Abstract
The present study aimed to identify a way for educators to improve the accuracy of their praise and reprimand reflections to ultimately improve their ability to set, monitor, and evaluate their use of praise and reprimand. To do this, teachers’ natural use of praise and reprimand (in the absence of intervention) were compared with their perceived use. A 20-min direct observation was collected from 66 middle and high school teachers to obtain praise and reprimand rates. Following the observation, teachers reported their perceived use of praise and reprimand. A t test and analysis of variance (ANOVA) were used to determine differences between praise and reprimand types. Correlations were used to determine the relation between perceived and actual praise and reprimand use. Statistical results indicated teachers used more general praise (GP) than behavior-specific praise and more mild reprimand than any other reprimand type. Teachers’ actual and perceived use of GP were positively correlated, as were teachers’ actual and perceived use of mild, gestural, and total reprimand. Furthermore, teachers with a greater difference between their actual and perceived praise also had a greater difference between their actual and perceived reprimand use. Future research and implications of these findings are discussed.
Effective classroom management, which includes evidence-based practices (EBPs), is positively related to student participation and academic success (Reinke et al., 2008). Proactive EBPs include praise, precorrection, and opportunities to respond (Smith et al., 2011). Proactive EBPs increase the probability and strengthen the occurrence of appropriate behavior. Because teachers report needing additional support and training related to classroom management (Reinke et al., 2011), it is important to find ways to bolster teachers’ training related to EBPs, such as praise.
Praise and Reprimand
There are two types of praise, general praise (GP) and behavior-specific praise (BSP; Jenkins et al., 2015). Both types signal approval (Reinke et al., 2008); however, GP is nonspecific, whereas BSP identifies what was approved (Floress & Jenkins, 2015). The statement, “Excellent!” is an example of GP because it expresses approval without identifying what was excellent. However, “Excellent job providing evidence” is an example of BSP because approval “Excellent job” is paired with what was approved “providing evidence.” When used effectively, BSP increases student-appropriate behavior (Pas et al., 2015) because students learn what behavior led to approval. BSP also prevents inappropriate behavior from occurring and can alter a situation before a problem escalates (Clunies-Ross et al., 2008).
High rates of teacher praise and low rates of teacher reprimand can positively affect classroom climate and classroom management (Caldarella et al., 2020; Spilt et al., 2016). Teacher reprimands are verbal comments or gestures indicating disapproval (Reinke et al., 2008). An explicit or mild reprimand is concise, brief, and delivered in a tone typical for the setting (e.g., “You need to sit,” spoken to a student who is expected to be seated; Reinke et al., 2015). A medium reprimand is delivered in a sarcastic or critical tone (e.g., Did I say to talk to a friend?). A harsh reprimand can also be sarcastic or critical but is delivered in a loud or harsh tone that lasts at least 30 s (Reinke et al., 2015). Last, a gestural reprimand is nonverbal disapproval (e.g., finger to lips to communicate “quiet!” Floress et al., in press).
Unfortunately, teachers use more reprimands than praise (Reinke et al., 2013; Stormont et al., 2007), especially with challenging students (Pas et al., 2015; Shook, 2012). This is particularly problematic because reprimands do not teach students what to do. Furthermore, reprimands can exacerbate inappropriate behavior because when problem behavior is readily attended to, it may be inadvertently strengthened (Downs et al., 2019; Gable et al., 2009).
Teacher BSP Training and Natural Use of Praise
When teacher BSP increases, student on-task behavior increases (Caldarella et al., 2020; Downs et al., 2019) and student disruptive behavior decreases (Reinke et al., 2008; Sutherland et al., 2000). In addition, when teachers receive BSP training, their use of BSP improves (Zoder-Martell et al., 2019). However, how teachers are trained to use praise may influence their accurate, reliable, and sustained use of BSP (Dufrene et al., 2014; Hiralall & Martens, 1998). In the absence of professional development (PD), teachers praise infrequently, and rates decline as grade levels increase (Floress et al., 2018; Jenkins et al., 2015). Therefore, it is important to find easy-to-implement, sustainable ways to improve teachers’ use of BSP, and examining natural praise (in the absence of intervention) may guide training recommendations.
Few studies have examined teachers’ natural use of praise and reprimand in general education (Jenkins et al., 2015) and most have occurred in elementary classrooms (Floress & Jenkins, 2015; Floress et al., 2018). Scott et al. (2011) examined teacher praise and reprimand across two elementary and two high schools; however, rates were combined across schools, so secondary rates (alone) were not reported. Other studies suggest secondary teachers praise infrequently, and rates decline as grade levels increase; however, these studies are dated. For example, White (1975) reported middle school teachers used 17.1 praises and 28.1 reprimands per hour (0.61–1) and high school teachers used 8.4 praises and 15.0 reprimands per hour (0.56–1). Heller and White (1975) and Wheldall et al. (1989) reported similar ratios.
