Abstract
This study examined the efficacy of an intervention package consisting of small group instruction, online instructional modules, and telecoaching to increase planning skills of eight young adults (ages 17–26) with autism spectrum disorder (ASD) and other related neurological disorders. Telecoaching sessions used existing video conference platforms (e.g., FaceTime, Google Hangouts, Skype, Zoom) with young adults as the direct recipients of the coaching. The primary dependent variable was number of planning steps described on fictional vignettes, and a supplementary dependent variable was self-reported participation in weekly social activities. A multiple baseline across participants design with participants assigned to three cohorts was utilized. Results indicated all participants increased knowledge of steps required to plan activities. However, implementation of the planned activities varied with minimal gains demonstrated by five of the eight participants. Implications for practice and research related to technology-based interventions to increase social engagement of young adults with ASD is discussed.
Autism spectrum disorder (ASD) is a neurological disorder associated with impairments in social communication skills and repetitive patterns of behavior, interests, and activities (American Psychiatric Association, 2013). The core symptoms of ASD have been associated with quality of reciprocal friendships maintained by individuals with ASD, with greater severity associated with fewer and lower quality friendships (Mazurek & Kanne, 2010). Individuals with ASD commonly experience a lack of social activities with same-age peers (Orsmond et al., 2013) and these challenges with social relationships likely continue into adolescence and adulthood (Howlin, 2000). Specifically, approximately 50% of young adults with ASD experience no or very limited social activities with friends on a regular basis and only one third engage with same-age peers in community-based social activities (Shattuck et al., 2011). This is concerning because engagement with peers buffers the impact of perceived stress on quality of life (Bishop-Fitzpatrick et al., 2017).
Despite the challenges associated with the disorder, young adults with ASD often report a desire for meaningful relationships with others (Müller et al., 2008). For example, individuals with ASD may exhibit highly specialized interests and activities referred to as circumscribed interests (Turner-Brown et al., 2011) or special interest areas (Bross & Travers, 2017). Specialized interests in nonsocial activities are common and these interests can interfere with both individual and group activities (Turner-Brown et al., 2011). According to parent report, young adults with ASD spent an average of 28.41 hr per week in solitary activities and an average of 11.81 hr using the computer or playing video games (Neary et al., 2015). The number of hours parents estimated young adults with ASD spent in structured social activities was only 1.64 hr a week and 5.81 hr in unstructured social activities (Neary et al., 2015). As a result of these factors, fostering social connections with others is a critical component of the successful transition to adulthood for individuals with ASD (Carter et al., 2013).
Potential shared social activities for young adults with ASD are no different for any young adult regardless of disability status and can include activities such as calling and/or texting peers and participating in leisure activities. Individuals with ASD have reported high levels of enjoyment when interacting with friends (S.-F. Chen, Chien, et al., 2016). However, individuals with ASD have also reported difficulty initiating social interactions (Müller et al., 2008) and often experience in-the-moment anxiety when attempting to engage in social contexts (S.-F. Chen, Chien, et al., 2016). Many individuals with ASD report social anxiety as a contributing factor for the lack of social engagement (Maddox & White, 2015). For instance, 21% of young adults with ASD have reported mild social anxiety, 29% reported moderate social anxiety, and 21% reported severe social anxiety (Maddox & White, 2015). Despite a desire for social interaction, comorbid anxiety and fear of negative evaluations by others appears to contribute to challenges with positive social experiences. Young adults with ASD may also be inhibited due to challenges with planning and executing social activities.
Executive Functioning Skills and Interventions to Teach Planning
Participating in social activities with others requires advance planning (e.g., where and when to meet). The skills required to plan a social activity may best be described as executive functioning skills, which are mental processes and cognitive skills that serve independent, purposeful, and goal-directed behavior (Lezak et al., 2012). Individuals with ASD often encounter executive functioning difficulties (Kiep & Spek, 2017), with some difficulties seemingly interrelated to specific symptoms of ASD (de Vries & Guerts, 2012). For example, Leung and colleagues (2015) reported executive functioning skills (e.g., initiation, working memory, planning, organization) predicted social functioning only in children and adolescents with ASD but not in typically developing peers. Similarly, poorer executive functioning of children with ASD is associated with higher levels of isolation on the playground and less engagement with peers (Freeman et al., 2017). Although executive functioning skills of individuals with ASD are variable over time and often change with development (Pellicano, 2010), research suggests youth with ASD continue to experience deficits in working memory and planning skills across their life span (Y. W. Chen, Bundy, et al., 2016). This is supported by youth with ASD who have reported an increased difficulty in planning skills as they mature to adolescence (van den Bergh et al., 2014). Executive functioning skills appear particularly important to develop in transition-aged youth with ASD given demands of adult life, such as structuring daily activities, using a calendar, organizing materials, and maintaining attention (Conley, 2008).
Similar to the findings on contextually based executive functioning interventions for children with ASD (Kenworthy et al., 2013), interventions designed to increase planning skills of young adults with ASD should occur in convenient and natural environments. These interventions should target specific social skills such as initiating activities with others, selecting activities, and successfully implementing the planned activities. Fortunately, there are several interventions with empirical support that can be used to promote planning skills and, in turn, potentially increase participation in social activities of young adults with ASD. Technology-based interventions are particularly advantageous given the various forms of technology readily available for adolescents and young adults in today’s modern society. A brief overview of interventions utilized in the current study is provided below.
