Evaluating the Use of Alternative Seating with Kindergarteners at Risk for Emotional and Behavioral Disorders

Abstract

Characteristics of emotional and behavioral disorders (EBD) include learning difficulties that cannot be explained by intellectual, sensory, or health factors and difficulties in building or maintaining interpersonal relationships with peers and teachers. Children with or at risk for an EBD often have a tendency to have negative experiences in school and engage in challenging behavior in the classroom including out-of-seat behavior. One possible antecedent manipulation, alternative seating, may reduce challenging behavior and involves exchanging the typical seating in classrooms for different types of seating options. The purpose of this study was to evaluate the use of stability stools and scoop rocker chairs on in-seat behavior and on-task behavior in classrooms with kindergarten students who engaged in challenging behavior and were at risk for EBD. All three participants demonstrated improvements in in-seat behavior using both types of alternative seating compared to a standard classroom chair. On-task behavior improved for all students but was variable for two students. Teachers indicated a preference for the stability stool, whereas results were mixed between the stool and the rockers for student preference.

Students with or at risk for an emotional and behavioral disorder (EBD) have difficulty adjusting to classroom expectations and often engage in challenging behavior in the classroom. They may have difficulties attending to classroom instructions and complying with group directions and engage in higher out-of-seat behavior and lower on-task behavior (Corrin et al., 2015; Garwood, 2018; Gottfried & Harven, 2015; Kelly & Shogren, 2014). Additional characteristics and behaviors seen in children classified with or at risk for EBD include hyperactivity, aggression or self-injurious behavior, withdrawal, immaturity, and learning difficulties (Council for Exceptional Children, 2018). According to the Individuals with Disabilities Education Act (2004), young students at risk for EBD commonly display low engagement levels and high levels of challenging behavior, but due to their age are often not given an EBD classification because of the criteria for the occurrence of challenging behavior for a long period of time (e.g., across several months) that affects their educational performance.

Over 40% of students who have an EBD classification are educated in segregated settings because teachers have difficulty addressing these students’ behavioral and academic needs (McLeskey et al., 2010). Unfortunately, a high rate of suspension and expulsion is also common for students with EBD (Kauffman & Landrum, 2009; Sullivan et al., 2014). In addition, students with EBD report having negative experiences in school, which could diminish their motivation to succeed in the classroom and result in poor academic outcomes (Toms et al., 2018). Therefore, interventions focused on decreasing challenging behavior and increasing engagement for young students at risk for an EBD classification are warranted to potentially prevent these students from having negative experiences in school and being placed in noninclusive self-contained classroom environments.

Previous behavioral strategies that have been explored for students with or at risk for EBD include choice making (Wehby et al., 1995), teacher praise (Allday et al., 2012; Sutherland et al., 2000), group contingencies (Denune et al., 2015), and self-management (Digangi & Maag, 1992). Sutherland et al. (2000) used a reversal design to examine how the rate of behavior-specific praise from teachers affected the on-task behavior of nine students with EBD. During the first intervention phase, results showed that on-task behavior, measured in 1-min intervals, increased from 48.7 to 85.6 when the teacher provided behavior-specific praise. Following a return to baseline conditions, these results were replicated with the reintroduction of behavior-specific praise during the second intervention phase. In addition, Denune et al. (2015) combined an interdependent group contingency and self-monitoring strategy to address on-task behavior and disruptive behavior in a sixth-grade classroom for children with EBD. They found that the self-monitoring component of the intervention did not enhance the effectiveness of the interdependent group contingency in terms of increasing on-task behavior and decreasing disruptive behavior. Using a social validity questionnaire, both the teacher and the students indicated the desire to continue and preference for the interdependent group contingency in isolation.

Although strategies such as teacher praise and interdependent group contingencies have been effective for decreasing challenging behavior in the classroom and increasing on-task behavior for children with or at-risk for EBD, teachers may find it difficult to implement more complex classroom interventions consistently and with high implementation fidelity given the range of competing demands for teachers (Noell, 2008). For example, teachers in traditional elementary school classrooms are tasked with high teacher to student ratio’s and increased academic requirements for students. Thus, teachers may prefer interventions that require less time, expertise, and training to implement (Elliott et al., 1984). One evidence-based intervention strategy leveraging positive behavioral approaches that may require little effort or time to implement involves manipulating antecedent stimuli to evoke a desired behavior (Dunlap et al., 2001), which has been shown to prevent or decrease challenging behavior that interferes with learning in the classroom (Dunlap et al., 2001; Parcells et al., 1999; Wingrat & Exner, 2005). One antecedent manipulation to enrich the classroom environment is offering students alternative seating choices.

Alternative seating in classrooms involves exchanging typical classroom chairs and tables for different types of seating options with various sizes and functions. Parcells et al. (1999) examined the possible mismatch between the individual body dimensions of students and the classroom furniture they use. This study used anthropometric measurements (e.g., elbow height, shoulder height, upper arm length, knee height, popliteal height, buttock-popliteal length, student stature) to evaluate the match and mismatch of each student’s measurements across three styles of chairs and three styles of desks. When compared with the classroom furniture available to these students, there was a substantial degree of mismatch with the students’ dimensions, suggesting that the furniture already in these classrooms was not conducive to learning and students should have different sizes of desks and chairs available that better fit their bodies.