It is unclear whether today’s secondary teachers’ praise and reprimand rates would be consistent with past research (Heller & White, 1975; Wheldall et al., 1989; White, 1975). Considering the push to implement Positive Behavior Intervention Supports (PBIS), a framework intended to promote EBPs (e.g., using more praise than reprimand; Loveless, 1996), it is important to know how often secondary teachers use praise. Examining natural praise and reprimand may guide training recommendations, like having teachers reflect on their use of praise and reprimand to support continued BSP use. Therefore, the aim of this study is twofold: (a) to extend the literature among secondary teachers’ use of praise and reprimand and (b) to identify a way for educators to improve the accuracy of their praise and reprimand reflections to ultimately improve their ability to set, monitor, and evaluate their praise and reprimand use.
Teachers’ Reported and Actual Use of Recommended Strategies
Asking teachers about their classroom management practices is one way to guide intervention recommendations (Witt et al., 1996). However, research suggests discrepancies between teacher-reported and actual use of strategies (Wickstrom et al., 1998). Noell et al. (2005) examined teachers’ treatment plan implementation following consultation and teachers uniformly reported high levels of treatment integrity, which was inconsistent with direct assessment. In another study, teachers reported using more positive statements following consultation; but this was inconsistent with direct observations (Robbins & Gutkin, 1994).
These studies emphasize that teacher-reported use and actual use of a strategy may be inconsistent; therefore, reported measures should be used with additional outcome variables. Robbins and Gutkin (1994) indicated discrepancies may exist because it is difficult for a teacher to admit (in a postconsultation interview) they are no longer using a recommended strategy. Wickstrom et al. (1998) concluded differences in their study were influenced by the rigor of the strategy rather than the severity of the problem, treatment acceptability, or collaboration.
Few studies have examined teacher-reported and actual strategy use in the absence of intervention (Kim & Stormont, 2016). Examining what EBP teachers (not seeking consultation) use is important because intervention research depends on teachers with low baseline rates, which may not provide a clear picture of teachers’ use of EBPs in general. Furthermore, studying natural praise and reprimand may guide sustainable training activities, such as reflecting on EBPs, a less intensive training approach compared with ongoing, in-classroom supports.
PD and Perceptions of Praise and Reprimand
Performance feedback (PF) and self-monitoring are commonly used to train teachers to use BSP. Zoder-Martell et al. (2019) synthesized 28 single-case research studies examining training used to increase teachers’ BSP. Meta-analytic results indicated training had a strong effect on teachers’ use of BSP, and PF was used in 84% of the studies. Furthermore, when teachers received feedback regarding treatment plan implementation, they continued to implement plans accurately compared with teachers who did not receive feedback (Noell et al., 2005). Interestingly, teachers reported high levels of satisfaction and treatment integrity regardless of condition. This was inconsistent with direct measures, which indicated implementation and student behavior change varied substantially across conditions. These findings again suggest teachers’ perceptions may be inconsistent with direct measures.
Antecedent strategies such as self-management via PF may be a superior form of praise training (Zoder-Martell et al., 2019) because teachers become aware of their current performance through self-evaluation and feedback in comparison with a set criterion (Bandura, 1991). Furthermore, teachers can continue to self-monitor their use of praise and reprimand by planning for generalization in the classroom setting (when consultation is over; Horner et al., 1984). Teacher–child interaction therapy (TCIT), a modified version of parent–child interaction therapy (McNeil & Hembree-Kigin, 2010), uses in vivo coaching and feedback to increase teachers’ use of TCIT skills. Teachers are trained to mastery (e.g., 10 BSP; 10 descriptions; 10 reflections within 5 min) with the intention that overlearning will assist in generalizing skills to the classroom (Lyon et al., 2009). Along with overlearning, using relevant stimuli and planning for generalization assist in implementing high rates of praise. Floress et al. (2020) describes using praise on a fixed-interval schedule, as an antecedent-based practice (e.g., setting a periodic timer on a smartwatch for 3 min and delivering BSP when the watch vibrates).
Kim and Stormont (2016) is one of the few studies to compare teacher-reported and actual use of EBPs outside of consultation. Thirty-four Korean early-childhood educators reported to implement more proactive strategies (i.e., precorrection and BSP) than reactive (i.e., redirection or reprimand). However, educators used more redirection than precorrection and were observed to use BSP the least. It is unclear whether these results would be replicable to secondary educators in the United States. It is also unclear whether having teachers reflect on their praise and reprimand use may assist in setting, monitoring, and evaluating their use of BSP overtime.