Social skills training
Social skills training involves one-on-one or group instruction designed to teach social interaction skills in a variety of contexts (Wong et al., 2015). Social skills training can include a variety of instructional activities such as role-play, practice with feedback, social games, and natural interactions with peers. While implementation of social skills training is variable based on different factors (e.g., resources, individual needs, finances), both group social skills training (Hotton & Coles, 2016) and virtual social skills training (Craig et al., 2016) have been found effective. In addition, individuals with ASD who participated in social skills training reported gains in their social knowledge and mental health outcomes such as a decrease in anxiety, depression, and loneliness (Gantman et al., 2012).
Instructional modules
There are numerous features of online learning that appear to align well with the characteristics associated with ASD. For example, the structure of online courses and modules often provide consistent learning formats, distinct organization of content, and visual presentation of material (Sabella & Hart, 2014). Promising evidence exists to support video-based instruction to enhance social and communication skills of individuals with ASD (DiGennaro-Reed et al., 2011). For example, Wainer and Ingersoll (2011) reviewed 14 studies that utilized innovative multimedia programs to teach language, emotion recognition, or social skills. The authors concluded interactive multimedia programs are a promising strategy for providing direct intervention to individuals with ASD.
Telecoaching
Telecoaching is a coaching or therapy session that occurs using a telephone or video conference platform. Telecoaching is a viable intervention option when in-person delivery of services is not available for a variety of reasons (e.g., rural area, professional with large number of clients). The existing telecoaching literature for individuals with ASD is limited due to a small number of studies (Neely et al., 2017) and has primarily focused on younger children with parents as the recipients of online interventions (e.g., Simacek et al., 2017). For example, telehealth has been used to provide applied behavior analytic services to three young children with ASD or Rett syndrome (Simacek et al., 2017) with parents serving as the intervention agents. Similarly, a cognitive-behavioral intervention for youth with ASD and anxiety, ages 7–19 years, was delivered via telehealth in a group format with families present (Hepburn et al., 2016). However, a major weakness in the telecoaching literature is a lack of reported procedural fidelity, with only 10 of the 19 studies in a systematic literature review having reported fidelity of coaching procedures (Neely et al., 2017). Additional research on telecoaching with procedural fidelity appears necessary. In addition, despite the high prevalence of solitary computer activity among young adults with ASD (Neary et al., 2015), no studies to our knowledge have implemented telecoaching with individuals with ASD without a parent or therapist present. Therefore, the investigation of telecoaching with the inclusion of procedural fidelity as an intervention provided directly to young adults with ASD without parental or therapist supervision is a viable area of inquiry. Further, many young adults with ASD likely have access to existing online video conference platforms that could be used for telecoaching.
Accordingly, the purpose of this study was to examine the efficacy of an intervention package titled ASD On The Go to develop planning skills of young adults with ASD. The intervention package consisted of in-person small group instruction, online instructional modules, and telecoaching. We were guided by the following research questions: (a) To what extent do young adults with ASD increase the number of planning steps described on vignettes to demonstrate knowledge of planning social activities as a result of the ASD On The Go intervention package? (b) What is the social validity of the ASD On The Go intervention package as reported by young adults with ASD? Finally, a supplementary question based on self-report was (c) To what extent do young adults with ASD increase their participation in weekly social activities as a result of the ASD On The Go intervention package?
Method
Participants and Setting
This study was approved by a university institutional review board prior to recruitment and data collection. All participants attended a young adult social skills class twice a month at a nonprofit organization that provided a range of services (e.g., licensed therapy, summer camp, parent workshops) to individuals with ASD and their families. The social skills class was taught by a licensed special education teacher and covered basic social skills such as taking turns during a conversation, solving problems, and recognizing emotions. Participants met the following eligibility criteria: (a) were a minimum of age 16, (b) had a diagnosis of ASD or other related neurological disorder as confirmed by records at the nonprofit organization, (c) expressed an interest in increasing their planning skills and participation in weekly social activities, and (d) agreed to complete online instructional modules and participate in telecoaching sessions outside the regularly scheduled social skills classes. Participation was voluntary, and young adults received a US$150 stipend for their participation.
Seven males and one female, ages 17–26, provided signed consent to participate. Table 1 provides demographic information for all participants, including their hobbies and interests. Participants were randomly assigned to cohorts using a number generator. Cohort 1 consisted of Jeremiah and Hannah; Cohort 2 consisted of Bruce, Kai, and Michael; and Cohort 3 consisted of Charles, Adam, and Martin. Three participants had a diagnosis of ASD alone: Jeremiah, Bruce, and Charles. Four participants had a comorbid diagnosis of ASD and another disorder such as social anxiety disorder (Hannah), generalized anxiety disorder (Adam and Martin), Obsessive Compulsive Disorder (OCD; Kai), or attention-deficit hyperactivity disorder (Martin). Michael was the only participant without ASD. However, Michael had a diagnosis of OCD and Tourette syndrome, exhibited social skill support needs, and expressed an interest in participating. Therefore, he was included in the study.