Due to issues with traditional seating, alternative types of seating have been evaluated for improving the on-task and in-seat behavior of students (Brennan & Crosland, 2021; Hoofman, 2018; Krombach & Miltenberger, 2020; Schilling & Schwartz, 2004; Schilling et al., 2003). Different types of alternative seating allow students to move, bounce, rock, or sway when seated instead of staying stationary in typical chairs. From a behavioral perspective, one might conceptualize alternative seating and the mechanisms that might account for behavior change in a couple of ways. First, alternative seating, such as stability balls or stools, might alter the reinforcing value (abolishing operation) of out of seat behavior by allowing students to move in more socially appropriate and less disruptive ways while seated, which may provide continuous access to reinforcement that in turn results in decreases in out-of-seat behavior (Morrison et al., 2011; Piazza et al., 2000; Ringdahl et al., 1997). Second, the hard plastic or wood standard desk chair may be uncomfortable to sit on, resulting in squirming around and elopement (i.e., getting up and leaving the chair). Other types of seating may be more comfortable (e.g., therapy balls have more cushion than traditional plastic desk chairs), and therefore, out-of-seat behavior may be less likely when students are seated in more comfortable chairs.

Teachers are often open to trying alternative seating in their classrooms, and they consistently report its implementation as easy (Gochenour & Poskey, 2017; Hill & Nickels, 2018; Schoolcraft, 2018). In addition, alternative seating options have typically been shown to improve attending for students with attention difficulties (see review by Gochenour & Poskey, 2017). A wide range of alternative seating options exist, including commercial and do-it-yourself (DIY) options, all which range in price. Alternative seating can be as simple as modifying the tables (i.e., raised or lowered) or introducing standing desks. In addition, alternative seating can encompass area rugs, cushions, bean bags, floor pillows, DIY bungee cord bouncing bands, therapy balls, and wobble chairs (Hill & Nickels, 2018).

The use of therapy balls has been extensively evaluated (Fedewa & Erwin, 2011; Krombach & Miltenberger, 2020; Schilling & Schwartz, 2004; Schilling et al., 2003; Taipalus et al., 2017). Therapy balls, also known as yoga balls or exercise balls, are large inflatable balls which allow students to gently bounce while seated and encourage good posture (Schilling & Schwartz, 2004). Schilling et al. (2003) used therapy balls in a classroom with three elementary school students with ADHD. Their findings showed an increase on in-seat behavior and legible word productivity when the students were sitting on therapy balls compared to typical desk chairs (Schilling et al., 2003). Similar results have been demonstrated for the on-task and in-seat behavior of elementary school students with ADHD (Fedewa & Erwin, 2011), as well as the engagement, attendance to instruction, and in-seat behavior of students with autism spectrum disorder (ASD; Krombach & Miltenberger, 2020; Schilling & Schwartz, 2004).

Although results for therapy balls in classrooms have been mostly positive, social validity results have been mixed. Responses from teachers and therapists indicated dissatisfaction with therapy balls; in particular, stakeholders found the bouncing to be distracting, difficult to use, and dangerous (Hoofman, 2018; Krombach & Miltenberger, 2020; Taipalus et al., 2017). Schoolcraft (2018) conducted interviews, observations, and surveys to understand the viewpoints of students and teachers regarding the use of therapy balls instead of desk chairs. Relative to the students’ viewpoints, some students believed moving while seated was helpful, whereas others found the movement distracting. As a potential solution to the concerns related to the use of therapy balls as a seating option, therapy ball chairs include a base for the ball to sit on with wheels that lock; thus, the ball remains stationary while still allowing students to bounce while seated. This modification to the therapy ball was evaluated by Brennan and Crosland (2021) in a clinic setting with three children diagnosed with ASD. Results showed that both on-task behavior and in-seat behavior increased for all three participants, with the greatest increase in on-task behavior. The therapists rated the therapy ball chairs positively indicating that the addition of the stationary aspect might be more acceptable than therapy balls that allow for movement around a classroom.

Therapy stools have also been growing in popularity as a form of alternative classroom seating. Also known as wobble stools or stability stools, this alternative seating option allows children to rock, sway, tilt, and fidget. The stool’s base allows movement while preventing tipping. Hoofman (2018) evaluated the effects of therapy balls versus stability stools on in-seat and on-task behavior for children with a variety of disabilities. For one of the three participants in this study, in-seat behavior was higher when seated on the stability stool compared to the therapy ball. For the other two participants, both therapy balls and stability stools resulted in similar improvements on in-seat and on-task behavior compared to the traditional chair. When provided a choice between seating options, teachers in this study preferred the stability stools over the balls, and the students also picked the stools more often. Finally, another type of alternative seating, scoop rocking chairs allow children to sit on the floor with the flexibility to rock and recline as they work. There has been no known research that evaluates scoop rocking chairs, or scoop rockers, on in-seat behavior although they are used in the classroom and are less expensive to purchase compared to stability balls or stools.

Alternative seating has been evaluated mostly with typically developing children (Castellucci et al., 2017; Parcells et al., 1999; Schoolcraft, 2018; Wingrat & Exner, 2005), children with ASD (Brennan & Crosland, 2021; Hoofman, 2018; Krombach & Miltenberger, 2020; Matin Sadr et al., 2017; Schilling & Schwartz, 2004; Umeda & Deitz, 2011), and children with ADHD (Fedewa & Erwin, 2011; Schilling et al. 2003; Taipalus et al., 2017). Research on alternative seating with children with or at risk for EBD is needed as many of these children struggle to remain seated and attend to classroom instruction. Therefore, the purpose of this study was to address the following research questions: (1) to what extent will the use of stability stools and scoop rocker chairs affect in-seat behavior with students at-risk for EBD; (2) to what extent will each of these seating options change on-task behavior for students at-risk for EBD; and (3) will alternative seating be acceptable and preferred by teachers and students.

