Abstract
Behavior intervention plans (BIPs) based on a functional behavior assessment are supported by a large body of research showing their potential for positively impacting student behavior; however, research also indicates that many classroom teams struggle to implement BIPs with fidelity. We conducted a statewide survey of over 600 teachers to examine teacher-reported experiences with BIP implementation, including 13 implementation barriers previously identified in the literature. Selecting from the list provided, teachers reported the most prominent barrier to be “the cause of student problem behavior cannot be addressed through a BIP.” In written comments, teachers stated that many behavior problems were at least partially caused by factors beyond teacher control, such as a difficult home life, and perceived BIPs as less effective in these cases. Teachers from urban schools and schools with higher percentages of minority students reported encountering this barrier significantly more frequently. In general, teachers from schools (a) in urban settings, (b) with higher percentages of students receiving free or reduced lunch, and (c) with higher percentages of minority students reported significantly greater challenges to BIP implementation as well as lower fidelity and effectiveness of BIPs. Other prominent barriers identified by teachers included inconsistent implementation of BIPs across staff, inadequate resources to implement BIPs, and ineffective BIPs. Implications for improving implementation of BIPs in schools are discussed.
Keywords
A behavior intervention plan (BIP) is a document comprising individualized behavior supports for students whose behavior interferes with learning (Individuals with Disabilities Education Act, 2004). The use of BIPs is supported by an extensive body of research showing that individualized interventions incorporating functional behavior assessment (FBA), antecedent, skill-building, and reinforcement strategies are highly effective in decreasing problem behaviors and increasing adaptive behaviors (Cho & Blair, 2017; Lloyd & Kennedy, 2014). Although a large evidence base has established BIPs based on FBA as effective, research indicates that the fidelity of BIP implementation in school settings is often low (Bambara et al., 2009, 2012) and, as a result, may fail to impact student outcomes (Durlak & DuPre, 2008). For example, school-level positive behavior support (PBS) leaders surveyed on the fidelity with which BIPs were implemented by teachers reported an average BIP fidelity of 68% with a standard deviation of almost 20% (Cook et al., 2012). Even when BIPs are initially implemented as planned, fidelity of behavior interventions tends to significantly decrease within 7–10 days of initiation (Johnson et al., 2014; Noell et al., 2005). Other studies of BIP implementation under natural school conditions have found school-based BIPs to be inadequate or of low quality (Blood & Neel, 2007; Van Acker et al., 2005). Taken together, these findings suggest that many BIPs may be poorly implemented under natural school conditions and unlikely to improve student behavior.
In light of these findings, factors that promote and hinder implementation of BIPs need to be further examined. Implementation science is the study of methods to promote the systematic uptake of research findings into practice to improve the quality and effectiveness of services (Horner et al., 2017). To date, the majority of implementation research conducted on PBS has focused on Tier 1, or the universal level, of school-wide PBS (SWPBS). These investigations have found major implementation barriers to include lack of effective collaboration across staff (Coffey & Horner, 2012); lack of administrator support (McIntosh et al., 2013); philosophical differences with a PBS approach (Andreou et al., 2015; Feuerborn et al., 2016); and lack of staff knowledge, skills, professional development, and technical support in SWPBS (Yeung et al., 2016).
Less is known about (a) how these barriers are manifested in the implementation of PBS with students for whom Tier 1 (universal) and Tier 2 (targeted group) supports are insufficient, that is, at the Tier 3 (individualized) level; (b) barriers to PBS implementation with students with more significant disabilities; or (c) barriers unique to BIP implementation. Bambara and colleagues investigated barriers to BIP implementation using interview (Bambara et al., 2009) and survey (Bambara et al., 2012) methods, and found three primary barriers: insufficient time for staff to collaborate, lack of training in PBS principles, and beliefs about behavior that contradicted a PBS approach. Administrative support, specifically leadership and buy-in at the principal level, was identified as a facilitator that promoted BIP implementation by creating time structures during which school teams could meet and collaborate. Along these lines, Katsiyannis et al. (2008) surveyed 75 special education district-level administrators on their district’s implementation of FBA procedures. Participating administrators generally reported that FBA increased the effectiveness of behavior interventions and were typically conducted by teams; however, districts waited for problem behaviors to become moderately serious prior to conducting an FBA and relied on less accurate, indirect data collection methods (e.g., interviews) to inform the FBA.
A number of researchers have experimentally tested interventions aimed at improving teacher implementation of BIPs. Intervention procedures have included self-monitoring of implementation (Mouzakitis et al., 2015; Pinkelman & Horner, 2017), performance feedback (Mouzakitis et al., 2015), progress monitoring of student behavior and implementation fidelity (Pinkelman & Horner, 2017), and advanced logistical planning for implementation and responses to anticipated implementation barriers (Sanetti et al., 2014). Across studies, teacher-focused implementation support strategies were associated with improved fidelity of BIPs and improved student behavior, showing implementation supports can improve the use of BIPs.