1. Participant Demographics.
Note. ADHD = attention-deficit hyperactivity disorder; AS = Asperger Syndrome; HS = high school; HFA = high-functioning autism; F = female; M = male; N/A = not applicable; NASA = National Aeronautics and Space Administration; OCD = obsessive compulsive disorder; PDD-NOS = Pervasive Developmental Disorder–Not Otherwise Specified.
Some of the participants had highly specialized interests, such as cowboys (Michael), action figures (Charles), or National Aeronautics and Space Administration (NASA; Adam). Participants were engaged in a variety of secondary or postsecondary activities. Three participants (Jeremiah, Bruce, and Kai) were enrolled in their final year of high school at the time of the study and received special education services in the category of educational autism. The other five participants had graduated high school and no longer received special education services. Of these five participants, one was pursuing postsecondary education at a local community college (Adam), two were competitively employed (Michael and Charles), and two were unemployed and not currently participating in postsecondary educational activities (Hannah and Martin). Adam was pursuing an associate degree in physics. Michael was employed at a grocery store, and Charles was employed at a courier delivery service company. Hannah was actively seeking employment related to her bachelor’s degree in biology. Similarly, Martin was actively seeking employment related to his associate degree in fine arts.
All participants were administered the Behavior Rating Inventory of Executive Function (BRIEF; Gioia et al., 2000) or BRIEF–Adult Version (BRIEF-A; Roth et al., 2005) and Social Responsiveness Scale, 2nd edition (SRS-2; Constantino & Gruber, 2012) for participant description purposes (see Table 2). Results of the BRIEF or BRIEF-A indicated Jeremiah and Martin exhibited the highest impairments in executive functioning. All other participants exhibited moderate levels of impairment in executive functioning, while Charles scored within a mild range. Results of the SRS™-2 indicated Hannah’s social behavior was in the severe range as reported by herself and moderate range as reported by her parent. Jeremiah and Kai’s social behavior was also in the moderate range as reported by their parents. Results of the SRS™-2 for Michael indicated that his social behavior was within standard limits, consistent with his lack of an ASD diagnosis. The SRS-2 indicated all other participants scored in the mild range of social functioning. All participants used verbal language as their primary means of communication.
Participant Standardized Assessment Results.
Note. BRIEF-A or BRIEF scores reported are the Global Executive Composite score percentile, where higher percentile scores indicate greater support needs. SRS™-2 scores reported are the total score results, where a score of 59 or below indicates social behavior within normal limits, 60–65 indicates a mild range of social deficiencies, 66–75 a moderate range, and 76 or higher a severe range. — indicates participant completed alternate forms based on requirements as determined by age. * indicates participant did not return form. BRIEF-A® = Behavior Rating Inventory of Executive Function–Adult Version; BRIEF® = Behavior Rating Inventory of Executive Function; SRS™-2 = Social Responsiveness Scale™, 2nd edition.
Settings for the study varied according to the intervention package components. Small group instruction occurred in a classroom at the nonprofit organization that contained a table, whiteboard, and chairs. Instructional modules were administered online through a web-based platform. Participants completed the modules at both the nonprofit organization classroom and their homes. Telecoaching sessions occurred online through video conference platforms with participants primarily at home during the telecoaching sessions. Social activities occurred at locations within the communities where participants lived (e.g., restaurants, movie theaters).
Interventionists and Materials
The interventionist for small-group instructional sessions was the first author who was a special education doctoral student at the time of the study and licensed special education teacher. The first author, second author (counseling psychology doctoral student), and third author (research assistant) served as the coaches during all telecoaching sessions. All interventionists were experienced working with young adults with ASD in one-on-one and small group formats.
Online instructional modules
Participants completed four password-protected online instructional modules related to (1) planning activities, (2) using a personal calendar, (3) self-monitoring, and (4) developing routines (ASD On The Go Modules, 2019). These modules consisted of visual and auditory information presented via slideshows and video delivered through smart phones, handheld tablets (e.g., iPad), or laptops owned by participants. Checks for understanding in the form of multiple-choice and short answer questions were incorporated throughout the modules. One module typically required 20–25 min to complete. The planning module focused on how to get organized to plan and participate in chosen activities. The using a calendar module focused on how to set up an electronic calendar and enter appointments and reminders into the calendar. The self-monitoring module focused on how self-monitoring can potentially assist to reach goals and become more independent. Finally, the developing routines module focused on how to develop positive habits and routines. All modules are available for free at asdonthego.ku.edu.
Planning Activity Checklist
A paper checklist consistent with content presented in the planning module was used to teach the following steps required to plan an activity: (1) time, (2) cost, (3) location, (4) transportation, and (5) materials. Space was provided to the right of each step for participants to write details regarding these steps according to their chosen activity. For example, if participants planned to see a movie, they may have written on the Planning Activity Checklist: (1) 7:10 p.m., (2) US$12.00, (3) address of the movie theater, (4) my friend will pick me up, and (5) I will bring my wallet, cell phone, and jacket.
