Value of Compliance Journal-Central Texas College .

Received: 29 August 2016 Revised: 5 April 2017 Accepted: 7 July 2017 DOI: 10.1002/bin.1494 BRIEF REPORT Embedding a functional analysis of compliance in small group instruction Blair P. Lloyd | Emily S. Weaver Department of Special Education, Vanderbilt University, Nashville, TN, USA Correspondence Blair Lloyd, Box 228 Peabody College, Vanderbilt University, Nashville, Tennessee 37203, USA. Email: blair.lloyd@vanderbilt.edu | Johanna L. Staubitz Research on adaptations to standard functional analyses for use in classroom settings has increased in recent years. However, few studies have focused on procedural variations specific to assessing noncompliance in the context of academic instruction. In the current study, we trained a special education teacher to embed a functional analysis of compliance in small group instruction. The goal of the functional analysis was to identify an effective reinforcement contingency for compliance for a second grade student with an intellectual disability. Results suggested a combined escape + tangible contingency increased compliance to instructional prompts relative to other conditions. The functional analysis of compliance represents a variation on previous functional analyses of noncompliance with potential to increase ecological and social validity of assessment procedures for classroom settings. 1 | I N T RO DU CT I O N Noncompliance is a barrier to accessing effective instruction in school. Teachers consistently identify “following directions” as a pivotal skill as students begin early elementary school (e.g., Lin, Lawrence, & Gorrell, 2003). In addition to predicting low academic achievement, chronic noncompliance has been shown to predict disciplinary sanctions, restrictive educational placements, and more severe forms of problem behavior (e.g., Agostin & Bain, 1997; Kalb & Loeber, 2003; Skiba, Peterson, & Williams, 1997). To inform intervention strategies to increase compliance, functional analyses of noncompliance have been applied in clinical and educational settings. In young children with and without disabilities, noncompliance has been found to be maintained by negative reinforcement in the form of escape from tasks (e.g., McKerchar & Abby, 2012), positive reinforcement in the form of adult attention (e.g., Kern, Delaney, Hilt, Bailin, & Elliot, 2002; Rodriguez, Thompson, & Baynham, 2010), and positive reinforcement in the form of access to preferred activities (e.g., Wilder, Allison, Nicholson, Abellon, & Saulnier, 2010). With very few exceptions (i.e., Noell, VanDerHeyden, Gatti, & Whitmarsh, 2001), prior studies on the assessment and treatment of noncompliance have been conducted in controlled clinic rooms by experimenters (e.g., Bouxsein, Roane, & Harper, 2011; Kern et al., 2002; McKerchar & Abby, 2012; Rodriguez et al., 2010; Wilder, Harris, Reagan, & Rasey, 2007). In many cases, assessment conditions included This research was supported by a Peabody College Small Research Grant. We would like to thank Claire Diekman for her assistance with data collection and analysis. Behavioral Interventions. 2017;32:427–433. wileyonlinelibrary.com/journal/bin Copyright © 2017 John Wiley & Sons, Ltd. 427 428 LLOYD ET AL. demands to complete nonacademic tasks, such as picking up or throwing away trash (e.g., Bouxsein et al., 2011; Rodriguez et al., 2010; Wilder et al., 2007). Though such procedures maximize experimental control, they might also compromise the extent to which results are relevant to the academic settings in which students display noncompliance. Though procedural variations designed to improve the ecological and social validity of functional analyses in schools have increased in recent years (Lloyd, Weaver, & Staubitz, 2016), few studies have focused on variations specific to assessing noncompliance in the context of academic instruction. One potential variation is to focus on identifying reinforcers for compliance, rather than noncompliance, in the functional analysis. For the following reasons, this approach may provide an efficient path to identifying effective intervention strategies. First, compliance and noncompliance can be considered two concurrently available alternatives that differ in response effort. Because response effort has been shown to be an influential factor in compliance for young children (e.g., Wilder, Fischetti, Myers, Leon‐Enriquez, & Majdalany, 2013), a reinforcer maintaining noncompliance may not be sufficient to reinforce compliance as a higher effort alternative (Neef, Shade, & Miller, 1994). In other words, identifying a reinforcer for noncompliance via functional analysis may not inform the conditions in which compliance is effectively reinforced (Holden, 2002). Second, each condition of a functional analysis of compliance represents a potential intervention, thus combining assessment and intervention into one endeavor. An additional anticipated advantage of this variation is that teachers and other school staff involved in these assessments may be more likely to understand and accept the logic of a functional analysis when targeting a desirable behavior. To maximize the ecological and social validity of assessment procedures, we trained a special education teacher to complete a functional analysis of compliance for a student who was noncompliant during small group instruction. All assessment conditions were embedded in small group instruction to identify an effective reinforcement contingency for compliance in this routine. 