Because matching implementation strategies to implementation barriers is likely to increase the effectiveness of these strategies, there remains a need to identify the most pervasive and impactful barriers to BIP implementation. In addition, little is known about how teacher and school contextual variables, such as years of experience and percentage of student body receiving free or reduced lunch (FRL), impact BIP implementation (Bambara et al., 2012). Finally, little research on BIP implementation has included teachers of students with more significant disabilities. As a result, a detailed theoretical framework for how barriers to BIP implementation occur in schools, how they impact implementation, and how best to address these barriers is still lacking.
Purpose and Research Questions
The purpose of the present study was to better understand the primary barriers impacting BIP implementation in schools across one state through an exploratory statewide survey. Our research questions were as follows: (1a) Which barriers do teachers identify as most prominent to implementing BIPs? (1b) What teacher and school demographics are associated with the most prominently identified barriers? (1c) What are teachers’ qualitative experiences with the most prominently identified barriers? (2a) What do teachers report as the overall effectiveness and fidelity of their BIPs? and (2b) What demographics are associated with reported BIP effectiveness and fidelity?
Method
Participants
A total of 602 respondents completed the online survey. The majority were special education teachers (94%), followed by administrators (2%), and general educators or paraeducators (less than 1% each). Most had a master’s degree (71%), some graduate school (16%), or a bachelor’s degree (12%). The sample was predominantly Caucasian (94%) with 2% identifying as African American. Most participants (36%) had fewer than 5 years of experience, the next largest group of participants had 5–10 years of teaching experience (27%), and the third largest group had over 10 years of experience (25%). The largest proportion of respondents taught children in Grades 3–6 (38%), followed by Grades 7–8 (25%), Grades 9–12 (20%), and Grades K-2 (17%). Most of the respondents (76%) taught in public school, another 14% taught in private schools, and 10% taught in charter or other types of schools. The sample was 41% suburban, 31% urban, and 27% rural. The mean percentage of the student population reported as (a) receiving FRL was 60% (SD = 33%) and (b) racial minority was 37% (SD = 33%). The majority of respondents (73%) taught in self-contained classrooms (defined as classrooms in which all students were identified with disabilities, for example, life skills); the largest proportion of teachers primarily taught students with autism spectrum disorder (23%), followed by teachers of students with intellectual disabilities (20%), specific learning disabilities (12%), and emotional and behavioral disorders (11%).
Measures
Survey development
To better understand teacher perceptions of the most prominent barriers to BIP implementation, the research team developed the Barriers to BIPs Survey. The survey was developed as part of a set of research activities designed to improve the administration of one state’s alternate assessment. The research team implemented a five-step survey development process consisting of a review of the literature on BIP implementation, creation of a survey draft based on this literature, iterative review of the survey draft by a team of content experts, additional survey review and confirmation by a national expert, and a pilot of the survey with pre- and in-service teachers.
First, a literature review was conducted to identify relevant research pertaining to school personnel’s experiences with implementation of BIPs. In particular, literature regarding barriers to implementing BIPs (e.g., Bambara et al., 2009, 2012; Cook et al., 2012; Mouzakitis et al., 2015; Sanetti et al., 2014) and behavior interventions more broadly (e.g., Durlak & DuPre, 2008; Long et al., 2016; Pas et al., 2015; Perepletchikova & Kazdin, 2005; Yeung et al., 2016) was reviewed by the first author, after which an initial survey draft and list of seven barriers was developed (i.e., BIP not effective, BIP too complicated, lack of teacher training, lack of teacher input, cause of student behavior cannot be addressed through a BIP, lack of resources, and lack of implementation support). This initial draft was then reviewed by the research team.
The research team consisted of two PhD-level content experts in PBS and BIPs, a PhD-level survey methods specialist, two advanced special education doctoral students, and one expert school-based Board Certified Behavior Analyst. The PhD-level content experts each had over 15 years’ experience developing, implementing, and supporting classroom teams in implementing BIPs with students with significant behavioral challenges. Team members accessed the related research and provided feedback on the initial survey draft and list of barriers through iterative process meetings. As a result, several barriers were reworded to improve readability, six additional barriers were developed (i.e., BIP implemented inconsistently across staff, BIP not based on an FBA, BIP not necessary, BIP not a good fit for classroom, BIP not individualized to student, and no one checks BIP implementation), and questions regarding teacher and school demographics were added. The survey was also modified to include two questions per barrier: one question related to the frequency with which teachers encountered the barrier and one related to the degree of impact the barrier had on BIP implementation. Finally, the research team added an open-ended question in relation to each of the 13 barriers in which teachers could describe their experiences with that particular barrier. Decisions on survey modifications were made via group consensus.