Literature Summary and Impact of Proposed Research
Teachers’ knowledge and sustained use of EBPs, such as praise, are critical to student success (Reinke et al., 2008). Traditional PD is limited in that educators lack the knowledge to implement EBPs or abandon their use over time. Although educators are encouraged to use more praise than reprimand (i.e., 4:1; Loveless, 1996), existing studies suggest teachers use praise infrequently, and praise to reprimand ratios (PRRs) are low. Even with training, BSP improves but rates are not sustained. Helping educators engage in low-intensity, sustainable activities, such as reflecting on their praise and reprimand use, may increase effective implementation. However, reflections would only be effective if they were valid. This study aims to extend the literature among secondary teachers’ natural use of praise and reprimand and to identify a way to improve educators’ accuracy of praise and reprimand reflections to ultimately improve setting, monitoring, and evaluating this EBP. The following questions were posed:
Method
Participants and Setting
To ensure observations were consistent across settings and participants, inclusion criteria included teaching a general education class with at least 20 min of lecture-based instruction (e.g., math, English). Sixty-nine teachers agreed to participate; however, the final sample consisted of 66 middle and high school, general education teachers from Illinois (95.6% response rate) from seven middle, six high, and two middle/high schools. Three teachers were excluded because one taught a class entirely in Spanish, one did not return survey forms, and one’s use of reprimand exceeded the survey form. The school communities ranged from rural to microurban (see Table 1A in supplemental materials). Of the 66 participants, 25 were middle school teachers and 41 were high school teachers (see Table 1). Participant age ranged from 23 to 67 years (M = 39 years). Most participants were female (71%), Caucasian (98%), and held a master’s degree (68%). Teaching experience was well distributed across the sample and approximately half of teachers reported they took a preservice, classroom management course. Six schools implemented PBIS and six implemented positive behavior supports, but not PBIS specifically (see Table 2A in supplemental materials). None of the schools used the School-Wide Evaluation Tool or a similar tool to evaluate their PBIS or positive behavior supports. The first 40 participants received US$5 gift cards and all others received chocolate.
Teacher and Classroom Demographics for the Study.
Note. Classroom difficulty rating—teachers were asked to “rate the behavioral difficulty of the class observed (as a whole) compared to other classes you have taught in the past” using a 1 (much less difficult) to 5 (much more difficult) scale. Behavior Management Class—teachers were asked “Have you taken an undergraduate or graduate course that focuses on managing student behavior.”
Materials and Instruments
Praise and reprimand data collection form
This form was created by the authors and was used to collect praise and reprimand frequency data during a 20-min observation. The form had 20, 1-min intervals and within each interval, praise and reprimand were broken down by type. Praise was recorded as BSP or GP and reprimand was recorded as mild, medium, harsh, or gestural. Observers used a cued audio tape that provided a verbal prompt (e.g., 1, 2, 3) to observe each of the 20, 1-min intervals. Praise and reprimand frequencies were recorded within each interval, along with the verbatim statement/gesture. Praise and reprimand frequencies were added across the 20 intervals.
Operational definition: Praise
Praise was coded as GP or BSP. GP included any nonspecific verbalization or gesture that expressed a favorable judgment on an activity, product, or attribute of the student (e.g., Great, thumbs-up). BSP included any specific verbalization or gesture that expressed a favorable judgment on an activity, product, or attribute of the student (e.g., I like your detailed writing; Floress & Jenkins, 2015).
Operational definition: Reprimand type
Reprimand was coded mild, medium, harsh, or gestural. Mild was any concise (brief) verbal comment (using a normal speaking tone) indicating disapproval. A verbal comment giving an instruction following student misbehavior or “redirection” was coded mild. Disagreeing with a student without sarcasm or a critical tone was coded mild (e.g., No, not now; Reinke et al., 2015). Medium was any verbal comment (using a sarcastic or critical tone) indicating disapproval. A question that was disapproving/had a mocking, rude, or critical tone was coded medium. Medium was also coded if the teacher disagreed with the child using a critical or sarcastic tone (e.g., Is that your best work? Did you even study? Floress et al., in press). Harsh was any verbal comment (using a louder than typical tone for the setting) indicating disapproval. Harsh was coded if the reprimand implied negative consequences or a prolonged discussion (30 s or longer) about misbehavior (e.g., Once more, and I’m sending you to the office; Reinke et al., 2015). Gestural was any nonverbal signal indicating disapproval (e.g., hands on hips), including physically guiding a student (e.g., student refuses to get up and teacher touches elbow to indicate “get up”; Floress et al., in press).
Interobserver agreement (IOA)
Of the 66 observations, 38% were collected using two observers. Interval-by-interval comparisons were completed using percentage agreement (i.e., the number of agreements within each interval was divided by the number of agreements within each interval plus the number of disagreements within each interval; Mudford et al., 2009). Average IOA for BSP was 98% (range = 90%–100%), GP 92% (range = 60%–100%), and total praise 95% (range = 80%–100%). Average IOA for mild reprimand was 95% (range = 78%–100%), medium 98% (range = 86%–100%), harsh 100% (range = 95%–100%), gestural 98% (range = 90%–100%), and total 98% (range = 90%–100%). Percentages indicated acceptable reliability among observers.
