Behavior Support Strategies in Singapore Preschools: Practices and Outcomes

PW-PBS Universal Practices

The universal tier of PW-PBS consists of two levels: building nurturing and responsive relationships with children and families, and creating high quality supportive environments (Fox et al., 2003). Class-wide positive behavior support practices that have been shown to help teachers build nurturing and responsive relationships with children and families include greeting every child at the door by name, responding to children’s conversations, offering praise and encouragement, conducting home visits, and routinely sharing information about children with parents (Fox et al., 2003; Fox, Hemmeter, Synder, Binder, & Clarke, 2011; Powell, Dunlap, & Fox, 2006; Steed & Pomerleau, 2008). Similarly, practices that teachers can adopt to create high quality supportive classrooms include organizing the materials in the classrooms to minimize clutter, defining expectations (globally stated standards of conduct, such as “be safe,” “be kind”), identifying behavioral expectations that align with each classroom routine (e.g., “walk to our cubbies and sit on mat” upon arrival), teaching behavioral expectations, gaining the attention of all students at the beginning of a lesson or transition from one activity to the next, acknowledging children’s appropriate behaviors, and responding appropriately to inappropriate behaviors (Fox et al., 2003; Fox et al., 2011; Powell et al., 2006; Steed & Pomerleau, 2008).

Researchers have provided guidelines for the use of some of the practices to achieve optimal effects. Classroom rules should be clearly defined, stated in simple and positive language with visual supports, publicly posted at children’s eye level, and kept to a maximum of five (Steed & Pomerleau, 2008). Behavioral expectations need to be actively taught to children several times throughout the year by modeling, practicing, and acknowledging the expected behaviors (Reinke, Herman, & Sprick, 2011; Steed & Pomerleau, 2008). The amount of acknowledgment given to children for displaying expected behaviors should be more than the amount of corrections, reprimands, or disapprovals given for misbehaviors (Steed, Pomerleau, & Horner, 2008). The recommended ratio of positive to negative interactions is 4:1 (Reavis, Jenson, Kukic, & Morgan, 1993). In addition, specific praise should be used more than general praise (Good & Brophy, 2003).

Research on teachers’ use of universal PW-PBS practices at a classroom-wide level is limited. Carter, Van Norman, and Tredwell (2011) investigated teachers’ use of universal practices in an early childhood program in the United States before and after PW-PBS implementation using the Preschool-Wide Evaluation Tool (Pre-SET; Steed et al., 2008) and found that although approximately 70% of the teachers acknowledged children’s appropriate behaviors at the preimplementation phase, only 50% defined expectations for their classrooms. Furthermore, only 25% created an organized and predictable classroom environment and taught behavioral expectations. At the post implementation phase, the percentage of teachers acknowledging children’s appropriate behaviors dropped to 52%. However, close to 90% of the teachers defined expectations for their classrooms; 30% of them taught behavioral expectations and about 60% were reported to have created an organized and predictable classroom environment.

Benedict, Horner, and Squires (2007) also used the Pre-SET to evaluate 15 preschool classrooms in the United States where the lead teachers had attended a workshop in positive behavior support and received support from a behavior consultant and reported that about 67% of the classrooms had an organized and predictable environment. However, no more than 40% of the classrooms had publicly posted expectations and a matrix of classroom rules for classroom routines. In addition, children were taught behavioral expectations in only 22.67% of the classrooms and acknowledged for their appropriate behaviors in only 30%. These rates, however, contrast favorably with findings of Kim and Stormont (2012) who examined strategies that 34 South Korean teachers of children 3 to 5 years of age used to address problem behavior in natural settings. Specifically, precorrection and behavior-specific praise were used at low rates, and redirection and reprimand/punishment were used at high rates. In addition, the ratio of positive to negative interactions was nonoptimal at 1:7.

Universal PW-PBS practices for early childhood settings do not include practices related to academic instruction. However, they are important to include in Singapore and some Asian countries, where the preschool curriculum is academically demanding. Key universal academic instruction practices for elementary school students that can be included in PW-PBS at the preschool level are providing students with numerous opportunities to respond (OTR) to academic content, and using effective error correction (i.e., telling, showing, or demonstrating the correct answer rather than saying “no” or “wrong”) (Missouri Schoolwide Positive Behavior Support, 2016; Simonsen, Fairbanks, Briesch, Myers, & Sugai, 2008).

Rate of Disruptive Behavior and Level of Academic Engagement

To date, researchers have not established the rates of disruptive behavior that are acceptable, or the optimal levels of academic engagement, in a preschool classroom. Although researchers have examined disruptive behaviors of preschoolers with high rates of externalizing behaviors, no study has systematically and comprehensively examined the rates of disruptive behavior or the levels of academic engagement of all children in a preschool general education classroom (e.g., Feil et al., 2015; Walker et al., 1998).