The survey was then provided to a nationally recognized PhD-level content expert, who confirmed the barriers identified in the survey and recommended minor revisions of the wording and format, which were made. After consultation with a Qualtrics survey specialist, the survey was put on Qualtrics. The electronic survey was piloted with a group of 15 master’s-level graduate students, including special education teachers, behavior specialists, and trainees in Applied Behavior Analysis. The group provided feedback on the length and wording of the survey, which resulted in further revisions to improve readability. Finally, the survey was shared with a local district Director of Special Education who approved the wording and format of the questions with no recommended changes.
Barriers to BIPs survey
The final survey contained approximately 60 questions (depending on responses provided) and consisted of four sections. The first section collected participants’ demographic information. The second section asked participants to report the frequency and impact of the 13 barriers to implementation (see Table 1 for a full list of barriers in the survey). To differentiate between (a) how frequently teachers encountered a barrier and (b) the size, or impact, of the barrier (e.g., a barrier could be infrequent but highly impactful when it is present, or frequent but have a relatively minor impact on implementation), teachers were asked about both frequency and impact of each barrier. Specifically, participants were asked to rate the frequency with which they encountered each barrier on a 4-point Likert-type scale by selecting Always or almost always, Often, Occasionally, or Never (e.g., “How frequently have you encountered the following barrier to implementing students’ BIPs: The BIP is not implemented consistently across staff members”). Participants encountering the barrier at least occasionally were then asked to rate the impact of the barrier on their BIP implementation by selecting it as a Major, Minor, or Not a barrier (e.g., “When the BIP has not been implemented consistently across staff members, was it a major barrier, a minor barrier, or not a barrier to implementing BIPs”). A major barrier was defined as one that makes implementation of the BIP impossible or extremely difficult, whereas a minor barrier was one that did not greatly hinder BIP implementation but may require some adaptation of the BIP. Participants were also given space to comment on how each barrier had or had not been an issue in their BIP implementation. The final two questions in this section asked participants to rate how often, in general, they faced any barriers to BIP implementation and whether those barriers were most often major or minor.
Reported Frequency and Impact of Barriers to BIP Implementation (N = 602).
Note. BIP = behavior intervention plans; FBA = functional behavior assessment.
The third section asked teachers about the effectiveness and implementation fidelity of their BIPs. Specifically, teachers were asked to rate the effectiveness of their BIPs for (a) decreasing problem behavior and (b) increasing appropriate behavior on a 4-point Likert-type scale by selecting Extremely effective, Fairly effective, Somewhat effective, or Not effective. Next, teachers indicated the percentage fidelity with which their BIPs were typically implemented by selecting from a range of 0–100. The fourth section of the survey asked teachers to check all that apply from a list of potential barriers and facilitators to BIP implementation that included more specific items not asked about in section two, such as access to paraeducators, materials, and behavior specialists. Teachers were provided with a final opportunity to leave comments and describe strategies they use to work around barriers to BIP implementation. The survey took approximately 30 min to complete and is available from the first author upon request.
A Cronbach’s alpha reliability analysis (Bonett & Wright, 2015) was performed across the items that required an answer from all respondents: how frequently the 13 barriers were encountered and the five overall implementation and effectiveness items listed above. Reliability for these items was adequate (α = .80).
Procedures
A link to the anonymous electronic survey was sent via email to all special education teachers in one Northeastern Mid-Atlantic state listed as having students who took the state’s alternate assessment in reading or math. The initial email list, totaling 4,387 potential respondents, was a convenience sample shared by the test vendor through a preexisting research-practice partnership designed to improve the administration of the state’s alternate assessment and approved by the authors’ institutional review board. Of these email addresses, 143 bounced back, resulting in 4,244 potential participants. Because the survey was only relevant for school staff who currently or previously worked with a student with a BIP, it was expected that some proportion of teachers receiving the survey link would not be members of the target population and would not complete the survey. Therefore, a snowball recruitment method (Heckathorn, 2011) was also used by asking teachers to forward the survey to additional school staff with at least one student with a BIP. The email included a cover letter explaining the goals of the study, outlining the main parts of the survey to the participants, and detailing participants’ rights to consent. After receiving the email, 750 (17.6%) potential participants opened the email, 694 (16.4%) began the survey, and 602 (14.2%) answered “yes” to having had a student with a BIP and continued with the full survey. Respondents were entered into a drawing to win a $100 gift card for their participation. The email addresses for the gift card drawing were collected via a separate survey link allowing participants’ responses to remain anonymous. Reminder emails for participants to compete the survey were sent at 3 weeks and 7 weeks into the survey period, and the survey was closed 8 weeks after it was initiated. Due to the survey distribution method, it was not possible to assess how many respondents were from the initial distribution as opposed to receiving the email as a forward; however, evidence suggests the vast majority of respondents were from the initial distribution (e.g., the majority of responses occurred immediately after the initial distribution or reminder emails; 99.9% taught in the target state; 94% were special educators).