Teacher demographic questionnaire
After the observation, the teacher completed the demographic questionnaire, which included 13 items. Teachers indicated their sex, age, race, years of teaching experience, education level, type of teaching certificate, specialized training, grade/subject of the class observed, class makeup (e.g., only general education students), and a rating of classroom behavioral difficulty compared with other classes taught. Teachers were also asked, “Did you take a course in classroom management when you were pursuing your teaching certificate?”
Teacher perception of praise and reprimand form
After the observation, the teacher received the teacher perception of praise and reprimand form. This form was created by the authors to measure teacher’s perceived praise and reprimand use in 20 min. Prior to each rating, teachers read a definition for each praise or reprimand type. Definitions were the same as operational definitions used to collect data in the classroom. After each definition, teachers rated how many times they used the praise (or reprimand) type within a 20-min lecture by circling the frequency (0–20) on a number line.
Direct Observation Training
Five research assistants (two undergraduate and three graduate students) were trained to collect observation data. Assistants learned the operational definitions for praise and reprimand, reviewed examples and nonexamples, and asked questions. Each assistant coded three training videos and were required to demonstrate at least 80% IOA with a previously trained assistant before coding live. Before coding independently, at least 80% IOA in the classroom with a previously trained assistant was required. To combat observer drift, assistants attended weekly meetings led by the first author. Assistants reported IOA, shared coding issues, and asked questions. If IOA fell below 80%, discussion took place to ensure definitions were understood.
Procedures
Institutional review board (IRB) and school administrator approval were secured. Teachers received a flyer describing the study via email. To minimize reactivity, participants were told that the purpose of the study was to examine teachers’ natural use and perceptions of classroom management strategies (rather than praise and reprimand). Prior to the observation, participants indicated optimal times they were likely to lecture for at least 20 min. Most observations were completed in a single observation (one included two, 10-min observations). After the observation, assistants left the demographics and perceived use of praise and reprimand forms in the teacher’s mailbox. Assistants followed up with teachers via email to ensure they received the forms and reminded them to return the sealed/completed forms to the office at their convenience.
Data Analysis
To answer the first research question (What are the praise and reprimand rates among secondary, general education teachers?), praise and reprimand data were collected via direct observations. Frequency counts for praise and reprimand type were totaled from each 20-min observation. Praise and reprimand per minute and per hour were calculated. A t test for dependent means was used to determine whether teachers used more GP than BSP, and an analysis of variance (ANOVA) for repeated measures was used to determine whether middle and high school teachers used more mild reprimand compared with other reprimand types.
The second question (Are teachers’ perceived and actual use of praise consistent?) was analyzed using Pearson’s r correlational statistic. Pearson’s r is a correlation coefficient that is used to determine whether there is a relation between two variables (i.e., teachers’ perceptions of their use of praise and their actual use of praise). The correlation coefficient can range from a negative relation (−1) to a positive relation (1) depending on the type of relation between the two variables (Taylor, 1990). This analysis was used with each type of praise and total praise to determine whether there was a relation between perceived and actual praise among teachers.
The third question (Are teachers’ perceived and actual use of reprimand consistent?) was analyzed using Pearson’s r correlational statistic. This analysis was used with each type of reprimand and total reprimand to determine whether there was a relation between perceived and actual reprimand among teachers.
The final research question (Is there a relation between teachers’ PRR and their praise and reprimand accuracy?) was also analyzed using Pearson’s r correlational statistic. This analysis was used to determine whether there was a relation between three variables: actual and perceived praise difference, actual and perceived reprimand difference, and PRR. Praise difference and reprimand difference were computed by finding the absolute value between each teacher’s total perceived and total actual praise and reprimand. PRR was calculated by finding the greatest common divisor (gcd) between each participant’s total actual praise and total actual reprimand. PRRs were calculated by dividing each praise and reprimand actual total to the computed gcd. For example, one participant had nine total actual praises and three total actual reprimands. The gcd was 3; therefore, PRR was 3:1 (9/3 and 3/3).
Results
Observations
Sixty-six middle and high school teachers’ praise and reprimand frequencies were collected during 20-min observations for a total of 1,320 min (22 hr). A total of 496 incidents of praise and reprimand were recorded. Across the 66 teachers, there were 186 incidents of GP and 44 incidents of BSP. There were 197 incidents of mild, 28 medium, nine harsh, and 32 gestural reprimand. One teacher was excluded from the analysis because her mild reprimand rate exceeded the range on the form, and her total reprimand (50 mild, three medium, 14 harsh, 20 gestural) was much higher than any other participant (see “Limitations” section for a discussion).
Praise and Reprimand Rates
To answer Research Question 1 (What are the praise and reprimand rates among secondary, general education teachers?), praise and reprimand frequencies were collected from each 20-min observation. Across 66 teachers, an average rate of 10.65 total praise per hour and an average rate of 12.09 total reprimand per hour were observed. Table 2 further breaks down praise and reprimand rates by type. Across 66 participants, the average PRR was 0.86 to 1. Twenty participants had more praises than reprimands, three had ratios reflecting the recommended 4:1, and four had ratios exceeding 4:1. To determine whether middle and high school teachers used more GP than BSP, a t test for dependent means was conducted. At an alpha level of .05, GP (M = 2.82, SD = 3.41) was used significantly more than BSP (M = 0.67, SD = 1.71), t(65) = 5.37, p < .001 (one tailed), d = 1.26 (large effect).