Telecoaching platforms
Telecoaching sessions were implemented using existing video conference applications (apps), software, or platforms. The four types of video conference platforms utilized were FaceTime, Google Hangouts, Skype, and Zoom. FaceTime is a video chat app operated by Apple and is available on supported iOS mobile devices and computers that run Mac OS X Version 10.6.6 and later. Google Hangouts is a communication service that is built into Google+ and Gmail email services. Mobile Google Hangout service apps are available for both iOS and Android devices. Skype is a video conference platform that is built for one-on-one and group conversations and is available via mobile, PC, Xbox, and Alexa. Finally, Zoom is a communications software that combines video conferencing, online meetings, and chat features. Research staff asked participants their platform preference for telecoaching sessions based on their previous usage, device capabilities, internet connectivity, and overall comfort. All video conferences were held using desktops, laptops, handheld tablets, and smart phones that participants owned. Handheld tablets would have been provided to participants without a device; however, all participants owned devices with sufficient capabilities to participate in the study.
Dependent Variables
Planning vignettes
The primary dependent variable was participant scores on planning vignettes modeled after the story-based items in the self-regulation domain of the ARC’s Self-Determination Guide (Wehmeyer, 1995). The first author created the planning vignettes and other research staff provided feedback to ensure consistency across all vignettes used in the study. The vignettes were selected as a means to measure planning skills because of the self-regulated behaviors required to plan activities. In addition, vignettes could be completed within the established class structure at the nonprofit organization. Participants were presented with the beginning of a story (e.g., “I want to have coffee with my friend this weekend”) and corresponding end to the same story (e.g., “My friend and I went to a coffee shop on Saturday morning”). Participants wrote the middle of the story to demonstrate their knowledge of specific steps required to plan and participate in the sample activity. Stories were scored by two research staff on a 5-point scale according to the five planning steps presented on the Planning Activity Checklist. Specifically, participants were awarded 1 point for each step of the Planning Activity Checklist described in sufficient detail (e.g., Transportation: My brother will drive me to the coffee shop). Research staff offered to serve as a scribe for participants during all conditions of the study as an accommodation. However, no participants opted to use a scribe.
Social activity sample
A supplementary dependent variable was the number of weekly social activities the young adults participated in as measured by self-report. Participants were asked to complete a log regarding the number and type of social activities participated in the prior week during the regularly scheduled social skills class. The log consisted of a simple form with dates of the previous week and space for participants to write specific activities they participated in during that week. This log represented a sample as participants reported only 1 week per month of their social activities prior to the scheduled social skills class. The mean number of activities participated in per week was calculated for each participant before and after implementation of the ASD On The Go intervention package. Therefore, the social activity sample represented a pre- and postmeasure to supplement the primary dependent variable.
Experimental Design
The efficacy of the intervention package was examined using a multiple baseline across participants design (Gast et al., 2018) with participants assigned to three cohorts. This design and cohort model was selected because there were eight participants who attended the same social skills class and received the intervention package in a staggered manner. The study took place over a 28-week period from October 2017 to May 2018 with some breaks in data collection that corresponded with common school year breaks (e.g., winter break, spring break). Participants received the intervention package according to their assigned cohort (i.e., Cohort 1 received the intervention first, followed by Cohort 2, and finally Cohort 3). The intervention condition for each cohort was approximately 6 weeks in duration. Small group instruction occurred with members of the same cohort, and telecoaching sessions occurred individually between participants and research staff. The 5-point scale planning vignette scores were converted to percentages for each participant for reporting and graphing purposes.
The ASD On The Go intervention package was introduced to each cohort in a systematic manner following a minimum of 5 data points and once a pattern of responding during baseline was clearly established. After the primary intervention condition for one cohort ended, that cohort faded (i.e., decreased) the number of telecoaching sessions and the next cohort began the intervention condition. Fading consisted of no small group instruction and telecoaching sessions implemented once a week or every other week rather than twice a week. The fading component was incorporated as an integral component of the intervention condition to promote increased autonomy of all participants given one of the purposes of the study was for young adults to be causal agents (Shogren et al., 2017) in their own lives without reliance on parents or professionals (e.g., research staff, teachers). Phase changes were determined by visual analysis while also taking into account the schedule at the nonprofit organization. Three maintenance probes were conducted on the planning vignettes for all participants to evaluate the potential sustained effects of the intervention package.
Procedures
Baseline
Participants did not engage in small group instruction or telecoaching sessions during baseline, nor did they have access to the password-protected online instructional modules. Rather, young adults participated in the social skills class instructed by the special education teacher at the nonprofit organization as usual with no topics related to planning skills. To measure participant knowledge of planning steps and social activity engagement during baseline, participants completed a minimum of five planning vignettes and completed a minimum of two social activity samples. All baseline measures were collected during the regularly scheduled social skills classes. No instruction regarding planning activities was provided while participants completed these baseline measures. In particular, participants were not provided with the Planning Activity Checklist and independently completed the planning vignettes.
Technology check
A technology check was conducted after the baseline condition and before the first telecoaching session. Participants indicated convenient days and times to participate in telecoaching sessions. Research staff called participants during the scheduled technology check and answered any questions regarding video conference platforms or display device used. Research staff and participants ensured the selected video conference platform operated correctly for both parties. All participants were agreeable to telecoaching sessions in which video played, allowing participants and research staff to see each other’s faces, with the exception of Hannah who expressed a preference for regular telephone calls. Given Hannah’s diagnosis of social anxiety disorder, this request was accommodated throughout the study. During the technology check, participants also indicated a secondary video conference platform preference in the event their first preference malfunctioned.