Research Design
This exploratory, mixed-methods study was designed to investigate teachers’ experiences with barriers to BIP implementation and associated demographics through a one-time survey created for the purpose of this study. Both quantitative (Likert-type-type scales) and qualitative (open-ended) data were collected. The first stage of analysis used descriptive quantitative data to identify barriers to BIP implementation teachers reported as (a) occurring most frequently and (b) having the greatest impact on implementation, as well correlations between those responses and participant demographics. Next, qualitative data regarding the most prominent barriers were analyzed phenomenologically (Saldaña, 2015) to understand teachers’ subjective experiences with these barriers. Findings from quantitative and qualitative analyses were integrated at the discussion level (Creswell & Plano Clark, 2011) to provide implications for BIP implementation.
Data Analysis Plan
Relative ranking of barriers
The research team used descriptive and inferential statistics to analyze the quantitative survey results. Descriptive statistics were calculated for the overall frequency and impact of implementation barriers, effectiveness of BIPs, and implementation fidelity of BIPs. For the survey’s 13 individual barriers, we calculated the total number and percentage of respondents who selected each response option, identifying the most prominent individual barriers by frequency and impact. Because (a) the Always or almost always category for several barriers was not endorsed sufficiently to reach the minimum threshold required to perform a chi-square analysis (n = 5; Gravetter & Wallnau, 2013) and (b) Always or almost always and Often categories were qualitatively and interpretively similar, the decision was made to collapse the categories rather than exclude those responses from the analysis. Barriers were ranked by frequency based on the percentage of respondents indicating a barrier occurred Always/Often. If two barriers had the same percentage of respondents selecting Always/Often, the barrier with the greater percentage of respondents selecting Occasionally was ranked higher. Barriers were ranked by impact based on the percentage of respondents indicating a barrier had a Major impact on their implementation of BIPs. If two barriers had the same percentage of respondents selecting Major, the barrier with the greater percentage of respondents selecting Minor was ranked higher. Overall rank was determined by creating an index value (Babbie, 2016) calculated by summing frequency and impact ranks for each barrier, with smaller sums indicating higher overall rank. When summed scores were the same across barriers, those barriers received the same overall rank.
Chi-square analyses
Chi-square analyses (Gravetter & Wallnau, 2013) were used to examine whether a statistically significant relationship between teacher and school demographics and categorical questions on implementation barriers existed. In this context, a significant chi-square test indicated that a specific categorical response for one group was more likely than expected under the null hypothesis that responses across participant groups would be equivalent. Cramer’s V (Gravetter & Wallnau, 2013) was calculated for all significant results to give a measure of the effect size for the association.
First, chi-square analyses were conducted for all teacher and school demographics and responses to the five overall BIP questions (i.e., how often barriers to implementation were encountered in general, whether barriers were usually major or minor, overall effectiveness of BIP for decreasing problem behavior, overall effectiveness of BIPs for increasing appropriate behavior, average fidelity of BIPs). To reduce the total number of statistical tests and chances of type I error, only those school and teacher demographics significantly associated with overall BIP questions were used in chi-square analyses with the four individual barriers reported to be most frequent and impactful across the total sample. To keep the method of analysis consistent, continuous variables (percentage FRL, percentage minority, and percentage fidelity) were binned into four categorical variables (high, medium high, medium low, low) based on +/- one standard deviation from the mean value. Because the analysis was descriptive and the variables were not dichotomized, these categorizations avoided the primary concerns associated with categorizing continuous variables (Altman & Royston, 2006). Demographics not meeting chi-square assumptions (e.g., cells < 5) were not included in the analysis. For each significant chi-square, standardized residuals greater than 2.0 indicated where significant differences occurred (Hinkle et al., 2003) and Cramer’s V indicated the effect size of the association.
