Journal of School Psychology 53 (2015) 137148

Contents lists available at ScienceDirect

Journal of School Psychology

journal homepage: www.elsevier.com/locate/jschpsyc

Proles of classroom behavior in high schools: Associations with

teacher behavior management strategies and
classroom composition
Elise T. Pas a,, Anne H. Cash b, Lindsey O'Brennan a, Katrina J. Debnam a, Catherine P. Bradshaw c
a
b
c

Johns Hopkins Bloomberg School of Public Health, Center for the Prevention of Youth Violence, United States
Johns Hopkins University, School of Education, United States
University of Virginia, Curry School of Education, United States

a r t i c l e

i n f o

Article history:
Received 14 February 2014
Received in revised form 22 December 2014
Accepted 23 December 2014
Available online xxxx
Keywords:
Student behavior
Classroom management
High schools
Observation data
Latent prole analysis

a b s t r a c t
Although there has been considerable attention to the issue of classroom management and processes in educational reform models, there has been relatively limited research on these factors
in high schools. The current study utilized observational data from 1262 classrooms in 52 high
schools to examine teacher classroom management strategies and ratings of student compliance,
engagement, and social disruption. Latent prole analysis (LPA) was conducted to examine specific patterns of classroom-wide student behavior in relation to teachers' use of classroom management strategies and classroom composition. The LPA revealed three distinct classroom behavioral
proles where students consistently met behavioral expectations (71%), inconsistently met expectations (23%), and were noncompliant (6%). Analyses indicated a functional association between patterns of student behavior and teachers' classroom management. In classrooms where
students consistently met expectations, teachers provided more opportunities to respond and
less disapproval and reactive behavioral management. Classrooms with noncompliant students
had teachers who used the most disapproval and reactive behavior management. In addition,
classrooms characterized as consistent had fewer males and more White students than classrooms characterized by inconsistent and noncompliant behaviors. These ndings highlight the
link between student patterns of behavior and teacher classroom management and have important implications for screening and professional development.
2014 Society for the Study of School Psychology. Published by Elsevier Ltd. All rights reserved.

1. Introduction
Disruptive student behavior fosters a classroom environment that is not conducive to learning, limits time for instruction, and
contributes to negative peer interactions in the classroom, therefore creating a cause for concern in schools. Much research has
established that behavior problems such as classroom disruption often co-occur with poor academic functioning and low school
connectedness (Bradshaw, Buckley, & Ialongo, 2008; Konishi, Hymel, Zumbo, & Li, 2010; Malecki & Elliott, 2002). However, disruption
can also be problematic from an ecological perspective. A classroom environment characterized by a high rate of negative behaviors
presents the risk that such disruption becomes normative, leading otherwise nondisruptive students to also engage in these behaviors
(Barth, Dunlap, Dane, Lochman, & Wells, 2004). Although literature establishes the association between disruptive behavior and individual student demographics such as gender and race/ethnicity (Kellam, Ling, Merisca, Brown, & Ialongo, 1998; Kewel Ramani,
Corresponding author at: Department of Mental Health Johns Hopkins University, Bloomberg School of Public Health Johns Hopkins Center for the Prevention of
Youth Violence 415 N. Washington Street, Ofce 507, Baltimore, MD 21231.
E-mail address: epas@jhu.edu (E.T. Pas).
Action Editor: Sterett Mercer

http://dx.doi.org/10.1016/j.jsp.2014.12.005
0022-4405/ 2014 Society for the Study of School Psychology. Published by Elsevier Ltd. All rights reserved.

138

E.T. Pas et al. / Journal of School Psychology 53 (2015) 137148

Gilbertson, Fox, & Provasnik, 2007; Pas, Bradshaw, Hershfeldt, & Leaf, 2010; Pas, Bradshaw, & Mitchell, 2011; Skiba, Michael, Nardo, &
Peterson, 2002), less is known about how classroom-level characteristics relate to these behaviors. This focus on the classroom
behavior prole as a whole is largely absent from extant research.
Research shows that teachers can promote a positive classroom environment through the use of specic classroom management
strategies (e.g., Epstein, Atkins, Cullinan, Kutash, & Weaver, 2008), as well as interventions to improve classroom management
(e.g., Ialongo, Poduska, Werthamer, & Kellam, 2001); although much of this research has focused on elementary school classrooms.
Therefore, the current study examined the association between specic proles of classroom behavior in high school settings in relation to teachers' use of classroom management strategies, teacher demographics, and classroom composition. This research will inform our understanding of different high school classroom environments, where students engage in positive versus negative
behaviors, and the classroom management strategies that characterize these settings. The characteristics of classrooms with differing
behavioral proles can be later targeted through professional development models.
1.1. Promoting positive and preventing negative behaviors in the classroom
The foundation for effective teaching is classroom behavior management, which maximizes time for academic instruction, student
engagement, and achievement, and instills proactive behavior management practices and clear expectations (Sugai & Horner, 2002).
Prior research on classroom management has identied critical components which are associated with enhanced conditions for student learning and good behavior (e.g., Bryk & Schneider, 2002; Greenberg et al., 2003). Specically, research regarding classroom
management demonstrates the importance of clear expectations, consistent responses to behavioral infractions, adequate opportunities for students to respond, checking for student understanding, use of effective praise for positive behaviors, utilizing group behavioral contingency methods, and a classroom layout that allows for active movement around the room (Armendariz & Umbreit, 1999;
Colvin, Flannery, Sugai, & Monegan, 2009; Evertson, 1985; Evertson & Emmer, 1982; Ialongo et al., 2001; Partin, Robertson, Maggin,
Oliver, & Wehby, 2010). Taken together, research generally emphasizes the importance of proactive, rather than reactive, behavioral
management as a means for promoting positive and preventing negative student behaviors within the classroom.
Intervention studies targeting teacher behavioral management as a means of improving student outcomes have provided further
evidence of the link between student behavior and teacher practice. One widely-disseminated model, Positive Behavioral Interventions and Supports (PBIS; Sugai & Horner, 2006), is a noncurricular prevention model that improves school systems and procedures
for promoting positive behavior and responding to behavioral infractions through the use of data-based decision-making. Randomized controlled trials examining PBIS have demonstrated school-level impacts on student ofce discipline referrals, suspensions, behavior problems, and school climate (Bradshaw, Koth, Bevans, Ialongo, & Leaf, 2008; Bradshaw, Koth, Thornton, & Leaf, 2009;
Bradshaw, Mitchell, & Leaf, 2010; Bradshaw, Waasdorp, & Leaf, 2012; Horner et al., 2009); however experimental impacts have
only been demonstrated at the elementary school level. Emerging non-experimental studies also provide support for PBIS at the secondary level (e.g., Flannery, Fenning, Kato, & McIntosh, 2014). Similarly, the Good Behavior Game (GBG; Barrish, Saunders, & Wolf,
1969) uses social learning principles of peer and teacher reinforcement within classrooms to improve student behavior, demonstrating evidence of decreasing disruptive behavior and improving academic performance in elementary classrooms (e.g., Bradshaw,
Zmuda, Kellam, & Ialongo, 2009; Ialongo et al., 1999; Kellam et al., 1998). Extant studies of behavior management focus on schoolwide outcomes rather than classroom-level indicators though (i.e., the focus of the current study).
1.2. Support for the use of specic classroom management strategies
Praise and opportunities to respond (OTRs) are two related and specic teacher strategies for fostering appropriate student behavior that are frequent targets of research and training in classroom management (Conroy, Sutherland, Snyder, Al-Hendawi, & Vo, 2009;
Partin et al., 2010) and were assessed in the current study. Teacher praise, when given purposively and when behavior-specic rather
than general, can increase on-task behavior (Sutherland, Wehby, & Copeland, 2000) as well as reduce ofce discipline referrals
(Peterson Nelson, Young, Young, & Cox, 2009). OTRs also can be useful for increasing students' appropriate academic and social behaviors (Partin et al., 2010) and decreasing disruptive behavior (Armendariz & Umbreit, 1999; Haydon, Mancil, & Van Loan, 2009).
There are limitations to the conclusions drawn regarding praise and OTRs, however. For example, praise is not effective in all classroom settings or for every student (Brophy, 1981), and some consideration of the mode of delivery may be needed (Blaze, Olmi, Mercer,
Dufrene, & Tingstom, 2014). Students' preferences related to receiving praise may also change as they grow older (Elwell & Tiberio,
1994). Likewise, much of the literature on OTRs refers to students at risk for emotional or behavior disorders and utilizes case study
methodology (e.g., Haydon et al., 2009; Sutherland & Wehby, 2001; Sutherland et al., 2000); thus the most effective way of engaging
high school students in general education settings is still unclear. The inclusion of praise and opportunities to respond in this study,
which focuses on the high school general education setting, will expand our understanding of these management methods.
An additional area of signicance for student behavior is studentteacher relationships and studentpeer relationships (Pianta,
Hamre, & Allen, 2012). Studies have shown that improved classroom organization and management have been associated with positive student interactions with peers, as measured by teacher reports of aggression in the classroom as well as observed prosocial behavior (Luckner & Pianta, 2011). In fact, extant research studies in the middle and high school settings have shown that opportunities
for autonomy and trusting relationships are important and perhaps more so for highly disruptive students (Gregory & Ripski, 2008;
Hafen et al., 2012). In addition, there is also some evidence that high school teachers who emphasize relationships and connection
with students experience lower levels of student deance and that this association is explained by student trust in teacher authority
(Gregory & Ripski, 2008). Taken together, these studies demonstrate the associations between relationships and behavior for older

