Evaluation of Alpha and Beta Commands on Latency to Comply During Transitions with Young Children: A Preliminary Analysis.
AbstractThe present study evaluated the effects of different types of teacher commands on response latency for pre-kindergarten and kindergarten students. Two different types of commands were assessed: alpha commands and beta commands. Research on instructional time in schools shows that loss of instructional time during transition periods may result in decreased academic achievement (Berliner, 1990). The current study evaluated the effects of both alpha and beta commands on response latency within a multiple baseline design across participants for students who did not respond adequately to teacher-led transitions. Results indicated that both alpha and beta commands were effective for all participants for reducing response latency, including a participant with Autism Spectrum Disorder. Initial results also demonstrated generalization of response latency to non-targeted commands in the classroom.
Keywords: commands, alpha, beta, instructional time
**********
Across the past few decades in the field of education, the concept of instructional time and time spent by students actively engaged in classroom tasks has been a highlighted concern. In 1963, Carroll outlined a model that calculated the time students spent learning (by actively engaging in classroom tasks) divided by the overall amount of time needed to learn, resulting in the amount of learning that occurred. This model suggested that the amount of time spent engaged in academic learning is related to the amount of knowledge acquired and subsequently related to overall academic achievement. In more recent literature, researchers have made distinctions between allocated time, instructional time, and academic learning time in schools. Heward (1994) stated that allocated time in schools refers to the amount of time on the school calendar that a school is open and available for instruction (e.g., time between scheduled school arrival and dismissal). The Pennsylvania State Education Association (2010) defined instructional time as the amount of time allocated to instruction itself, excluding classroom disruptions, (e.g., number of minutes that reading instruction is delivered per day), whereas academic learning time, as defined by Heward (1994), is the actual amount of time dedicated to moderately difficult academic tasks in which students are actively engaged.
Increased instructional and academic learning time has been shown to be linked to higher academic achievement, whereas lost instructional time contributed to lower academic achievement (Heward, 1994; Patall, Cooper, & Allen, 2010). For example, Heward (1994) demonstrated that lessons with increased active student responding (i.e., increased academic learning time and opportunities to respond) resulted in higher rates of learning than lessons with fewer active student responding when all teaching and learning variables (e.g., motivation, curricula delivery) were held constant. Similarly, Jez and Wassmer (2015) showed that the addition of 15 minutes of instructional time per school day was found to improve average student achievement by 1 to 1.5 percent. However, a review of time-related instructional variables by Hollowood, Salisbury, Rainforth, and Palombaro (1994) suggested that, in general, schools allow less than half of each school hour for instruction. Students are only engaged in academic tasks for about 70% of that time and academic learning time only accounts for 33% of the available instructional time, thus resulting in an overall reduction in the time that learners actually spend engaged in an instructional activity.
One period of the school day that leaves students at a heightened risk of distractibility and loss of instructional time is that of transitions, which is defined as the non-instructional time before and after some instructional activity (Berliner, 1990). To reduce transition time and therefore increase instructional time and time spent actively engaged in instruction, one common method utilized within the classroom has included class-wide behavioral management strategies (e.g., Hine, Ardoin, & Foster, 2015; Martella & Marchand-Martella, 2015). For example, Hine and colleagues (2015) utilized a computerized system to provide explicit instruction on behavioral expectations and follow-up performance feedback for four elementary school students who struggled with transition times. Results indicated that the combination of explicit instruction and performance feedback effectively reduced transition time and increased time on task for all participants. Similarly, transition time was reduced across all subjects in a second-grade classroom upon implementation of a packaged intervention consisting of direct instruction of behavioral expectations, an explicit timing procedure, public performance feedback, and an interdependent group contingency (Yarbrough, Skinner, Lee, & Lemmons, 2004).
One population that may be more sensitive to lost instructional time due to transitions is preschoolers and kindergarteners, as they may spend 20% to 35% of their time transitioning between activities (Olive, 2004). Given the novelty of behavioral expectations and the large number of transitions utilized in preschool classrooms, students of this age may find it more difficult to transition from one activity to another, thereby increasing their risk for lost instructional time. Several methods have been utilized in early childhood education classrooms to decrease transition times and increase student familiarity with classroom structure and activities. For example, Register and Humpal (2007) showed that musical transitions (e.g., the use of pre-selected songs) were effective at decreasing the transition time for three young children. Similar results were obtained with other methods such as "beat the buzzer" (e.g., Wurtele & Drabman, 1984), clicker cues (e.g., Goetz, Avala, Hatfield, Marshall, & Etzel, 1983), and self-monitoring (e.g, Connell, Carta, & Baer, 1993). In some cases, these results generalized or maintained when intervention components such as teacher praise were absent (e.g., Goetz et al., 1983; Wurtele & Drabman, 1984). Although these strategies may be useful for most students, they may be necessary but not sufficient for others.