Method

Participants and Setting

This study was conducted in three general education classrooms in a large, metropolitan public-school district in the southeast. Participants were three male kindergarten students nominated by their classroom teachers due to concerns with them being out of their seats and not engaged in academic tasks. Roman was a 6-year-old Caucasian boy diagnosed with ADHD and was considered at risk for a diagnosis of EBD due to reported aggression and impulsivity. Roman’s kindergarten teacher was Mrs. Rita, a Caucasian female with 25 years of teaching experience. Juan was a 5-year-old Hispanic boy with no formal diagnosis. He was reported to engage in challenging behavior that included impulsivity and aggression, which placed him in the at risk for an EBD classification category. Juan’s kindergarten teacher was Mrs. Eleanor, a Caucasian female with 17 years of teaching experience. Paul was a 6-year-old Caucasian boy diagnosed with a communication/ speech disorder, and he received Exceptional Student Education (ESE) services for this diagnosis. Paul was at risk for an EBD due to exhibiting challenging behavior that included impulsivity, inability to focus, and inability to stay on task. Paul’s kindergarten teacher was Mrs. Lucy, a Caucasian female with 11 years of teaching experience. All three participants used vocal-verbal communication, were in general education classrooms, and were reported to struggle to attend to classroom instruction and remain in their seats. Paul and Roman were in the same school but in different classrooms, whereas Juan was in a classroom at a separate school in the same county. Because all three participants were in kindergarten, an official EBD classification from the school could not be provided because they did not have documented behavioral issues for a long period of time as described by IDEA (2004). However, all participants were identified as at-risk for EBD by their school personnel, including the school psychologist, and were recommended for additional services, due to challenging and disruptive behaviors observed in the classroom.

Teacher participants were recruited with a flyer that was distributed to each school by the district, and throughout each school by each school’s principal. Teacher participants contacted the first author via email to indicate their interest in participating and that they had at least one student in their class that was at risk for an EBD. Flyers were then sent home to potential students and once parental consent was obtained students were enrolled in the study. Data were collected in each participant’s normal classroom during morning independent desk work, which was the time of day when all three participants were reported by their teachers to have difficulty staying in their seat and staying on-task. Academic subjects and topics for morning work varied, ranging from writing, reading, and math. An initial observation was conducted that verified teacher reports that the students had low levels of in-seat and on-task behaviors during independent work time.

Materials

Standard desk chairs, stability stools, and scoop rocker chairs were the three seating options utilized in this study. The standard chair was the typical desk chair already in the classroom; this chair had four legs and was appropriate for the size of a typical kindergartener. The stability stool (Kore Wobble Chair) was an oblong stool with a gentle-rounded base to prevent tipping over. Kore wobble stools can be purchased in a variety of online stores and range in price from $50 to $60 per stool (Kore Design LLC, 2022). The scoop rocker was a small rocker that sat on the floor. Scoop rocker chairs can be purchased in a variety of online stores and are sold in sets of six. Prices range from $35 to $48 for a pack of six. When using the scoop rocker chair, the participants were also provided a clip board for completing work given the height discrepancy to a desk with the rocker sitting on the floor. Researchers used the Countee application on a smart phone or tablet for data collection.

Dependent Variables

All sessions were 5-min because teachers indicated that students were expected to work for 5 min before taking a brief break followed by working again for 5 min during the morning independent work time routine in which data were collected. The morning independent work time routine lasted between 10 and 15 min each day. Data were collected using duration of on-task and in-seat behavior. Researchers sat in the classroom in view of the participant and recorded if the participant was in his seat and on task using the Countee app. Data were collected by the researcher clicking the on-task behavior and in-seat behavior buttons when the participant was on-task and in-seat, respectively, and clicking the buttons again to stop duration to signify when the participant was off-task and out-of-seat, respectively. At the end of each session, the total duration of each target behavior (on-task, in-seat) was converted to a percentage of session time by dividing the number of seconds the target behavior occurred by the total number of seconds (i.e., 300 sec as each session was 5 min) multiplied by 100. Frequency of teacher prompts was also recorded during each session as a secondary measure, and it was reported as a total count.

In-Seat Behavior

In-seat behavior for standard chairs was defined as any portion of the participants’ buttocks in contact with the seat portion of the chair and at least one foot in contact with the floor. In-seat behavior while on the stability stools was defined as any portion of the participants’ buttocks in contact with the stool and the stool simultaneously remaining in contact with the floor; this also included having at least one foot in contact with the floor or in contact with the base of the stool (Hoofman, 2018). In-seat behavior for the scoop rocker chair was defined as any portion of the participants’ buttocks in contact with the inside curved part of the chair and the chair remaining in contact with the floor. This also included having at least one foot in contact with the floor. If the participant rocked hard enough where both feet were off the floor, the participant was scored as out-of-seat.

On-Task Behavior

On-task behavior was defined as the participant oriented toward the teacher or the appropriate task and materials. This included appropriate interaction with the materials such as reading the paper and writing on worksheets, responding to the teacher if asked a question, and/or looking at the teacher when spoken to (Brennan & Crosland, 2021; Hoofman, 2018; Krombach & Miltenberger, 2020; Matin Sadr et al., 2017; Schilling et al., 2003). As soon as the participant was not appropriately interacting with the materials or oriented to the teacher, the duration timer was stopped. This included the participant spinning around on the seating or looking away from the materials. The duration timer was resumed when the participant began interacting with the materials appropriately and/or oriented to the teacher again.