Qualitative analysis
To better understand participating teachers’ experiences with each barrier, the research team asked an optional open-ended question (i.e., “Please describe how this issue has or has not been a barrier in implementing BIPs”) in relation to each of the 13 barriers. After quantitative analyses identified the four most prominent barriers, the first and second authors reviewed respondents’ comments regarding these four barriers using a qualitative phenomenological approach (Saldaña, 2015). In alignment with phenomenology, the goal was to understand the lived experiences of the teachers and the meanings they placed on each of the four barriers. Therefore, no a priori codes were developed and instead themes emerged directly from respondents’ descriptions of their experiences with each barrier. In phenomenological analysis, trustworthiness is achieved by recognizing and setting aside the researcher’s a priori knowledge and assumptions, approaching the data with an open mind, consulting with colleagues, and writing analytic memos (Starks & Brown Trinidad, 2007). Accordingly, the first and second authors each independently (a) read all teacher comments on the four barriers, (b) described the “essence” or core experience conveyed on each barrier through iterative analytic memos, and (c) selected representative quotes. Afterwards, the two authors met to share their analyses, discuss their interpretations, and identify points of convergence and divergence between their analyses. A consensus model was then used to find agreement on the overall narratives and illustrative quotes that best described teachers’ experiences with each of the four most prominent barriers.
Results
Barriers to BIP Implementation
Table 1 presents the rankings of the 13 individual barriers by frequency, impact, and an overall rank based on the sum of frequency and impact ranks. Overall rankings indicated the most prominent BIP barriers to be (a) The cause of student problem behavior cannot be addressed through a BIP, (b) The BIP is implemented inconsistently across staff, (c) I or my classroom team have not been provided with adequate resources to implement the BIP, and (d) Even when implemented correctly, the BIP is not effective. In response to questions asking teachers to assess the overall frequency and impact of barriers, the majority of teachers reported encountering barriers to BIPs Occasionally (58%) and that these barriers were usually Minor (78%).
Demographics and most prominent barriers
No teacher or school demographics were significantly associated with teacher reports of the impact of the four most prominent barriers; however, a number of demographics were significantly associated with the frequency with which teachers reported encountering these barriers. For the most prominent barrier, The cause of the student’s problem behavior cannot be addressed through a BIP, teachers from (a) urban settings, χ2(3, N = 527) = 16.315, p = .003, V = 0.12, and (b) schools with higher percentages of minority students reported this barrier to occur more frequently, χ2(3, N = 502) = 24.085, p = .0005, V = 0.16. Regarding the barrier of Inconsistent implementation across staff, teachers from urban schools, χ2(2, N = 578) = 11.804, p = .019, V = 0.10, and schools with higher percentages of students receiving FRL, χ2(3, N = 561) = 20.781, p = .002, V = 0.14, reported this barrier to occur more frequently, whereas teachers from private schools encountered this barrier less frequently, χ2(2, N = 559) = 9.908, p = .042, V = 0.13. For the barrier of having Inadequate resources to implement the BIP, teachers from (a) urban settings, χ2(2, N = 524) = 18.167, p = .001, V = 0.13; (b) schools with higher percentages of students with FRL, χ2(3, N = 509) = 17.194, p = .009, V = 0.13; and (c) schools with a higher percentage of minority students, χ2(3, N = 499) = 15.053, p = .020, V = 0.12, reported encountering this barrier more frequently. Teachers in inclusive and resource settings reported this barrier to occur less frequently, χ2(1, N = 463) = 8.077, p = .018, V = 0.13. No teacher or school demographics were significantly associated with the frequency of the barrier, Even when implemented correctly, the BIP is not effective. In addition, no demographics were significantly associated with reported frequency of implementation barriers overall; however, teachers from urban schools reported encountering more major barriers to BIP implementation, χ2(2, N = 476) = 6.31, p = .043, V = 0.12.
Qualitative analysis of teacher comments
Qualitative comments on each of the four most prominent barriers are summarized below to examine teachers’ understandings of these barriers.
Cause of problem behavior cannot be addressed through a BIP
When asked to describe how this barrier had or had not been an issue for them, 193 teachers (32%) left comments. The majority of respondents wrote that many of their students’ behavior problems were at least partially caused by factors beyond teacher control, such as a difficult home life or internal characteristics. Teachers reported that because they did not have control over these factors, they were limited in the degree to which they could improve their students’ behavior. For example, one teacher wrote, “The BIP doesn’t modify the behavior of the parents, change the students’ home life, or take them to the doctor. It also cannot address, control or manipulate internal behavior triggers, only external.” In general, when teachers perceived student behavior as driven by factors unrelated to school environments, they found BIPs less helpful in addressing student behavior. Primary among these factors was a student’s home environment. Most teachers who described home life as a major influence in student behavior reported that because they were unable to impact a student’s home environment, they were less able to improve the student’s behavior at school: “Since we cannot control issues at home, we are limited on how we can help the child and shape his/her behaviors.”