In these cases, one possibility for decreasing transition times and maximizing instructional time may be command types. For example, Forehand (1977) and Forehand and McMahon's (1981) research on command compliance and command type yielded improvements in the general compliance of children during interactions with teachers and parents. These researchers outlined two primary command types: alpha commands and beta commands. Alpha commands were defined as commands that are highly specific in nature and allow the opportunity for a child to respond (e.g., "Pick up the marker"), whereas beta commands were defined as commands that are vague and often do not allow for opportunities to respond (e.g., "Stop that"). Across multiple settings, alpha commands have been shown to be more effective and efficient at improving the compliance rates of children (Forehand, 1977; Houlihan, Vincent, Ellison, & Jones, 1994; Starkweather-Matheson & Shriver, 2005; Williams & Forehand, 1984). For example, Williams and Forehand (1984) found that, when utilized with 56 mother-child pairs, alpha commands yielded greater rates of compliance for children ranging in age from 2.5 to 8 years old. Similarly, within an educational setting, Starkweather-Lund (2001) and Starkweather-Matheson and Shriver (2005) showed that when teachers delivered effective commands, alpha commands that included only one verb in the command, student responding increased by almost 20% for some participants.
The success of alpha commands is likely due to the specificity of the command that is directed to the child and the target behavior, and these types of commands have been shown to be effective in a variety of ways. For example, Allday, Bush, Ticknor, and Walker (2011) found that greeting students with a positive statement and encouraging task engagement at the start of a class period reduced student latency to task engagement, suggesting that the speed at which students begin to engage with tasks can decrease when teacher prompts are altered to be more positive and specific to targeted activities. Similarly, Hamlet, Axelrod, and Kuerschner (1984) showed that compliance with specific commands was more effective when eye contact from the student was obtained prior to the delivery of a command.
Taken together, strategies for cueing transitions and delivering commands have demonstrated improvements in instructional time and compliance with teacher directives. However, unknown are the effects of these strategies on task initiation during common transition times observed in pre-kindergarten and kindergarten classrooms. Additionally, it remains unknown whether improvements in task initiation with a targeted transition period generalize to non-targeted transition periods. Therefore, the purpose of this study was to evaluate the effects of adding alpha or beta commands to common transition practices in a pre-kindergarten/kindergarten classroom on compliance for students who were non-responsive to typical classroom transition routines, and to evaluate whether compliance generalized to non-targeted transitions occurring within close proximity to the targeted transition.
Method
Participants
Three students and two teachers were recruited for the study from a pre-kindergarten/kindergarten classroom at a lab school on a university campus in the Southern Plains region of the United States. Student participants were recruited for the study based on teacher recommendations, which included teacher perceptions of student engagement, moderate to high levels of off-task behavior, and low rates of compliance with teacher commands. Following teacher recommendations, written informed consent was obtained from parents and legal guardians for the students' participation in the study. Because the teachers were also participating in the study (i.e., delivering the interventions), written informed consent was obtained from the teachers.
Jake was a typically developing male student of approximately four years of age who attended the pre-kindergarten level in the classroom. Emma was a typically developing female student of approximately four years of age who attended the pre-kindergarten level in the classroom. Jenny was a female student of approximately five years of age who was diagnosed with Autism Spectrum Disorder and attended the kindergarten level in the classroom. Both teachers were female.
Before data collection began, the first author observed the classroom and the recruited participants in an observation room above the classroom for one week (i.e., five consecutive school days) to measure preliminary response latency to transition commands. The authors of the present study conducted this initial observation period to ensure that response latency was high for the participants and to ensure that alpha commands were not primarily being used by the teachers in the classroom. Each observation session was approximately 25 min in duration. During these pre-study observations, researchers observed participants playing or working at centers in the classroom and then transitioning to their morning time, which consisted of singing songs, greeting one another, reviewing the calendar, and reviewing the class schedule for the day. Results of this initial observation period showed that the participants' response latency averaged approximately 210 s. Anecdotally, participants typically continued to play at their centers during transition times or wandered around the room after a transition cue was delivered in the classroom. The number of beta commands and alpha commands issued by teachers during this time period averaged 15:1 (beta-to-alpha ratio).
