Factor Validation of the Staff Perceptions of Behavior and Discipline (SPBD) Survey

Abstract

The implementation of schoolwide positive behavior interventions and supports (SWPBIS) requires a shift from a reliance on reactionary discipline to prevention-oriented supports, and it requires that staff adopt and regularly apply the practices throughout the building. This level of systemic change in staff thinking and practice is challenging to achieve and sustain, but data can assist teams in navigating the process. The Staff Perceptions of Behavior and Discipline (SPBD) is a needs assessment survey developed to measure staff beliefs, needs, and concerns for discipline and behavioral support practices. This tool can help teams make data-informed decisions, and target professional development to fit the needs and concerns of staff in their building. This study reexamines the internal consistency and factor structure of the SPBD using an expanded data set of 147 schools, including elementary, middle, and high schools across several geographic regions in the United States. The results corroborate the SPBD’s existing factor structure over alternative factor structures and support the internal consistency of each subscale. Implications for practice and future research are provided.

Keywords

schoolwide positive behavior supports PBIS school discipline system change needs assessment teacher perceptions

Implementing schoolwide positive behavior interventions and supports (SWPBIS) can yield meaningful outcomes for schools, including decreased student behavioral violations leading to exclusionary discipline, and increased student academic outcomes, organizational health, and teacher well-being (e.g., Bradshaw, Koth, Thornton, & Leaf, 2009; Bradshaw, Reinke, Brown, Bevans, & Leaf, 2008; Lassen, Steele, & Sailor, 2006; Ross, Romer, & Horner, 2012). However, these outcomes are not always attained as schools can struggle to achieve full and sustained implementation (McIntosh, Mercer, Nese, Strickland-Cohen, & Hoselton, 2016). Implementation of SWPBIS typically requires that staff reach shared agreement and commit to the key SWPBIS elements, such as a common set of behavior expectations, methods for acknowledging students meeting expectations, and procedures for addressing behavioral violations when they occur (Sugai & Horner, 2006). Yet within any school, staff bring a wide range of knowledge, skills, beliefs, and perspectives, often making it difficult for leadership teams to achieve building-wide agreements, and ultimately, reach a shared vision for implementation (Valenti & Kerr, 2014).

Over the past decade, research has underscored the impact that staff perceptions can have on SWPBIS implementation. SWPBIS team members, coaches, and coordinators have all reported prominent implementation barriers associated with staff perceptions, including staff who disagreed with the central tenets of SWPBIS and staff who perceived SWPBIS to be unnecessary, ineffective, and insufficiently resourced (e.g., Kincaid, Childs, Blase, & Wallace, 2007; Lohrmann, Forman, Martin, & Palmieri, 2008; Lohrmann, Martin, & Patil, 2013). Given this importance, it is beneficial for leadership teams to understand the perceptions and needs of staff in their building, and then use this understanding to guide action planning and achieve greater fidelity of implementation.

SWPBIS is not a manualized or packaged program; rather, it is a framework with a set of guiding practices to be adjusted to fit the unique needs and resources of each building. As such, it is considered best practice in implementation science to achieve a contextual fit between SWPBIS and its application (Sugai & Horner, 2006). This contextual fit may be achieved through team-led actions such as professional development that is targeted to meet staff where they are. Thus, one of the first steps to achieving contextual fit between SWPBIS and implementers is a sound mechanism for collecting perceptual data from staff in the building. These data can lead to an understanding of staff needs and perspectives, and help teams achieve contextual fit through data-informed practices such as differentiating professional development, and proactively addressing staff priorities and concerns (Bohanon & Wu, 2014). The end goal in gathering staff perceptual data is to meet implementers’ needs, so that they implement with consistency and longevity, and thereby better support all students.