E.T. Pas et al. / Journal of School Psychology 53 (2015) 137148

139

students. Further, a study of one teacher professional development model focused on quality of student-teacher interactions in secondary schools demonstrated that teachers' participation in the intervention was associated with increases in students' observed positive social interactions (Mikami, Gregory, Allen, Pianta, & Lun, 2011).
When high school students were asked to identify teachers who promoted student learning and motivation, students identied
teachers who were aware of and responsive to classroom events, monitored student understanding, maintained an appropriate
pace, moved around the classroom, and provided varied opportunities for student participation (Anderman, Andrzejewski, & Allen,
2011). Although that particular study is based on student report, it coincides with other studies supporting the use of active movement and tracking, provision of opportunities to respond, and the importance of student relationships reviewed earlier.
1.3. A classroom prole approach
The extant literature demonstrates that teacher management is intricately related to student behavior (Bradshaw et al., 2010;
Reinke, Herman, & Stormont, 2013). For example, there is evidence in elementary schools that the social structure and behavioral
norms of an entire classroom can inuence the escalation of disruptive behavior for individual students prone to aggression (Barth
et al., 2004; Roland & Galloway, 2002). This association is best established in elementary schools, although some research has
shown that behavior management and relationship-building strategies are relevant for improving the emotional climate of secondary
classrooms (Eccles & Roeser, 2011). For instance, Frenzel, Pekrun, and Goetz (2007) examined factors inuencing secondary students'
reported enjoyment or boredom of math class. They found students rated the class more favorably when there was high quality
teacher instruction and they perceived their peers to value the class and rate it as cool. Given the increasing importance of peer inuence during adolescence (Hawkins, Catalano, & Miller, 1992; Laursen, Hafen, Kerr, & Stattin, 2012), classroom norms and the escalation of student behavior are of concern in high school classrooms. Finally, as established earlier, literature demonstrates that
students with specic student demographics (e.g., males) are more likely to engage in disruptive behavior (Kellam et al., 1998;
Kewel Ramani et al., 2007; Pas et al., 2010; Pas et al., 2011; Skiba et al., 2002), although it is unknown whether classroom composition
relate to student behavior.
1.4. Observational measures of teacher and student behaviors
Reliable and valid measurement of student behaviors and classroom management techniques is challenging. Although surveys and
administrative data are efcient and low cost ways to assess these constructs, the use of such measures can introduce concerns regarding bias (e.g., Irvin et al., 2006; Pas et al., 2011). The use of externally-conducted observations can serve as an alternative, providing a more objective assessment of the classroom. The use of such measures requires more time and resources than surveys do and
therefore is not included in many school-based studies; use of observations has been particularly limited in secondary school settings
(Cash & Hamre, 2013).
1.5. Current study
Taking into consideration the conclusions as well as gaps in the extant literature, the current study aimed to identify patterns or
proles of high school student behavior at the classroom level and examine their association with teachers' observed use of both positive and negative classroom management strategies and with classroom composition (e.g., subject, grade level, and student composition). We expected at least two proles: (1) classrooms with high levels of student compliance and engagement and low levels of
socially disruptive behaviors as compared to (2) classrooms with low levels of compliance and engagement and higher levels of socially disruptive behavior. Based on the literature linking teacher classroom management to student behaviors (Epstein et al.,
2008), we expected to see greater teacher use of positive behavioral strategies in classrooms where students behaved well
(i.e., more proactive management, approval, and opportunities to respond) and fewer instances of negative strategies
(i.e., disapproval and reactive behavior management). Exploratory analyses of student behavior proles and classroom composition
were hypothesized to support prior research showing that male students are more likely to engage in disruptive behavior in the classroom (Kellam et al., 1998; Pas et al., 2010). Based on concerns regarding disproportionality in exclusionary discipline procedures
(e.g., Skiba et al., 2011, 2002), we also examined whether the classroom racial composition was related to student behavior. Given
developmental changes in behavior during adolescence (Eccles & Roeser, 2010, 2011), we tested whether the classroom proles
were different across grade levels.
2. Method
2.1. Participants
Data come from 1262 unique classrooms across 52 high schools in 10 school districts in Maryland. All schools were involved in a
large-scale study of school climate in high schools; however, the data reported here are from the fall 2011 baseline data collection
(i.e., collected during a 1-month period) prior to intervention training (see Bradshaw et al., 2014 for an overview of the trial). The participating schools included a diverse student population; the average rate of ethnic minorities in this sample of schools was 48%
(SD = 25.34). The mean student enrollment of these schools was 1254 (SD = 476.53). Observers were trained to conduct observations in 25 classrooms that rst included all language arts teachers and then a random sampling of other classrooms after all language

140

E.T. Pas et al. / Journal of School Psychology 53 (2015) 137148

arts teachers had been assessed. Therefore, data were collected in 1262 classrooms and a project-wide average of 24.27 (SD = 2.77)
classrooms observed per school. This sample represents 37.6% of all teachers in these schools (range 24% to 96% in a given school).
Following the classroom selection procedure, about half of the classroom observations were language arts classrooms (49%); the
remaining observations were conducted in math (18%), history (17%), and science (17%) classrooms. Approximately, 61% of the
observed classroom teachers were women. See Table 1 for a detailed description of classroom and school demographic characteristics.
2.2. Measure
The Assessing School Settings: Interactions of Students and Teachers (ASSIST; Rusby, Taylor, & Milchak, 2001) observational measure is comprised of event-based tallies (i.e., counts of specic behaviors) and global ratings (i.e., scale scores) as indicators of social
processes occurring in the high school classrooms (see Rusby, Crowley, Sprague, & Biglan, 2011). Event-based tallied behaviors included teacher use of (1) Proactive Behavioral Management, (2) Opportunities to Respond, (3) Approval, (4) Disapproval, and (5) Reactive
Behavior Management. Specically, Proactive Behavioral Management included all verbal (e.g., explaining, reminding, commanding,
prompting) and physical (e.g., modeling) demonstrations of behavioral expectations prior to a behavior becoming a problem (i.e., not
in response to misbehavior). Opportunities to Respond (OTRs) were dened as behavioral or instructional prompts requiring an
immediate response to the teacher, a peer, or a publicly-written response (e.g., on a white board at the front of the room); it did
not include privately-written responses (e.g., in a notebook). Approval was dened as recognizing students' performance by providing
a tangible item, verbal praise, approving gestures (e.g., thumbs up), or physical contact (e.g., pat on the back). Disapproval was the
threat or use of a tangible punitive consequence (e.g., detention), verbal criticism or sarcasm, or gestural or physical contact to demonstrate dissatisfaction with behavior. Reactive Behavior Management included teacher cues (e.g., touch, gesture, proximity, and
comment) to redirect inappropriate behavior (excluding disapproval). The descriptive statistics for the event-based tallied behaviors
are presented in the center section of Table 1.
Table 1
Classroom and school demographics.
Classroom characteristics (N = 1262 classrooms)

N (%)

Male teachers
Subject area
Language arts
Math
Science
Social studies
Grade
9th grade
10th grade
11th grade
12th grade
Mixed
Unknown