Purpose of Study

The purposes of this study were to find out more about Singaporean teachers’ use of universal PW-PBS practices at the classroom-wide level, and the rate of disruptive behavior and level of academic engagement in Singapore preschool general education classrooms for children 3 to 6 years of age. Direct observational methods were mainly employed, as previous research has demonstrated that teacher self-reports often do not reflect actual practices (Noell et al., 2005). The research questions were as follows:

Method

Measures

Classroom Ecology Checklist (Consultation Version)

The Classroom Ecology Checklist (Consultation Version; Reinke et al., 2011) consists of 25 items reflecting elementary school teachers’ use of universal classroom management practices and student behaviors in the classrooms. As many universal PW-PBS practices are similar to elementary school universal classroom management practices, this measure is appropriate for use in preschool setting with the exclusion of one item, the system for turning in completed work and retrieving graded materials. In addition, only items relating to teachers’ use of universal classroom management practices (19 items) were used. The practices are categorized into six domains: classroom structure, behavioral expectations, instructional management, interacting positively, and responding to appropriate and inappropriate behaviors. With two exceptions, data on all domains were obtained through direct observation. The item on teacher actively teaching classroom rules and expectations several times throughout the year was assessed via teacher self-report if it was not directly observed, and the amount of class time allocated to academic instruction was assessed based on review of the classroom schedule. Activities that were considered academic instruction in this study consisted of large group academic instruction, arts and crafts, music and movement, second language instruction, and current affairs discussion. All items are categorical, with most requiring either a yes, no, or sometimes/somewhat response.

Classroom Check-Up 10-Minute Classroom Observation Form

The Classroom Check-Up 10-Minute Classroom Observation Form (CCU 10-Minute Observation Form; Reinke et al., 2011) records the number of OTRs provided by the teacher to a student or group of students, correct academic responses from students, teacher use of general praise, behavior-specific praise, explicit reprimands, harsh reprimands, and student disruptive behaviors, during a 10-min academic instruction period. Operational definitions of these variables are provided in Table 1. The form also enables the computation of the percentage of correct academic responses and a ratio of positive to negative interactions and has been found to be sensitive to change over time. The overall reliability of this instrument was found to be 88% before intervention (Time 1) and 90% after initial intervention (Time 2). All variables on this instrument were also related to one another in the expected direction with correlation ranging from r = −.29, p < .01 to r = .99, p < .01 at Time 1 and from r = −.52, p < .01 to r = .98, p < .01 at Time 2 (Reinke, Stormont, Herman, Wachsmuth, & Newcomer, 2015).

OPEN IN VIEWER

Table 1.

Operational Definitions of Variables in CCU 10-Minute Observation Form.

Variable	Definition
Opportunities to respond	An academic question, statement, or gesture made by the teacher that seeks a student or group of students’ response, for example, “what shape is this?” “Mary, what letter is this?” and teacher gestures to a student for a response.
Correct academic response	A correct response provided to an OTR.
General praise	A verbal statement or gesture that indicates approval of behavior and does not name specific behavior, for example, “well done,” “good job,” high-five, tokens, stickers.
Behavior-specific praise	A verbal statement with or without gestures that indicates approval of behavior and name a specific behavior, for example, “thank you for raising your hand” with a sticker given.
Explicit reprimand	A brief verbal comment or gesture that indicates disapproval of behavior, for example, “class, eyes on me” (when students are talking), “stop” gesture.
Harsh reprimand	A verbal comment or gesture using loud, harsh, sarcastic, or critical tone that indicates disapproval of behavior, for example, “how many times do I have to tell you all about the rules,” rolling eyes at students after a reprimand.
Disruptive behaviors	Any statement or action by an individual student or group of students that could be predicted to disrupt or interfere with ongoing classroom activities for the teacher and/or one or more peers; disruptions are to be coded as discrete events.

Note. CCU = Classroom Check-Up; OTR = opportunities to respond.

Classroom Check-Up 5-Minute Academic Engagement Observation Form

The Classroom Check-Up 5-Minute Academic Engagement Observation Form (CCU 5-Minute Academic Engagement Observation Form; Reinke et al., 2011) measures the amount of time the students in the classrooms are engaged over a 5-min period during an academic instruction period. It uses a momentary time sampling procedure. One student is observed every 5 sec, and all students in the classroom are observed sequentially to determine whether they are on-task or off-task at that moment. This continues, repeating students as necessary for 5 min. In this study, every observation started with the child on the left and ended on the right for children seated in U shape or circle. For children seated in rows, observation started with children in the first row going from left to right and ended with the last row.

Results

Research Question 1: What Universal PW-PBS Practices Are Teachers Using at a Classroom-Wide Level?