Teacher Prompts

The frequency of teacher prompts in response to the child having out-of-seat and/or off-task behavior was recorded during each session. These prompts included verbal prompts, (e.g., “Go back to your seat”), gestural prompts (e.g., pointing at the seat), and/or full physical prompts (e.g., walking the student back to their seat). Both redirects and reprimands that were related to out-of-seat or off-task behavior were included as teacher prompts.

Interobserver Agreement

Secondary data collectors recorded the same duration data independently across sessions for each participant. The data collectors were trained on the data collection procedures using behavioral skills training (BST; Miltenberger et al., 2004). The researcher described the definitions for on-task behavior and in-seat behavior then modeled collecting data for one session. Next, the data collectors rehearsed the skills by collecting data with the study participants and received feedback from the researcher until they reached 80% or higher agreement. If interobserver agreement (IOA) fell below 80%, secondary data collectors were retrained by clarifying behavioral definitions. Data were collected in person by both the primary and secondary data collectors using the Countee app on separate devices. IOA was calculated manually using the data from the Countee app for both in-seat and on-task behavior. The total number of seconds in-seat and on-task were compared across the two observers to determine agreement. To calculate IOA, the lower number of seconds was divided by the larger number of seconds and multiplied by 100%.

A second observer collected data for 20% of baseline sessions, 35% of intervention sessions, and 35% of all choice sessions. During baseline, IOA averaged 94.5% (range: 85%–100%) for in-seat behavior and 88% (range: 71%–100%) for on-task behavior across participants. In intervention, IOA averaged 98% (range: 93%–100%) for in-seat behavior and 93% (range: 73%–100%) for on-task behavior across participants. Finally, in the choice phase, IOA averaged 97% (range: 90%–100%) for in-seat behavior and 93% (range: 80%–100%) for on-task behavior across participants. IOA was not collected for teacher prompts due to the few teacher prompts observed during the initial observation to determine participant eligibility. The primary data collector recorded teacher prompts as a secondary measure to obtain a record of prompts in case these changed from baseline to intervention. Please refer to Table 1 for individual IOA averages for each participant and behavior.

Table 1.

Mean Percentages of Interobserver Agreement

Phase	Dependent Variable	Overall M	Roman M	Juan M	Paul M
Baseline (% Collected)		25%	23%	39%	17%
	In-seat	98%	83%	93%	95%
	On-task	83%	80%	88%	80%
Intervention (% Collected)		38%	32%	41%	42%
	In-seat	98%	98%	96%	99%
	On-task	89%	95%	85%	85%
Choice Phase (% Collected)		35%	38%	33%	33%
	In-seat	98%	97%	98%	98%
	On-task	92%	96%	97%	83%

Social Validity

Social validity was assessed utilizing brief questionnaires designed for the participating classroom teachers as well as the student participants. Teacher social validity included a five-item questionnaire which was scored on a five-point Likert-type scale. Questions were related to how the teacher felt about the intervention and their perspective on how the intervention affected the student targeted behaviors. The student social validity questionnaire included a simplified scale comprised of three response options: yes, maybe, and no with corresponding facial expressions (e.g., smiling, neutral, and frowning) and images of the chairs. In addition to the student social validity measure, the choice phases embedded within the study served as an additional measure examining the type of seating most preferred by each participant.

Treatment Integrity

Treatment integrity was assessed using a researcher developed task analysis. The steps of the treatment integrity task analysis included: Teacher telling the student to sit on the predetermined or chosen seating; teacher presenting an independent work task to complete and a writing utensil, and after five minutes the teacher removing the work task and necessary materials and telling the student they could have a break. To measure treatment integrity, a research assistant recorded if the teacher correctly executed each of the steps during the intervention and choice phases. Treatment integrity was calculated by dividing the number of correct steps by the total number of steps and multiplying by 100. Treatment integrity was measured across 40%, 44%, and 42% of sessions for Mrs. Rita, Mrs. Eleanor, and Mrs. Lucy, respectively. Treatment integrity results were 100% for all teachers.

Experimental Design

A non-concurrent multiple baseline across participants with an embedded alternating treatments design was used to evaluate the different seating options. The intervention phase consisted of alternating between two conditions: intervention with the stability stool and intervention with the scoop rocker chair. A choice condition followed the alternating conditions treatment phase. Approximately two to three sessions occurred per day, once or twice a week for all three participants except when absences or school holidays occurred.

Procedures

During all phases of the study, data were collected during the morning independent work time as described previously. Students were expected to complete morning independent work and were allowed to take breaks every 5 min. For each session, the researcher provided the type of seating that was to be used by the student, and the teacher then instructed the student to use that seating. Teachers received training that included instructions, modeling, and feedback on each step of the task analysis (described above in the treatment integrity section) to implement the intervention correctly. In addition, for all conditions, teachers were instructed to respond to appropriate student behavior and challenging behavior as they typically would in their classroom, including prompting students to remain on-task, seated, or to take a break as the standard practice in their classrooms. Prior to the first session of each alternative seating condition, BST (Miltenberger et al., 2004) was used to teach participants to appropriately sit on the stability stool and scoop rocker chair. BST included instructions, modeling, rehearsal, and feedback on correct sitting. Due to size constraints, modeling could not be conducted during training for the scoop rocker as only children can fit in this type of chair.

Baseline

During baseline, no changes to the classroom setting were made. That is, participants sat in the typical chair that was already provided. The researcher did not intervene or prompt the teacher anytime during baseline.

Stability Stool

During this intervention condition, participants were provided the independent work task and any necessary materials and were instructed to sit on the stability stool at their normal desk.