484 (38.7)
611 (48.6)
221 (17.6)
201 (16.7)
213 (16.9)
201 (15.9)
222 (17.6)
166 (13.2)
132 (10.5)
239 (20.0)
302 (24.0)
M (SD)

Number of students in classroom

% male students
% White students
ASSIST observations
Tally: proactive behavior expectations
Tally: approval
Tally: disapproval
Tally: reactive behavior management
Tally: opportunities to respond
Global: meaningful participation
Global: compliance
Global: socially disruptive behaviors

20.3 (6.1)
50 (20)
50 (30)

School characteristics (N = 52 schools)

M (SD)

% Attendance
% Racial/ethnic minority students
% Suspension
School enrollment
% students receiving FARMS

93.0 (1.7)
48.0 (25.3)
23.1 (11.3)
1, 253.8 (476.53)
38.3 (16.2)

4.9 (4.4)
1.8 (2.5)
0.3 (0.7)
4.0 (4.6)
13.9 (12.0)
1.9 (0.9)
3.0 (0.7)
0.4 (0.4)

Note. ASSIST = Assessing School Settings: Interactions of Students and Teachers. FARMS = free and reduced priced meals. The means for ASSIST tallies represent the average count of occurrences whereas
the mean of global scales were on a scale of 0 to 4. The attendance variable represents the percentage
of students present in school for at least half the average school day during the school year. The suspension variable is calculated by dividing the total number of suspension events by total student
enrollment.

E.T. Pas et al. / Journal of School Psychology 53 (2015) 137148

141

The global ratings include the following student subscales: Meaningful Participation (9 items including students have opportunities to take leadership roles in the classroom and students share ideas/opinions; = .85), Compliance (7 items including students
consistently follow rules appropriate to settings and students cooperate; = .92), and Socially Disruptive Behaviors (7 items
including students are irritable or sarcastic toward the teacher and students physically harass and/or bully others; = .66).
Coefcient alpha values reported are from the current study and sample. Items were marked by observers using a 0 to 4-point
Likert-type scale, ranging from never to almost continuously. Scales were calculated by averaging scores across items. Higher scores
on each scale reected more of the construct, such that it was positive to have high scores on Meaningful Participation and Compliance
and negative to have high scores for Socially Disruptive Behaviors. High scores on the Meaningful Participation scale suggest that students not only contribute to the lessons but they also had opportunities to make choices and demonstrate leadership. The subscale
descriptive statistics are presented in the center section of Table 1.
Classroom characteristics were recorded at the start of the observation, and included teacher gender, the subject of the lesson
being taught, the grade level of students, total number of students in the class, student gender, and the number of White students
in the class (as an indicator of classroom diversity).
2.3. Procedure
2.3.1. Research procedures
The collection of the ASSIST measure was part of a larger study of school climate in high schools (see Bradshaw et al., 2014), whereby school participation was voluntary. The Maryland State Department of Education rst approached districts to participate. Upon district approval, school principals were approached to obtain interest and commitment to participating in the study. Teachers within
participating schools were notied that observers may come into their classroom and could opt out of the observation. The ASSIST
data were collected anonymously, such that no identiers about the teacher or students were used and thus the researchers' Institutional Research Board granted this study exempt status.
2.3.2. ASSIST training
The ASSIST observers received didactic and on-site training sessions. Specically, each observer received a manual containing the
ASSIST classroom procedures, operational denitions of all codes, and step-by-step observational recording procedures. During didactic sessions, observers watched a series of videos and vignettes, which included coded behaviors and possible responses for application. After the training, observers practiced observing at nonproject schools in classrooms with an expert observer who provided
feedback and answered questions. Following practice sessions, observers were assessed for calibration in these nonproject schools
and were required to meet 80% interobserver agreement in three classrooms prior to observing independently in study schools. Observers engaged in calibration testing until the criterion was met. Interobserver agreement was calculated by dividing the total number of agreements by the total number of agreements and disagreements (Barlow & Hersen, 1984). For the current study (i.e., in fall
2011 in 52 high schools), average interobserver agreement for three classroom observations conducted in nonproject schools during
initial training was 89%.
Reliability tests were again conducted during the active data collection. Specically, every observer was joined on at least one random occasion during active data collection for an on-site recalibration session. Recalibration again involved three classroom observation cycles. If observers did not calibrate at 80% or above, ongoing re-calibration took place until 80% criterion was reached. The expert
observer data were used when available for the analyses in this study. Average interobserver agreement during active data collection
was 92%. Intraclass correlation coefcients (ICCs) were also calculated using the fall recalibration data; ICCs for the tallied behaviors
were .82 (Proactive Behavior Management), .95 (Approval), .25 (Disapproval), .68 (Reactive Behavior Management), and .84 (OTRs).
ICCs for the global ratings were .71 (Meaningful Participation), .85 (Compliance), and .53 (Student Social Disruption). The lowest ICCs
occurred for low base rate behaviors indicated by zero occurrences (tallies) and scores (global ratings). A recent generalizability-study
of the ASSIST indicated that less than one percent of the variance in classroom event codes was attributable to rater, suggesting a high
level of reliability among observers. A corresponding decision-study supported the use of a single rater over using two raters (Abry,
Cash, & Bradshaw, 2014).
Observers were informed that they were collecting data for a study relating to student experiences in high schools in Maryland. No
other information about the broader study context was shared, as the observers were kept blind to the true purpose of the study as
well as assigned treatment status. As noted earlier, the data for this study were from a baseline data collection and thus no intervention targeting school climate (see Bradshaw et al., 2014) was taking place, and as a result, observers were not expected to rate teachers
on core elements of the intervention.
2.3.3. ASSIST observation and sampling procedures
Nineteen trained observers assessed an average of 24 teachers in each school (i.e., all language arts teachers, followed by a random
sample of other subjects). Individual teachers were not to be observed more than once. This sample represents 97% of the targeted
classroom sample of 25 classrooms per school (i.e., 1262 out of 1300 the target classrooms). Tallies and global ratings were completed
for every observation conducted; therefore, complete data for the 1262 classrooms were available for these analyses.
Teachers were informed in advance that observations may be conducted in their classrooms and that no personally identiable
information would be collected about the staff or students. Upon arriving at the school, observers used a class schedule to identify
the required language arts teachers and then randomly selected a class period for each language arts teacher to observe. After