2 2.1 METHOD | | Participants and setting Juan was an 8‐year‐old Latin‐American boy with an intellectual disability and autism. His primary mode of communication included pointing to pictures that were presented in a small array. Juan was nominated to participate in the study by his special education teacher, who reported he engaged in low levels of compliance during small group instruction, despite her attempts to promote compliance (e.g., repeating prompts, providing reminders of upcoming activities, and using choice boards) in this context. Juan0 s teacher was a 34‐year‐old Black woman with a Master0 s degree in Elementary Education. She had 6 years of experience teaching students with disabilities and had worked with Juan for 1 year. All sessions were conducted in Juan0 s special education classroom at a kidney‐shaped table during small group lessons. In a typical lesson, the teacher read a story or modeled an activity (e.g., sorting by color) for a group of three students including Juan. Throughout the lesson, she provided opportunities for each student to engage with the content of the story (e.g., “Touch the cat”) or practice a skill related to the activity (e.g., “Put the bear in the red cup”). A paraprofessional was occasionally present at the table to assist one of the other two students in the small group, but did not interact with Juan during sessions. Additional students were present in other areas of the classroom, but worked on different tasks with one‐on‐one support from other classroom staff. 2.2 | Response definitions We defined teacher instructional prompt (IP) as a prompt directed to Juan to complete an action related to the present academic activity. All IPs specified motor actions accompanied by gestural or model prompts (e.g., “Turn the page like this” and “Put the bear in the red cup” [points to bear and red cup]). We did this to minimize differences in prompt difficulty across trials and to minimize the likelihood that Juan0 s noncompliance was due to receptive language LLOYD ET AL. 429 deficits. Teacher play prompts (PPs) were defined identically with the exception that the prompt was to complete an action involving a preferred item rather than one related to the academic activity. We defined compliance as the initiation of the prompted action within 5 s of a teacher IP or PP. The 5‐s latency was selected based on (a) prior descriptive data indicating that when Juan did comply, it was within 5 s and (b) teacher report that this latency was appropriate for the routine. Complian
ce was coded even if the action was completed incorrectly. For example, if the teacher pointed to a cat on the page of a book and said “Touch the cat,” compliance would be coded if Juan touched any part of the page. Or, if the teacher presented a red cup and a blue cup and said “Put the bear in the red cup,” compliance would be coded if Juan put the bear in either cup. The decision to code such attempts as compliance was based on teacher report and prior observations indicating that Juan0 s noncompliance was characterized by either no response at all or, in some cases, attempts to leave the table. Thus, we coded noncompliance if Juan made no attempt to engage in the prompted action within 5 s or if Juan attempted to leave the table. 2.3 Data collection procedures and reliability | We used tablets with Multi‐Option Observation System for Experimental Studies software (Tapp, Wehby, & Ellis, 1995) to collect timed‐event data on student and teacher responses from video‐recorded sessions. Inter‐observer agreement (IOA) data were collected for a minimum of 30% of randomly selected sessions per condition (range, 31–50%). We calculated point‐by‐point agreement based on a 5‐s window of agreement by dividing the number of agreements by the sum of agreements and disagreements and multiplying by 100. Mean IOA across sessions was 97% (range, 85–100%) for teacher prompts and 96% (range, 75–100%) for compliance. We collected paper‐and‐pencil data to assess reliability of independent variables during 100% of sessions. We measured the occurrence of the following programmed teacher behaviors: (a) wears the discriminative