Parent Coaching in Natural Communication Opportunities Through Bug-in-Ear Technology

Abstract

We examined whether there was a functional relation between BIE eCoaching with parents of young children with language disorders and parent provision of natural communication opportunities and whether parents maintained their provision of natural communication opportunities when the intervention was faded. Data also were collected to assess the social validity of the intervention. The results indicate that BIE eCoaching with parents of young children with language disorders was effective in increasing parent provision of natural communication opportunities. Parents maintained their provision of natural language opportunities above baseline levels after intervention was withdrawn. All participants strongly agreed that BIE eCoaching was helpful for changing their communication practices with their child and would recommend it to other parents of children with language disorders.

Keywords

parent training naturalistic instruction language disorders autism

Language disorders are the most common type of developmental delay, causing language delays in up to 40% of children under 5 years of age (American Academy of Pediatrics, 2021). Language disorders are identified as “difficulties in the acquisition and use of language due to deficits in the comprehension or production of vocabulary, sentence structure, and discourse” (American Psychiatric Association, 2014, Communication Disorders section, para. 2). They have been associated with difficulty communicating wants and needs, learning difficulties, trouble forming friendships, and behavior concerns (Rosenbaum & Simon, 2016). Children with language disorders often do not acquire language through natural interactions with caregivers, instead requiring repeated opportunities to practice and use language, which, research shows, can be provided through customized early intervention programs (Rosenbaum & Simon, 2016).

Parent Training

Best practices in early intervention, including early childhood special education, are family-centered interventions that build parent capacity, and value the collaboration of family members and professionals (Division for Early Childhood, 2014). Because caregiver-child interactions are the most influential for early language acquisition, parent engagement in intervention leads to better outcomes for children (Heidlage et al., 2020; Meadan et al., 2016). Parents are typically with their children for much of the day and can provide multiple experiences to learn, practice, and generalize language skills (Rosenbaum & Simon, 2016). Research indicates that language acquisition interventions are most effective when embedded into play and daily routines, especially for children with significant delays (Heidlage et al., 2020). However, parents often need to be taught how to implement evidence-based interventions, but with the proper support they can acquire the necessary skills (Meadan et al., 2016).

Naturalistic instruction

Embedding learning opportunities in daily activities is part of naturalistic instruction (NI; Meadan et al., 2016). NI supports positive outcomes for young children with disabilities and lends itself to parent implementation (Snyder et al., 2015). It is distinguished by four main features: (a) occurs in the child’s natural routines and environment; (b) centers on learning targets that allow the child to engage in their environment; (c) teaching episodes follow the child’s lead or their interests, and the child’s response is followed by a consequence; and (d) intervention is delivered by familiar adults (Snyder et al., 2015). Embedding intervention into naturally occurring routines creates meaningful and fun opportunities for children to engage in repeated practice of skills (Heidlage et al., 2020).

A large portion of NI focuses on acquiring meaningful communication skills that allow children to get their needs met and to interact with people in their environment (Snyder et al., 2015). NI promotes reciprocal interaction skills that lead to increased language, including verbal and non-verbal turn taking, responding to children’s attempts to gain adult attention, modeling, and expanding language (Heidlage et al., 2020).

Challenges to parent training

Despite best efforts by parents and early intervention practitioners, teams may encounter barriers to implementing parent training, such as work commitments or language barriers (Keilty & Trivette, 2017). In some areas of the country, there is more demand for early intervention services than there are practitioners available (Wacker et al., 2013). Most recently, during the COVID-19 pandemic, early intervention practitioners encountered new challenges due to health and safety concerns. All these challenges taken together indicate that alternatives to the traditional model of face-to-face early intervention may be beneficial. Researchers found that technology enhanced parent training can ameliorate some of the challenges associated with face-to-face parent training, while continuing to build parent capacity (Wacker et al., 2013).

Technology enhanced training has been conducted using a vast range of equipment, including eBooks, text messages, avatars, computer modules, and Bluetooth technology (Bigelow et al., 2020; Feil et al., 2020; Rock & Holden, 2019; Troseth et al., 2020). In one study, researchers used text messages to remind parents of specific skills targeted during home visits (Bigelow et al., 2020). An increase in text frequency was associated with increased parent implementation of the skill. In another study, parents read an eBook with their young child that had an avatar on every page who prompted the parent to ask the child a question (Troseth et al., 2020). Data showed that parents who read the book with real-time prompting from the avatar asked significantly more questions and engaged in more conversational exchanges when the avatar was removed than their counterparts who were never provided with the avatar. Lending more data to the benefit of real-time coaching, two studies using intervention packages designed to support parents or educators in implementing language and play strategies increased in efficacy only after practice and real-time feedback from a coach was introduced (Feil et al., 2020; Lorio & Woods, 2020). These studies support previous findings that synchronous, real-time coaching improves evidence-based intervention implementation and fidelity (Coogle et al., 2016).

Bug-in-ear eCoaching

eCoaching is defined as technology enhanced coaching. One type of eCoaching, Bug-in-ear (BIE), allows for coaching to occur discreetly and in real-time by connecting the coach and coachee via a headset for two-way communication (Rock & Holden, 2019). It has been shown to increase evidence-based teacher behaviors (Rock et al., 2009). One benefit of the immediate feedback given during BIE eCoaching is the ability for the coachee to change her behavior and practice the corrected skill multiple times within the session, rather than practice errors before receiving feedback (Horn et al., 2021; Scheeler & Lee, 2002).