142

E.T. Pas et al. / Journal of School Psychology 53 (2015) 137148

completing the language arts teachers' observations, the observers then randomly selected an equal number of teachers from the
other three class subjects and a class period in which to observe until the target of 25 teachers was met.
At the start of each observation, the observer found an unobtrusive place to view the classroom teacher and all students and then
waited approximately three minutes to acquaint him or herself with the current classroom activities. After the initial acclimation window, the observer used a handheld tablet to complete demographic information about the classroom and activated the timer to begin
the 15-minute tally period. For 15 min, the observers tallied specic teacher behaviors. Each behavior could only count as one tallied
event; no behavior was tallied in two or more categories simultaneously. Once tallies were completed, the observer left the classroom
and immediately completed the global ratings of the classroom environment (i.e., subscales detailed above). In contrast to the tallies,
global ratings of the classroom were based on all behaviors observed during the 15-minute assessment window. ASSIST data were
transferred remotely to the project computer server for download and analysis.
2.3.4. Analyses
Latent prole analysis (LPA; Hagenaars & McCutcheon, 2002; Muthn & Muthn, 19972012) was conducted in Mplus 7.1 to create
latent proles of classrooms based on the global rating scales of student behaviors, where means and variances were allowed to vary
across proles. LPA is a latent class approach used with continuous data, whereby nonobservable proles are generated based on similar response patterns across multiple indicators using a set of continuous manifest indicators. The global ratings included represented
the average of the categorical items for the scale; thus, the variables were treated as continuous. The number of proles was determined using three t indices and two statistical tests: the Akaike Information Criteria (AIC), Bayesian Information Criterion (BIC;
Schwartz, 1978), sample size adjusted BIC, LoMendellRubin likelihood ratio test (LMR; Lo, Mendell, & Rubin, 2001), and VuongLMR likelihood ratio test (Muthn & Muthn, 19972012). The latent class models were built iteratively, adding one prole at a
time. Improved t was determined through the examination of a decreasing AIC, BIC, and adjusted BIC as well as a statistically significant (p b .05) LMR and Vuong-LMR. A nonsignicant LMR and Vuong-LMR indicates that the addition of a class does not result in a
statistically signicant improvement in model t and the more parsimonious model (i.e., with fewer classes) should be retained. In
addition, entropy scores close to 1.00 and posterior probabilities greater than .70 were considered favorable (Nagin, 2005;
Ramaswamy, Desarbo, Reibstein, & Robinson, 1993). Finally, the size and conceptual and theoretical meaning of the classes were considered (Muthn, 2004; Nylund, Asparouhov, & Muthn, 2007). Clustering of teachers and classrooms within schools was modeled
using the HuberWhite sandwich estimator to adjust the standard errors (Muthn & Muthn, 19972012). Previous studies have
used a similar approach to create proles of classroom teachers (Pas & Bradshaw, 2014).
Once the LPA model was nalized, the auxiliary (e) function in Mplus was rst used to determine whether observed classroom
management strategies differed in classrooms based on the student behavior prole observed (Clark & Muthn, 2009). The same analytic approach was used to secondarily determine whether the student behavioral proles were equally distributed across subject
content and grade level and to determine whether the composition of the classrooms (i.e., student demographics) varied based on
the student behavioral proles observed. The auxiliary (e) function in Mplus allows the user to include additional variables in the
model, without allowing them to inform the generation of the latent proles. Mean differences on these targeted variables between
the proles identied in the LPA were examined with this function and tests of statistical signicance (i.e., p b .05) using the chi-square
difference tests are provided.
3. Results
3.1. Latent proles of student behavior
A series of models with up to ve latent classes was t using the global rating subscale scores of student Compliance, Meaningful
Participation, and Socially Disruptive Behavior in 1262 classrooms. The best t for the latent prole analysis of student behaviors
included three proles of student behaviors (LMR p b .05 for 3-class solution, entropy = .85). See Table 2 for t statistics and Fig. 1
Table 2
Fit statistics for latent prole analysis of student behavior.
Classes

Log likelihood

AIC

BIC
(n adj BIC)

LMR (VLMR)
p values

Entropy

Class
proportions

3600.526

7213.051

3209.082

6438.163

3035.695

6099.389

.02
(.21)
.03
(.03)

.895

7243.894
(7224.835)
6489.568
(6457.803)
6171.356
(6126.885)

2959.807

5955.614

6048.142
(5990.142)

.24
(.23)

183 (14.5%)
1079 (85.5%)
292 (23.1%)
901 (71.4%)
69 (5.5%)
66 (5.2%)
650 (51.5%)
283 (22.4%)
263 (20.8%)

.847

.731

Note. AIC = Akaike Information Criteria, BIC = Bayesian Information Criterion, n adjusted BIC = sample size adjusted BIC, LMR = LoMendellRubin likelihood ratio
test, and VLMR = Vuong-LMR likelihood ratio. Bold indicates best tting model.

E.T. Pas et al. / Journal of School Psychology 53 (2015) 137148

143

Fig. 1. Latent proles of student behavior. Note. Values on the y-axis represent average scores on the 04 point Likert-type global ratings scale. The x-axis includes the
meaningful participation, compliance, and socially disruptive behaviors global rating scales.

for a graphical depiction of the three-prole model. Although the 4-class model indicated improved AIC and BIC t indices, the nonsignicant LMR tests and the poor entropy (0.73) indicated that the model was not statistically adequate. Conceptually, the additional
class also did not add meaning and thus the more parsimonious solution was chosen.
The posterior probabilities for the 3-class solution were .86, .95, and .92, and the smallest group was 5.5% of the total sample. The
largest prole (comprising 71.4% of observed classrooms) included classrooms where students consistently met expectations.
Observers reported that students complied a lot of the time or almost continuously, engaged in nearly no socially disruptive behavior,
and there was evidence of some meaningful participation. We refer to this latent prole as Consistent. The next largest class comprised 23.1% of observed classrooms and was characterized by students inconsistently meeting expectations. Some compliance was
observed, students were seldom engaged in socially disruptive behavior, and meaningful participation was reported as occurring
seldom to some of the time in these classrooms. We refer to this latent prole as Inconsistent. Finally, 5.5% of the classrooms were
characterized as noncompliant because students in these classrooms were sometimes compliant, socially disruptive, and seldom participated meaningfully. We refer to this latent prole as Noncompliant. Interestingly, all three groups had similar ratings of meaningful
participation (i.e., that it occurred seldom to some of the time), and no classroom prole included very high or very low ratings of
participation. The two-class solution included the Consistent and Noncompliant classes (but not Inconsistent), and the four-class
solution added an additional class of students meeting expectations at an even higher level.
3.2. Association of student behavior and teacher classroom management strategies
Using the auxiliary function in Mplus, the association between the classroom prole of student behavior and tallied teacher use of
classroom management strategies was examined. Results revealed signicant relations between student behavior proles and teacher
use of OTRs, Approval, Disapproval, and Reactive Behavior Management. Specically, the use of OTRs occurred about once per minute
in classrooms where students consistently met expectations. This rate was signicantly higher as compared to classrooms where students inconsistently met expectations, 2 = 31.90, p = .00, and students were noncompliant, 2 = 6.10, p = .01. In the Inconsistent
and Noncompliant classrooms, OTRs occurred once every 1 to 1 minutes. The total OTRs observed did not differ across classrooms
where students inconsistently met expectations and noncompliant (an average of 11 and 12 times total in 15 min, respectively).
Approval was highest in classrooms where students consistently met expectations, 2 = 7.79, p = .01, and were noncompliant,
2
= 4.83, p = .03, as compared to classrooms where students inconsistently met expectations. Disapproval was used at the highest
rate in classrooms where students were noncompliant, 2 for noncompliant versus inconsistent = 13.69, p = .00, and 2 for noncompliant versus consistent = 31.15, p = .00, followed by classrooms with inconsistent student behavior, 2 for inconsistent versus
consistent = 21.25, p = .00, and then classrooms where students consistently met expectations. It should be noted that although
signicant differences were detected on Approvals and Disapprovals, on average, these rates were low for each prole. Disapprovals
were used 1 or fewer times and Approvals were used close to 2 times across proles during the 15-minute observation.
Reactive Behavior Management was also highest in classrooms where students were noncompliant, 2 for noncompliant versus
inconsistent = 17.82, p = .00 and 2 for noncompliant versus consistent = 74.63, p = .00, followed by classrooms with inconsistent
student behavior, 2 for inconsistent versus consistent = 78.82, p = .00, and then classrooms with students consistently meeting expectations. There were no signicant associations between student behavior proles and the use of Proactive Behavior Management.
See Table 3 for a full listing of the means and standard error values for each outcome and results of chi-square tests.
3.3. Association of student behavior and classroom characteristics
The auxiliary function was used to determine whether the student composition, classroom subject and grade, and teacher demographics differed signicantly across the three latent proles of student behavior. The chi-square tests revealed that student

144

E.T. Pas et al. / Journal of School Psychology 53 (2015) 137148

Table 3
Means (standard error) and chi-square statistics for teacher use of classroom management strategies and classroom composition variables.
Student behavior latent proles

Tallied teacher management

Proactive
OTRs
Approval
Disapproval
Reactive
Percent of students who are
Male
White
Percent of classrooms with a male teacher
Percent of classrooms teaching
English
Math
Science
Social studies
9th graders
10th graders
11th graders
12th graders

Consistent 1

Inconsistent 2

Noncompliant 3

4.75 (0.15)
15.142,3 (0.44)
1.932 (0.09)
0.16 2,3 (0.02)
2.862,3 (0.13)

5.03 (0.28)
10.631 (0.63)
1.441,3 (0.14)
0.411,3 (0.05)
6.161,3 (0.34)

5.64 (0.69)
11.931 (1.22)
2.232 (0.33)
1.011,2 (0.15)
9.911,2 (0.81)

0.472,3 (0.01)
0.562,3 (0.01)
0.40 (0.02)

0.511 (0.01)
0.491 (0.02)
0.37 (0.03)

0.531 (0.02)
0.411 (0.04)
0.32 (0.06)

0.48 (0.02)
0.18 (0.01)
0.16 (0.01)
0.17 (0.01)
0.17 (0.01)
0.20 (0.02)
0.15 (0.01)
0.12 (0.01)