Small group instruction
Figure 1 provides an implementation time line for all intervention components. The intervention package began with small group instruction delivered by the first author in the small classroom at the nonprofit organization. Only the assigned cohort designated to receive the intervention package joined the first author in the small classroom. The three 50-min small group instructional sessions occurred 2 weeks apart during the regularly scheduled social skills class. The small group instruction focused specifically on increasing planning skills and social activities aligned to the purposes of the study. Lesson plans are available upon request from the first author and included the following components: (1) using the Planning Activity Checklist to plan social activities, (2) watching online modules and discussing key concepts, (3) completing two practice planning vignettes, (4) reviewing social activities participated in the prior 2 weeks and setting goals to increase participation, (5) problem-solving challenges related to implementing planned activities, (6) assigning action steps to complete independently before the next small group session, and (7) providing a reminder about upcoming telecoaching sessions. Participants had access to the online instructional modules and Planning Activity Checklist during the first and second small group instructional sessions. During the third small-group session, participants completed a minimum of five planning vignettes for data collection purposes without the Planning Activity Checklist.

Timeline of ASD On The Go intervention package components.
Participants watched the planning module during Session 1 of the small group instruction and the using a calendar module during Session 2. These two instructional modules were displayed on a laptop in the classroom setting. The self-monitoring and developing routines modules were assigned as homework for participants to complete independently with a device of their choice. Modules reported completed independently by participants were checked using the administrator feature, with all participants having completed the modules assigned as homework.
Telecoaching sessions
After Session 1 of the small group session, telecoaching sessions began on a twice per week schedule for all participants in the designated cohort. Each telecoaching session began with the coach greeting the participant and asking if they had entered new activities on their calendar or participated in social activities. If yes, the coach provided behavior-specific praise (e.g., “I like the way you planned the activity and used your Google calendar to enter the activity”) and discussed the activity with the participant (e.g., “Did you have fun going to the baseball game?”). If the participant had not entered a new activity on their calendar or participated in a social activity, the coach focused the discussion on planning for the upcoming week (e.g., “That’s okay, what activity would you like to plan for this week?”). During all telecoaching sessions, participants selected an activity of their choice and planned it using the Planning Activity Checklist. After verbally stating their selected activity, participants described each step in detail (e.g., “For cost, I will bring US$10”). Participants obtained relevant information needed to plan the activity (e.g., times of operation for a museum) by searching the internet. Participants were encouraged to plan different activities during each telecoaching session, and activities often related to participants’ hobbies and interests.
Total duration in minutes for each telecoaching session was recorded, plus the activity planned. If a participant did not join the video conference during the scheduled telecoaching session, research staff texted or called to provide a reminder. Similarly, if participants cancelled a session, they were required to reschedule within the same week. Total number of telecoaching sessions and duration of each session was monitored throughout the study to ensure young adults participated in a minimum of 10 telecoaching sessions with each session an approximate duration of 10 min. Specifically, all participants engaged in an average of 13 (range = 10–16) telecoaching sessions during the study. Each telecoaching session was a mean duration of 11 min (range = 6–20 min). The average cumulative duration of telecoaching sessions per participant was 150 min (range = 125–184 min).
Fading
Telecoaching sessions were faded in a systematic manner that corresponded to the scheduled social skills classes. As described, the fading component was a part of the primary intervention and was incorporated to promote autonomy of skills learned with less intensive assistance from a coach. Telecoaching sessions were implemented twice a week for 6 weeks during the primary intervention condition, then once a week for 4 weeks, and finally, once every other week for 2 weeks.
Maintenance
Maintenance probes occurred 2 weeks following completion of the telecoaching fading component for each cohort. Participants completed three additional planning vignettes not previously used in the study. Participants completed maintenance vignettes independently without the Planning Activity Checklist or assistance from research staff.
Reliability
Interrater reliability (IRR)
To ensure IRR of the primary dependent variable, the second and third authors independently scored all planning vignettes. The middle section of vignettes was scored on a 5-point scale corresponding to the five steps of the Planning Activity Checklist. The second and third authors separately marked each step of the Planning Activity Checklist as + or − to indicate whether participants had included that step in their written response. The two authors’ scores were then compared for consistency of scoring. IRR was calculated for a mean of 39% of vignettes for all participants across all conditions of the study, with 98.6% (range = 80%–100%) agreement. No IRR was collected on the social activities self-reported by the participants as these activities occurred in community settings and research staff did not attend the activities.
Procedural fidelity
The lesson plan for Session 1 of the small group instruction included 12 steps, and Session 2 included 10 steps. The first author checked each step of the lesson plan as it was completed for all small group sessions. In addition, a 10-step procedural fidelity checklist was utilized to ensure fidelity of the telecoaching sessions. The checklist entailed steps such as greeting the participant, asking the participant to select an activity to plan, discussing the five steps of the Planning Activity Checklist, and giving a reminder for the next telecoaching session. The checklist was completed for 100% of the telecoaching sessions by research staff. Procedural fidelity was calculated using the formula: number of steps performed correctly/total number of steps × 100% (Cooper et al., 2020). Mean procedural fidelity was 99% (range = 80%–100%) across all telecoaching sessions for all participants.