BIE in early childhood special education has been delivered in 10-min eCoaching sessions to effectively increase teacher use of communication strategies (Coogle et al., 2015). BIE has also been used with success to coach teachers on implementing naturalistic communication strategies during 6-min eCoaching sessions (Coogle et al., 2018). Both studies employed a common practice of eCoaching: prompting with short phrases once a minute, as necessary, and following targeted teacher behaviors with praise or feedback.

Most published studies report on BIE eCoaching with pre-service and in-service educators, but BIE eCoaching has been introduced to the parent training literature to increase child communication in response to challenging behavior (Akemoglu et al., 2020). Investigations have included BIE eCoaching as an embedded component of a larger intervention package aimed at teaching parents to proactively increase language, but with limited elaboration on its specific implementation (Akemoglu et al., 2020; Vismara et al., 2013). BIE eCoaching has the potential to be used on its own to teach parents to use evidence-based interventions within daily routines to proactively increase communication skills before challenging behaviors develop.

This study sought to coach parents to use NI during naturally occurring play opportunities during the day to proactively increase their child’s communication skills. NI was structured in this study as a natural communication opportunity consisting of one of two components: communication temptations or prompting (Kaiser & Hampton, 2014). The research questions we sought to answer were:

RQ1. Is there is a functional relation between Bug-in-Ear eCoaching with parents of young children with language delays and parent provision of natural communication opportunities?

RQ2. Do parents feel that Bug-in-Ear eCoaching is a socially valid intervention?

Method

Research Design

A single-subject concurrent multiple baseline design across participants was used to determine whether there was a functional relation between eCoaching via Bug-in-Ear technology and an increase in parental provision of natural communication opportunities. The study met the What Works Clearinghouse (2020) Standards 4.1 Without Reservations: (a) data were available in graphical form; (b) the independent variable was systematically manipulated; (c) inter-assessor agreement was conducted on all sessions in the study, which is more than the 20% required; (d) intervention was introduced in a staggered fashion and the study was unlikely to have residual treatment effects; and (d) there were more than six phases in which there was an attempt to demonstrate an effect, each with at least five data points.

Participants and Setting

Participants were selected using purposeful criterion-based sampling. The five mother/son dyads fit the following criteria: (a) child was between 24 and 60 months of age when diagnosed with a language disorder due to developmental delay or autism spectrum disorder, (b) spoke English as their primary language, (c) had a laptop or tablet that is capable of videoconferencing, (d) had high-speed wireless internet, (e) committed to two half-hour sessions a week of video-conferencing, (f) gave permission to record the sessions, (g) had a primary goal of increasing their child’s functional language, and (h) their child communicated through gestures, single words, or just starting to combine words. All dyads participated through completion of the study, and all were awarded a $250 stipend. All five children were male and were between 3 and 4.5 years of age (see Table 1). Four out of five children (Laurence, Barry, Ned, and Terrence [pseudonyms]) had a diagnosis of ASD and were receiving direct intervention at the start of the study. Michael (pseudonym) had a diagnosis of apraxia and was at home with his mother at the start of the study. The children all lived at home with their father, mother, and sibling(s). The mothers participated in the study and spoke English (n = 3) or were bilingual in English and Spanish (n = 2); all coaching occurred in English (see Table 2).

Table 1.

Beginning Demographics of Sons Participating in Study.

Demographic	Laurence	Barry	Michael	Ned	Terence
Age	3 years	3 years	3 years	4.5 years	4.5 years
Diagnosis	ASD	ASD	Apraxia	ASD	ASD
Early intervention	Since 18 months	Not reported	Since 17 months	For 2 years	Not reported
Current intervention	ABA in-home	Virtual ECSE	None	ABA in-home	Virtual ECSE
Primary comm.	60 words	Gestures^a; 50 words	Signs/gestures^a; 2 words	10 words; Signs	Pulling^a; A few words
Lives with	Mom, dad, younger sister	Mom, dad, younger sister	Mom, dad, older sister	Mom, dad, twin brother	Mom, dad, older siblings
Other?				Tantrums, SIB & aggression	Starting to read

Note. ABA = applied behavior analysis; ECSE = early childhood special education; SIB = self-injurious behaviors.

Primary mode of communication.

Table 2.

Demographics of Mothers Participating in Study.

Demographic	Mariana	Sara	Rebecca	Adina	Lori
Age	42	34	41	42	42
Race	Hispanic	API	White	Hispanic	White
Primary language	Bilingual^a	English	English	Bilingual^a	English
Education	Some college	Post-graduate degree	Post-graduate degree	College degree	Some college
Career	Stay at home parent	Works full-time^b	Works part-time	Stay at home parent	Works part-time
Parent goal for child^c	Build independence	Functional	Functional	Functional	Functional

Note. API = Asian/Pacific Islander.

Bilingual in English and Spanish.

Also in school.

Communication focused.

The coach was a PhD candidate and board certified behavior analyst with a master’s degree in infant and early childhood special education. She had 18 years of experience working with children with language disorders and their families. She had experience coaching families and school staff in-person to create natural communication opportunities during their day.