Leaders in the fields of organizational and systems change have long attested to the value of needs assessments in the installation of new practices (e.g., Curtis, Castillo, & Cohen, 2008; Hall & Hord, 2011; Nagle & Gagnon, 2008). Needs assessments are conducted to understand the needs, concerns, and perspectives of key stakeholders, particularly those who will be charged with implementing new practices. While there are tools available to measure staff commitment and perceived implementation of SWPBIS (see Filter, Sytsma, & McIntosh, 2016; Sugai, Horner, & Todd, 2003), we are aware of no other tools to assess staff needs and perceptions of behavior and discipline in the context of SWPBIS. The Staff Perceptions of Behavior and Discipline (SPBD) survey was developed in response to this gap in research and practice, and was developed to be consistent with best practices in SWPBIS implementation science (e.g., involvement of stakeholders, the use of data to inform decisions, focus on proactive action). The primary purposes of the SPBD are threefold: (a) involve staff and create greater ownership in the implementation process; (b) gather data from staff to better understand their perspectives, needs, and concerns; and (c) leverage this understanding to develop a data-informed implementation plan that supports them and prevents problems in implementation such as resistance and/or the eventual abandonment of the practices.

Research and Development

The SPBD was created to assess staff-related factors that may act as barriers and facilitators to implementation. These factors are based on themes found in both the SWPBIS and systems change literature, including staff perceptions of the effectiveness of and need for the innovation/SWPBIS; commitment and effectiveness of leadership; availability of resources such as time and training; climate, including staff cohesion, trust, and shared vision; and philosophical congruence with the innovation (e.g., Adelman & Taylor, 2007; Curtis et al., 2008; Hall & Hord, 2011; Kincaid et al., 2007; Lohrmann et al., 2008; Rogers, 2003; Sugai & Horner, 2006).

The SPBD includes 23 Likert-type scale items that assess these themes. Most items contain four response options: strongly disagree, disagree, agree, and strongly agree. A 4-point scale was selected for most items to reduce ambivalent response biases, a problem associated with the measurement of perceptions (e.g., Nowlis, Kahn, & Dhar, 2002). Six items contain five response options, however, allowing for a response that is akin to an “I do not know” response—to account for circumstances such as new staff in the building, for example, I do not know my colleagues well enough to answer this question.

In addition to the 23 core items of the SPBD, other items provide supplementary information relevant to SWPBIS and useful to leadership teams. These items ask respondents to identify their job role, report hours of professional development received, assess their knowledge of SWPBIS, report their level of commitment to SWPBIS, and rate the quality of communication in their building. Last, three open-ended items qualitatively assess staff concerns and perceived needs and existing capacities. For greater detail on the development of the SPBD, the readers are referred to Feuerborn, Tire, and King (2015).

Use of the SPBD Survey

Schools commonly administer the SPBD once a year, in the autumn or spring, to inform planning and implementation decisions, and to assess changes in staff perceptions, needs, and concerns as implementation evolves. The SPBD is available online and free of charge at http://spbdsupport.com. To begin the process, a school leader completes an online survey request and selects a time frame for survey data collection. After the SPBD request is submitted, leaders receive an email that contains their school’s SPBD survey link, and they then disseminate this link to all certified and classified staff working with students in the building. Typically, staff are informed of the purpose and intent of the SPBD during a staff meeting. The SPBD requires about 15 min to complete, although the precise duration is dependent on the amount of time a user invests in completing the open-ended items. Identifying information is not collected, so that staff may feel comfortable offering authentic responses.

After staff have completed the survey and the survey has closed, the survey requestor automatically receives their school’s SPBD report. The report is generated via a computer software process developed specifically for the SBPD. The report includes quantitative data in the form of numbers, charts, and tables and qualitative data from the open-ended items. The report also identifies potential barriers and facilitators to SWPBIS implementation, and offers teams recommendations and considerations for action planning. For example, if the majority of staff agree to an item in the philosophy domain, “When problem behaviors occur, we need to get tougher,” the report offers recommendations for the team in further investigating the extent to which staff in the building have a reliance on punitive responses to student behavior issues. Also, teams are encouraged to inspect their qualitative data for contextual and diagnostic information frequently found in the staff responses to the open-ended items. The SPBD report also disaggregates the quantitative perceptual data by certified and classified staff, allowing teams to target professional development and coaching support by staff role as appropriate. Sample reports are available on the SPBD website indicated above.