Scoop Rocker Chair

During this intervention condition, participants were provided with a clipboard with their morning work already attached for them to complete and were instructed to sit on the scoop rocker chair in a designated location within the classroom. Teachers selected an appropriate location for the participant to utilize the scoop rocker to facilitate class participation and limit hindering class instruction. Mrs. Rita and Mrs. Lucy had their students, Roman and Paul, sit toward the back of the room, whereas Mrs. Eleanor had Juan sit in the front of the room on a rug.

Choice

In the choice condition, participants chose which seating method they wanted to use during the same morning independent work time used in the baseline and the alternating treatments conditions. All three types of seating were available: chair, stability stool, and scoop rocker chair. Images of all three seating options were presented to participants by the researcher to facilitate choice selection. The participant could either respond verbally or physically point to the picture of the seating type during this condition. After the participant selected a seating option, the teacher provided the chosen seating option, told the participant to sit in the chosen seating option, and provided all necessary work materials. The participant was expected to sit in the chosen seating for the entire 5-min session. If participants indicated they wanted to change their seating option, they were told they could change their seat later, but needed to sit in their chosen seat for the remainder of the 5-min session. For example, if the participant chose the standard classroom chair during the choice phase, they were expected to sit in the standard classroom chair for the duration of the 5-min session.

Results

Figures 1 and 2 display the effects of alternative seating on in-seat and on-task behavior, respectively, for all three participants. In addition, Tau effect size estimates for in-seat and on-task behaviors with the stability stool and scoop rocker chairs are displayed in Table 2. The Tau effect size compares aggregate responses from baseline to intervention across all participants, with the Tau statistic ranging from 0 (no differentiation between baseline and intervention) to 1 (complete differentiation between baseline and intervention; Vannest et al., 2016). The Tau effect size estimates for in-seat and on-task behavior are considered moderate effects for both the stability stool and the scoop rocker chair.

Fig. 1.

Changes in In-Seat Behavior Based on Alternative Seating Option. Note. Asterisks indicate a medication change

Fig. 2.

Changes in On-Task Behavior Based on Alternative Seating Option. Note. Asterisks indicate a medication change

Table 2.

Tau Effect Sizes for In-Seat and On-Task Behavior

		Tau (Baseline vs. Intervention)	Var-Tau	Z	p-value	95% CI
In-Seat Behavior
	Stability Stools	0.8261	0.1304	6.3374	> .001	[0.5706, 1]
	Scoop Rocker Chairs	0.8372	0.1305	6.4169	> .001	[0.5815, 1]
On-Task Behavior
	Stability Stools	0.7942	0.1304	6.0927	> .001	[0.5387, 1]
	Scoop Rocker Chairs	0.7222	0.1305	5.5356	> .001	[0.4665, 0.9780]

Note. The Tau effect size represents the aggregated change between baseline and each intervention. Var-Tau indicates the variance in the Tau statistic, and Z represents the z-score. p-values of less than 0.05 indicate statistically significant results. The 95% confidence interval (CI) indicates the range of Tau effect sizes that could be expected when replicated. All calculations in this table were conducted with Vannest et al. (2016). Tau estimates between 0 and 0.31 indicate small effects, 0.32–0.84 indicate moderate effects, and 0.85–1 indicate large effects (Parker et al., 2011).

In the top panels, Roman’s in-seat behavior during baseline averaged 17% of the session with moderate to high variability (range: 0%–99%), and his on-task behavior was also variable during baseline with an average of 34% of the session (range: 0%–68%). Mrs. Rita, Roman’s teacher, averaged 1.1 prompts per session with a range of 0 to 4 prompts during baseline. During intervention, Roman’s in-seat behavior increased to an average of 89% (range: 70%–100%) in the stability stool and 92% (range: 77%–100%) in the scoop rocker chair with less variability. Roman’s on-task behavior also increased during intervention to an average of 89% (range: 70%–100%) in the stability stool and 94% (range: 77%–100%) in the scoop rocker chair with less variability when compared to baseline. In addition, Mrs. Rita’s teacher prompts decreased to an average of 0.8 prompts per session, whereas the range remained from 0 to 4 prompts per session during intervention. During the choice analysis, Roman chose to sit on the standard desk chair 1 time, the stability stool 10 times, and the scoop rocker chair 2 times. His average in-seat behavior and on-task behavior in the standard chair increased to 99% when compared to baseline levels. In-seat behavior remained at 89% (range: 69%–100%) on the stability stool and increased to 96% (range: 92%–100%) on the scoop rocker chair in the choice phase. On-task behavior increased to 90% (range: 58%–100%) on the stability stool and 95% (range: 90%–99%) in the scoop rocker chair. Teacher prompts continued to decrease in the choice phase to an average of 0.6 prompts, ranging from 0 to 2 prompts per session.