Classroom structure

Figure 1 displays the percentage of classrooms in which the teachers adopted practices related to classroom structure. More than half of the classrooms (65.6%) had traffic patterns that were clearly defined and allowed movement without disrupting others. About half of the classrooms (53%) had furniture arrangements that allowed students to be seen at all times and teachers to have easy access to all areas of the classroom. However, less than a third of the classrooms (28.1%) had materials that were clearly labeled, easily accessible, and organized to minimize clutter.

OPEN IN VIEWER

Figure 1.

Percentages of teachers adopting practices related to classroom structure.

Classroom expectations

Almost all (90.6%) of the classrooms had simple expectations, but none had the routines and expectations displayed. Nevertheless, 84.4% of the teachers were reported to teach rules and expectations several times throughout the year. These teachers were also observed to teach such expectations either three or four out of four times that they were observed.

Instructional management

Figure 2 shows the percentage of teachers who used specific practices of instructional management. Only one teacher (3.1%) provided an optimal number of OTRs (i.e., four to six opportunities per min for new materials and nine to 12 per min for drill and practice) to children (both individual and group combined). The mean rate of OTR provided per min to students (both individual and group combined) across 32 teachers was 3.64 (SD = 0.99), ranging from 1.88 to 5.15 per min. About 66% of the teachers provided more than three OTRs per min. In providing OTRs, less than half of the teachers (37.5%) elicited individual responses from the majority of children when asking for both group and individual responses.

OPEN IN VIEWER

Figure 2.

Percentages of teachers using specific practices of instructional management.

Most of the teachers (81.30%) used effective error corrections although less than half (46.90%) consistently gained the attention of all students at the beginning of a lesson or transition from one activity to the next. Finally, with regard to the classroom schedule, a high percentage of the teachers (93.80%) had 70% or more of their class time allocated to academic instruction.

Interacting positively

A large proportion of the teachers (68.80%) did not provide noncontingent attention to every student in the classrooms during the observation period. In addition, the majority of the teachers (96.90%) provided a positive to negative interactions ratio (ratio of overall praise to reprimands) of less than 4:1. The mean ratio across 32 classrooms was at 1:3.03. Only one teacher provided the optimal ratio of 4:1.

Responding to appropriate behavior

None of the teachers had systems for documenting and rewarding appropriate child behavior. The mean rate of behavior-specific praise given per min across 32 teachers was very low, at 0.06 (SD = 0.09) with a range of 0 to 0.45. Reviewing the data across the whole sample revealed that more than half of the teachers were delivering behavior-specific praise at the rate of 0 to 0.05 per min, and none of the teachers were using behavior-specific praise more than general praise. In addition, even the use of general praise was low, at 0.32 per min (SD = 0.23) with a range of 0.03 to 0.90. About 66% of the teachers gave general praise at a rate of less than 0.32 per min.

Responding to inappropriate behavior

None of the teachers had a documented system to address specific behavioral violations, and only one teacher used a continuum of consequences to discourage rule violations including praising others, planned ignoring, explicit reprimand, proximity, nonverbal cue, reteach, providing choice, and logical consequences. The majority of the teachers (96.9%) were not consistent in delivering reprimands to children who display inappropriate behaviors.

The mean rate of explicit reprimands used by the teachers was higher than either behavior-specific praise or general praise, at 0.57 per min (SD = 0.35) with a range of 0.08 to 1.55 (see Figure 3). Overall, the use of harsh reprimands was low, at 0.01 per min (SD = 0.03) with a range of 0 to 0.18 (see Figure 3).

OPEN IN VIEWER

Figure 3.

Rate of praise and reprimand delivered by teachers.

Discussion

The aim of this study was to assess the state of classroom management in Singapore preschools, specifically, teachers’ use of universal PW-PBS practices at the class-wide level, the levels of disruptive behavior and academic engagement, and the relationships between the rates of disruptive behavior, and praise and reprimands. PW-PBS has not been systematically implemented in Singapore preschools, but the majority of the teachers have received training in PW-PBS universal practices. Overall, the findings indicated that while a few of the universal practices were adopted in the majority of the classrooms, many of the core practices were not evident. The results also showed relatively high rates of disruptive behavior, which were associated with high rates of reprimands, consistent with previous studies (Van Acker et al., 1996; Wehby et al., 1995). Contrary to many previous studies (e.g., Sutherland et al., 2000), this study did not show a relationship between rates of disruptive behavior and praise, possibly because teachers in this study largely delivered praise to children for their correct academic responses rather than appropriate behaviors. Findings also revealed that although the overall level of academic engagement across all classrooms was relatively high, a third of them had low academic engagement levels. In sum, the observation data suggest that preschools in Singapore are likely to benefit from PW-PBS implementation and that existing teacher training on universal PW-PBS practices might need to be enhanced.