To date, SPBD research includes qualitative, quantitative, and mixed-methods studies. In qualitative and mixed-methods research, the SPBD has been utilized to explore the perceptions of secondary teachers (Feuerborn, Wallace, & Tire, 2016), the concerns of staff in opposition to SWPBIS (Tire & Feuerborn, 2017), and the perspectives and concerns of classified staff (Feuerborn, Tire, & Beaudoin, 2017). Quantitative research has provided preliminary support for the SPBD’s potential to identify the unique perspectives of staff in planning and implementing schools (Feuerborn & Tire, 2016). As mentioned previously, the psychometric properties of the SPBD were examined in Feuerborn et al. (2015)—providing evidence for reliability (Cronbach’s alpha = .80) and concurrent validity—through the study of significant relationships between the SPBD and school and staff factors relevant to SWPBIS such as implementation level, school level, and staff knowledge and training. This research identified a five-factor structure, as listed here along with coefficient alphas per factor (Feuerborn et al., 2015):

Factor I: Teaching and acknowledging expectations: Effectiveness and need (.67)

Factor II: Systems: Resources, supports, and climate (.73)

Factor III: Implementation integrity (.79)

Factor IV: Philosophical views of behavior and discipline (.62)

Factor V: Systems: Cohesiveness and openness to change (.58)

Purpose

Although the findings of Feuerborn et al. (2015) offered support for the psychometric properties of the SPBD, the previous analyses were conducted with data that were relatively limited in terms of location and school level. The previous analyses were conducted with 1,210 responses from 36 schools in nine school districts limited primarily to Washington State. These data included only eight middle schools and three high schools. As these initial analyses were conducted, a larger, more diverse data set in terms of locale, geographic region, and school level has been accrued. Hence, the purpose of this study was to examine the extent to which the original factor structure of the SPBD found in Feuerborn et al. (2015) was confirmed with a broader sample. The research questions were as follows:

Research Question 1: Is the existing SPBD factor structure confirmed with a broader sample, or is an alternative factor structure more psychometrically sound?

Research Question 2: Also, what is the internal consistency of each factor of the SPBD?

Method

Participating Schools

Schools included in these analyses requested and used the SPBD survey through the open access, free online system described previously. After school leaders learned about the SPBD via sources such as conference presentations, published research, and the recommendations of colleagues, they elected to use it in their schools. As part of the online SPBD survey request process, schools consented to the use of their data for research purposes, per the approved process of the institutional review board for human subjects research. They also reported their level of SWPBIS implementation at this time. Participating schools included 147 schools reporting to be either implementing or preparing to implement SWPBIS. They included 78 elementary schools, 36 middle schools, and 18 high schools. The remaining 15 schools were preschool or other combined school levels (e.g., K–8). By location, 22 schools were in rural areas, 24 in towns, 65 in suburbs, and 30 in cities, with six school locales unspecified—as classified by the National Center for Education Statistics (https://nces.ed.gov/). The schools enrolled an average of 586 students with an average of 64.12% receiving free or reduced-price lunch. Table 1 includes more demographic information for the participating schools.

Table 1.

Demographic Information for Participating Schools (n = 147) and Staff Response Rates for the SPBD Survey (n = 5,349).

Demographics	Elementary	Middle	High	Other^a
Demographics	n (%)	n (%)	n (%)	n (%)
School level	78 (53.06)	36 (24.48)	18 (12.24)	15 (10.20)
	M	M	M	M
Response rate
%Classified staff	41.77	36.58	34.22	38.40
%Certified staff	76.38	69.38	68.33	78.40
	n (%)	n (%)	n (%)	n (%)
Implementation level
Planning	15 (19.23)	9 (25.00)	5 (27.77)	5 (33.33)
Partial	37 (47.43)	16 (44.44)	10 (55.55)	9 (60.00)
Full	25 (32.05)	11 (30.55)	3 (16.66)	1 (6.66)
Unknown	1 (1.28)	0	0	0
Region
Northeast	1 (1.28)	1 (2.77)	0	2 (13.33)
Midwest	2 (2.56)	0	0	0
South	13 (16.66)	5 (13.88)	4 (22.22)	3 (20.00)
West	62 (79.48)	30 (83.33)	14 (77.77)	10 (66.66)
Locale
Rural	12 (15.38)	2 (5.55)	5 (27.77)	3 (20.00)
Town	10 (12.82)	8 (22.22)	2 (11.11)	4 (26.66)
Suburb	36 (46.15)	19 (52.77)	8 (44.44)	2 (13.33)
City	18 (23.07)	7 (19.44)	3 (16.66)	2 (13.33)
Unknown	2 (2.56)	0	0	4
	M	M	M	M
Students enrolled	492.30	698.77	931.22	263.80
%F/RPL	64.32	61.63	59.72	83.25

Note. SPBD = Staff Perceptions of Behavior and Discipline; F/RPL = free or reduced-price lunch.