0.48 (0.003)
0.17 (0.02)
0.19 (0.03)
0.17 (0.02)
0.22 (0.03)
0.20 (0.03)
0.16 (0.03)
0.13 (0.02)

0.57 (0.06)
0.18 (0.05)
0.13 (0.04)
0.12 (0.04)
0.20 (0.06)
0.23 (0.06)
0.12 (0.05)
0.13 (0.05)

Note. OTRs = opportunities to respond. Subscripts following a mean score represent a statistically signicant difference on chi-square test of independence at the
p b .05 level between the column class and the subscript denoted (i.e., 1 = inconsistent class, 2 = inconsistent class, and 3 = noncompliant class).

composition variables but not teacher demographics, classroom subject, or grade level taught were signicantly related to the student behavior proles. Specically, the proportion of males present was lowest in classrooms where students consistently met expectations as compared to classrooms where students inconsistently met expectations, 2 = 10.44, p b .01, and were noncompliant,
2 = 9.17, p b .01. Similarly, the proportion of White students was the highest in classrooms where students consistently met expectations as compared to those where students inconsistently met expectations, 2 = 7.93, p b .01, and were noncompliant, 2 = 14.86,
p b .01.
4. Discussion
The current study aimed to identify specic high school student behavior proles and determine whether these student behavior
proles were signicantly related to teachers' use of positive and negative classroom management strategies. Using observational
data from 52 high schools across the state of Maryland, results revealed three distinct classroom proles based on observers' ratings
of student participation, compliance with classroom rules, and displays of socially disruptive behavior. Specically, the results of the
LPA demonstrated classrooms where students (1) consistently met expectations, (2) inconsistently met expectations, and (3) were
noncompliant. Signicant differences in teachers' behavior management as well as the classroom-level student demographics were
also noted across classroom behavior proles.
4.1. Student behavior patterns across classrooms
Students in the classrooms demonstrating consistent meeting of expectations tended to display very few disruptive behaviors
(e.g., off-task conversations, verbal aggression, and bullying). Conversely, less engagement and more social disruption were displayed
by students in classrooms with the Inconsistent prole and Noncompliant prole. Differences in compliance and disruptions can have
implications for student achievement as student misbehavior often co-occurs with poor achievement and low school connectedness
(Bradshaw, Buckley, & Ialongo, 2008; Konishi et al., 2010; Malecki & Elliott, 2002). The issue of compliance and disruption may be of
particular importance at the secondary level, where the mastery of a specic content area and accumulation of credits in these areas is
the focus of each classroom. On the other hand, Meaningful Participation occurred only some of the time across all of the high school
classroom proles in this study.
Interestingly, though the three proles were characterized by similar levels of Meaningful Participation, the rate at which teachers
used OTRs did differ signicantly in the classrooms where students consistently met expectations as compared to the other two
proles. It is possible that OTRs that were rote in nature were used in all classrooms, whereas OTRs that are engaging, meaningful,
and more challenging are needed to ensure a higher level of student participation. High scores on the Meaningful Participation
scale require not only that students are responsive to teacher questions but also that teachers provide students with opportunities
for leadership in the classroom. Alternatively, more OTRs are likely needed in all classrooms in order to facilitate greater student participation; the highest average rate across classroom proles reected one OTR per minute. Research regarding OTRs with students in
elementary schools implies much higher rates (e.g., 4 to 6 OTRs per minute when new material is presented) are needed than were
demonstrated here (Council for Exceptional Children, 1987). Future research should examine the frequency and extent to which the
co-occurrence of OTRs and opportunities for student leadership engage secondary students. The nding that Meaningful Participation

E.T. Pas et al. / Journal of School Psychology 53 (2015) 137148

145

occurred only some of the time across all proles is consistent with other studies that show low levels of participation and engagement in middle and high schools (e.g., Hanson & Kim, 2007). The low level of student involvement may be reective of common teaching practices in the upper grades (e.g., lecturing and individual testing), as opposed to interactive lesson plans encouraging student
engagement with staff and their peers (Pianta et al., 2012; Skinner & Belmont, 1993).
The distribution of positive (i.e., Consistent), at-risk (i.e., Inconsistent), and difcult classroom behaviors (i.e., Noncompliant
classrooms, likely in need of intensive interventions) maps well on to the public health lens, where it is expected that 80% of individuals will be responsive to interventions targeting social behavior whereas 15% demonstrate additional risk and 5% have intensive
needs (Hawken, Vincent, & Schumann, 2008; Walker et al., 1996). Specically, about 71% of classrooms were characterized by consistent student behavior, 23% had inconsistent behaviors, and students were noncompliant in 6% of classrooms. The identication and
use of these classroom-level student behavior proles may in fact translate well into interventions, whereby teachers in at-risk
and difcult classrooms are identied for additional tiered professional development and supports. For example, the use of tailored
coaching targeting classroom management (e.g., the Classroom Check-Up [CCU]; Reinke, 2006) could be used in classrooms where
students display Inconsistent or Noncompliant behavior, with the specic goal of altering certain teacher management such as praise
and reprimands (Reinke, Lewis-Palmer, & Merrell, 2008). Further investigation into classroom proles and tailored intervention
supports is an area ripe for future research.

4.2. Relationship of student behavior with classroom management and context

In classrooms where students consistently behaved, teachers generally provided students with more OTRs, more positive recognition for behavior, few disapproving statements, and limited use of reactive behavior management strategies. This nding is similar
to the ndings of intervention research which has demonstrated associations of teacher classroom management with student disruptive behavior, on-task behaviors, and peer interactions (e.g., Bradshaw et al., 2010; Colvin et al., 2009; Evertson, 1985; Flannery et al.,
2009; Ialongo et al., 2001; Mikami et al., 2011). OTRs were used the most frequently in classrooms where students consistently met
expectations; this nding is consistent with prior research, suggesting that a greater number of OTRs corresponds with increased student on-task behavior and engagement in materials (Sutherland, Adler, & Gunter, 2003). Of the ve classroom management strategies
tallied, OTRs were the most frequently used. Again, further research is needed to determine the optimal number of OTRs needed to
maximize engagement and achievement. In addition, an exploration of the preparation of secondary teachers with regard to
specic classroom management and instructional strategies, such as OTRs, is needed.
Teachers in classrooms characterized as Noncompliant used roughly three times as many reactive strategies as compared to the
teachers in classrooms with the Consistent prole. It is important to note, however, that reactive strategies in this study were not
only punitive responses to student behavior; rather, they included cueing and redirection when a student was observed off-task. Because these data are cross-sectional, we cannot determine if teachers' use of reactive strategies caused a greater amount of disruptive
behavior to occur or if teachers only used these strategies because of the presence of student disruptive behavior. Longitudinal intervention research is needed to understand the directionality of this association over time.
Interestingly, proactive behaviors (i.e., explanation or modeling of expected behavior) did not signicantly differ across the three
student behavior proles. Teachers' explicit explanation of expectations at the high school level may not result in noticeable behavior
changes because students are typically well-versed in the social milieu of the classroom by this time (Crosnoe, 2011). Alternatively,
our ndings may be in part the result of the denition of Proactive Behavior Expectations in the current study, which included all examples of verbal or physical demonstration of expectations and not just examples that specically targeted the prevention of student
misbehavior. A broad denition was used to decrease the cognitive load in coding Proactive Behavior Expectations; this measurement
decision may have implications for the ndings and, therefore, is an area for future research to take into consideration.
Teachers' indications of approval and disapproval generally occurred at low rates. The data indicated that teachers in the Noncompliant classrooms dually used relatively high rates of verbal, tangible, or both types of reinforcers to increase the frequency of positive
behaviors as well as disapproval to address undesired behaviors. The coding of Approval included many forms of praise, and it is
unclear whether teachers in classrooms with Consistent, Inconsistent, and Noncompliant students differ in the types of approval
they use. For example, it would have been important to know whether teachers in these different classrooms use vague praise versus
behavior-specic praise because the latter has been found effective for decreasing disruptive behavior (Reinke, Lewis-Palmer, &
Martin, 2007; Sutherland et al., 2000). Similarly, both verbal and nonverbal signs of approval were coded, and it is unknown whether
one type was more prevalent, and in turn, if either would be more or less effective in secondary classrooms. Some research suggests
that both types of praise should be emphasized (Blaze et al., 2014), but additional research in large secondary samples is needed.
A secondary aim of the study was to examine the association between the student behavior proles and classroom and teacher
demographics. The proles of student behavior were consistent across classroom subjects, grades, and teacher gender but varied
based on student gender and classroom racial composition. Consistent with previous research that indicates that male students
engage in higher levels of disruption than female students (e.g., Kellam et al., 1998; Pas et al., 2010) as well as research that shows
minority students are more likely to be referred to the ofce for discipline problems and receive less favorable teacher ratings
(Kewel Ramani et al., 2007; Pas et al., 2011; Skiba et al., 2002), we found that classrooms characterized as Consistent had a lower
proportion of male and higher proportion of White students. Although there may be a number of factors contributing to the disproportionately higher rates of discipline problems in diverse classrooms, there is a growing body of research suggesting that teachers
need training in cultural prociency in order to better connect and form authentic and supportive relationships with culturally diverse
students (Bottiani et al., 2012).