Setting
The lab school at which the study was conducted taught students from 12 months to 6 years of age. All data collection procedures for the present study occurred during the mid- to late-spring of the school year and were conducted in a single pre-kindergarten/kindergarten classroom, where the two participating classroom teachers co-taught.
For the present study, general classroom activities and procedures were those observed during the pre-study observations and included participants playing at center-based activities (e.g., dramatic play, blocks, and a teacher-led activity) and transitions for cleaning up their center activities to start the class's "morning meeting." Toward the end of the center-based activities, one of the classroom teachers provided a verbal warning indicating a few minutes left of play before the transition. At the end of the center-based activities, a timer produced audible beeps, and one of the teachers or teacher's aides began playing a clean-up song/music video on the classroom's SMARTboard via YouTube. The combination of timer and song served as a general transition cue for students to clean up their centers. Immediately following the clean-up song, one of the teachers or teacher's aides played a "come to the carpet" song followed by a sit/"crisscross applesauce" song. At the end of the three-song sequence and when the majority of students were seated on their designated carpet spot (i.e., marked with a piece of tape and the student's name), the classroom teachers began the class's morning meeting, which included classroom greetings and reviews of the weather, recent topics learned in class, the calendar, and the classroom activities scheduled for the day.
During data collection, participants were observed in an observation room directly above the pre-kindergarten/kindergarten classroom. The observation room had two-way mirrors, which allowed the researchers to collect data hidden from the view of students and teachers in the classroom. Microphones located on the ceiling of the classroom allowed the researchers to hear activity in the classroom via headphones in the observation room.
Response Definitions and Data Collection
Response latency. Response latency was the dependent variable measured in this study, which was defined as the duration of time between the end of the command delivered and the initiation of compliance. Task initiation was defined as the child beginning to move in the direction of the task. Specific definitions were developed for each command. For the clean-up command, response latency was defined as the duration of time between the simultaneous presentation of the end of the timer beeps and delivery of the command and the initiation of cleaning up/putting away items at center activities. Task initiation for cleaning up was defined as the child beginning to move in the direction of the task, which included touching an object that needed to be put away. For the come to the carpet command, response latency was defined as the duration of time between the initiation of the come to the carpet song and when the child stood on the classroom carpet. For the sit/"crisscross applesauce" command, response latency was defined as the duration of time between the initiation of the sit/"crisscross applesauce" song and when the child's buttocks was on the child's designated carpet spot.
For all command types, response latency data were collected with the use of M Stopwatch, an iPhone application, and were calculated by converting all data from minutes and seconds to total seconds.
Data collection rules. First, a ceiling of 2 min was established in which the researchers stopped the stopwatch if compliance was not initiated by the student within a two-minute time frame. Forehand (1977) suggested that a reasonable amount of time for compliance is approximately 5 s. Thus, the researchers chose a ceiling of 2 min to provide a sufficient amount of time for opportunities to comply.
Second, data were not collected when specific situations occurred. These situations included: (a) the delivery of multiple commands, defined as the teacher delivering more than one verbal command within 5 s of the previous command, (b) task initiation by the participant before the command was delivered, (c) the absence of an intervention component (e.g., delivery of beta or alpha commands) before the timer sounded indicating the transition for the clean-up command, (d) participants missing the commands for come to the carpet and sit/"crisscross applesauce" because cleaning up their centers took too long, and (e) a lack of materials to clean up during the clean-up command.