Other category includes preschools and combined grade-level schools.

SPBD Survey Participants

A total of 5,362 SPBD survey responses were included in these analyses. Response rates were calculated by dividing the number of staff employed in the school by the number of staff respondents per school. The average response rate per school was 60.9%, with a range of 21% to 100% across the participating schools. The average response rate for certified staff was 73.1%, and the average response rate for classified staff was 37.7% (see Table 1 for average response rates by school level). Of the total responses, 53.06% were received from elementary schools, 24.48% from middle schools, and 12.24% from high schools. The remaining 10.2% were received from staff in schools with other school levels (e.g., K–8, K–12). Of the staff responding to the SPBD survey, 65.8% were certified teachers, 18.1% were classified staff, and 14.5% were in certified support roles (e.g., school psychologist, counselor). The remaining 1.6% reported their role as “other” or “administrator.” Respondents had a wide range of experiences working in their school building, with 13.6% having 0 to 1, 16.1% having 2 to 3, 13.8% having 4 to 6, 14% having 7 to 10, and 42.4% having 10 or more years. Respondents were asked to assess their level of understanding of SWPBIS as part of the survey: 2.3% reported unfamiliar, 17.9% reported limited, 56.3% reported basic, and 23.2% reported high.

Procedures

From fall 2015 through fall 2016, SPBD survey data were collected via the online system. As described previously, a school leader completed the online SPBD survey request form, received their school’s SPBD survey link, and then distributed the link to all school staff regularly interacting with students, including certified and classified.

Analysis

Following screening, the data were analyzed with respect to factor structure and internal consistency. Exploratory factor analysis (EFA) is suitable when few, or no, hypotheses exist regarding the instrument’s internal structure. Conversely, confirmatory factor analysis (CFA) is more appropriate for examination of existing and distinct hypotheses about an instrument’s dimensionality (Furr, 2011). Therefore, to test the five-factor structure proposed by Feuerborn et al. (2015), the 23 measured SPBD core items were evaluated with CFA using MPlus 7 software (L. K. Muthén & Muthén, 1998–2012). Our data were obtained from participants nested in 147 schools, and consequently, the clustered observations may not have been independent. MPlus software amalgamates CFA and multilevel modeling, which allowed us to adjust for the standard errors and fit indices for school-based clustering. Standardized factor loadings were estimated using robust weighted least squares (WLSMV), while allowing for covariance between five latent factors. WLSMV method of B. Muthén, du Toit, and Spisic (1997) demonstrated high levels of accuracy in simulation studies involving ordinal data of varying degrees of complexity, nonnormality, and sample sizes (Flora, Curran, & West Stephen, 2004). Hutchinson and Olmos (1998) suggested the use of root mean square error of approximation (RMSEA) as one measure of fit for models involving nonnormal, Likert-type data. As such, we used RMSEA, along with the stringent criterion of .07, as recommended by Steiger (2007). We also included chi-square (χ²), comparative fit index (CFI), and Tucker–Lewis index (TLI). Commonly, chi-square test with p > .05, CFI > .90, and TLI > .90 indicates a satisfactory fit (Awang, 2012). However, chi-square test is sensitive to sample size, and nearly always rejects the model when the large samples are used (Bentler & Bonett, 1980). Furthermore, Bentler and Bonett (1980), as well as Rigdon (1996), asserted that incremental fit indices (such as CFI and TLI) are appropriate only in exploratory applications, and χ² statistic for the null model is unlikely to follow a χ² distribution unless the sample size is very small. For these reasons, we decided to evaluate model fit predominately on RMSEA index. Although the collected data consisted of only 0.48% of missing responses, due to the different possible interpretations of “don’t know” responses in the six items, these were coded as missing values. The WLSMV estimator handled the combined missing responses (3.96% of all responses) with pairwise present analysis (L. K. Muthén & Muthén, 1998–2012). Last, to further examine SPBD’s internal consistency, we used SPSS 19 to compute Cronbach’s alpha coefficients for each factor.