In the middle panels, Juan’s in-seat behavior during baseline was variable (range: 0%–95%) with an average of 22%, whereas on-task behavior began at a high level and decreased across subsequent baseline sessions until decreasing to 0% across the final two sessions. His on-task behavior averaged 26% (range: 0%–80%) across baseline. Mrs. Eleanor, Juan’s teacher, averaged 1.8 prompts per session with a range of 0 to 5 prompts in baseline. During intervention, Juan’s in-seat behavior increased to a moderate to high level across sessions with an average of 84% (range: 41%–100%) in the stability stool and 80% (range: 52%–100%) in the scoop rocker chair, although moderate variability continued. Juan’s on-task behavior also increased during intervention to an average of 74% (range: 52%–99%) in the stability stool and 65% (range: 19%–96%) in the scoop rocker chair. During intervention, Mrs. Eleanor’s teacher prompts decreased to an average of 1.2 prompts per session, ranging from 0 to 4 prompts per session. During the choice phase, Juan chose the stability stool six times and the scoop rocker chair three times. In this phase, Juan’s in-seat behavior increased, with low variability, to an average of 93% (range: 88%–100%) in the stability stool and an average of 92% (range: 88%–94%) in the scoop rocker chair. Juan’s on-task behavior also increased and remained stable in the choice phase to an average of 84.5% (range: 62%–99%) in the stability stool and an average of 83% (range: 66%–97%) in the scoop rocker chair. Average teacher prompts decreased in the choice phase to 0.8 prompts per session (range: 0–3 prompts).

In the bottom panels, Paul’s in-seat behavior during baseline occurred at a higher level on average than the other two participants (M = 72%; range: 0%–99%) whereas on-task behavior was lower and averaged 41% (range: 1%–90%). However, Paul’s in-seat and on-task behavior were highly variable in baseline. Mrs. Lucy, Paul’s teacher, averaged 0.4 prompts per session with a range of 0 to 2 prompts. During intervention, Paul’s in-seat behavior increased and was less variable across both types of seating with an average of 94% (range: 64%–100%) in the stability stool and 95% (range: 71%–100%) in the scoop rocker chair. Paul’s on-task behavior also increased, but remained highly variable during intervention, to an average of 67% (range: 39%–96%) in the stability stool and 62% (range: 0%–97%) in the scoop rocker chair. During intervention, Mrs. Lucy’s teacher prompts remained at 0.4 prompts per session, ranging from 0 to 2 prompts per session. In the choice phase, Paul chose to sit on the standard desk chair two times, the stability stool five times, and the scoop rocker chair five times. Paul’s in-seat and on-task behaviors decreased to an average of 63% (range: 53%–72%) and 21% (range: 16%–25%), respectively, when seated on the typical desk chair. Paul’s in-seat behavior decreased from the alternating treatments phase in the stability stool to 83% (range: 75%–93%) and increased to an average of 97% (range: 90%–100%) in the scoop rocker chair but remained stable for both types of seating. On-task behavior decreased from the alternating treatments phase to an average of 52% (range: 35%–67%) in the stability stool and 61% (range: 22%–85%) in the scoop rocker chair and were variable. Teacher prompts increased slightly to an average of 0.5 prompts per session with a range of 0 to 2 prompts per session in the choice phase. Please see Table 3 for the mean frequency of teacher prompts for each student and across each seating option per phase.

Table 3.

Mean Frequency of Teacher Prompts

Phase	Condition	Roman M (Range)	Juan M (Range)	Paul M (Range)
Baseline		1.08 (0–4)	1.78 (0–5)	0.41 (0–2)
Intervention
	Stability Stool	0.92 (0–4)	1.1 (0–4)	0.3 (0–1)
	Scoop Rocker	0.58 (0–2)	1.25 (0–3)	0.56 (0–2)
	Overall	0.76 (0–4)	1.18 (0–4)	0.42 (0–2)
Choice
	Standard Seating	0	-	1 (0–2)
	Stability Stool	0.8 (0–2)	1 (0–3)	0.4 (0–1)
	Scoop Rocker	0	0.33 (0–1)	0.4 (0–1)
	Overall	0.62 (0–2)	0.77 (0–3)	0.5 (0–2)

Note. A dash indicates that the seating option was never chosen. The zeros for the standard seating and the scoop rocker for Roman in the choice phase indicate that this seating option was chosen by Roman, and no teacher prompts were provided.

Teacher social validity and student social validity were assessed, and the results are reported in Tables 4 and 5, respectively. All teachers reported that their students had trouble staying in their seats and staying on task when seated in a typical classroom chair. All teachers agreed that providing students a choice of seating affected their in-seat and on-task behaviors, that alternative seating is an intervention that they can implement with fidelity, and that alternative seating is an easy intervention to implement. In addition, all teachers reported that they would use alternative seating in their classrooms, and they would recommend alterative seating to other teachers. Mrs. Rita believed that both the stability stool and scoop rocker chair improved in-seat and on-task behaviors for her student, Roman, that they were not distracting to other students, and that she would use both in her classroom. Mrs. Eleanor believed both stability stools and scoop rocker chairs helped her student, Juan, stay in his seat and on-task longer and reported that she would use both in her classroom. However, Mrs. Eleanor reported that both the stability stool and scoop rocker chair might have been a distraction to other students. Mrs. Lucy, Paul’s teacher, did not agree nor disagree that the stability stools and scoop rocker chairs helped her student stay in-seat or on-task longer. She also reported that she believed the stability stools and scoop rocker chairs were not a distraction to other students. All three teachers preferred the stability stools because students were able to sit at their assigned desks.

Table 4.