146

E.T. Pas et al. / Journal of School Psychology 53 (2015) 137148

4.3. Limitations
Although this study has much strength, such as the inclusion of over 1200 classrooms across 52 high schools and the use of an external observation which utilizes both tallied and global behaviors, there are limitations to be considered. The nature of these data is
such that teachers were completely anonymous in order to protect teachers' condentiality as well as make them more comfortable
with outside observers being present in their classroom. However, because of this anonymity, the observations cannot be linked to
any other classroom-level data, making this study reliant on a single data source. On the other hand, the use of an objective measure,
rather than a teacher report, provides valuable data which is likely less biased than teachers' own ratings or disciplinary referrals
(e.g., see Pas & Bradshaw, 2014). Similarly, because classroom-level data cannot be linked, the longitudinal examination of student
behavior and teacher classroom management cannot be conducted; furthermore, the data are correlational, which further limits conclusions regarding a possible causal association between student behavior patterns and teacher practice. Additional experimental research is needed to document a causal association between these factors as well as to determine whether distinct student behavioral
proles can be used to identify classroom teachers in need of additional professional development and supports. Future research
should incorporate a multi-method approach, which includes data from teachers and students, as well as longitudinal data; these
additions would allow for research to better capture directionality and causality. Further, although our focus on high schools is a
novel contribution to the eld, future studies should look at different school levels simultaneously to compare these settings. Finally,
as noted above, there are nuances in the measurement of student and teacher behaviors that may have affected our ndings; for
example, the broad denitions of Proactive Behavior Management and Approval leave questions unanswered about whether higher
rates of behavior-specic proactive prompting and praise are used in classrooms where students consistently meet expectations.
Similarly, the collection of low base rate behaviors (e.g., teacher disapproval and student socially disruptive behavior) has implications
for how reliably these data are collected, given the high frequency of nonoccurence or minimal occurrence.
4.4. Conclusions and implications for school psychologists
Research supports that teacher management strategies and student behavior are intertwined, though this association is typically
studied and discussed at the individual student level rather than considering the broader functioning of the whole classroom. The
ndings from the current study regarding student behaviors in the classroom are in line with the public health lens, suggesting
three tiers of classroom behavior: classrooms where behavior is generally positive, classrooms where student behavior is less
consistent; and classrooms where students are noncompliant. These ndings highlight the possibility of school personnel
(e.g., administrators and school psychologists) measuring students' compliance, meaningful participation, and socially disruptive
behaviors class-wide to identify and provide additional support to teachers within classrooms where student behavior matches the
Inconsistent or Noncompliant prole. The exhibition of such behavioral proles is likely to result in a higher level of teacher stress.
Teachers who report high levels of stress also report having less control of their classroom, lower commitment to the teaching
profession, and increased likelihood of quitting (Klassen & Chiu, 2011).
In terms of the association between proles of student behavior and teacher management techniques, the nearly equal ratings of
student meaningful participation across classroom proles may indicate that improvements are needed universally for engagement in
high schools. Given the importance of student engagement for persistence through high school (Balfanz, Herzog, & MacIver, 2007),
this area is an important target for further research study and attention in schools. An additional area in need of research is the
interplay between Meaningful Participation and OTRs, as it may relate to students' behavioral and academic functioning. Finally,
more research is needed in the areas of proactive behavior management and approval both from a measurement and practical
standpoint as these may vary across classrooms and in effect on students (e.g., Elwell & Tiberio, 1994). The inclusion of high, middle,
and elementary schools in the same study as well as longitudinal data will most enhance our knowledge of these areas.
In sum, the current study supports the widening of school psychologists' and other school personnel's lenses when interpreting
student behavior, as it likely inuenced by the overall classroom behavioral climate as well as teacher behavior management
practices. School personnel, such as psychologists, are in a position to observe teachers and identify at-risk teachers with regard
to student behavior and teacher classroom management and instructional strategies. Using an observational approach, psychologists
can help to enhance teacher skills as a means for promoting student outcomes. Interventions such as PBIS (Sugai & Horner, 2006) can
be used school-wide whereas interventions such as the GBG (Barrish, Saunders, & Wolf, 1969) and the CCU (Reinke, 2006) can be used
in classrooms to explicitly address the use of positive reinforcement versus negative and punitive responses to behavioral infractions.
References
Abry, T., Cash, A. H., & Bradshaw, C. P. (2014). Using generalizability theory to examine sources of variance in observed behaviors within high school classrooms. Manuscript
submitted for publication Psychology in the Schools.
Anderman, L., Andrzejewski, C. E., & Allen, J. (2011). How do teachers support students' motivation and learning in their classroom? Teachers College Record, 113,
9691003.
Armendariz, F., & Umbreit, J. (1999). Using active responding to reduce disruptive behavior in a general education classroom. Journal of Positive Behavior Interventions,
1, 152158.
Balfanz, R., Herzog, L., & MacIver, D. J. (2007). Preventing student disengagement and keeping students on the graduation path in urban middle-grades schools: Early
identification and effective interventions. Educational Psychologist, 42, 223235. http://dx.doi.org/10.1080/00461520701621079.
Barlow, D., & Hersen, M. (1984). Single case experimental design. New York: Pergamon Press.
Barrish, H. H., Saunders, M., & Wolf, M. M. (1969). Good behavior game: Effects of individual contingencies for group consequences on disruptive behavior in a
classroom. Journal of Applied Behavior Analysis, 2, 119124.