Teachers’ Mean and Range of Observed Rate of Praise and Reprimand Statements per Hour.
Note. Rate per minute is provided in parentheses. BSP = behavior-specific praise; GP = general praise.
To determine whether teachers used more mild reprimand than any other type of reprimand, an ANOVA for repeated measures was conducted. At an alpha level of .05, there was a significant difference in reprimand frequency across types, F(1, 65) = 35.23, p < .001, η2 = .35 (large effect). Multiple t tests with a Bonferroni correction demonstrated that mild reprimand (M = 2.98, SD = 4.83) was used significantly more than medium (M = 0.42, SD = 0.86), d = 0.75; harsh (M = 0.14, SD = 0.39), d = 0.64; or gestural (M = 0.48, SD = 0.77), d = 0.62 (medium effect for each comparison). Medium and gestural reprimand were used significantly more than harsh, d = 0.90 (large effect). There was no significant difference between medium and gestural reprimand; however, there was a medium effect (d = 0.67).
Between-group differences were also examined. Middle school teachers had significantly higher rates of reprimand than high school teachers, t(64) = 2.87, p = .003 (one tailed), d = 0.73. More specifically, they used more mild, t(64) = 2.48, p = .008 (one tailed), d = 0.63; and gestural reprimand, t(64) = 2.35, p = 0.01 (one tailed), d = .30 (see Table 3A). Teachers with 10 or less years of experience used significantly more gestural reprimand, t(64) = 2.69, p = .004 (one tailed), d = 0.68. Interestingly, teachers with no classroom management course had significantly higher rates of praise than those with a course, t(62) = 2.76, p = .004 (one tailed), d = 0.69. They gave more BSP, t(62) = 2.44, p = .008 (one tailed), d = 0.61, as well as more GP, t(62) = 2.21, p = .01 (one tailed), d = 0.55. In contrast, teachers with a classroom management course used significantly more reprimand than those with no course, t(62) = 1.69, p = .04 (one tailed), d = 0.42. They gave more mild reprimand, t(62) = 1.73, p = .04 (one tailed), d = 0.43.
Differences were also examined within middle and high school groups. Middle school teachers with a classroom management course used significantly more total reprimand than those with no course, t(21) = 2.30, p = .02 (one tailed), d = 0.95. In particular, they gave significantly more mild reprimand, t(21) = 2.20, p = .02 (one tailed), d = .92. High school teachers with no course had significantly more total praise than those with a course, t(39) = 2.14, p = .02 (one tailed), d = 0.72. They gave significantly more GP, t(39) = 1.73, p = .05 (one tailed), d = 0.54, and BSP, t(39) = 2.27, p = .01 (one tailed), d = 0.67. Middle school teachers with less experience (<10 years) gave significantly more gestural reprimands than those with more experience, t(23) = 3.84, p < .001 (one tailed), d = 1.65. Middle school teachers with more experience (10+ years) used significantly more harsh reprimands than those with less experience, t(23) = 1.75, p = .04 (one tailed), d = 0.74. High school teachers with less experience (<10 years) used significantly more GP than those with less experience, t(39) = 1.76, p = .04 (one tailed), d = 0.55.
Teacher Perceptions
To answer Research Question 2 (Are teachers’ perceived and actual use of praise consistent?), Pearson’s r correlation coefficients were calculated among actual and perceived praise types. At an alpha level of .05, there was a significant positive relation between actual and perceived GP, r(64) = .27, p = .01 (one tailed). In other words, participants who were observed to use more GP also reported to use more GP. This relation had a small (approaching medium) effect size. Actual GP in relation with perceived GP accounted for 7% of the variance between these two variables. BSP, r(64) = .06, p = .66 (two tailed), and total praise, r(64) = .20, p = .11 (two tailed), were not significant (both small effect sizes).
Pearson’s r correlation coefficients were also calculated for observed and perceived reprimand types (Are teachers’ perceived and actual reprimand use consistent?). At an alpha level of .05, there was a significant positive relation between actual and perceived mild reprimand, r(64) = .37, p = .002 (two tailed). Teachers who were observed to use more mild reprimand also reported to use more mild reprimand (medium effect). Actual mild in relation with perceived mild accounted for 14% of the variance between the two constructs.
At an alpha level of .05, there was a significant positive relation between actual and perceived gestural reprimand, r(64) = .38, p = .002 (two tailed). In other words, participants who were observed to use more gestural reprimand also reported to use more gestural reprimand (medium effect). Actual gestural in relation with perceived gestural accounted for 14% of the variance between the two variables.