Setting

All training and coaching sessions took place via BIE with both the coach and dyads connecting from their homes via a Pro Zoom account maintained by the first author. Participants were on wireless devices with a built-in camera and connected to the sound via a Bluetooth^® headset. Parents provided the wireless devices and signed on using a smart phone, laptop, or tablet. Participants chose between using personal Bluetooth earbuds (n = 2) or a coach-purchased Jabra Talk 25 Bluetooth^® headset delivered to the family’s home (n = 3). Each location within the home was specific to the dyad based on the play activity for the day. The coach was on her MacBook Pro and connected to the sound via Bluetooth^® earbuds.

Procedures

Prior to the start of the study, the authors obtained institutional review board approval and parent participants signed informed consent.

Technology set up

Prior to baseline, the coach sent a guide to parents via email with directions on syncing the headset with their device and installing and running Zoom. The coach engaged in a 15-minute practice session to check that both parties could hear each other and that the coach could see the dyad’s activities.

Pre-baseline

Three to 10 days prior to entering baseline (dependent upon when the family joined the study), the coach sent parents a study-created training manual via email to review starting. The manual contained information on how to provide the components of a natural communication opportunity (Kaiser & Hampton, 2014). The components, described in Table 3 were: (a) inadequate portions, (b) assistance, (c) pause in routine, and (d) choice making. Each component was explained in less than one page, with an example included. Prior to baseline, parent participants completed a questionnaire measuring social validity of BIE eCoaching.

Table 3.

Natural Communication Opportunities.

Component	Definition	Example
Communication temptations
Inadequate portions	Provide the child with a small amount of a desired item	Give child a piece of a cookie and wait for indication she wants more.
Assistance	Situation where child needs to ask for help or for someone to move.	Place cookie in closed container, wait for request it be opened.
Pause in routine	Pause in the middle of a familiar routine; wait for the child to fill in.	Say: “Ready, set,” then pause to see if child says, “Go.”
Choice making	Silently hold up two items	Hold up a cookie and a cracker, what for child to indicate choice.
Prompting
Time delay	An obvious, but non-verbal cue for a child to communicate.	Hold up an item and look expectantly at the child.
Open question	Question with no single correct answer nor yes/no answer.	Ask: “What should we take on our picnic?”
Choice question	Ask a choice question. This can be supported by visual cues if needed	Ask: “Do you want waffles or effs for breakfast?”
Model procedure	The parent provides a model of exactly what to say.	Say: “Want cookie,” or “Open.”

Note. All components above require the communication partner to wait at least 3 seconds for the child to communicate before following up.

Baseline

Baseline sessions took place over Zoom. The coach did not interact with the parent during the baseline sessions, aside from greetings and farewells. The parent engaged her child in a 1:1 play activity. After the initial minute of play, which allowed the parent and child to settle into their routine, the primary investigator collected partial interval data in 1 minute intervals for a 10-min session. Data were collected on the components of a natural communication opportunity and supplementary data were collected on child verbal behavior. The coach staggered the introduction of intervention, starting BIE coaching after the participant had at least five baseline sessions and the data for natural communication opportunities were stable or decreasing based on the initial partial interval data.

Bug-in-ear intervention

Following baseline, the coach met with each participant to discuss baseline data and select two goal areas from the four components of a natural communication opportunity that had been presented in the training manual. All five participants agreed that the data indicated they target the provision of communication temptations (manipulating the environment to elicit child communication) and prompts (signal for the child to say something).

eCoaching was delivered twice weekly (with a few exceptions due to travel or family commitments) in 15- to 30-min parent coaching sessions for 3 months. During intervention, parents signed into Zoom in the same manner as baseline. Each session started with a 5-min check-in during which the parent asked questions and discussed any issues that had arisen since the previous session. The parent then provided their child a choice of play activities to engage with that day, after which the 11-min parent practice with coaching began. Mariana’s activities included coloring, reading, playsets, puzzles, and sticker scenes. Sara’s activities included sensory, books, pretend play, sidewalk chalk, blocks, and Play-Doh. Rebecca offered playsets, Play-Doh, and vehicles. Adina used all close-ended activities (i.e., puzzles, Mr. Potato Head) during intervention and Lori used Play-Doh for all intervention sessions.

After the first minute, the parents were coached for 10 minutes to provide the natural communication opportunities of communication temptations or to deliver prompts. The coach used the empirically based strategies of prompting and immediate, specific positive or corrective feedback delivered in short, specific phrases approximately once every minute during the coaching sessions (Rock et al., 2009). Positive feedback specifically labeled the parent’s targeted action and included a positive phrase. For example, when the parent provided a model for their child, the coach said, “Nice model.” If the parent did not follow the prompt, the coach waited and provided another prompt in the next minute interval. When the parent naturally provided a communication opportunity the coach delivered positive feedback immediately following the behavior and did not provide a prompt in that minute.

The session ended with a structured 5-min wrap-up session. The coach and parent discussed the prompts and feedback delivered during the session and whether anything should be changed and the parent asked questions she had. The team discussed implementation during the coming week, mainly, what the parent could target.

Maintenance

Coaching ended when the parents demonstrated a consistent or accelerating trend in providing both communication temptations and prompting, based on the partial interval data. After coaching ended, parents again completed the social validity questionnaire. Two weeks later, maintenance was introduced with the same structure as baseline. During wrap-up, no questions were answered until after the final maintenance session. Each participant participated in three maintenance sessions except for Mariana, who had two maintenance sessions due to scheduling issues.