stimulus (i.e., lei) assigned to each condition, (b) delivers 10 IPs or PPs (depending on condition) to Juan, (c) delivers no additional prompts to Juan, (d) allows 5‐s wait time for compliance following each IP or PP, (e) delivers correct programmed consequences for compliance within 5 s, and (f) delivers correct programmed consequences for noncompliance within 5 s. Behaviors (d)–(f) were documented for each teacher prompt. Mean percentages of fidelity were 93% (range, 86–98%) for the baseline condition, 97% (range, 88–100%) for the play condition, 95% (range, 89–100%) for the contingent attention condition, 95% (range, 87–100%) for the contingent escape condition, and 97% (range, 84–99%) for the contingent escape + tangible condition. IOA data on procedural fidelity were collected for a minimum of 25% of randomly selected sessions per condition. Mean IOA on procedural fidelity data across conditions was 95% (range, 89–100%). 2.4 Experimental design | Following an initial baseline condition, we used a series of alternating treatments designs to compare levels of compliance in each test condition to those in a play condition. Each test condition represented a potential reinforcement contingency for compliance and included programmed consequences for both compliance and noncompliance. We tested one hypothesis at a time to minimize the likelihood of fidelity errors and multitreatment interference. We conducted one test and one play condition per day, using a blocked pairs random assignment procedure to determine the order of conditions. Once we identified a test condition that produced levels of compliance similar to those observed in the play condition, we briefly withdrew and then reimplemented the contingency to allow replications of effect. 2.5 2.5.1 Study procedures | | Teacher training and coaching After an initial open‐ended interview to collect information on Juan0 s noncompliance, we conducted two teacher training meetings. The first meeting took place prior to initiating baseline and consisted of reviewing procedures 430 LLOYD ET AL. for baseline sessions. The second training meeting took place after baseline data were collected. During this meeting, we reviewed procedures for the contingent attention, contingent escape, and play conditions. We role‐played each of these conditions until the teacher implemented each condition without errors. After this meeting, and before formal data collection continued, the teacher practiced each condition twice with Juan during small group lessons. Research staff provided corrective feedback as needed. Finally, prior to initiating the escape + tangible condition, the teacher practiced implementing this condition with Juan and with corrective feedback from research staff. Throughout all phases, research staff provided (a) procedural reminders before sessions, (b) verbal prompts during sessions for programmed teacher behaviors that required monitoring time (e.g., when the next teacher prompt could be delivered [30 s following previous prompt]; when the window for compliance had elapsed [5 s following teacher prompt]), and (c) corrective feedback following sessions, as needed. The total time spent in teacher training activities was 3.2 hr (i.e., 30‐min interview, 20‐min baseline training, 1‐hr functional analysis training, and 1.4 hr spent in practice sessions). 2.5.2 | Experimental conditions All sessions were embedded in small group lessons and included 10 teacher prompts directed to Juan at approximately 30‐s intervals. Session durations varied to ensure 10 opportunities for compliance per session (M = 8.3 min; range, 5.6–11.1 min). Across test conditions, the teacher only delivered IPs (e.g., “Turn the page,” “Put the bear in the red cup,” “Give me the blue bear”); in the play condition, the teacher only delivered PPs (see examples in Play condition description below). Prompts were allowed to be repeated within session as long as they were not repeated across consecutive trials. Teacher consequences for compliance and noncompliance were selected in collaboration with the teacher to maximize the extent to which programmed events during experimental conditions represented those commonly occurring during instruction. For example, the teacher consistently delivered praise following instances of compliance across students. Although she was willing to evaluate additional potential reinforcers for Juan0 s compliance, she was unlikely to withhold praise when he did comply. Thus, across conditions, the teacher delivered a brief praise statement following instances of compliance in addition to the programmed consequences (e.g., “Nice matching, Juan” prior to a 30‐s break). To aid discrimination, the teacher wore different colored leis in each condition. The total time spent in experimental sessions was 5.8 hr. Baseline During baseline