At an alpha level of .05, there was also a significant positive relation between actual and perceived total reprimand, r(64) = .37, p = .002 (two tailed). In other words, participants who were observed to use more total reprimand also reported to use more total reprimand (medium effect). Actual total reprimand in relation with perceived total reprimand accounted for 14% of the variance between the two constructs. Medium reprimand, r(64) = .17, p = .17 (two tailed), and harsh reprimand, r(64) = .12, p = .33 (two tailed), were not significant.
Teacher Perceptions and PRRs
For the fourth research question (Is there a relation between teachers’ PRR and their praise and reprimand accuracy?), Pearson’s r correlation coefficients were calculated among actual and perceived praise difference, actual and perceived reprimand difference, and PRR. At an alpha of .05, results indicated that there was a significant positive relation between actual and perceived praise difference and actual and perceived reprimand difference, r(64) = .30, p = .008 (one tailed). The greater the teacher’s misperception between actual and perceived praise, the greater the misperception between actual and perceived reprimand (medium effect). Teachers with a greater difference between actual and perceived praise were more likely to have a greater difference between actual and perceived reprimand. Praise difference in relation to reprimand difference accounted for 9% of variance between the two variables. However, at an alpha level of .05, PRR in relation to actual and perceived praise difference was not significant, r(64) = .05, p = .34 (one tailed). At an alpha level of .05, there was no significant difference among PRR in relation to actual and perceived reprimand difference, r(64) = .04, p = .39 (one tailed).
Discussion
This study extends the literature on secondary teachers’ natural use of praise and reprimand. It also explores whether reflecting on praise and reprimand may offer educators a strategy to assist in their accuracy, reliability, and sustainability of this EBP. Teachers in this sample had an average total PRR of 0.86 to 1, which was higher than the ratio (0.58–1) reported by White (1975). These findings are consistent with the literature in that teachers tend to use more reprimand than praise in the classroom (Kim & Stormont, 2016; Reinke et al., 2013; Stormont et al., 2007). In this study, teachers used mild reprimand more than any other type, and teachers used more GP than BSP. Also, teachers who used more GP also reported to use more GP; however, no significant relation was found between teachers’ use and reported use of BSP. There were also positive correlations between teachers’ use and reported use of mild, gestural, and total reprimand. Finally, teachers with a larger difference between their actual and perceived praise had a larger difference between their actual and perceived reprimand.
First, teachers used significantly more GP than BSP (i.e., 2 BSP per hour, 8.45 GP per hour, or 0.24 to 1 BSP to GP ratio), which is consistent with prior research. Floress and Jenkins (2015) examined natural GP and BSP rates among four general education kindergarten teachers, and found teachers used 8.8 BSP per hour and 38.5 GP per hour (0.23–1). Among 28 general education elementary teachers, Floress et al. (2018) found teachers naturally used 5.9 BSP per hour and 28.9 GP per hour (0.20–1). This is an important finding because BSP is considered a superior form of praise (Simonsen et al., 2008), yet teachers commonly use GP over BSP.
Floress et al. (2018) maintained teachers may use GP more frequently because it is easier. For instance, GP is often given automatically as a social nicety (e.g., recognizing a child’s compliance by saying “good” or “thank you”) and does not require much thought or strategy. Arguably, BSP is more effortful than giving a quick, “thumbs up” across a noisy classroom. When using BSP, an educator must think quickly and clearly about a student’s behavior to identify and strategically strengthen that behavior (e.g., Thank you for staying focused and working hard on finishing the assignment). This is particularly difficult when working with students with problem behaviors, who may display more inappropriate than appropriate behaviors. Future research should examine the importance of the GP to BSP ratio and whether having teachers reflect on their use of BSP increases their use of BSP relative to GP.
Second, teachers in this study used more reprimand than praise and more mild reprimand than any other type. This is consistent with past research in that teachers are more likely to correct behavior than acknowledge appropriate behavior (Clunies-Ross et al., 2008). Pointing out minor student misbehaviors may also be reinforcing to teachers, because most students (95%) comply when mildly reprimanded (Maag, 2001). Reprimanding may be more intuitive compared with strategically “growing” (i.e., praising) behavior. However, it is important for teachers to understand the detrimental trajectory of using reprimands to reduce student misbehavior, especially among at-risk students. Downs et al. (2019) found students who were at risk of behavior problems were more sensitive to reprimand, in that, their disruptive behavior increased more readily than their non–at-risk peers. Moreover, focusing on appropriate behavior creates a positive climate, where compliance is more likely because learning is enjoyable (Skinner, 1972).
Unfortunately, even when teachers are trained and aware of proactive strategies, they may not alter their previous (reactive) strategies or utilize proactive strategies (Shook, 2012). It is unlikely teachers intentionally avoid using EBPs, even when they have the knowledge. Rather, teaching is complex and trying to plan and implement EBPs without quality support or training sets teachers up to fail (Stormont & Reinke, 2009). Research should focus on how teachers are taught to praise. How and when teachers are trained may help explain why teachers (in our sample) who did not take a classroom management course praised more than teachers who took a course. Teachers often report receiving insufficient behavior management preservice training (Reinke et al., 2011). It is possible that they receive better training (e.g., PBIS) in the field.