Measures

Parent behavior data

Initial data were recorded for minutes 2 through 11 of baseline, intervention, and maintenance with partial interval recording divided into minute intervals. The coach used this initial data to determine movement between phases for each participant. During intervention, the coach observed that partial interval recording did not accurately capture parents’ independent use of the strategies, as they often provided more than one during an interval. Therefore, following maintenance, the coach coded all 115 videos from baseline, intervention, and maintenance to capture frequency data through event recording; these data were used to report outcomes of the study. Intervention data were recorded on independent, partially prompted (short phrase) and fully prompted (longer explanation) use of both communication temptations and prompting (Coogle et al., 2017). This datasheet was used for interrater reliability.

Questionnaire

The pre- and post-questionnaires were ranked on a seven-point Likert scale and questions fell into three areas: (a) feelings toward bug-in-ear technology to receive immediate feedback, (b) parent confidence, and (c) child communication (Coogle et al., 2017). The pre- and post-survey included the same questions, and the post-survey included an additional four open ended questions about the BIE coaching and immediate feedback.

Supplemental data

The coach recorded secondary data to track child vocal behavior using partial interval recording in 1-minute intervals. Data were recorded in real-time during all 10-min parent coaching sessions, which occurred twice a week. Vocal behavior was defined as vocalizing or singing, using words/word approximations, or phrases. Communication initiations were coded when a child spontaneously vocalized. Child responses were coded when a child used vocal behaviors to respond to a parent’s comment, question, or prompt. Vocal responses by the child were classified as independent, prompted, or in response to the parent. Responses to the question, “What is it?” were not coded, as this was considered a test behavior not engagement. The percentage of intervals that independent initiations and responses occurred were calculated and compared across participants and phases.

Interobserver Agreement

An early childhood educator who was trained on the coding definitions served as the second person coder. Point-by-point agreement was used to calculate interobserver agreement (IOA) on the frequency data collected for parents’ independent use of strategies (Ayers & Ledford, 2014). The formula is a count of the number of agreements (occurrences), divided by number of agreements plus the number of disagreements (occurrence marked by one observer and not marked by the other) and the total multiplied by 100. Initial IOA yielded 78% accuracy across 100% of the 115 sessions. Following the initial coding of IOA data, the second observer and the first author independently watched the videos for sessions that had agreement under 80% to confirm correct coding. The second coder and the first author then discussed disagreements across sessions and came to a consensus for each of them. The data were adjusted to reflect that consensus was reached.

Procedural Fidelity

A retired educator collected data on 33% of the sessions from baseline, intervention, and maintenance using author created checklists. Baseline and maintenance required that the coach did not verbally communicate with the parent during the 11-min data collection session and that conversation before and after was limited to pleasantries or questions related to next steps. During intervention, the coach was to give one prompt or positive statement per minute. Additionally, the coach had five follow-up questions to ask following the coaching session. Data recorded by the second observer for 33% of the overall sessions indicated procedural fidelity across all phases was 97.56%. The intervener maintained a 93.91% fidelity for baseline. This increased to 99% in intervention and 100% in maintenance.

Data Analysis

Data were visually examined for patterns or changes in level, trend, variability, immediacy of effect, data overlap between phases, and consistency (Kratochwill et al., 2010). Tau-U was calculated to analyze the nonoverlap between phases combined with within phrase trends (Vannest et al., 2016). The supplemental child data were analyzed for the percentage of intervals that independent initiations of communication and independent responses to mothers occurred.

The mean of the five parents’ scores for each of the 11 questions on the social validity questionnaire was compared between pre- and post-intervention. Additionally, answers to the pre- and post-survey Likert-scale questions were analyzed per participant to assess whether there was an increase in individual responses following intervention.

Results

Research Question 1

Overall outcomes of the study showed that all five participants demonstrated an increase in level between baseline and treatment in their spontaneous provision of natural communication opportunities, and levels remained above baseline during maintenance (see Figure 1). Four out of the five participants showed an accelerating trend line in treatment. Variability in baseline and intervention were moderate to high for all participants. All participants demonstrated an immediacy of effect moving from baseline to treatment.

Figure 1.

Frequency of parent-provided natural communication opportunities per session.

Mariana

Mariana’s provision of natural communication opportunities was low during baseline (range = 0–7) with a decelerating trend. We observed an immediacy of effect when treatment was introduced. The level was higher than baseline (range = 0–22) with an accelerating trend and increased variability. Seven out of 15 data points overlapped with baseline, (Tau-U = 0.41 (p = .18), 90% CIs [−0.09, 0.92]) suggesting a moderate effect. Prompted responses showed a decelerating trendline. Also, the coach delivered fewer prompts as Mariana increased her spontaneous responses. Maintenance data were higher than and did not overlap with intervention data (range = 25–29).

Sara

Sara’s level of provision of natural communication opportunities ranged from 4 to 10 with an accelerating trendline and moderate variability. We observed an immediacy of effect when treatment was introduced accelerating trendline and more variability than during baseline (range = 6 – 30). Two of the fifteen datapoints in treatment overlapped with baseline (Tau-U = 0.91 (p = .001), 90% CIs [0.44, 1]), which suggests a large effect. Prompted responses showed a decelerating trendline. The coach delivered fewer prompts at the end of the phase. The maintenance levels were higher than baseline (range = 5–17).All data points in maintenance overlapped with intervention data.