Interobserver agreement (IOA)
A second research team member conducted interobserver data on a minimum of 20% of telecoaching sessions for each participant. Specifically, the secondary researcher observed the primary researcher who served as the coach and interacted with the young adult. Both researchers completed the procedural fidelity checklist independently and compared at the conclusion of the telecoaching session. Reliability of IOA data was calculated using a point-by-point formula where [(Agreements)/(Agreements + Disagreements) × 100 = Percent of Agreement] (Cooper et al., 2020). IOA was conducted for a mean of 38% of all telecoaching sessions across all participants with 98.6% (range = 50%–100%) agreement. The only session below 90% was the first session at 50%, and additional training was subsequently provided to this secondary data collector by the first author.
Social Validity
Participant satisfaction with the ASD On The Go intervention package was assessed via a 12-item survey completed at the conclusion of the study. The survey consisted of statements that participants rated their level of agreement using a 5-point Likert-type scale with anchors of strongly agree, agree, neutral, disagree, and strongly disagree. Statements were organized according to the three components of the intervention package and related to the feasibility (e.g., “The online instructional modules were easy to do”) and likeability of each component (e.g., “Overall, I liked the telecoaching sessions”). Participants were also asked to write the most effective component and give suggestions for improvement using an open-ended text box.
Data Analysis
Formative visual analysis was conducted within and across conditions throughout the study according to procedures described by Barton and colleagues (2018). Within condition, visual analyses were conducted by examining level, trend, and variability/stability for each participant and condition. Level for baseline and intervention conditions was reported as a mean and range for each participant. Trend direction was described as therapeutic if the data path was increasing or contra-therapeutic if the data path was decreasing. Variability/stability was described as stable or variable. Finally, summative visual analysis was conducted at the conclusion of the study to determine a potential effect between the intervention package and increased scores on the planning vignettes for each participant. Current single-case design standards recommend a minimum of three demonstrations of effect to establish a function relation (Barton et al., 2018). Given there were three cohorts in a multiple baseline across participants design, there were three opportunities for demonstrations of effect across different points in time.
Results
Results of the ASD On The Go intervention package are displayed in Figure 2. Overall, participants increased their knowledge of specific steps to plan activities as measured on the 5-point scale aligned with the Planning Activity Checklist. Seven of the eight participants maintained their high scores on the vignettes during maintenance. However, only five of the eight participants reported minimal increased participation in weekly social activities. A summary of individual responding organized by cohorts is described below.

Planning vignette scores.
Cohort 1
Cohort 1 consisted of Jeremiah and Hannah. Jeremiah and Hannah demonstrated low levels of accurate responding during the baseline condition. Jeremiah scored a mean of 3% (range = 0%–20%), and Hannah scored a mean of 5% (range = 0%–20%) on the planning vignettes during baseline. Baseline data paths for Jeremiah and Hannah were contra-therapeutic and stable. Upon implementation of the intervention package, Jeremiah increased to a mean of 89% (range = 80%–100%) and Hannah to a mean of 96% (range = 80%–100%). Intervention data paths for Jeremiah and Hannah were therapeutic and stable. Jeremiah and Hannah maintained their knowledge of the planning steps as indicated by a mean of 100% on the planning vignettes during the maintenance condition.
Regarding the activity samples, Jeremiah reported participating in 1.3 social activities per week preintervention and increased to 2.0 social activities per week postintervention. Given Jeremiah’s participation in Boy Scouts of America, many of his planned activities during the telecoaching sessions related to activities with fellow boy scouts (e.g., camping, skiing). Hannah requested to plan activities with only her family, stating peers would be too overwhelming for her. As a result, Hannah reported participating in 3.0 social activities with her family preintervention and increased to 3.2 postintervention. Activities planned by Hannah related to her interest in animals and nature (e.g., walking at a nature center, visiting a pet store).
Cohort 2
Cohort 2 consisted of Bruce, Kai, and Michael. During the baseline condition, Bruce scored a mean of 28% (range = 0%–60%) on the planning vignettes. Similarly, Kai scored a mean of 24% (range = 0%–40%) on the planning vignettes during baseline. Finally, Michael scored a mean of 6% (range = 0%–20%). Baseline data paths for Cohort 2 participants were more variable compared to Cohort 1; however, trends were still contra-therapeutic. Upon implementation of the intervention package, Bruce and Kai increased to a mean of 100% on the planning vignettes. Michael increased to a mean of 98% (range = 80%–100%). Intervention data paths for Cohort 2 were therapeutic and stable. The three participants in Cohort 2 maintained their knowledge of the planning steps as indicated by a mean of 100% on the planning vignettes during the maintenance condition.
Regarding the activity samples, Bruce participated in 0.2 social activities weekly preintervention and increased slightly to 0.3 postintervention. Kai participated in 4.2 social activities weekly preintervention and decreased to 3.7 postintervention. Given Bruce and Kai were seniors in high school at the time of the study, their planned activities were primarily with classmates such as shopping at a local mall, going out to dinner, or playing pickleball. Michael participated in 0.3 social activities preintervention and remained the same at 0.3 postintervention. Many of Michael’s planned activities related to his special interest area of western lifestyle (e.g., shopping at a boot barn).