Teacher Social Validity Results

		Ms. Rita	Ms. Eleanor	Ms. Lucy	Average
Overall
	1. My students do not have problems with staying in their seats and being on task when seated in a typical classroom chair.	1	1	2	1.3
	2. I believe giving students a choice of seating impacted their in-seat behavior and on-task behaviors	4	5	4	4.3
	3. I believe using alternative seating is an intervention I can actively implement with fidelity	4	5	5	4.7
	4. Alternative seating is an easy intervention to implement	5	5	5	5
	5. I would use alternative seating in my class	5	5	5	5
	6. I would recommend alternative seating to other teachers	5	5	5	5
Stability Stools
	1. Stability Stools helped my students focus on their task	5	5	3	4.3
	2. My students were able to stay seated longer when seated on the stability stool	5	5	3	4.3
	3. Stability Stools are not distracting to other students in the classroom	5	2	4	3.7
	4. I would use stability stools in my classroom	5	5	5	5
Scoop Rocker Chair
	1. Scoop Rocking chair helped my students focus on their task	5	5	3	4.3
	2. My students were able to stay seated longer when seated on the scoop rocker chair	5	5	3	4.3
	3. Scoop rocker chairs are not distracting to other students in the classroom	5	2	4	3.7
	4. I would use scoop rocker chairs in my classroom	5	5	5	5
Preference		Stability Stool	Stability Stool	Stability Stool

Note. This table demonstrates the social validity questionnaire results from the teachers. 1 = Strongly Disagree; 2 = Somewhat Disagree, 3 = Neither Agree nor Disagree; 4 = Somewhat Agree; 5 = Strongly Agree.

Table 5.

Student Social Validity Results

	Roman	Juan	Paul	Average
1. I enjoy sitting in the normal desk and chairs.	0	0	0	0
2. I like using the stability stool.	2	2	2	2
3. I like using the scoop rocker chairs.	2	2	1	1.67
4. I think I worked harder when sitting on the stability stool.	2	2	2	2
5. I think I worked harder when sitting on the scoop rocker chair.	2	2	1	1.67
6. I liked getting to choose my seat.	2	2	2	2
Preference	Scoop Rocker Chair & Stability Stool	Scoop Rocker Chair	Stability Stool

Note. This table demonstrates the social validity results from the student participants. Preferences in this table are based on a questionnaire the researcher conducted with each participant, not on the choice analysis in the intervention phase. 0 = No; 1 = Maybe, 2 = Yes.

In addition, all three participants reported that they did not enjoy sitting in their normal desk chair. In contrast, all participants reported that they liked using the stability stools and felt they worked harder while working on the stability stool. Roman and Juan stated that they liked the scoop rocker chairs and felt they worked harder while working on the scoop rocker chair. Paul, however, was not sure if he liked the scoop rocker chairs and was not sure if he worked harder while sitting on the scoop rocker chairs. Roman reported that he liked both the scoop rocker chair and the stability stool. Juan reported that he preferred the scoop rocker chair, and Paul reported that he preferred the stability stool.

Discussion

The purpose of this study was to evaluate the effects of stability stools and scoop rocker chairs on in-seat and on-task behaviors in an academic setting for children at risk for EBD. Results showed that both in-seat and on-task behaviors improved for all participants when sitting in the scoop rocker chairs and the stability stools compared to regular desk chairs. The staggered implementation within the multiple-baseline design allowed for experimental control to be established between the standard chair condition and the two alternative seating conditions. Visual analysis rendered no noticeable differentiation of the stability stool or the scoop rocker chair data paths for any of the participants in the alternating treatments phase. However, when analyzing averages of in-seat behavior, all three participants had higher averages while seated in the scoop rocker chair when compared to the stability stool. As an alternative, when analyzing on-task behavior, Juan and Paul had higher averages when sitting on the stability stool, while Roman had a higher average of on-task behavior while seated in the scoop rocker chair.

Within the choice phase, the stability stool was selected most often by all participants, with the exception of Paul who chose the stability stool and scoop rocker an equal number of times. However, all students chose both the stability stool and scoop rocker chair at least twice, displaying no clear preference between the two seating options. In addition, Paul and Roman chose to sit in their regular desk chairs at least once. Paul’s in-seat behavior reduced below baseline levels when seated in the regular desk chair during the choice phase, which may add to the already established experimental control of baseline (standard chair condition) over the two seating alternatives within the intervention phase of the multiple-baseline design. In addition, the social validity results from the participants regarding their preference of seating were not consistent with the type of seating they chose most often. Roman chose to sit in the stability stool across 77% of all choice sessions but chose both the stability stool and scoop rocker as his preference when social validity was assessed via a questionnaire. Juan chose to sit in the stability stool across 67% of all choice sessions; however, he reported preference for sitting in the scoop rocker chair when asked via a questionnaire. Lastly, Paul chose to sit in the stability stool and scoop rocker chair equally across 42% of choice sessions but reported his preference to be the stability stool. Paul also reported that he was not sure if he liked sitting on the scoop rocker chair. Taken together, the ability to change seating often in the classroom could be a mitigating factor for maintaining high levels of in-seat and on-task behavior. Teachers may consider having a variety of seating options available and providing seating choices to students in the classroom.

When collecting social validity with the participants, Roman indicated that he thought the stability stool and scoop rocker chair made him work harder and they were very comfortable. Juan indicated that he enjoyed both chairs, stating that he enjoyed spinning on the stability stool and rocking all the way back in the scoop rocker chair. It should be noted that rocking all the way back in the scoop rocker chair was scored as out-of-seat behavior. Paul indicated that the regular chairs were uncomfortable, the scoop rocker chairs were kind of comfortable, and the stability stools were very comfortable. The verbal responses from students might support both potential behavioral mechanisms previously mentioned that could be responsible for behavior change. For example, Roman and Paul stated that the alternative seating options were more comfortable, which could result in less escape behavior from an uncomfortable seat. Juan stated that he could move around, spin, and rock with alternative seating options, which could indicate that these types of seating options may function as an abolishing operation (Morrison et al., 2011).