Rebecca

Rebecca’s baseline level of provision of natural communication opportunities ranged from 1 to 9 with a decelerating trend and variable data. We observed an immediacy of effect when treatment was introduced. The level during treatment was ranged from 8 to 25 with an accelerating trend and moderate variability. Two of the 11 data points in treatment overlapped with baseline (Tau-U = 0.92 (p = .001), 90% CIs [0.45, 1]), which suggests indicating a large effect. Prompted responses showed an accelerating trendline. The coach delivered more prompts at the end of treatment, despite Rebecca increasing the median number of strategies used during treatment. All data points in maintenance overlapped with intervention data (range = 8–11).

Adina

Adina’s baseline level of provision of natural communication opportunities ranged from 5 to 26 with some variability and a decelerating trendline. Data did not immediately change when treatment was introduced. The overall level of treatment ranged from 3 to 31 with an increase in variability and an accelerating trendline. Ten of the 12 data points in treatment overlapped with baseline (Tau-U = 0.38 (p = .13), 90% CIs [−0.032, 0.8]). The coach decreased the number of prompts given by the end of intervention. All data points in maintenance overlapped treatment and baseline (range = 1–23).

Lori

Lori’s level of provision of natural communication opportunities during baseline ranged from 0 to 26 with some variability and anaccelerating trendline. Her data did not have an immediate of effect when the treatment was introduced. The level of spontaneous provision of natural communication opportunities increased during treatment with a decelerating trend (range = 10–26). All nine data points in treatment overlapped with the baseline data (Tau-U = 0.59 (p = .03), 90% CIs [0.15, 1]) indicating no effect. The coach delivered a consistent number of prompts across treatment. All data points in maintenance overlapped treatment and baseline (range = 20–24).

Research Question 2

Changes in participant responses from the pre-intervention to the post-intervention survey indicated that all five mothers felt BIE eCoaching was a socially valid intervention. The area that showed the largest gain in ratings across all questions in the category was child communication (M = 2.13, range = 1–2.8).

Feelings toward Bug-in-ear technology

All participants strongly agreed immediate feedback via BIE was helpful in changing their communication practices with their child (M = 7). This response was an average increase of two points from the original responses, however, one participant moved six points, from strongly disagree to strongly agree, over the course of treatment. All five participants also strongly agreed that they would recommend immediate feedback via BIE to other parents of children with delays in communication (M = 7). This was an average increase of 1.6 points for each participant from the original survey response. All participants agreed or strongly agreed that BIE feedback was manageable (M = 6.6, range = 6–7).

Parent confidence

The mean score on the post-questionnaire for each of the five questions composing the parent confidence category fell in the agree range (all question means were between 6 and 6.4). Scores on the question I am confident in working and playing with my child ranged from 5 to 7 on the post-survey. This was a mean increase of 1.2 points from the pre-survey range of 1 to 7.

Child communication

This was the area with the largest change from pre- to post-intervention. On the post intervention survey, parents all slightly agreed, agreed, or strongly agreed that their child communicated regularly with words or word approximations (M = 5.8, range = 5–7). This was a mean increase of 2.6 from the pre-intervention survey, which ranged from 1 to 5. The statement my child communicates effectively using words or word approximations increased by 2.8, from a range of 1 to 5 to a range of 5 to 7. The statement my child does not get frustrated with communication rose by a mean of 1 from a pre-survey range of 1 to 3 to a post-survey score of 3 across participants.

Supplemental Child Data

While there is no experimental control for the child data, they do show promising trends.

Vocal initiations

Mean vocal initiations across children did not show a change between baseline (M = 47%, range = 18%–68%), treatment (M = 50%, range = 26%–83%), and maintenance (M = 51%, range = 30%–73%; see Figure 2). However, there was a slight (3 percentage point) increase between baseline and treatment and a slight increase again (1 percentage point) between treatment and maintenance. Three of the five children demonstrated an increase in mean vocal initiations from baseline to treatment.

Figure 2.

Graph by child of percentage of intervals with vocal initiations.

Vocal response

Vocal response behavior across children showed an increase of 29 percentage points from baseline (M = 36%, range = 18%–58%) to treatment (M = 65%, range = 45%–90%; see Figure 3). Another level increase was demonstrated between treatment and maintenance (M = 86%, range = 77%–100%). All children demonstrated an increase from baseline to treatment and then again from treatment to maintenance.

Figure 3.

Graph by child of mean percentage of intervals with vocal responses.

Discussion

Bug-in-Ear eCoaching was effective when used to proactively coach parents of young children with language disorders for three of five participants. Through eCoaching, parents increased their spontaneous provision of natural communication opportunities and maintained these levels after intervention was withdrawn. In turn, three of the five children increased their number of vocal initiations and all of the children increased their responses to their mother’s communication. Parents identified eCoaching as a socially valid intervention. All participants strongly agreed that BIE feedback was helpful in changing their communication practices with their child and would recommend it to other parents of children with language delays. Parent confidence levels when working and playing with their children showed an increase from pre- to post- of the social validity measure. These results both support and extend previous research on BIE eCoaching in early childhood.

Supports Previous Research

Three of the five participants demonstrated a moderate increase in levels when the treatment was introduced supporting previous research indicating BIE eCoaching in 1 minute intervals during short coaching sessions is an effective way to increase the use of targeted skills (Coogle et al., 2015). However, the data showed variability across conditions and participants, likely due to a variety of factors that parents encountered during intervention. The results support outcomes seen in an eCoaching study conducted with early childhood special educators, who were coached to embed learning opportunities, leading to an increase in child communicative behavior (Coogle et al., 2017).