Cohort 3
Cohort 3 consisted of Charles, Adam, and Martin. Charles scored a mean of 7% (range = 0%–40%) on the planning vignettes during baseline. Adam scored a mean of 10% (range = 0%–40%), and Martin scored a mean of 5% (range = 0%–20%) during baseline. Baseline data paths for the three participants in Cohort 3 were contra-therapeutic and stable. Upon implementation of the intervention package, Charles and Adam increased to a mean of 100%. Martin increased to a mean of 93% (range = 60%–100%) during the intervention condition. Charles and Adam maintained their knowledge of the planning steps as indicated by a mean of 100% on the planning vignettes during the maintenance condition. However, Martin demonstrated a lack of maintained planning skills as indicated by low and variable responding during the maintenance condition.
Regarding the activity samples, Charles participated in 0.4 weekly social activities preintervention and decreased to 0 activities postintervention. Example activities planned by Charles were shopping for action figures, bowling, and hosting a cookout. Adam participated in 1.5 activities preintervention and increased to 2 activities postintervention. Adam primarily planned activities with peers at his community college such as visiting an art museum or attending college-sponsored events (e.g., star gazing at a conservatory). Finally, Martin participated in 0.4 activities preintervention and increased to 1 postintervention. Given Martin’s interest in comedy and theater, he often planned to attend a comedy show.
Social Validity Survey
Table 3 presents all Likert-type survey items administered with percentage of participant responses according to each anchor. Participants reported overall high satisfaction with the ASD On The Go intervention package. Regarding feasibility, 50% of participants strongly agreed the small group sessions and online instructional modules were easy to do. Similarly, 50% agreed telecoaching was easy to do. Regarding effectiveness of the intervention components, 62.5% agreed the small group sessions and online instructional modules helped to plan more activities with friends or family. Telecoaching was rated slightly more effective as 50% of participants strongly agreed telecoaching helped to plan more activities with friends or family. Finally, regarding likability, 75% of participants agreed they liked the small group sessions and online instructional modules, while 50% reported liking the telecoaching sessions. Hannah wrote in the text box, “It was helpful to be shown exactly what steps I should follow to plan an activity and have people available to assist me if I wasn’t sure exactly what to do.” Michael wrote, “I liked the telecoaching sessions because I had to stick to a routine to make sure I didn’t miss a call.” Bruce wrote, “I liked the modules we did in class, and the most helpful was the planning module.” Some suggestions for improvement were “I think a little more variety in the telecoaching session would help” (Hannah) and “Maybe add more modules” (Martin).
Social Validity Survey Results.
Note. Percentages were calculated from the number of participant responses within a category divided by 8 (e.g., four responses = 50%) where 1 = strongly disagree, 2 = disagree, 3 = neutral, 4 = agree, and 5 = strongly agree.
Discussion
The purpose of this study was to evaluate the efficacy of the ASD On The Go intervention package consisting of small group instruction, online instructional modules, and telecoaching to increase planning skills of young adults with ASD and other related neurological disorders. Results indicated all participants increased their knowledge of steps required to plan social activities. However, increased participation in weekly social activities was minimal. This suggests although participants developed the skills to plan social activities, actual implementation of the activities was more challenging.
Regarding our first research question, participants increased their planning skills as indicated by improved vignette scores. Consistent with previous research that has noted individuals with ASD often encounter executive functioning difficulties (Kiep & Spek, 2017), no participants scored above 60% on the vignettes during baseline, suggesting explicit instruction regarding how to plan social activities was necessary. Following implementation of the intervention package, all participants scored a minimum of 80% or higher on the vignettes. Specifically, an effect was present for all three cohorts, which indicated a functional relation between the ASD On The Go intervention package and planning skills. Seven of the eight participants maintained their planning skills during the maintenance condition. However, Martin’s planning skills were maintained at considerably lower levels.
Regarding our second research question, participants reported overall high satisfaction with the intervention package. In particular, participants reported they liked the technology-based components of the package (i.e., online modules and telecoaching). Online learning may align well with the learning needs of individuals with ASD, such as visual presentation of material and direct organization of content (DiGennaro Reed et al., 2011; Sabella & Hart, 2014). In addition, participants navigated the online modules and video conference platforms with overall ease and competency. Participants accessed their personal computers, handheld devices, and smart phones regularly for a variety of functions and required little to no technical assistance from research staff. Notably, no devices were purchased for purposes of the study; all participants used devices owned prior to their participation in the study. Given all participants regularly used technology in their lives, technology-based interventions appeared to have high social validity for young adults with ASD.
Regarding our supplemental research question, only five of the eight participants reported increased participation in social activities postintervention. Planning activities with assistance of research staff during telecoaching sessions appeared more feasible than authentic implementation of activities in community settings with peers. Although disappointing, this finding was perhaps not surprising given the lack of social engagement frequently experienced by this population (Eaves & Ho, 2008). Variability in implementation of activities can be attributed to an array of potential factors such as family characteristics, current level of programming in the individual’s life, and access to community resources and transportation, among others. For example, the only participant pursuing postsecondary education at the time of the study, Adam, reported high levels of support provided by his community college instructors. Conversely, Martin had no ongoing activities in his life as he was unemployed, not pursuing postsecondary education, and lived at home with his parents. As a result of these differences, planning one or two activities with same-age peers was a significant feat for Martin compared to Adam who interacted with classmates on a regular basis at his community college. Additional research is needed to evaluate innovative interventions to support young adults with ASD with diverse levels of programming and resources as these factors are influential to authentic participation in social activities.