Results

CFA resulted in the following model fit information: χ²(220) = 4,357.24, p < .001; RMSEA = .059, 90% confidence interval (CI) = [0.058, 0.061]; CFA = .85; and TLI = .83. Chi-square, CFA, and TLI indices did not meet the recommended cutoff values. However, RMSEA indicated that the hypothesized model provided an acceptable fit to the observed data. As such, we decided to maintain our criteria for interpretation of fit indices, as explained in the Analysis section, and rely on RMSEA index as the primary evidence of model fit. All indicators showed significant positive factor loadings, with standardized coefficients ranging from .44 to .86. There were also significant positive correlations among all five factors, ranging from .26 to .75. Coefficient alphas were in the range of .68 to .73, each comparable to or exceeding the coefficient alphas obtained from prior research.

Post Hoc Analysis

After examination of modification indices, we reassigned the item “Behavior plans do not work well in our school” from Factor V “systems: cohesiveness and openness to change” to Factor I “teaching and acknowledging expectations: effectiveness and need.” The new five-factor model was reanalyzed, and showed to be a better fit to the original model: χ²(220) = 3,500.33, p < .001; RMSEA = .053, 90% CI = [0.051, 0.054]; CFA = .88; and TLI = .87. All factor loadings were significant and positive, with standardized coefficients ranging from .53 to .81, providing support for the instrument’s convergent validity. The detailed breakdown of SPBD items, factor loadings, and descriptive statistics is provided in Table 2. Similar to the original analysis, we found significant positive correlations among all five latent factors (see Table 3).

Table 2.

Items, Descriptive Statistics, and Standardized Coefficients of the SPBD Scale (n = 5,362).

Item	n		M	SD	Factor loadings
Item	Valid	DK	M	SD	Factor	β
Schoolwide behavior support may work in other schools, but I doubt it will work in ours.	5,331	—	3.29	0.61	1	.79
I don’t have time to teach the schoolwide behavioral expectations.	5,329	—	3.14	0.65	1	.65
I resent being asked to do one more thing in my classroom.	5,322	—	3.10	0.67	1	.56
We should not have to teach students how to behave at school.	5,355	—	3.27	0.75	1	.55
I feel that rewarding students is the same as bribing them.	5,341	—	3.05	0.64	1	.57
Behavior plans do not work well in our school.	4,812	528	2.89	0.75	1	.70
The climate at this school is positive.	5,349	—	2.83	0.78	2	.69
I have trust in my administrator’s ability to lead us through change.	5,351	—	3.03	0.90	2	.55
I believe our school has (or will have) the necessary resources to support schoolwide positive behavior support.	5,349	—	2.82	0.76	2	.70
Overall, I am satisfied with my job.	5,342	—	3.15	0.70	2	.64
Schoolwide behavior support is likely to be yet another fad that comes and goes in this school.	5,339	—	2.86	0.79	2	.73
Currently, I teach the agreed upon schoolwide behavior expectations to students.	4,985	330	3.26	0.59	3	.81
Currently, I acknowledge/reward students for meeting the agreed upon schoolwide behavior expectations.	5,041	282	3.15	0.59	3	.78
Currently, I apply the agreed upon schoolwide disciplinary consequences	4,614	699	3.09	0.55	3	.73
The students at this school need to be held more responsible for their own behavior.	5,325	—	1.78	0.69	4	.57
When problem behaviors occur, we need to get tougher.	5,336	—	2.39	0.74	4	.63
I believe we should reserve rewards for students exceeding expectations, not simply for meeting them.	5,335	—	2.51	0.84	4	.55
Parents in the community don’t seem to care about how their children behave at school.	5,339	—	2.68	0.73	4	.63
If students aren’t disciplined at home, they aren’t likely to accept any discipline at school.	5,332	—	2.57	0.83	4	.63
I suspect that my colleagues will not (or are not) consistently implementing the agreed upon schoolwide behavior plan.	5,336	—	2.55	0.76	5	.70
The staff at this school tends to resist change with concerns such as “We don’t do it that way here.”	5,335	—	2.77	0.71	5	.54
My colleagues and I share a common philosophy for behavior and discipline.	4,522	829	2.78	0.70	5	.66
This school has successfully implemented change efforts in the past.	3,725	1,615	2.96	0.64	5	.70

Note. SPBD = Staff Perceptions of Behavior and Discipline; DK = number of “don’t know” responses that were treated as missing.