Inconsistent BIP implementation across staff
When asked to describe how this barrier had or had not been an issue for them, 317 teachers (53%) left comments. Many teachers reported that their BIPs were implemented differently by different staff and in different settings. For example, one teacher stated that “everyone is running their own version of the BIP.” Added another, “Staff use their own methods instead of following the BIP.” Many reported that staff in other settings did not have the training or time to implement the BIP: “Art, home ec, music teachers not trained in behavior support,” and “Regular education teachers have large classes and no time or resources.” Respondents described frustration with lack of consistency across staff and felt consistent BIP implementation was critical in improving student behavior. Many teachers reported that inconsistent implementation decreased the effectiveness of BIPs and could contribute to increases in challenging student behavior: “Lack of consistency often leads to inadvertently reinforcing behaviors that are targeted for reduction and/or extinction.” Teachers also found that inconsistent implementation made it impossible to determine whether the BIP was effective. As one respondent wrote, “Data is not valid and you don’t know if the strategies are a failure due to teacher not implementing it or because of the student failure to respond.”
Inadequate resources to implement the BIP
When asked to describe how this barrier had or had not been an issue for them, 182 teachers (30%) left comments. The majority of respondents described not having the amount of time or trained staff needed to handle all aspects of BIP development, implementation, and monitoring. As one respondent wrote, “We are chronically understaffed, so classroom support is often not possible. Also, the teachers do not have contracted planning time.” Many teachers similarly described working conditions that were not sufficient for developing and implementing effective BIPs and needing to do this work outside of school hours: “I often do my FBA, IEP and BIP writing during my nights and weekends since I do not get a plan period each day.” Many teachers described how lack of planning time also made it difficult to collaborate with other school staff regarding the BIP: It is important for teachers to have time to observe the children to get a greater understanding of the student’s behavior. It is also important for teachers to have time to design the plan with all members of the team.
Some teachers described lacking materials needed to implement BIPs and having to purchase or create their own: “Little money for resources so most I purchase out of my own wallet (over $2000 a year).” A number of teachers recognized these conditions as unsupportive of their work in general: “This is an ongoing issue and affects every aspect of our profession, and BIPs are no exception.” Many teachers also described a lack of effective paraprofessionals: “Paraprofessionals are not always easy to train. Often we are short paraprofessionals and either have to go without or have a variety of subs.”
BIP is ineffective
In describing how this barrier had or had not been an issue for them, 252 teachers (42%) left comments. Many teachers described implementing BIPs as written and continuing to see student behavior problems: “The behavior continues and the plan doesn’t work.” Many teachers described ineffective BIPs as resulting in plan failure: “When a plan does not work at all that’s clearly an obstacle. There might as well be no plan.” Teachers also described ineffective BIPs as leading to decreased implementation: “If it is not effective there is no reason to continue because behavior does not change.” These teachers often reported that the BIP was created by someone outside of the immediate classroom team, such as a consultant or school psychologist, who did not know the student or the classroom context well enough to create a realistic and effective BIP: “The people writing them barely know the student and have been out of the classroom for too long (or never were a classroom teacher) and put in unrealistic expectations for both staff and the student.” Alternatively, many other teachers described encountering ineffective BIPs and surmounting this barrier by revising BIPs to increase effectiveness. Many of these teachers emphasized that they write their own BIPs and feel adept at revising the plan to maintain effectiveness: “If a BIP is not effective then we request an update and implement more effective strategies so it is not an obstacle for long.”
BIP Effectiveness and Fidelity
The majority of teachers reported that their BIPs were Somewhat (43%) or Fairly (36%) effective in decreasing problem behavior, and Somewhat (43%) or Fairly (39%) effective in increasing appropriate behavior. Teachers also reported a mean BIP fidelity of 69% with a standard deviation of 20% and a range from 0% to 100%.
Demographics and BIP effectiveness and fidelity
Chi-square analyses were conducted to examine the association between teacher responses regarding BIP effectiveness and fidelity and teacher and school demographics. Teacher-reported effectiveness of BIPs in decreasing problem behavior was associated with a number of demographics (see Table 2 in supplemental online material). All school demographics were significantly associated with effectiveness of BIPs in decreasing problem behavior, with teachers from urban schools, χ2(2, N =527) = 15.36, p = .018, V = 0.12; schools with more students receiving FRL, χ2(3, N =524) = 28.47, p = .001, V = 0.14; and schools with more minority students, χ2(3, N =510) = 36.48, p < .001, V = 0.15, reporting BIPs as less effective in decreasing problem behavior. In addition, teachers from private schools reported BIPs as more effective in decreasing problem behavior, χ2(1, N =495) = 25.34, p < .001, V = 0.23. Urban schools and schools with a higher percentage of students receiving FRL were also associated with BIPs being less effective at increasing appropriate behavior. Alternatively, teachers from suburban and private schools described BIPs as more effective in increasing appropriate behavior. In addition, teachers working in inclusive and resource settings, χ2(1, N =433) = 9.88, p = .02, V = 0.15, and teachers with more than 10 years of experience perceived BIPs as less effective in decreasing problem behavior whereas teachers with less than 5 years of experience identified BIPs as more effective, χ2(2, N =480) = 12.89, p = .045, V = 0.12.