Some universal PW-PBS practices were evident in more than 80% of the participating classrooms. These included the teaching of rules and expectations to the children several times throughout the year, use of effective error corrections, and having more than 70% of class time allocated to academic instruction. More teachers in this study taught rules and expectations as compared with the study by Carter et al. (2011) at both pre- and post implementation phases and by Benedict et al. (2007), where the teachers had been trained in PBS. The higher implementation rate in this study could be because there is a strong culture in Singapore preschools to teach children rules and expectations. Nonetheless, teacher training should still emphasize the need to review rules and expectations with children and remind children of them daily (Hemmeter, Fox, & Synder, 2014).

Perhaps unsurprisingly, given the absence of relevant PW-PBS training, a number of core universal practices were used by less than 40% of the teachers. None of the teachers had a system for documenting and for rewarding appropriate child behavior. Also, none of the teachers had a documented system to address specific behavioral violations. These practices were not as widely implemented in the current study as in Carter et al. (2011) and Benedict et al. (2007). Although greater than in Singapore, the adoption rates of these practices in the Carter et al. study were less than 70%, even at the post PW-PBS implementation phase. We suggest, therefore, that most teachers are likely to need support in implementing these practices in the classrooms. Given that dissemination of knowledge/information (e.g., theory, descriptions of skills) alone has been shown to be insufficient to result in adoption of effective practices by classroom teachers, it may be beneficial for teacher training to employ additional methods that are known to help increase adoption in the classrooms, such as practice, performance feedback, or coaching (Fullerton et al., 2009; Gage, MacSuga-Gage, & Crews, 2017; Joyce & Showers, 2002; Mathews, McIntosh, Frank, & May, 2014).

In addition, only one teacher was observed to use a continuum of consequences to manage rule violations, and less than a third of the teachers provided noncontingent attention to every child in the classroom. This may in part be a function of observations in this study being restricted to large group instructional time. Further research exploring instructional and interactional patterns during other classroom routines is clearly warranted.

The use of general praise was also found to be low, and delivery of specific praise, a recommended practice, even lower. The overall use of praise was also much lower than the use of reprimands; and thus, with one exception, all teachers provided nonoptimal positive to negative interaction ratios. Despite methodological differences, this finding is similar to that of Kim and Stormont (2012) in a study conducted in South Korea. However, compared with preschool teachers in the United States, praise rates of Singaporean and South Korean teachers have been found to be much lower (Floress, Berlinghof, Rader, & Riedesel, 2017), suggesting that there may be cultural differences. In Asian cultures, it is common not to acknowledge appropriate behavior in children, as educators have the philosophy that it is expected. Besides using instructional procedures that have been shown to increase teachers’ praise rates during teacher training, such as performance feedback, video self-monitoring, and coaching (Hawkins & Heflin, 2011; Myers, Simonsen, & Sugai, 2011; Thompson, Marchant, Anderson, Prater, & Gibb, 2012), it may also be important to explore and address teachers’ acceptability of using praise.

A third of these preschool teachers did not provide the children with at least three OTRs per min, a guideline recommended for elementary school students by MacSuga-Gage and Simonsen (2015). Researchers have yet to recommend a guideline for preschoolers, but it is possible that more than three OTRs are required to maintain engagement, as younger children have a shorter attention span. Furthermore, in providing OTRs to the children, about 62% of the teachers did not provide the majority of children with individual OTR. To help teachers meet the recommended guidelines, teacher training could include certain effective strategies, such as using a tracking system to ensure that all children are provided with individual OTR, asking drill-and-practice questions, and breaking complex problems into smaller chunks (Reinke et al., 2011). Researchers are encouraged to continue studying the cultural compatibility of specific instructional procedures like increased delivery of noncontingent attention, praise, and OTR.

One limitation of this study is that observations were only carried out during large group instruction time, which might have limited the opportunities to observe for teachers’ use of various practices. Future work could consider carrying out observations during other class routines. It might also be useful to carry out observations early in the school year, when use of certain practices (e.g., teaching expectations) might be more intense. Another limitation of this study is the small sample size in terms of the number of participating teachers, which may therefore limit the generalizability of the findings. The participating schools and teachers being volunteer participants may also limit the generalizability of the findings. Furthermore, no validity studies have yet been conducted to assess measurement invariance across cultural contexts for the measures used in the study; however, this limitation does not apply to the direct observational measures used. Despite these limitations, this study provides a starting point to understanding and consequently improving classroom management in Singapore preschools. Our findings suggest that PW-PBS implementation in Singapore preschools is likely to be beneficial and that there is a need to enhance current teacher training in universal practices. This study also adds to the literature of PW-PBS by providing insight into the implementation of a range of universal PW-PBS practices at a class-wide level in typical classrooms.