sessions, the teacher delivered a brief praise statement following compliance and a physical prompt following noncompliance, then continued with the lesson for 30 s (i.e., presented content or modeled activities to the group and delivered IPs to one of the other two students) until the next programmed IP for Juan. These programmed consequences were selected to match those the teacher typically used during instruction. Play The play condition was designed to minimize the establishing operation (EO) of the IP while preserving the opportunity to engage in compliance. We hypothesized that the play condition would represent a “best case scenario” for Juan0 s compliance, such that any test condition producing similar levels of compliance should be evaluated as a potentially effective reinforcement contingency. The items and prompts used in the play condition were selected based on teacher report of Juan0 s preferred items and appropriate ways to engage with these items. Examples included an electronic toy with buttons to activate sounds (“Press the button”), a drum toy (“Tap the drum”), a plastic tube toy that expanded and contracted (“Pull”), and beads (“Put the beads in the jar”). Following compliance with PPs, the teacher delivered brief praise, set the preferred item aside, and continued the group lesson for 30 s (i.e., presented content or modeled activities to the group and delivered IPs to one of the other two students) until the next programmed PP for Juan. Following noncompliance, the teacher physically prompted Juan to complete the action, set the item aside, and co
ntinued with the lesson (30 s). LLOYD ET AL. 431 Contingent attention In this test condition, following compliance with IPs, the teacher delivered attention in the form of enthusiastic, descriptive praise (e.g., “Wow Juan, look at you! Great job turning the page!”), followed by a new IP. Contingent on noncompliance, the teacher provided escape by removing academic materials and delivering prompts to another student for 30 s. Contingent escape This test condition was designed to represent reversed contingencies from the contingent attention condition (Rodriguez et al., 2010). Contingent on compliance with IPs, the teacher provided a brief praise statement, then escape by removing academic materials and delivering prompts to another student for 30 s. Contingent on noncompliance, the teacher provided attention in the form of verbal encouragement, a narration of the response as she physically prompted Juan to complete the action, then a new IP. Contingent escape + tangible In this condition, contingent on compliance with IPs, the teacher delivered an enriched break by giving Juan a preferred item, removing academic materials, and delivering prompts to another student for 30 s. The preferred items in this condition were identified by the teacher and were distinct from those used in play sessions. Contingent on noncompliance, the teacher physically prompted Juan while narrating the requested action, then removed instructional materials and delivered instruction to another student for 30 s. Preferred items were rotated such that no single item was presented more than three times per session. We also asked the teacher to restrict access to these items prior to each lesson. 2.5.3 | Social validity questionnaire Following the functional analysis of compliance, Juan0 s teacher completed a questionnaire addressing the acceptability and ecological validity of assessment procedures (adapted from the Intervention Rating Profile‐15; Albin, Lucyshyn, Horner, & Flannery, 1996; Martens, Witt, Elliott, & Darveaux, 1985). The questionnaire included 18 statements, each of which was rated on a Likert‐type scale from 1 (strongly disagree) to 6 (strongly agree). Teacher ratings of social acceptability and ecological validity of assessment procedures were high (M = 5.6; range, 4–6). Example statements rated as 6 (i.e., strongly agree) were as follows: (1) Overall, these assessments fit the daily routines in my classroom; (2) it was easy to transition from assessment activities to my typical teaching activities; and (3) I would suggest the use of these assessments to other teachers who support students with similar behavior challenges. The only item rated as 4 (i.e., slightly agree) was as follows: These assessments are a fair way to handle the student0 s problem behavior. 