E.T. Pas et al. / Journal of School Psychology 53 (2015) 137148

147

Barth, J. M., Dunlap, S. T., Dane, H., Lochman, J. E., & Wells, K. C. (2004). Classroom environment influences on aggression, peer relations, and academic focus. Journal of
School Psychology, 42, 115133. http://dx.doi.org/10.1016/j.jsp.2003.11.004.
Blaze, J. T., Olmi, D. J., Mercer, S. H., Dufrene, B. A., & Tingstom, D. H. (2014). Loud versus quiet praise: A direct behavioral comparison in secondary classrooms. Journal of
School Psychology, 52, 349360. http://dx.doi.org/10.1016/j.jsp.2014.05.006.
Bottiani, J. H., Bradshaw, C. P., Rosenberg, M. S., Hershfeldt, P. A., Pell, K. L., & Debnam, K. J. (2012). Applying double check to response to intervention: Culturally
responsive practices for learning disabilities. Insight on Learning Disabilities: Prevailing Theories to Validated Practices, 9, 93107.
Bradshaw, C. P., Buckley, J., & Ialongo, N. (2008). School-based service utilization among urban children with early-onset educational and mental health problems: The
squeaky wheel phenomenon. School Psychology Quarterly, 23, 169186. http://dx.doi.org/10.1037/1045-3830.23.2.169.
Bradshaw, C. P., Debnam, K. J., Lindstrom Johnson, S., Pas, E. T., Hershfeldt, P., Alexander, A., et al. (2014). Maryland's evolving system of social, emotional, and behavioral interventions in public schools: The Maryland Safe and Supportive Schools Project. Adolescent Psychiatry, 4, 194206.
Bradshaw, C. P., Koth, C. W., Bevans, K. B., Ialongo, N. S., & Leaf, P. J. (2008). The impact of school-wide Positive Behavioral Interventions and Supports (PBIS) on the
organizational health of elementary schools. School Psychology Quarterly, 23, 462473.
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 group-randomized effectiveness trial. Prevention Science, 10, 100115. http://dx.doi.org/10.1007/s11121-008-0114-9.
Bradshaw, C. P., Mitchell, M. M., & Leaf, P. J. (2010). Examining the effects of schoolwide positive behavioral interventions and supports on student outcomes: Results
from a randomized controlled effectiveness trial in elementary schools. Journal of Positive Behavior Interventions, 12, 133148. http://dx.doi.org/10.1177/
1098300709334798.
Bradshaw, C. P., Waasdorp, T. E., & Leaf, P. J. (2012). Effects of school-wide positive behavioral interventions and supports on child behavior problems. Pediatrics, 130,
11361145. http://dx.doi.org/10.1542/peds. 2012-0243.
Bradshaw, C. P., Zmuda, J. H., Kellam, S. G., & Ialongo, N. S. (2009). Longitudinal impact of two universal preventive interventions in first grade on educational outcomes
in high school. Journal of Educational Psychology, 101, 926937. http://dx.doi.org/10.1037/a0016586.
Brophy, J. (1981). Teacher praise: A functional analysis. Review of Educational Research, 51, 532. http://dx.doi.org/10.3102/00346543051001005.
Bryk, A. S., & Schneider, B. (2002). Trust in schools: A core resource for improvement. New York, NY: Sage.
Cash, A. H., & Hamre, B. K. (2013). Evaluating and improving studentteacher interactions. In J. A. C. Hattie, & E. M. Anderman (Eds.), International guide to student
achievement (pp. 119122). New York, NY: Routledge.
Clark, S. L., & Muthn, B. (2009). Relating latent class analysis results to variables not included in the analyses. Unpublished manuscript. Retrieved from http://hbanaszak.
mjr.uw.edu.pl/TempTxt/relatinglca.pdf
Colvin, G., Flannery, K. B., Sugai, G., & Monegan, J. (2009). Using observational data to provide performance feedback to teachers: A high school case study. Preventing
School Failure: Alternative Education for Children and Youth, 53, 95104.
Conroy, M. A., Sutherland, K. S., Snyder, A., Al-Hendawi, M., & Vo, A. (2009). Creating a positive classroom atmosphere: Teachers' use of effective praise and feedback.
Beyond Behavior, 18, 1826.
Council for Exceptional Children (1987). Academy for effective instruction: Working with mildly handicapped students. Reston, VA: Author.
Crosnoe, R. (2011). Fitting in, standing out: Navigating the social challenges of high school to get an education. New York, NY: Cambridge University Press.
Eccles, J. S., & Roeser, R. W. (2010). An ecological view of schools and development. In J. L. Meece, & J. S. Eccles (Eds.), Handbook of research on schools, schooling, and
human development (pp. 622). New York, NY: Routledge.
Eccles, J. S., & Roeser, R. W. (2011). Schools as developmental contexts during adolescence. Journal of Research on Adolescence, 21, 225241. http://dx.doi.org/10.1111/j.
1532-7795.2010.00725.x.
Elwell, W. C., & Tiberio, J. (1994). Teacher praise: What students want. Journal of Instructional Psychology, 21, 322.
Epstein, M., Atkins, M., Cullinan, D., Kutash, K., & Weaver, R. (2008). Reducing behavior problems in the elementary school classroom: A practice guide (NCEE #2008-012).
Washington, DC: National Center for Education Evaluation and Regional Assistance, Institute of Education Sciences, U.S. Department of Education (Retrieved from
http://ies.ed.gov/ncee/wwc/publications/practiceguides).
Evertson, C. M. (1985). Training teachers in classroom management: An experimental study in secondary school classrooms. Journal of Educational Research, 79,
5158.
Evertson, C. M., & Emmer, E. T. (1982). Effective management at the beginning of the school year in junior high classes. Journal of Educational Psychology, 74, 485498.
http://dx.doi.org/10.1037/0022-0663.74.4.485.
Flannery, K. B., Fenning, P., Kato, M. M., & McIntosh, K. (2014). Effects of school-wide positive behavioral interventions and supports and fidelity of implementation on
problem behavior in high schools. School Psychology Quarterly, 29, 111124. http://dx.doi.org/10.1037/spq0000039.
Flannery, K. B., Sugai, G., & Anderson, C. M. (2009). School-wide positive behavior support in high school: Early lessons learned. Journal of Positive Behavior
Interventions, 11, 177185. http://dx.doi.org/10.1177/1098300708316257.
Frenzel, A. C., Pekrun, R., & Goetz, T. (2007). Perceived learning environment and students' emotional experiences: A multilevel analysis of mathematics classrooms.
Learning & Instruction, 17, 478493. http://dx.doi.org/10.1016/j.learninstruc.2007.09.001.
Greenberg, M. T., Weissberg, R. P., Utne O'Brien, M., Zins, J. E., Fredericks, L., Resnik, H., et al. (2003). Enhancing school-based prevention and youth development
through coordinated social, emotional, and academic learning. American Psychologist, 58, 466474.
Gregory, A., & Ripski, M. B. (2008). Adolescent trust in teachers: Implications for behavior in the high school classroom. School Psychology Review, 37, 337353.
Hafen, C. A., Allen, J. P., Mikami, A. Y., Gregory, A., Hamre, B., & Pianta, R. C. (2012). The pivotal role of adolescent autonomy in secondary school classrooms. Journal of
Youth and Adolescence, 41, 245255.
Hagenaars, J. A., & McCutcheon, A. L. (2002). Applied latent class analysis. Cambridge, UK: Cambridge University Press.
Hanson, T. L., & Kim, J. -O. (2007). Measuring resilience and youth development: The psychometric properties of the Healthy Kids Survey. (Issues & Answers Report, REL
2007-No. 034) Washington, DC: U.S.: Department of Education, Institute of Education Sciences, National Center for Education Evaluation and Regional Assistance,
Regional Educational Laboratory West (Retrieved from http://ies.ed.gov/ncee/edlabs).
Hawken, L. S., Vincent, C. G., & Schumann, J. (2008). Response to intervention for social behavior: Challenges and opportunities. Journal of Emotional and Behavioral
Disorders, 16, 213225.
Hawkins, J. D., Catalano, R. F., & Miller, J. Y. (1992). Risk and protective factors for alcohol and other drug problems in adolescence and early adulthood: Implications for
substance abuse prevention. Psychological Bulletin, 112, 64105. http://dx.doi.org/10.1037/0033-2909.112.1.64.
Haydon, T., Mancil, G. R., & Van Loan, C. (2009). Using opportunities to respond in a general education classroom: A case study. Education and Treatment of Children, 32,
267278.
Horner, R. H., Sugai, G., Smolkowski, K., Eber, L., Nakasato, J., Todd, A. W., et al. (2009). A randomized, wait-list controlled effectiveness trial assessing school-wide
Positive Behavior Support in elementary schools. Journal of Positive Behavior Interventions, 11, 133144.
Ialongo, N., Poduska, J., Werthamer, L., & Kellam, S. (2001). The distal impact of two first-grade preventive interventions on conduct problems and disorder in early
adolescence. Journal of Emotional and Behavioral Disorders, 9, 146160.
Ialongo, N. S., Werthamer, L., Kellam, S. G., Brown, C. H., Wang, S., & Lin, Y. (1999). Proximal impact of two first-grade preventive interventions on the early risk behaviors for later substance abuse, depression, and antisocial behavior. American Journal of Community Psychology, 27, 599641. http://dx.doi.org/10.1023/a:
1022137920532.
Irvin, L. K., Horner, R. H., Ingram, K., Todd, A. W., Sugai, G., Sampson, N. K., et al. (2006). Using office discipline referral data for decision making about student behavior
in elementary and middle schools: An empirical evaluation of validity. Journal of Positive Behavior Interventions, 8, 1023.
Kellam, S. G., Ling, X. G., Merisca, R., Brown, C. H., & Ialongo, N. S. (1998). The effect of the level of aggression in the first grade classroom on the course and malleability
of aggressive behavior into middle school. Development and Psychopathology, 10, 165185.
Kewel Ramani, A., Gilbertson, L., Fox, M., & Provasnik, S. (2007). Status and trends in the education of racial and ethnic minorities (NCES 2007-039). Washington, DC:
National Center for Educational Statistics, Institute of Education Sciences, U.S. Department of Education (Retrieved June 17, 2010, from http://nces.ed.gov/
pubs2007/2007039.pdf).