Table 3.

Factor Intercorrelations of the SPBD.

Factor Number	Factor I	Factor II	Factor III	Factor IV	Factor V
Factor I	—	—	—	—	—
Factor II	.71	—	—	—	—
Factor III	.60	.39	—	—	—
Factor IV	.76	.51	.26	—	—
Factor V	.41	.59	.30	.29	—

Note. SPBD = Staff Perceptions of Behavior and Discipline; Factor I = Teaching and acknowledging expectations: Effectiveness and need; Factor II = Systems: Resources, supports, and climate; Factor III = Implementation integrity; Factor IV = Philosophical views of behavior and discipline; Factor V = Systems: Cohesiveness and openness to change.

Alpha coefficients showed evidence of internal consistency for each SPBD dimension. Cronbach’s alphas above .6 can be considered acceptable for scales that are experimental and not used for high-stakes decisions (Loewenthal, 2004), and Nunnally and Bernstein (1994) indicated that a minimum Cronbach’s alpha of .7 is acceptable for similar purposes. All five scales herein exceeded .6, and the majority of factors exceeded .7—with two approaching the latter threshold. The five factors along with their Cronbach’s alpha were distributed as follows: Table 4 offers more thorough coverage of the psychometric properties of SPBD factors:

Factor I: Teaching and acknowledging expectations: Effectiveness and need (.72)

Factor II; Systems: Resources, supports, and climate (.73)

Factor III: Implementation integrity (.73)

Factor IV: Philosophical views of behavior and discipline (.68)

Factor V: Systems: Cohesiveness and openness to change (.66)

Table 4.

Psychometric Properties of the Factors Composing the SPBD Scale (n = 5,362).

Factor	No. of items	M	SD	Cronbach’s α
Teaching and acknowledging expectations: Effectiveness and need	6	3.13	0.68	.72
Systems: Resources, supports, and climate	5	2.94	0.79	.73
Implementation integrity	3	3.17	0.58	.73
Philosophical views of behavior and discipline	5	2.38	0.77	.68
Systems: Cohesiveness and openness to change	4	2.79	0.71	.66

Note. SPBD = Staff Perceptions of Behavior and Discipline.

Discussion

The purposes of this study were to determine the extent to which the existing SPBD factor structure was replicated in a broader sample (as opposed to an alternative factor structure) and to determine the internal consistency of each factor. The current results statistically confirmed the internal consistency and overall factor structure in a manner consistent with the SPBD’s development. As compared with the sample used in previous research (i.e., Feuerborn et al., 2015), the current sample included data from more diverse geographical regions, more secondary schools (37 cf. 11), schools with a higher proportion of students receiving free and reduced-price lunch (67% cf. 56%), and contained more schools overall (147 cf. 36). Despite the differences in the two samples, the current findings are consistent with the findings of previous research.

In light of the present findings, we refined the factor structure of the SPBD by moving one item from Factor V to Factor I, resulting in an improvement in both fit and factor loadings. Besides a better statistical fit, the new structure provided a better theoretical fit because the degree to which staff believe behavior plans work well in their school is more conceptually relevant to Factor I (i.e., perceptions of the effectiveness of, and need for teaching and acknowledging expectations) than Factor V (i.e., perceptions of staff cohesion and openness to change).

As compared with the previous findings (Feuerborn et al., 2015), the current findings not only supported the hypothesized structure of the instrument but also revealed similar to more robust internal consistency and convergent validity. Respectively, structure and internal consistency were supported by the acceptable fit index along with equivalent or higher Cronbach’s alphas. Convergent validity was indicated by significant factor loadings; that is, each item significantly loaded onto the hypothesized factor, implying its theoretical connection to the underlying construct the item was intended to measure.