Command type. The independent variable in this study was the command type delivered by the teacher. Three different levels of commands were provided: (1) group beta commands, (2) individual beta commands, and (3) individual alpha commands. Beta and alpha command types were classified based on the properties and definitions set forth in studies conducted by Forehand (1977), Forehand and McMahon (1981), and Starkweather-Lund (2001). Generally, properties included specificity in command delivery, with more specificity for alpha commands and less specificity for beta commands. Specifically, group beta commands were defined as commands addressed to the class, as a whole, that did not offer a clear opportunity to respond. For this study, group beta commands included commands that were part of the typical classroom structure (e.g., sounds of a timer, song played on the classroom SMARTboard). Individual beta commands were defined as commands delivered to an individual student that did not offer a clear opportunity to respond. For this study, individual beta commands included stating the child's name and a general instruction (e.g., "Emma, let's clean up."). Individual alpha commands were defined as commands addressed to an individual student that provided a specific, direct, and concise instruction. For this study, individual alpha commands included stating the child's name and a specific instruction (e.g., "Emma, put the markers in the marker box.").
Interobserver Agreement
Two independent observers scored response latency to obtain a measure of interobserver agreement (IOA). IOA was calculated using a simple ratio method outlined by Allday et al. (2011), which consisted of dividing the lower latency time observed (in seconds) by the higher latency time observed (in seconds) and multiplying by 100%. For example, if the primary rater's observed latency to compliance was 120 s for the clean-up command and the secondary rater's latency to compliance was 110 s for the clean-up command, calculations would include dividing 110 s (the lower latency time observed) by 120 s (the higher latency time observed) and multiplying by 100%, resulting in 92% agreement. IOA was collected for all command types across 15% of sessions for Jake, 21% of sessions for Emma, and 31% of sessions for Jenny. For Jake, mean IOA for the clean-up, come to the carpet, and sit/"crisscross applesauce" commands was 89.5% (range, 84-95%), 88.9% (range, 87.8-90%), and 98.9% (range, 97.8-100%), respectively. For Emma, IOA for the clean- up, come to the carpet, and sit/"crisscross applesauce" commands averaged 97.8% (range, 94.5-100%), 98.8% (range, 95.2-100%), and 86.9% (range, 47.4-100%), respectively. For Jenny, mean IOA for the clean-up, come to the carpet, and sit/"crisscross applesauce" commands was 93.1% (range, 84.3-99.6%), 94.7% (range, 81.4-100%), and 95.4% (range, 76.9-100%), respectively.
Experimental Design
A multiple baseline design across participants was utilized in this study to determine the effects of command types on response latency during the clean-up command because it was the first directive the participants were given to initiate the transition to their morning meeting. Baseline data for response latency (in seconds) were obtained for all three participants. Phases in the study were changed when stability was established for each phase. When baseline data were stable, the first intervention was implemented for Jake, while Emma and Jenny remained in the baseline phase. Additionally, response latency was obtained for additional commands (come to the carpet and sit/"crisscross applesauce") to determine if the initial clean-up command and the type of command used (group beta vs. individual beta vs. individual alpha) resulted in spontaneous generalization of response latency for the come to the carpet and sit/"crisscross applesauce" commands.
Procedures
General procedures. Each session occurred once every school day, Monday through Friday, when participants were present in the classroom. Sessions occurred during the classroom's morning center time before the teachers started the class's morning meeting, which began around 9:15 a.m. The first and second author arrived at the observation room for data collection around 8:50 a.m. in the morning and left at approximately 9:15 a.m., which allowed for observations during the transition period between classroom centers and the morning meeting. Across all phases, if a participant did not comply with the teacher's command, no specific, programmed consequences were delivered. At times, the teacher redirected a child and stated the command again (this only occurred during the baseline phase) if the child did not comply; however, this was not a consistent phenomenon in the classroom during the baseline phase. If participants did comply to a teacher's command, general praise statements were sometimes delivered.
Baseline phase (group beta commands). During baseline, group beta commands that were naturally provided to the class in the form of an audible timer and/or songs were delivered to the entire class. Specifically, for the clean-up command, a teacher or teacher's aide set the timer at the beginning of the morning centers period. When the timer beeped, it was stopped by a teacher while a second teacher initiated the clean-up song which began the transition from the morning centers activity to the morning meeting. This command signaled the class to clean up the activities or work they were playing with during the morning centers. In addition to the timer and song, one or both classroom teachers frequently provided general vocal instructions to the class as a whole (e.g., "It's time for us to clean up."). For the come to the carpet command, a teacher initiated the come to the carpet song when it appeared that most of the class was cleaning up. This command signaled the class to move to their designated spot on the carpet, which was marked with their name on a piece of tape. For the sit,/"crisscross applesauce" command, a teacher initiated the sit/"crisscross applesauce" song when it appeared that the majority of the class was on their way to the carpet. This command signaled the class to sit on their designated spot on the carpet with their legs crossed. Following the transition from morning centers to the morning meeting (completion of all three group beta commands and with most of the class in their designated spots), the morning meeting was initiated by the teachers. Because group beta commands were typical procedures within the classroom, the teachers continued to deliver these group beta commands throughout all intervention phases.