Teacher-reported fidelity of BIP implementation was significantly associated with both school setting and percentage of students receiving FRL. Specifically, teachers from urban schools, χ2(2, N =527) = 15.182, p = .02, V = 0.12, and schools with higher percentages FRL, χ2(3, N = 524) = 22.299, p = .008, V = 0.12., reported lower BIP fidelity. In contrast, teachers from schools with lower percentages FRL reported higher BIP fidelity.
Discussion
Although the research base supporting the use of BIPs for students with persistent disruptive behavior is strong, school and classroom teams face many implementation challenges in this area. Previous research investigating barriers and facilitators to PBS implementation has found collaboration across staff, administrative support, beliefs about behavior management, and skill and professional development in PBS to be prominent factors impacting implementation. The purpose of the present study was to extend previous literature by (a) further investigating the most prominent barriers to BIP implementation, (b) examining how demographic characteristics of teachers and schools may affect experiences with implementation and effectiveness of BIP, and (c) describing teachers’ qualitative experiences with prominent BIP barriers.
When asked to reflect on their overall experience with barriers to BIPs, teachers reported encountering barriers Occasionally (58%) and that these barriers were usually Minor (78%); however, questions on the frequency and impact of specific barriers revealed four barriers that prominently affected some teachers’ implementation of BIPs: The cause of the student’s problem behavior cannot be addressed through a BIP, Inconsistent implementation of BIP across staff, Inadequate resources to implement the BIP, followed by Even when implemented correctly, the BIP is not effective. These findings corroborate the prominence of several individual barriers identified by Bambara et al. (2009, 2012) and add detail regarding how and for whom these barriers may manifest. First, Bambara et al. found a major barrier to BIP implementation to be beliefs about behavior that contradict a PBS approach. The belief that the cause of student behavior cannot be addressed through a BIP can be considered a belief that contradicts a PBS approach (Andreou et al., 2015) and was reported in the present study as the greatest barrier to BIP implementation. A foundational belief to PBS is that behavior is affected by the immediate environment, including antecedents and consequences to behavior, and that by changing the environment one can change behavior. Many teachers in our survey rejected this notion, at least in relation to some students, qualitatively identifying problematic home environments and internal student factors as causing problem behavior in school. Because these factors could not be directly manipulated by teachers, they reported that they were limited in the degree to which they could improve these students’ behavior. Previous research has found this belief to be fairly common among teachers (Wang et al., 2015) and linked to the notion that a proactive approach to behavior management rewards difficult students for misbehavior (Pinkelman et al., 2015). As a result, attributing the cause of student behavior to unmalleable factors may represent a specific, prominent belief pattern held by some teachers that can interfere with implementation of BIPs.
In addition concerning is the finding that teachers from urban schools and schools with higher percentages of minority students, but not higher percentages of FRL, reported encountering that the cause of student problem behavior could not be addressed through a BIP significantly more frequently. One potential explanation for this finding is that implicit racial bias may have affected teacher judgments about the cause of student misbehavior. Implicit bias refers to attitudes or stereotypes that affect understanding, actions, and decisions without an individual’s awareness or intentional control (Staats et al., 2017). All people possess implicit biases, which may be favorable or unfavorable toward different constructs. Implicit racial bias has been found to affect teachers’ expectations and responses to challenging student behavior, in that some teachers may be more likely to expect (Gilliam et al., 2016) and respond punitively to (Okonofua & Eberhardt, 2015) challenging behavior from Black children. In the context of our survey, teachers reported encountering causes of challenging student behavior that were beyond their control (e.g., home environment, mental health needs) significantly more frequently when the teachers were from predominantly minority schools. The fact that teachers from schools with higher percentages of students receiving FRL did not reach significance in this barrier further supports the implication that race may have impacted teacher judgments about student misbehavior in ways that are different from socioeconomic status alone (McIntosh et al., 2014).