Two of the five parents (Mariana and Lori) also maintained their levels above treatment levels when the treatment was faded. Previous research has suggested stronger maintenance effects when using BIE with early childhood teachers (Coogle et al., 2018). Parents reported feeling more confident in their skills of working and communicating with their children, and they reported that their child’s behavior benefitted. For Ned, there was an inverse relationship between the number of vocal responses to his mother during intervention and maintenance sessions and the tantrum and aggression behaviors that were observed during 70% of baseline sessions. This outcome suggests that the coaching feedback and cues provided to Ned’s mother during intervention positively impacted Ned’s behavior and that the parent continued this learned behavior of communicating with her child during maintenance. These findings align with reports from parent coaching studies (Feil et al., 2020; Meadan et al., 2016) and BIE eCoaching studies (Ottley et al., 2015) that demonstrate coaching a caregiver has positive effects on children.

Our results provided support to families in implementing the best practice of embedding language strategies into natural interactions with their child. It extends the current literature focused on BIE eCoaching with professionals to include coaching parents. It also extends findings that parent coaching can be conducted from a distance through video review (Feil et al., 2020; Meadan et al., 2016).

The immediacy of effect seen when BIE eCoaching was introduced in 1-min intervals has been seen previously with early childhood teachers who were coached to deliver naturalistic communication strategies (Coogle et al., 2018). In this study, it was expanded to parents who learned to provide natural communication opportunities in their natural environment through BIE eCoaching in 1-min intervals. The current study further expanded on previous research using BIE with parents by using eCoaching as the main intervention rather than incorporating it as part of a package (Vismara et al., 2013). Additionally, the parents in this study only needed to commit to 15 minutes sessions twice a week, whereas previous studies using BIE with parents required 1.5-hr sessions (Vismara et al., 2013).

Previously, in-the-moment modeling by an avatar was shown to be an effective and socially valid way to coach parents (Troseth et al., 2020). BIE eCoaching extends these findings to show that in the moment feedback responsive to parent behavior meets the parents where they are and builds their capacity in facilitating communication. Responses to the social validity questionnaire indicated that parents found this meaningful and beneficial. One parent stated the “experience was awesome. I have seen how the ABA therapist of my son works, but [being] able to practice like this has been so amazing for me. I feel a lot more confident now.” All participants said they strongly agreed with the statement that immediate feedback via BIE [was] helpful in changing my communication practices with my child and the statement that immediate feedback via BIE is something that I will recommend to other parents of children with delays in communication. Based on these responses, eCoaching is a practice that could be implemented by professionals to support parents in working with their children with language disorders.

Limitations

As with all research studies, this study has limitations. Therefore, although the results of the study are promising, the limitations of the study must be kept in mind while interpreting the results. Initial data for the study were taken using partial interval recording. These data indicated when to move participants from one phase of the study to the next; however, it underrepresented parent behaviors. Following the study, the first author reanalyzed the data using event recording to better represent parent behavior, but baseline was started for two participants when they were on a downward trend. While this could be due, in part, to the child interacting for longer with the materials and spontaneously engaging his parent, it is a limitation in the data reporting.

Due to parents’ various levels of knowledge and learned behaviors in interacting with their children, some parents had to be taught priming behaviors, such as pausing after asking a question before they could properly execute prompts and communication temptations. As a result of the study being conducted in the natural environment, it was more difficult to control for extraneous variables, likely leading to more variability in the data. In three out of the five families, siblings were present during sessions, which distracted the mother and led to a few sessions having to end early. In all five families, children were tempted to engage in an activity other than the ones that the mothers offered as choices, or the child would choose a preferred activity and then lose interest or get distracted by something else in the environment.

In keeping with the natural environment, parents were not limited in the types of play activities they could offer, rather they selected activities they believed would be of interest to their children. The inconsistency across participants for the type and amount of variation in types of activities likely accounts for variation in outcomes between participants. For example, Mariana did not incorporate close-ended activities (i.e., puzzles and sticker scenes) until intervention Session 9, after which her level of independent strategy use jumped from 3.75 for the first eight sessions to 15.43 for the last seven sessions. Adina and Lori, who both showed a moderate effect size, used the same or similar materials throughout the study, which may have led to complacency for both the parents and contributed to the lower effect size for their intervention.

Additionally, as the children progressed and their language and play improved, the children required less prompting from their parent. They showed more spontaneous engagement in activities for longer periods of time, which meant they did not need to ask for items. They instead engaged their parents in other ways that were not captured under the researchers’ definition of natural communication opportunity. This likely impacted both parent data and supplemental child data, especially nearing the end of the study. The supplemental child data should also be interpreted with caution. These data were collected using partial interval recording, which has the potential to underrepresent the child’s vocal behavior. It was not the main focus of data analysis and therefore, some data may have been missed while being recorded in real time. There is also the potential that not all the child’s vocal behavior was picked up by the microphone or that it could have been masked by a sound in the environment.

Poor wireless connections and difficulty with Bluetooth connections impacted nine of the 115 sessions, causing the coach difficulty in seeing or hearing participants (indicated with TT on Figure 1). Technology troubles mainly impacted Mariana’s sessions, interrupting five out of her 15 intervention sessions. Four of those five sessions showed a corresponding drop in frequency. The last 2 minutes of Session 10 were impacted when Laurence left the chosen activity and Mariana shared some of her frustrations with the coach (indicated with SI on Figure 1). Technology issues impacted 2 minutes of two of Rebecca’s baseline sessions and one each of Adina and Lori’s baseline sessions. Lori’s fifth baseline session was interrupted for 1 minute due to Terence needing to use the bathroom (indicated with SI on Figure 1). One of Rebecca’s intervention sessions was interrupted by a tantrum from Michael and in that session, the frequency data dropped 15 points from the previous session. The following session the data increased by five.