Implications for Practice and Research
These findings have implications for professionals who strive to increase the quality of life of young adults with ASD. Special education teachers, therapists, and adult service providers may consider telecoaching a viable method to provide services to young adults with ASD who have limited direct contact with such professionals. For example, telecoaching may be particularly relevant in rural areas. Or, as utilized in the current study, telecoaching can complement traditional, in-person delivery of services. Additional research examining the efficacy of telecoaching is needed with individuals with ASD as direct recipients of the coaching given much of the telecoaching studies have been implemented with a parent or therapist present (Neely et al., 2017). Furthermore, we encourage future researchers to consistently collect procedural fidelity measures when implementing telecoaching studies.
An additional implication is professionals should consider utilizing the special interest areas of young adults with ASD to increase social engagement. As described, some participants had highly specialized interests (e.g., Western lifestyle, NASA), and all participants planned activities related to their personal hobbies and interests. Future researchers may consider evaluating if participation in social activities increases when activities relate to a special interest area. Future researchers may also consider pairing or matching participants with peers who share similar interests. Special interest areas of young adults with ASD is a relatively understudied area of inquiry that, with better understanding, may lead to improved adult outcomes.
A final implication for practice and research relates to the comorbidity between ASD and anxiety. Three participants, Hannah, Adam, and Martin, were diagnosed with either social anxiety disorder or generalized anxiety disorder in addition to their ASD diagnoses. In particular, Hannah’s social anxiety disorder appeared to affect her daily functioning and ability to plan and participate in social activities (e.g., planning activities with family rather than same-age peers, requesting telecoaching sessions with no video feature). Martin had a diagnosis of generalized anxiety disorder and reported he often felt nervous to ask peers to participate in social activities for fear they would decline his invitation. Finally, although Adam had a formal diagnosis of generalized anxiety disorder, he never reported anxiety-specific symptoms during his participation in the study. The experiences of Hannah, Martin, and Adam highlight the challenges of navigating ASD and anxiety, which can often be interrelated (Maddox & White, 2015). Future researchers should examine how interventions aimed to increase social engagement can meet the complex needs of young adults with ASD and comorbid anxiety.
Limitations
This study has several limitations that should be considered. First, the primary dependent variable was merely one means to measure planning skills. The planning vignettes were selected because they could be easily administered within the structure of the social skills classes at the nonprofit organization. The planning vignettes also aligned with the content presented in the instructional modules and Planning Activity Checklist. However, this variable only measured knowledge of steps to plan social activities rather than actual implementation of social activities. Although there are certainly benefits to developing knowledge necessary to plan activities, measuring actual participation in social activities is likely the most valid indicator of the intended outcomes of this study. Future researchers should consider other dependent variables that capture authentic social participation such as number of social activities participated in per week.
Second, there was no reliability or confirmation of activities participated in as reported by participants. Therefore, these results should be interpreted with caution and, hence, why this variable was described as a supplementary measure. Future researchers should consider alternative methods, such as peers and parents, to verify self-report. For example, two studies implemented structured social planning with college students with ASD and incorporated either peer mentors (Ashbaugh et al., 2017) or observers (Koegel et al., 2013) to verify participation in social activities. Future researchers may also consider the use of technology-based tracking and monitoring systems that allow the verification of location while simultaneously maintaining independence while participating in activities.
Third, there was no reliability conducted during the small group instructional sessions due to logistics at the nonprofit organization. The first author checked each step of the lesson as it was completed but a secondary observer was not present to conduct reliability. A final limitation is the most effective component of the intervention package is unknown without conducting a component analysis (Cooper et al., 2000). Therefore, it remains unclear if the small group instruction, online instructional modules, or telecoaching sessions had the greatest impact on participant responding. Regardless, participants reported they appreciated aspects of a packaged intervention, such as having a coach to check in with remotely between the small group instructional sessions. As more sustainable and durable interventions are researched for this population, future researchers may consider conducting a component analysis if implementing a packaged intervention. In particular, we encourage future researchers to examine the efficacy of telecoaching in conjunction with other interventions given its feasibility to provide services remotely.
Conclusion
Participating in shared activities with others contributes to higher overall quality of life (Bishop-Fitzpatrick et al., 2017). One way to enhance social engagement of transition-aged youth with ASD is to teach skills necessary to plan and participate in social activities with others. Technology-based interventions, such as the instructional modules and telecoaching used in this study, appear to be promising methods to develop planning skills of young adults with ASD. While previous telecoaching studies have focused on parents or other family members as recipients of the coaching, young adults with ASD participated as the direct recipients of telecoaching services in this study. Results indicated participants increased their planning skills as measured on fictional planning vignettes but demonstrated little gains in authentic implementation of social activities with peers outside of the telecoaching sessions. Additional research is clearly needed to refine and evaluate technology-based interventions, including telecoaching, to enhance the social experiences and outcomes of individuals with ASD during their transition to adulthood.
Footnotes
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) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This study was funded by the National Institute on Disability, Independent Living, and Rehabilitation Research (NIDILRR, Grant Number #90DP0058)