Limitations and Implications for Future Research and Practice

There are several limitations to the current study that warrant additional exploration and research: First, the response rates in the current study varied across schools (21%–100%), and may not represent the views of all staff in the buildings. Although the exact cause for the variation in response rates is unknown, it could be due to the conditions under which the survey link was presented to staff, differing norms of staff involvement and participation across buildings, and competing demands on staff time. Future researchers could use a sample with consistently high response rates by monitoring the survey administration process, and assuring that staff understand the intent of the SPBD and have time to complete it. Second, although the current sample includes more secondary schools than the previous sample, their representation is still proportionately low as compared with that of elementary schools. If staff responses to the SPBD are markedly different in secondary schools as compared with elementary schools, the results of factor analysis could be affected. Future research should include a sample with a higher proportion of middle and high schools. Third, less than ideal alpha coefficients are a current limitation that could be improved through future research and refinement of the SPBD. Last, self-selection may be a limitation of this study, in that schools electing to use the SPBD may not represent all schools planning to implement or implementing SWPBIS. Schools in the current sample may have placed greater value on the insights of their staff, and/or they may have experienced more staff-related challenges. Future researchers could actively recruit schools to participate in the SPBD.

These limitations notwithstanding, the results of this study further substantiate the psychometric properties of the SPBD, a tool designed to assist teams in understanding the perspectives, needs, and concerns of staff in their building. Equipped with this research-based tool, teams may be better prepared to make data-informed decisions that are inclusive and considerate of the needs of all stakeholders charged with implementation. The SPBD can help teams tailor professional development to fit the needs of staff, identify and build on existing capacities, and proactively address staff concerns and priorities. Although more research is needed to better elucidate the effective application of the SPBD, this survey tool holds promise to help teams establish greater contextual fit between the framework of SWPBIS and the ways in which they install it in their schools. With this contextual fit, schools may benefit from higher rates of staff involvement, ownership, and implementation of SWPBIS, and thus achieve more meaningful outcomes for all students.

Footnotes

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

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

References

Adelman

H. S.

Taylor

(2007). Systemic change for school improvement. Journal of Educational and Psychological Consultation, 17, 55–77.

Awang

(2012). Structural equation modeling using AMOS graphic. Penerbit Universiti Teknologi MARA. UiTM Press.

Bentler

P. M.

Bonett

D. G.

(1980). Significance tests and goodness of fit in the analysis of covariance structures. Psychological Bulletin, 88, 588–606.

Bohanon

(2014). Developing buy-in for positive behavior support in secondary settings. Preventing School Failure, 58, 223–229.

Bradshaw

C. P.

Koth

C. W.

Thornton

L. A.

Leaf

P. J.

(2009). Altering school climate through school-wide positive behavioral interventions and supports: Findings from a randomized effectiveness trial. Prevention Science, 10, 100–115.

Bradshaw

C. P.

Reinke

Brown

Bevans

Leaf

(2008). Implementation of school-wide positive behavioral interventions and supports (PBIS) in elementary schools: Observations from a randomized trial. Education and Treatment of Children, 31, 1–26.

Curtis

M. J.

Castillo

J. M.

Cohen

R. M.

(2008). Best practices in systems-level change. In Thomas

Grimes

J. P.

(Eds.), Best practices in school psychology V (pp. 887–902). Bethesda, MD: National Association of School Psychologists.

Feuerborn

L. L.

Tyre

A. D.

(2016). How do staff perceive schoolwide positive behavior supports? Implications for teams in planning and implementing schools. Preventing School Failure, 60, 53–59.

Feuerborn

L. L.

Tyre

A. D.

Beaudoin

K. M.

(2017). Classified staff perceptions of behavior and discipline: Implications for schoolwide positive behavior supports. Journal of Positive Behavior Interventions. doi:1098300717733975

10.

Feuerborn

L. L.

Tyre

A. D.

King

J. P.

(2015). The Staff Perceptions of Behavior and Discipline survey: A tool to help achieve systemic change through schoolwide positive behavior support. Journal of Positive Behavior Interventions, 17, 116–126.

11.

Feuerborn

L. L.

Wallace

Tyre

A. D.

(2016). A qualitative analysis of middle and high school teacher perceptions of schoolwide positive behavior supports. Journal of Positive Behavior Interventions, 18, 219–229.

12.

Filter

K. J.

Sytsma

McIntosh

(2016). A brief measure of staff commitment to implement school-wide positive behavioral interventions and supports. Assessment for Effective Intervention, 42, 18–31.

13.