Implications for Research and Practice

Multiple studies point to the need for more generalization research in the BIE literature conducted with early childhood educators (Coogle et al., 2016; Ottley et al., 2015; Schaefer & Ottley, 2018). Because the current study aimed to be practical for the parents and engaging for the children, generalization was unintentionally programmed, which could potentially strengthen the parents’ use of the strategies in the long run. However, examining ways to intentionally program for generalization is a direction for future research.

Isolating the impact of different variables, such as the type of play activity, length of a coaching session, or comparing individual versus group coaching, would add valuable information in future studies. These could also lead to better generalization of the skills children acquire.

Finally, future research should examine the best ways to capture data on the impact BIE eCoaching of parents has on the child. As of a 2018 review of literature, only four studies systematically recorded changes in student behavior (Schaefer & Ottley, 2018). One way to do this would be to record the Mean Length of Utterance for children as a pre- and post-measure. Another possibility would be to look at the topography of the communication, such as whether the child vocalized, verbalized, gestured, or signed (Meadan et al., 2016).

The implications for using BIE eCoaching provides functional interventions to support families. Families could connect with interventionists around the country and the world, through BIE eCoaching, if both parties have access to the internet and the audio/visual technology. This would allow families to work with interventionists who speak their native language and to access interventions outside of normal business hours. It could also change supports for students who are homebound, and their families, who may not have a robust schedule of private or school-based interventions due to medical concerns or other access issues.

Conclusion

BIE eCoaching is an empirically validated intervention frequently used to coach pre-service and in-service teachers. There is a growing body of research for using BIE to coach early childhood educators (e.g., Coogle et al., 2016, 2017). However, this study fills a gap in the research for the use of BIE eCoaching with parents to increase their use of NI to target communication skills. The results of the study indicate that BIE eCoaching was effective for in increasing parent provision of natural communication opportunities for three of the five parents and two parents maintained their provision of natural language opportunities when the treatment was faded. Participant responses indicated that BIE eCoaching was a socially valid intervention for changing their communication practices with their child. Our resutls show eCoaching to support parent’s use of NI and children’s language development is promising and additional research is needed to develop a comprehensive understanding of BIE eCoaching across populations (e.g., parents, teachers).

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

ORCID iD

Rachel J. Hamberger

References

Akemoglu

Muharib

Meadan

(2020). A systematic and quality review of parent-implemented language and communication interventions conducted via telepractice. Journal of Behavioral Education, 29, 282–316. https://doi.org/10.1007/s10864-019-09356-3

American Academy of Pediatrics. (2021, April 30). Language delays in toddlers: Information for parents. Author. https://www.healthychildren.org/English/ages-stages/toddler/Pages/Language-Delay.aspx

American Psychiatric Association. (2014). Diagnostic and statistical manual of mental disorders (5th ed.). Author. https://doi.org/10.1176/appi.books.9780890425596

Ayers

K. A.

Ledford

J. R.

(2014). Dependent measures and measurement systems. In Gast

D. L.

Ledford

J. R.

(Eds.), Single case research methodology: Application in special education and behavioral sciences (pp. 124–153). Routledge.

Bigelow

K. M.

Walker

Jia

Irvin

Turcotte

(2020). Text messaging as an enhancement to home visiting: Building parents’ capacity to improve child language-learning environments. Early Childhood Research Quarterly, 51, 416–429. https://doi.org/10.1016/j.ecresq.2019.12.010

Coogle

C. G.

Ottley

J. R.

Rahn

N. L.

Storie

(2018). Bug-in-ear eCoaching: Impacts on novice early childhood special education teachers. Journal of Early Intervention, 40, 87–103. https://doi.org/10.1177/1053815117748692

Coogle

C. G.

Ottley

J. R.

Storie

Rahn

N. L.

Burt

(2017). eCoaching to enhance special educator practice and child outcomes. Infants and Young Children, 30, 58–75. https://doi.org/10.1097/IYC.0000000000000082

Coogle

C. G.

Rahn

N. L.

Ottley

J. R.

(2015). Pre-service teacher use of communication strategies upon receiving immediate feedback. Early Childhood Research Quarterly, 32, 105–115. https://doi.org/10.1016/j.ecresq.2015.03.003

Coogle

C. G.

Rahn

N. L.

Ottley

J. R.

Storie

(2016). eCoaching across routines to enhance teachers’ use of modeling. Teacher Education and Special Education, 39, 227–245. https://doi.org/10.1177/0888406415621959

10.

Division for Early Childhood. (2014). DEC recommended practices in early intervention/early childhood special education 2014. Author. http://www.dec-sped.org/recommendedpractices

11.

Feil

E. G.

Baggett

Davis

Landry

Sheeber

Leve

Johnson

(2020). Randomized control trial of an internet-based parenting intervention for mothers of infants. Early Childhood Research Quarterly, 50(1), 36–44. https://doi.org/10.1016/j.ecresq.2018.11.003

12.

