N.

McNaughton

Light

(2013). Online training for paraeducators to support the communication of young children. Journal of Early Intervention, 35, 223–242. doi:10.1177/1053815114526782

14.

French

N. K.

(2001). Supervising paraprofessionals: A survey of teacher practices. The Journal of Special Education, 35, 41–53. doi:10.1177/002246690103500105

15.

Ganz

J. B.

Simpson

R. L.

(2004). Effects on communicative requesting and speech development of the Picture exchange communication system in children with characteristics of autism. Journal of Autism and Developmental Disorders, 34, 395–409. doi:10.1023/b:jadd.0000037416.59095.d7

16.

Gast

D. L.

Lloyd

B. L.

Ledford

J. R.

(2014). Multiple baseline and multiple probe designs. In Gast

D. L.

Ledford

J. R.

(Eds.), Single subject research methodology in behavioral sciences (pp. 251–296). New York, NY: Routledge.

17.

Giangreco

M. F.

Doyle

M. B.

Suter

J. C.

(2012). Constructively responding to requests for paraprofessionals: We keep asking the wrong questions. Remedial and Special Education, 33, 362–373. doi:10.1177/0741932511413472

18.

Individuals With Disabilities Education Improvement Act, 20 U.S.C. § 1400 (2004).

19.

Snyder

Hemmeter

M. L.

Meeker

K. A.

Kinder

Pasia

McLaughlin

(2012). Characterizing key features of the early childhood professional development literature. Infants & Young Children, 25, 188–212.

20.

Tager-Flusberg

Kasari

(2013). Minimally verbal school-aged children with autism spectrum disorder: The neglected end of the spectrum. Autism Research, 6, 468–478.

21.

Thunberg

Sandberg

A. D.

Ahlsén

(2009). Speech-generating devices used at home by children with autism spectrum disorders: A preliminary assessment. Focus on Autism and Other Developmental Disabilities, 24, 104–114.

22.

U.S Department of Education (2013). IDEA Section 618 data products. Retrieved from https://www2.ed.gov/programs/osepidea/618-data/index.html