Third, in the current sample, there was a significant, positive relation between teachers’ actual and perceived GP. Teachers who used higher rates of GP reported to use higher rates of GP. There were also positive correlations between teachers’ actual and perceived use of mild, gestural, and total reprimand. This is interesting considering past consultation research, which found discrepancies between teacher-reported and implemented strategies (Robbins & Gutkin, 1994). Even in the absence of consultation, educator-reported and observed strategies were discrepant (Kim & Stormont, 2016).
Wickstrom et al. (1998) suggested that observed and self-reported discrepancies may be influenced by the rigor of the strategy, which may be supported by the current findings. As mentioned above, GP and mild reprimand require less effort to implement compared with BSP. This is consistent with the current finding in that no correlation was found between teachers’ actual and perceived BSP. Many teachers reported using BSP (1–15 times in 20 min), but used no BSP during the 20-min observation. It is possible teachers know they should be using BSP and report using BSP when, in fact, they are not. Robbins and Gutkin (1994) offered this explanation for discrepancies in reported and observed strategies following consultation. Even when teachers have knowledge of EBPs, it may not translate into practice due to the complexities of teaching and implementing EBPs simultaneously (Stormont & Reinke, 2009). Teachers likely need to overlearn strategy implementation until it is rote. These findings support universal, explicit praise training where teachers learn to praise by receiving PF, overlearning, and self-monitoring. Because praise tends to decline after training supports are removed, teachers should be taught to reflect on their ongoing use of praise and reprimand so they can monitor their use of these strategies over time. Teachers should also receive examples for how they can purposefully plan to use BSP in the classroom and other school settings (see Floress et al., 2020).
Last, praise and reprimand differences in relation to PRR were not significant. The hypothesis that teachers with higher PRR would be more accurate in their use of praise and reprimand than teachers with lower PRR was not supported. Future research might examine whether teachers who receive explicit praise training via overlearning, PF, and self-monitoring are more accurate in their perceived use of praise and reprimand compared with untrained teachers. Teachers in this sample did not receive praise training; therefore, a lack of PD related to praise may have influenced participants’ perceptions of praise and reprimand in that they were less accurate (regardless of whether they had higher or lower PRR).
The significant positive relation between actual and perceived praise differences and actual and perceived reprimand differences was surprising. Teachers with larger differences between actual and perceived praise also had larger differences between actual and perceived reprimand. This finding suggests teachers who are less accurate in their use of one strategy may be less accurate in their use of other strategies. Future research should examine whether training teachers to praise via overlearning, PF, and self-monitoring assists teachers in effectively using praise while teaching. To do this, teachers need to strategically use praise and accurately determine whether praise was effective (i.e., strengthened appropriate behavior).
Study Limitations
This study is the first to look at teacher perceptions of their praise and reprimand use; however, there are limitations. One limitation is the participant demographic and sample size. Most participants were Caucasian and from rural Illinois, which limits the generalizability of the results. Furthermore, Illinois’ long-standing presence of PBIS training and technical assistance may have influenced this sample. Six of the 15 schools (40%) reported implementation of PBIS and may have received PD, technical assistance, or resources in consultation with Midwest PBIS. Results may also differ for teachers working in suburban and urban settings, working in other U.S. regions, or teachers from different racial backgrounds. For example, research suggests students from low social–economic and racially diverse backgrounds tend to receive differentiated patterns of behavior management treatment and more severe infractions than their Caucasian peers (Skiba et al., 2002). This may have been the case with the outlier data removed from the sample. This teacher was employed at a micro-urban (i.e., a city of 250,000 or less with certain urban characteristics normally found in large metropolitan centers) middle school that was undergoing significant personnel, administrative, and system-level changes. It was understood that working at this school was stressful, which may have influenced this teacher’s use of reprimand (Clunies-Ross et al., 2008). Future research should examine rates of praise and reprimand in urban schools and the influence of stressful environments on teachers’ PRR. To obtain a larger, more diverse sample, researchers should consider using video technology (Floress et al., in press).
Another study limitation was the length and setting of the 20-min observation. To ensure consistency across observations, teachers were observed during lecture-based instruction. However, this means teachers’ use of praise and reprimand during transitions or other class time (e.g., independent seat work) were not captured. It is possible that praise and reprimand rates may have been different if other class times were included. Teachers were observed once for 20 min, and rates may have been different if additional observations were conducted for each teacher. The time of the year that observations take place may also influence praise and reprimand rates. Observations were conducted over four academic semesters. Student behavior and/or teachers’ use of strategies may vary based on time of school year. Future research should examine whether student behavior and teacher strategies vary depending on different points in the year.