Flora

Curran

(2004). An empirical evaluation of alternative methods of estimation for confirmatory factor analysis with ordinal data. Psychological Methods, 9, 466–491.

14.

Furr

(2011). Scale construction and psychometrics for social and personality psychology. Thousand Oaks, CA: Sage.

15.

Hall

G. E.

Hord

S. M.

(2011). Implementing change: Patterns, principles, and potholes (3rd ed.). Boston, MA: Allyn & Bacon.

16.

Hutchinson

S. R.

Olmos

(1998). Behavior of descriptive fit indexes in confirmatory factor analysis using ordered categorical data. Structural Equation Modeling: A Multidisciplinary Journal, 5, 344–364.

17.

Kincaid

Childs

Blase

Wallace

(2007). Identifying barriers and facilitators in implementing school-wide positive behavior support. Journal of Positive Behavior Interventions, 9, 174–184.

18.

Lassen

Steele

Sailor

(2006). The relationship of school-wide positive behavior support to academic achievement in an urban middle school. Psychology in Schools, 43, 701–712.

19.

Loewenthal

K. M.

(2004). An introduction to psychological tests and scales (2nd ed.). Hove, UK: Psychology Press.

20.

Lohrmann

Forman

Martin

Palmieri

(2008). Understanding school personnel’s resistance to adopting school-wide positive behavior support at a universal level of intervention. Journal of Positive Behavior Interventions, 10, 256–269.

21.

Lohrmann

Martin

S. D.

Patil

(2013). External and internal coaches’ perspectives about overcoming barriers to universal interventions. Journal of Positive Behavior Interventions, 15, 26–38.

22.

McIntosh

Mercer

S. H.

Nese

R. N.

Strickland-Cohen

M. K.

Hoselton

(2016). Predictors of sustained implementation of school-wide positive behavioral interventions and supports. Journal of Positive Behavior Interventions, 18, 209–218.

23.

Muthén

du Toit

S. H. C.

Spisic

(1997). Robust inference using weighted least squares and quadratic estimating equations in latent variable modeling with categorical and continuous outcomes. Psychometrika, 45(1), 40–45.

24.

Muthén

L. K.

Muthén

B. O.

(1998–2012). Mplus user’s guide (7th ed.). Los Angeles, CA: Author.

25.

Nagle

Gagnon

(2008). Best practices in planning and conducting needs assessments. In Thomas

Grimes

(Eds.), Best practices in school psychology V (pp. 2207–2223). Bethesda, MD: National Association of School Psychologists.

26.

Nowlis

Kahn

Dhar

(2002). Coping with ambivalence: The effect of removing a neutral option on consumer attitude and preference judgments. Journal of Consumer Research, 29, 319–334.

27.

Nunnally

J. C.

Bernstein

I. H.

(1994). Psychometric theory. Journal of Psychoeducational Assessment, 17, 275–280.

28.

Rigdon

E. E.

(1996). CFI versus RMSEA: A comparison of two fit indexes for structural equation modeling. Structural Equation Modeling: A Multidisciplinary Journal, 3, 369–379.

29.

Rogers

E. M.

(2003). Diffusion of innovations (5th ed.). New York, NY: Free Press.

30.

Ross

S. W.

Romer

Horner

R. H.

(2012). Teacher well-being and the implementation of school-wide positive behavior interventions and supports. Journal of Positive Behavior Interventions, 14, 118–128.

31.

Steiger

J. H.

(2007). Understanding the limitations of global fit assessment in structural equation modeling. Personality and Individual Differences, 42, 893–898.

32.

Sugai

Horner

R. H.

(2006). A promising approach for expanding and sustaining school-wide positive behavior supports: Integration of multi-tiered system approaches. Exceptionality, 17, 223–237.

33.

Sugai

Horner

R. H.

Todd

A. W.

(2003). EBS Self-Assessment Survey (2.0). Eugene, OR: Educational and Community Supports.

34.

Tyre

A. D.

Feuerborn

L. L.

(2017). The minority report: The concerns of staff opposed to schoolwide positive behavior interventions and supports in their schools. Journal of Educational and Psychological Consultation, 27, 145–172.

35.

Valenti

Kerr

(2014). Addressing individual perspectives in the development of schoolwide rules: A data-informed process. Journal of Positive Behavior Interventions, 17, 1–9.