3 | RESULTS A ND DIS CUS SION Results of the functional analysis of compliance are displayed in Figure 1. During baseline, the percentage of compliance was variable (range, 30–80%). Results of the first comparison phase indicated that, relative to the play condition, percentages of compliance were consistently lower when attention was delivered contingent on compliance and escape was delivered contingent on noncompliance. This pattern suggested teacher attention was unlikely to be an effective reinforcer for compliance. In the second comparison phase, percentages of compliance remained lower relative to the play condition when escape was delivered contingent on compliance and attention was delivered contingent on noncompliance. This pattern suggested that escape alone was unlikely to be an effective reinforcer for compliance. In the third comparison phase, percentages of compliance increased and approximated percentages in the play condition when escape with access to tangibles was provided contingent on compliance, and escape without tangibles was contingent on noncompliance. We interpreted this pattern to suggest that contingent enriched breaks 432 LLOYD ET AL. FIGURE 1 Percentage of compliance across baseline, comparison, brief reversal, and verification phases (Esc + Tan = Escape + Tangible) was a potentially effective reinforcer for compliance. To verify this hypothesis, we conducted a brief reversal to the previous test condition (contingent escape), followed by a reintroduction of the escape + tangible condition to replicate the effect.1 Despite the overlap between the first escape + tangible condition and the brief reversal, there was an immediate decrease in level when the brief reversal was implemented, followed by an immediate increase in level when the final escape + tangible condition was reintroduced. We interpreted these results to support the hypothesis that contingent access to escape and tangibles was an effective reinforcement contingency for compliance. The current study extends prior work on functional analyses of noncompliance in three ways. First, it offers a model of embedding a functional analysis within a typical instructional routine that was successfully and efficiently implemented by a classroom teacher. All assessment procedures, including teacher training activities, were completed in less than 10 hr, with more than half of this time representing activities occurring in the context of the usual small group instruction routine. Second, it demonstrates potential advantages of targeting compliance rather than noncompliance in the functional analysis. Common reinforcers for noncompliance (i.e., escape and attention) were ruled out as effective reinforcers for compliance, which led to identifying a combined escape + tangible contingency as most likely to reinforce compliance during small group lessons. Third, the teacher who implemented the functional analysis considered the assessment procedures to be both acceptable and representative of Juan0 s usual instructional routine, further supporting the potential for this variation to be used in classroom settings. Three primary limitations of this study should be noted. First, the play condition did not represent a true control condition. Although we attempted to minimize the EO by presenting noninstructional prompts, we were not able to completely eliminate the EO without losing the opportunity to measure compliance. That is, Juan was still exposed to small group instruction and was prompted to engage in specific motor actions in the play condition. Second, though we attempted to control for prompt difficulty by ensuring all prompts specified motor actions that were modeled or gestured, differences in specific instructional prompts within conditions may have contributed to variability in compliance. Third, due to time constraints (i.e., end of school year), we were not able to assess the maintenance of intervention effects. In future evaluations of this approach, continuing intervention and systematically fading the schedule of reinforcement for compliance will be necessary to evaluate clinical significance. ENDNOTE 1 Due to the timing of the end of the school year, we chose to conduct a brief (single session) reversal to ensure multiple sessions in the final verification phase. LLOYD 433 ET AL. ORCID Blair P. Lloyd http://orcid.org/0000-0001-9444-7230 RE FE R ENC ES Agostin, T. M., & Bain, S. K. (1997). Predicting early school success with developmental and social skills screeners. Psychology in the Schools, 34, 219–228. https://doi.org/10.1002/(SICI)1520‐6807(199707)34:3 Albin, R. W., Lucyshyn, J. M., Horner, R. H., & Flannery, K. B. (1996). Contextual fit for behavioral support plans: A model for “Goodness of Fit.”. In L. K. Koegel, R. L. Koegel, & G. Dunlap (Eds.), Positive behavioral support: Including people with difficult behavior in the community (pp. 81–98). Baltimore: Paul H. Brookes. Bouxsein, K. J., Roane, H. S., & Harper, T. (2011). Evaluating the separate and combined effects of positive and negative reinforcement on task compliance. Journal of Applied Behavior Analysis, 44, 175–179. https://doi.org/10.1901/ jaba.2011.44‐175 Holden, B. (2002). Some limitations of functional analyses: Treatment of problem behavior based on the study of alternative responses.