Overall IOA averages for observations fell within the acceptable range; however, GP IOA fell as low as 60% and mild reprimand fell as low as 78%. If IOA dropped below 80%, the first author reviewed definitions and provided additional examples and nonexamples to research assistants. This occurred for GP and mild reprimand; however, the average IOA across observations for GP and mild reprimand (92% and 95%, respectively) were still acceptable.
Given these limitations and that few studies have examined teachers’ actual and perceived EBPs, additional research is sorely needed. Using a larger sample, researchers should examine whether higher reprimand rates among middle school teachers (compared with high school teachers) are replicated and whether demographic characteristics (e.g., preservice courses, in-service PD, and teaching experiences) predict praise and reprimand rates among secondary educators. Future research should also manipulate teacher BSP training to determine whether differences are found between rates of praise and reprimand and teachers’ ability to set, monitor, and evaluate these strategies. Training educators to use BSP by engaging in low-intensity, sustainable activities, such as reflecting on their use of praise and reprimand, may help improve and maintain educators’ use of BSP overtime. More intensive training methods (self-monitoring, overlearning, and PF) might be used for educators needing additional support.
Implications
Overall teachers in this study used more reprimands than praise; however, middle school teachers had significantly higher rates of reprimands than high school teachers. It is plausible middle school teachers are faced with more challenging behavior than high school teachers, and this may influence the strategies middle school teachers use in the classroom (Clunies-Ross et al., 2008). Surprisingly, middle school teachers who had taken a preservice classroom management course used more mild reprimand than those without a course. It is possible that preservice training in behavior management may not address EBPs or didactic training without additional support is insufficient. Regardless, these results suggest the need to target and prioritize middle school teachers for praise PD. One barrier to implementing praise at the secondary level is the misconception that praise is less appropriate or less effective with older students. To overcome this barrier, consultants can use language that increases acceptability of praise with older students (e.g., using the term “positive feedback” rather than “praise” or helping a teacher relate using praise to their own experience in the workforce). Asking an individual to recall positive interactions with supervisors who praised frequently (or negative interactions with supervisors who criticized frequently) is a relatable exercise.
Praise training is effective in increasing teachers’ use of BSP (Zoder-Martell et al., 2019). Video, self-monitoring (reflecting), and PF may be especially effective training methods because they are time and resource efficient. This is important at the secondary level because teachers may have different training needs depending on the behavioral challenges presented by students in various classes. Consultants should provide PD to teachers across classes (starting with less behaviorally challenging classes and moving to more challenging classes), so teachers learn to effectively implement BSP and teach simultaneously. One way to do this is with video feedback, where educators watch themselves and then reflect on their use of BSP and PRR. Thompson et al. (2012) reported that teachers were more aware of their use of praise (e.g., certain word or phrase) or where they directed their praise (e.g., favoring one side of the classroom) after receiving video feedback. Another option is visual PF (i.e., visual representation displaying the amount of BSP observed; Reinke et al., 2007), feedback via email (Barton et al., 2013), or “in the moment” feedback for teachers, with wireless technology (Scheeler et al., 2006). Having educators reflect and then make a plan to increase BSP (Floress et al., 2020) may help sustain this EBP.
In conclusion, few studies have examined teachers’ actual and perceived EBPs. This study extends the literature among secondary teachers’ natural use of praise and reprimand. In addition, this study identifies a way for educators to improve the accuracy of their praise and reprimand use via ongoing reflections to ultimately improve their ability to set, monitor, and evaluate their use of this EBP. Teachers in this study used more reprimand than praise and significantly more GP than BSP, suggesting secondary teachers, and possibly middle school teachers specifically, would benefit from PD targeting the integration of BSP within daily teaching practices. Given teachers were more accurate in their perceived use of GP and mild reprimand, but not their use of BSP, may suggest teachers are more likely to use strategies that require less effort. Considering BSP may be more effortful, training should target methods such as overlearning, PF, and self-monitoring (reflecting), so teachers can use BSP with ease across various classroom settings to ultimately enhance outcomes for all students.
Supplemental Material
sj-docx-1-pbi-10.1177_10983007211000381 – Supplemental material for Exploring Secondary Teachers’ Actual and Perceived Praise and Reprimand Use
Supplemental material, sj-docx-1-pbi-10.1177_10983007211000381 for Exploring Secondary Teachers’ Actual and Perceived Praise and Reprimand Use by Margaret T. Floress, Melissa M. Beaudoin and Ronan S. Bernas in Journal of Positive Behavior Interventions
Footnotes
Acknowledgements
The authors are grateful for the assistance of the participating teachers in addition to the graduate and undergraduate students (Emma Riedesel, Korie Poe, LeAnn Brown, Sara Caldwell, and Zachary Yehling) at Eastern Illinois University for their assistance with data collection.
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) received no financial support for the research, authorship, and/or publication of this article.
Supplemental Material
Supplemental material for this article is available on the Journal of Positive Behavior Interventions website with the online version of this article.
References
Supplementary Material
Please find the following supplemental material available below.
For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.
For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.