148

E.T. Pas et al. / Journal of School Psychology 53 (2015) 137148

Klassen, R. M., & Chiu, M. M. (2011). The occupational commitment and intention to quit of practicing and pre-service teachers: Influence of self-efficacy, job stress,
and teaching context. Contemporary Educational Psychology, 36, 114129. http://dx.doi.org/10.1016/j.cedpsych.2011.01.002.
Konishi, C., Hymel, S., Zumbo, B. D., & Li, Z. (2010). Do school bullying and studentteacher relationships matter for academic achievement? A multilevel analysis.
Canadian Journal of School Psychology, 25, 1939. http://dx.doi.org/10.1177/0829573509357550.
Laursen, B., Hafen, C. A., Kerr, M., & Stattin, H. (2012). Friend influence over adolescent problem behaviors as a function of relative peer acceptance: To be liked is to be
emulated. Journal of Abnormal Psychology, 121, 8894. http://dx.doi.org/10.1037/a0024707.
Lo, Y., Mendell, N. R., & Rubin, D. B. (2001). Testing the number of components in a normal mixture. Biometrika, 88, 767778.
Luckner, A. E., & Pianta, R. C. (2011). Teacherstudent interactions in fifth grade classrooms: Relations with children's peer behavior. Journal of Applied Developmental
Psychology, 32, 257266. http://dx.doi.org/10.1016/j.appdev.2011.02.010.
Malecki, C. K., & Elliott, S. N. (2002). Children's social behaviors as predictors of academic achievement: A longitudinal analysis. School Psychology Quarterly, 17, 123.
Mikami, A. Y., Gregory, A., Allen, J. P., Pianta, R. C., & Lun, J. (2011). Effects of a teacher professional development intervention on peer relationships in secondary
classrooms. School Psychology Review, 40, 367385.
Muthn, B. O. (2004). Latent variable analysis: Growth mixture modeling and related techniques for longitudinal data. In D. Kaplan (Ed.), The SAGE handbook of
quantitative methodology for the social sciences (pp. 345368). Thousand Oaks, CA: Sage.
Muthn, L. K., & Muthn, B. O. (19972012). Mplus user's guide. Los Angeles, CA: Muthn and Muthn.
Nagin, D. S. (2005). Group-based modeling of development over the life course. Cambridge, MA: Harvard University Press.
Nylund, K. L., Asparouhov, T., & Muthn, B. O. (2007). Deciding on the number of classes in latent class analysis and growth mixture modeling: A Monte Carlo simulation study. Structural Equation Modeling, 14, 535569.
Partin, T. C. M., Robertson, R. E., Maggin, D. M., Oliver, R. M., & Wehby, J. H. (2010). Using teacher praise and opportunities to respond to promote appropriate student
behavior. Preventing School Failure: Alternative Education for Children and Youth, 54, 172178.
Pas, E. T., & Bradshaw, C. P. (2014). What affects teacher ratings of student behaviors? The potential influence of teachers' perceptions of the school environment and
experiences. Prevention Science, 15, 940950. http://dx.doi.org/10.1007/s11121-013-0432-4.
Pas, E. T., Bradshaw, C. P., Hershfeldt, P. A., & Leaf, P. J. (2010). A multilevel exploration of the influence of teacher efficacy and burnout on response to student problem
behavior and school-based service use. School Psychology Quarterly, 25, 1327.
Pas, E. T., Bradshaw, C. P., & Mitchell, M. M. (2011). Examining the validity of office discipline referrals as an indicator of student behavior problems. Psychology in the
Schools, 48, 541555. http://dx.doi.org/10.1002/pits.20577.
Peterson Nelson, J. A., Young, B. J., Young, E. L., & Cox, G. (2009). Using teacher-written praise notes to promote a positive environment in a middle school. Preventing
School Failure: Alternative Education for Children and Youth, 54, 119125. http://dx.doi.org/10.1080/10459880903217895.
Pianta, R. C., Hamre, B. K., & Allen, J. P. (2012). Teacherstudent relationships and engagement: Conceptualizing, measuring, and improving the capacity of classroom
interactions. In S. L. Christenson, A. L. Reschly, & C. Wylie (Eds.), Handbook of research on student engagement (pp. 365386). New York, NY: Springer. http://dx.doi.
org/10.1007/978-1-4614-2018-7_17.
Ramaswamy, V., Desarbo, W. S., Reibstein, D. J., & Robinson, W. T. (1993). An empirical pooling approach for estimating marketing mix elasticities with PIMS data.
Marketing Science, 12, 103124.
Reinke, W. M. (2006). The classroom check-up: Advanced tools for effective coaching. In R. Sprick, J. Knight, W. M. Reinke, & T. McKale (Eds.), Coaching for positive
classrooms: Supporting teachers with classroom management (pp. 139167). Eugene, OR: Pacific Northwest Publishing.
Reinke, W. M., Herman, K. C., & Stormont, M. (2013). Classroom-level positive behavior supports in schools implementing SW-PBIS: Identifying areas for enhancement.
Journal of Positive Behavior Interventions, 15, 3950. http://dx.doi.org/10.1177/1098300712459079.
Reinke, W. M., Lewis-Palmer, T., & Martin, E. (2007). The effect of visual performance feedback on teacher use of behavior-specific praise. Behavior Modification, 31(3),
247263. http://dx.doi.org/10.1177/0145445506288967.
Reinke, W. M., Lewis-Palmer, T., & Merrell, K. (2008). The classroom check-up: A classwide teacher consultation model for increasing praise and decreasing disruptive
behavior. School Psychology Review, 37, 315332.
Roland, E., & Galloway, D. (2002). Classroom influences on bullying. Educational Research, 44, 299312. http://dx.doi.org/10.1080/0013188022000031597.
Rusby, J. C., Crowley, R., Sprague, J., & Biglan, A. (2011). Observations of the middle school environment: The context for student behavior beyond the classroom.
Psychology in the Schools, 48(4), 400415.
Rusby, J. C., Taylor, T., & Milchak, C. (2001). Assessing school settings: Interactions of students and teachers (ASSIST) observation system. Unpublished manual
Schwartz, G. (1978). Estimating the dimension of a model. The Annals of Statistics, 6, 461464.
Skiba, R. J., Homer, R. H., Chung, C. -G., Rausch, M. K., May, S. L., & Tobin, T. (2011). Race is not neutral: A national investigation of African American and Latino
disproportionality in school discipline. School Psychology Review, 40, 85107.
Skiba, R. J., Michael, R. S., Nardo, A. C., & Peterson, R. L. (2002). The color of discipline: Sources of racial and gender disproportionality in school punishment. The Urban
Review, 34, 317341.
Skinner, E. A., & Belmont, M. J. (1993). Motivation in the classroom: Reciprocal effects of teacher behavior and student engagement across the school year. Journal of
Educational Psychology, 85, 571581.
Sugai, G., & Horner, R. H. (2002). Introduction to the special series on positive behavior support in schools. Journal of Emotional and Behavioral Disorders, 10, 130135.
Sugai, G., & Horner, R. H. (2006). A promising approach for expanding and sustaining school-wide positive behavior support. School Psychology Review, 35, 245259.
Sutherland, K. S., Adler, N., & Gunter, P. L. (2003). The effect of varying rates of opportunities to respond to academic requests on the classroom behavior of students
with EBD. Journal of Emotional and Behavioral Disorders, 11, 239248.
Sutherland, K. S., & Wehby, J. H. (2001). Exploring the relationship between increased opportunities to respond to academic requests and the academic and behavioral
outcomes of students with EBD: A review. Remedial and Special Education, 22, 113121.
Sutherland, K. S., Wehby, J. H., & Copeland, S. R. (2000). Effect of varying rates of behavior specific praise on the on-task behavior of students with EBD. Journal of
Emotional and Behavioral Disorders, 8, 28.
Walker, H., Horner, R. H., Sugai, G., Bullis, M., Sprague, J., Bricker, D., et al. (1996). Integrated approaches to preventing antisocial behavior patterns among school-age
children and youth. Journal of Emotional and Behavioral Disorders, 4, 194209.