The overall results of this study are consistent with previous research that also indicated improvements on in-seat and on-task behaviors with alternative seating (Castellucci et al. 2017; Hoofman, 2018; Krombach & Miltenberger, 2020; Matin Sadr et al., 2017; Parcells et al., 1999; Schilling et al. 2003; Schilling & Schwartz, 2004; Schoolcraft, 2018; Taipalus et al., 2017; Umeda & Deitz, 2011; Wingrat & Exner, 2005). Roman’s results were consistent with previous literature that indicated alternative seating led to increases in on-task behavior and in-seat behavior for students with ADHD (Fedewa & Erwin, 2011; Schilling et al., 2003). In addition, it is important to note that Roman was considerably younger than the participants from previous research on alternative seating with students with ADHD (e.g., kindergarten versus fourth and fifth grade). The results for the stability stool were consistent with the findings from Hoofman (2018) related to both improvements on in-seat behavior and teacher preference for the stability stool. However, none of the prior studies evaluated the scoop rocker chair, which resulted in similar improvements for participants. Although the scoop rockers are one of the more affordable options for alternative seating, a limitation is how low they sit to the ground, which does not allow students to use their desks to complete work.

During the study, Roman was taking medication for his ADHD and impulsivity. It is unclear whether changes in his in-seat and on-task behavior could be related to medication adherence. There were some days in which it was reported that he was not given his medication; however, it is unknown if the parents reliably reported medication adherence to his teacher. Each change in medication that was reported to the researchers is indicated in the figures by an asterisk. On the occurrence where Roman chose the regular desk chair in the choice phase, it was reported that he did not receive his medication that day. During this session, he indicated that he was “ready to go back to his normal chair.” The overall role of medication seems to be minimal because he was taking medication throughout the study, but his on-task and in-seat behaviors were highly variable during baseline and stabilized at high levels during the alternative seating intervention condition.

Although the average percentage of in-seat and on-task behaviors increased for all participants, some sessions in the intervention and choice phases were consistent with baseline levels. This may indicate that alternative seating options alone may not be potent enough to result in reliable behavior change, as this conclusion is consistent with other antecedent-only interventions. The addition of reinforcement-based strategies might further increase in-seat and on-task behaviors. However, the ease of implementation of alternative seating was preferred by teachers in this study. Other extraneous factors in kindergarten classrooms could also result in variability on any particular school day, such as schedule changes, absences, activities, and student interactions.

All three teachers were amenable to the alternative seating and indicated a preference for the stability stools because they allowed students to remain in their assigned areas at their desks. Mrs. Eleanor was the only teacher who specifically noted that the alternative seating options were distracting to peers and indicated that she wanted to provide alternative seating options to all her students. Thus, Mrs. Eleanor along with Mrs. Rita indicated that they were already looking into funding through their school to provide alternative seating to more students in their classes.

Although the results of this study are promising, there were several limitations to this study. One limitation was that the session length was short (i.e., 5 min) to accommodate the teachers regular activity schedule; therefore, it is not known if longer session lengths would have resulted in different outcomes. It was surprising that sitting for 5 min in baseline was a struggle for all of the participants. Adding the alternative seating increased in-seat behavior; however, future studies might extend the length of time or slowly increase the amount of time students are expected to remain seated and on-task. Other limitations of being in a classroom included fire drills, unexpected administrative observations, and distractions from other students. For example, administrative observations could have affected teacher and student behavior such as the teacher increasing prompting for students to stay on task. In all three classrooms, other students asked the study’s participants why they had a different chair, or where they got the chair. Other students also attempted to take the chair and sit on it themselves and were redirected and/or prompted to move by the teachers. Students were easily redirected back to their work and only one teacher, Mrs. Eleanor, expressed classroom concerns regarding the distraction alternative seating had on the other students in her classroom. However, given the interest in alternative seating by some of the other students, this suggests that other students may also find alternative seating preferable, and teachers could consider offering different seating options to all students. Two of the three teachers in the current study indicated they were meeting with their principals to discuss funding to provide alternative seating options to all their students.

Future research should replicate and extend the current findings to examine how alternative seating can be used for other age groups, academic tasks or settings, and diagnoses. While the research questions for the current study did not assess task completion, future research should consider supplementing on-task behavior with permanent product data on academic task completion and accuracy. Studies could also examine if and how alternative seating may decrease other disruptive behavior during academic tasks. Additional future research areas include using the scoop rocker chair during carpet time, using alternative seating as a reinforcer, or using alternative seating to increase in-seat and on-task behaviors during group work. Research should also evaluate how choice and preference for alternative seating affects the maintenance of treatment effects. A key aspect of the current study is how the results from the choice analysis and student social validity questionnaire were not consistent. Students verbally said one form of alternative seating was their most preferred; however, during the choice analysis they chose another seating option more often. Future research and practitioners should consider having participants/clients choose between two or more equally effective interventions instead of solely relying on standard social validity measures, because these social validity measures may not adequately capture preference. In addition, preference may shift over time, thus enabling clients to choose their intervention daily or multiple times a day (i.e., per session) may lead to increased appropriate behavior or better maintenance of these effects. In conclusion, results of this study indicate a positive extension to using alternative seating for children at risk for EBD by comparing two types of alternative seating that previously had limited research. Alternative seating was a feasible and low effort procedure for teachers to improve the in-seat and on-task behaviors of their students.

Data Availability

The data that support the findings of this study are available from the corresponding author upon reasonable request.

Declarations

Conflict of Interest/Competing Interests

The authors have no relevant financial or non-financial interests to disclose. The authors have no competing interests to declare that are relevant to the content of the article.

Ethical Approval

This study was approved by the University of South Florida’s Internal Review Board and the county school board where the study was conducted.