Heidlage

J. K.

Cunningham

J. E.

Kaiser

A. P.

Trivette

C. M.

Barton

E. E.

Frey

J. R.

Roberts

M. Y.

(2020). The effects of parent-implemented language intervention on child linguistic outcomes: A meta-analysis. Early Childhood Research Quarterly, 50(1), 6–23. https://doi.org/10.1016/j.ecresq.2018.12.006

13.

Horn

A. L.

Layden

S. J.

Roitsch

Karadimou

(2021). Providing performance-based feedback to teachers in real-time using Bug-in-Ear technology. Coaching: An International Journal of Theory, Research and Practice, 14(1), 92–101. https://doi.org/10.1080/17521882.2020.1784972

14.

Kaiser

A. P.

Hampton

L. H.

(2014, April). Enhanced Milieu Teaching: Setting the foundation for communication [Conference session]. International Congress on Early Inter-vention, Antalya, Turkey. https://kidtalk.vkcsites.org/info-for-professionals/

15.

Keilty

Trivette

C. M.

(2017). Working with families: Today and tomorrow. In Trivette

C. M.

Keilty

(Eds.), Family: Knowing families, tailoring practices, building capacity (DEC Recommended Practices Monograph Series No. 3, pp. v–xi). Division for Early Childhood of the Council for Exceptional Children.

16.

Kratochwill

T. R.

Hitchcock

Horner

R. H.

Levin

J. R.

Odom

S. L.

Rindskopf

D. M.

Shadish

W. R.

(2010). Single-case designs technical documentation. https://files.eric.ed.gov/fulltext/ED510743.pdf

17.

Lorio

C. M.

Woods

J. J.

(2020). Multi-component professional development for educators in an Early Head Start: Explicit vocabulary instruction during interactive shared book reading. Early Childhood Research Quarterly, 50(1), 86–100. https://doi.org/10.1016/j.ecresq.2018.12.003

18.

Meadan

Snodgrass

M. R.

Meyer

L. E.

Fisher

K. W.

Chung

M. Y.

Halle

J. W.

(2016). Internet-based parent-implemented intervention for young children with autism: A pilot study. Journal of Early Intervention, 38(1), 3–23. https://doi.org/10.1177/1053815116630327

19.

Ottley

J. R.

Coogle

C. G.

Rahn

N. L.

(2015). The social validity of bug-in-ear coaching: Findings from two studies implemented in inclusive early childhood environments. Journal of Early Childhood Teacher Education, 36(4), 342–361. https://doi.org/10.1080/10901027.2015.1100146

20.

Rock

M. L.

Gregg

Thead

B. K.

Acker

S. E.

Gable

R. A.

Zigmond

N. P.

(2009). Can you hear me now? Evaluation of an online wireless technology to provide real-time feedback to special education teachers-in-training. Teacher Education & Special Education, 32(1), 64–82. https://doi.org/10.1177/0888406408330872

21.

Rock

M. L.

Holden

K. B.

(2019). Technology-enabled one-on-one, in-ear coaching: Framing professional development in real-time instruction. In Rock

M. L.

(Ed.), The eCoaching continuum for educators: Using technology to enrich professional development and improve student outcomes (pp. 78–110). ASCD.

22.

Rosenbaum

Simon

(Eds.). (2016). Speech and language disorders in children: Implications for the social security administration’s supplemental security income program. The National Academies Press. https://doi.org/10.17226/21872

23.

Schaefer

J. M.

Ottley

J. R.

(2018). Evaluating immediate feedback via bug-in-ear as an evidence-based practice for professional development. Journal of Special Education Technology, 33(4), 247–258. https://doi.org/10.1177/0162643418766870

24.

Scheeler

M. C.

Lee

D. L.

(2002). Using technology to deliver immediate corrective feedback to preservice teachers. Journal of Behavioral Education, 11(4), 231–241. https://doi.org/10.1023/A:1021158805714

25.

Snyder

P. A.

Rakap

Hemmeter

M. L.

McLaughlin

T. W.

Sandall

McLean

M. E.

(2015). Naturalistic instructional approaches in early learning: A systematic review. Journal of Early Intervention, 37(1), 69–97. https://doi.org/10.1177/1053815115595461

26.

Troseth

G. L.

Strouse

G. A.

Flores

Stuckelman

Z. D.

Johnson

C. R.

(2020). An enhanced eBook facilitates parent-child talk during shared reading by families of low socioeconomic status. Early Childhood Research Quarterly, 50(1), 45–58. https://doi.org/10.1016/j.ecresq.2019.02.009

27.

Vannest

K. J.

Parker

R. I.

Gonen

Adiguzel

(2016). Single case research: Web-based calculators for SCR analysis (Version 2.0) [Web-based application]. Texas A&M University. https://singlecaseresearch.org

28.

Vismara

L. A.

McCormick

Young

G. S.

Nadhan

Monlux

(2013). Preliminary findings of a telehealth approach to parent training in autism. Journal of Autism & Developmental Disorders, 43, 2953–2969. https://doi.org/10.1007/s10803-013-1841-8

29.

Wacker

D. P.

Lee

J. F.

Dalmau

Y. C. P.

Kopelman

T. G.

Lindgren

S. D.

Kuhle

Pelzel

K. E.

Waldron

D. B.

(2013). Conducting functional analyses of problem behavior via telehealth. Journal of Applied Behavior Analysis, 46, 31–46. https://doi.org/10.1002/jaba.29

30.

What Works Clearinghouse. (2020). WWC version 4.1 standards handbook. Author. https://ies.ed.gov/ncee/wwc/handbooks