Our second and third most prominent barriers, inconsistent implementation of BIP across staff and inadequate resources to implement BIP, were both described in qualitative comments as related to insufficient time for staff to collaborate: a barrier also identified by Bambara et al. (2009, 2012). Specifically, our findings indicate that a primary challenge in BIP implementation is getting school staff to implement BIPs consistently across staff members, and that resources dedicated to checking and ensuring fidelity across staff are critical for adequate implementation, yet frequently lacking. Resource-based barriers have been found by Bambara et al. and others (McIntosh et al., 2016) to be influenced by lack of administrative support for PBS implementation. In the present study, teachers from urban schools, schools with more students receiving FRL, and schools with more minority students were significantly more likely to report inconsistent implementation across staff and inadequate resources to implement BIP as frequent barriers to BIP implementation. In addition, these teachers were significantly more likely to report BIPs as implemented with lower fidelity and as less effective in changing student behavior. These differences may relate to the many challenges faced by urban schools with high-need student populations, such as lack of resources (Kozleski et al., 2014), higher teacher and administrator turnover (Jacob, 2007), and higher ratios of students with special needs per qualified special education teacher (Murray, 2004). In contrast, teachers from suburban and private school settings perceived BIPs as significantly more effective, highlighting the importance of school-level variables on classroom-level BIP implementation (Pas et al., 2015).
Limitations
The results of this study should be interpreted with caution due to its limitations. First, of the 4,244 teachers who received the survey, 16.4% began the survey and 14.2% met the inclusion criteria needed to complete the full survey. This percentage represents a low response rate and could indicate a sample that is different from the target population, requiring extreme caution when generalizing results. The low response rate may be due in part to the web-based nature of the survey, as web-based surveys tend to be associated with lower response rates (Lozar Manfreda et al., 2008). Second, teachers in the survey were all residents of one state who worked with students with significant disabilities. Compared to samples in other studies (e.g., Bambara et al., 2012), our sample may be more representative of teachers of students with more significant disabilities and behavioral difficulties. Third, the self-report methodology used in this study is inherently subjective and characterized by limitations, such as social desirability bias and reliance on participant recollections. Fourth, although the survey used in this study was developed based on previous literature on BIP implementation, verified via expert opinions, and possessed adequate internal consistency, it was designed specifically for this study. Therefore, its findings should be considered exploratory and require replication.
Implications for Research
Although exploratory, the present study may have implications for directing intervention research aimed at improving implementation of BIPs (e.g., Mouzakitis et al., 2015; Pinkelman & Horner, 2017; Sanetti et al., 2014) toward barriers teachers are most likely to encounter in practice. In particular, the present study found inconsistent implementation across staff, inadequate time and resources, and student behavior problems perceived as driven by out-of-school factors as most prominent barriers to BIP implementation, and that these barriers were reported to occur more frequently by teachers in lower resourced schools. Based on these findings, interventions targeting teachers’ communication across staff, time- and resource-management, and perceptions and skill in addressing more complex behavior problems may produce greater improvements in BIP implementation and may be particularly needed in under-resourced schools. However, these findings need to be replicated and extended to other samples through further descriptive research aimed at identifying prominent barriers to BIP. Survey or focus group methods in which teachers generate their own lists of barriers to BIP implementation, as opposed to the present study in which they selected from a previously developed list, could help identify authentic, prominent barriers to target for intervention.
Implications for Practice
Effective implementation of BIPs is challenging in many settings, but may be especially difficult in lower-resourced schools with students perceived by teachers as having more out-of-school factors affecting their in-school behavior. Many of the causes of student problem behavior that teachers reported as unable to address in a BIP could be considered distal setting events (Friman & Hawkins, 2006), such as not receiving medication that morning. Teachers may feel less equipped to respond to behavior related to setting events than behavior triggered by immediate antecedents (Conroy & Fox, 1994); however, there are research-informed strategies for mitigating the effects of distal setting events on classroom behavior that can be incorporated into a BIP (Robertson & Coy, 2019). Understanding how to address the effects of distal setting events within a BIP may increase teachers’ perceived utility and implementation of BIP.
As indicated by our second and third greatest barriers, BIP implementation is heavily impacted by resources devoted to checking fidelity of implementation across school staff. For this reason, administrators should protect school staff time for BIP fidelity checks, without which BIP may not be implemented consistently enough to improve student behavior. When such time is not available, teachers may need to be particularly resourceful by taking time to plan for BIP implementation (Sanetti et al., 2014), developing simple but effective BIP (Dunlap et al., 2010), and creating efficient methods of communication, progress monitoring, and fidelity checks across staff to improve BIP implementation.
Supplemental Material
Supplemental_File_Table_2 – Supplemental material for Barriers to Implementing Behavior Intervention Plans: Results of a Statewide Survey
Supplemental material, Supplemental_File_Table_2 for Barriers to Implementing Behavior Intervention Plans: Results of a Statewide Survey by Rachel E. Robertson, Anastasia A. Kokina and Debra W. Moore in Journal of Positive Behavior Interventions
Footnotes
Acknowledgements
The authors would like to thank Drs. Kaylee Wynkoop, Emily Sobeck, Rachel Schwartz, and Benjamin Thomas for their assistance with this project.
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.
Supplemental Material
Supplemental material for this article is available online.
References
Supplementary Material
Please find the following supplemental material available below.
For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.
For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.