Behavioral Interventions, 17, 191–209. https://doi.org/10.1002/bin.118 Kalb, L. M., & Loeber, R. (2003). Child disobedience and noncompliance: A review. Pediatrics, 111, 641–652. https://doi.org/ 10.1542/peds.111.3.641 Kern, L., Delaney, B. A., Hilt, A., Bailin, D. E., & Elliot, C. (2002). An analysis of physical guidance as reinforcement for noncompliance. Behavior Modification, 26, 516–536. https://doi.org/10.1177/0145445502026004005 Lin, H. L., Lawrence, F. R., & Gorrell, J. (2003). Kindergarten teachers0 view of children0 s readiness for school. Early Childhood Research Quarterly, 18, 225–237. https://doi.org/10.1016/ S0885‐2006(03)00028‐0 Lloyd, B. P., Weaver, E. S., & Staubitz, J. L. (2016). A review of functional analysis methods conducted in public school classroom settings. Journal of Behavioral Education, 25, 324–356. https://doi.org/10.1007/s10864‐015‐9243‐y Martens, B. K., Witt, J. C., Elliott, S. N., & Darveaux, D. X. (1985). Teacher judgments concerning the acceptability of school‐ based interventions. Professional Psychology: Research and Practice, 16, 191–198. https://doi.org/10.1037/0735‐ 7028.16.2.191 McKerchar, P. M., & Abby, L. (2012). Systematic evaluation of variables that contribute to noncompliance: A replication and extension. Journal of Applied Behavior Analysis, 45, 607–611. https://doi.org/10.1901/jaba.2012.45‐607 Neef, N. A., Shade, D., & Miller, M. S. (1994). Assessing influential dimensions of reinforcers on choice in students with serious emotional disturbance. Journal of Applied Behavior Analysis, 27, 575–583. https://doi.org/10.1901/jaba.1994.27‐575 Noell, G. H., VanDerHeyden, A. M., Gatti, S. L., & Whitmarsh, E. L. (2001). Functional assessment of the effects of escape and attention on students0 compliance during instruction. School Psychology Quarterly, 16, 253–269. https://doi.org/10.1521/ scpq.16.3.253. 19892 Rodriguez, N. M., Thompson, R. H., & Baynham, T. Y. (2010). Assessment of the relative effects of attention and escape on noncompliance. Journal of Applied Behavior Analysis, 43, 143–147. https://doi.org/10.1901/jaba.2010.43‐143 Skiba, R. J., Peterson, R. L., & Williams, T. (1997). Office referrals and suspension: Disciplinary intervention in middle school. Education and Treatment of Children, 20, 295–315. Tapp, J., Wehby, J., & Ellis, D. (1995). A multiple option observation system for experimental studies: MOOSES. Behavior Research Methods, Instruments, & Computers, 27, 25–31. https://doi.org/10.3758/BF03203616 Wilder, D. A., Allison, J., Nicholson, K., Abellon, O. E., & Saulnier, R. (2010). Further evaluation of antecedent interventions on compliance: The effects of rationales to increase compliance among preschoolers. Journal of Applied Behavior Analysis, 43, 601–613. https://doi.org/10.1901/jaba.2010.43‐601 Wilder, D. A., Fischetti, A. T., Myers, K., Leon‐Enriquez, Y., & Majdalany, L. (2013). The effect of response effort on compliance in young children. Behavioral Interventions, 28, 241–250. https://doi.org/10.1002/bin.1362 Wilder, D. A., Harris, C., Reagan, R., & Rasey, A. (2007). Functional analysis and treatment of noncompliance by preschool children. Journal of Applied Behavior Analysis, 40, 173–177. https://doi.org/10.1901/jaba.2007.44‐06 How to cite this article: Lloyd BP, Weaver ES, Staubitz JL. Embedding a functional analysis of compliance in small group instruction. Behavioral Interventions. 2017;32:427–433. https://doi.org/10.1002/bin.1494 Copyright of Behavioral Interventions is the property of John Wiley & Sons, Inc. and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder’s express written permission. However, users may print, download, or email articles for individual use.