Individual beta commands. During the individual beta command condition, teachers read from a cue card to deliver an individual beta command to each participant for the clean-up command only. Generally, the cue card said, "[Student's Name], it's time for us to clean up." This beta command was chosen because it was similar to the language that the teachers used with their students during their regular classroom instruction. While delivering the command, the teacher approached each student individually to deliver the command on the cue card. The command was delivered to each student simultaneous with the timer beeps sounding in the classroom, which occurred right before the initiation of the clean-up song. There was no specific order in which the teachers approached each participant. Following the clean-up command, all other procedures for the come to the carpet and sit/"crisscross applesauce" commands were conducted in the same manner as described for the group beta commands.
Individual alpha commands. During the individual alpha command condition, teachers read from a cue card to deliver an individual alpha command to each participant for the clean-up command only. Generally, the cue card said, "[Student's Name], put away the [whatever he or she is working/playing with]." This alpha command was chosen because of its similarity in language to what the teachers used with their students during their regular classroom instruction. All other procedures were identical to the procedures outlined in the individual beta commands phase (i.e., command delivered simultaneously with the timer beeps, no specific order for approaching participants) and the group beta commands phase (i.e., delivery of the come to the carpet and sit/"crisscross applesauce" commands).
Treatment Integrity
To ensure that teachers implemented the interventions with integrity, the authors of the study reminded the teachers of which interventions (individual beta commands or individual alpha commands) to use during data collection via email the day before. Additionally, treatment integrity was examined during the individual beta and individual alpha command conditions by measuring the accuracy of three treatment components: (1) teachers delivered the clean-up command before the final beeps of the classroom timer, (2) teachers delivered the exact command as stated on the cue cards, and (3) teachers delivered the command type (group beta, individual beta, or individual alpha) according to the relevant phase of the study. Treatment integrity was calculated by dividing the number of correct treatment components by the total number of treatment components and multiplying by 100%. Treatment integrity was collected across 100% of sessions for all participants, and averaged 100%, 96.5% (range, 66.6-100%), and 100% for Jake, Emma, and Jenny, respectively. IOA was not collected on treatment integrity for any of the participants.
Results
Across all participants, results of the evaluation for types of commands on response latency for the clean-up command are displayed in Figure 1 and summarized as the mean response latency in Table 1.
For Jake (Figure 1, top panel) during baseline, response latency was high with a downward trend, in which response latency stabilized around 70 s between days three and five (M = 114.90 s). It should be noted that on day two, the researcher collecting data did not stop the timer at the pre-established 2 min ceiling, resulting in the author recording the actual amount of time it took Jake to comply with the clean-up command (249 s). During the individual beta command phase, results showed an initial drop in level with an overall performance of about 35 s for response latency during the first couple of days. On days nine and 10, data were not collected because Jake was absent due to illness. When Jake returned to class on day 11, his response latency occurred at similar levels to previous days with the individual beta command. Overall, the trend of the data in this first intervention phase was stable with Jake's average response latency for cleaning up occurring at 31.68 s (range, 14.66 s to 45 s). During the individual alpha command phase, Jake's response latency further decreased and remained stable near the 5 s level (M = 6.28 s).
During the clean-up command, response latency for Emma (Figure 1, middle panel) was ascending during the baseline phase. On the third day, Emma was completing work during small-group instruction and did not have material to clean up, thus response latency data were not collected. On average, Emma's response latency during this phase was 67.83 s (range, 38.19 s to 120 s). During the individual beta command phase, Emma's response latency decreased to an average of 13.03 s (range, 7.28 s to 18.62 s). Data were not collected on days nine, 10, and 14 because Emma was absent due to illness (days 9 and 10), and she responded to the classroom timer before the teacher delivered the individual beta command (day 14). During the individual alpha command phase, Emma's response latency decreased slightly and averaged 11.14 s (range, 6.91 s to 18.70 s). Data were not collected for days 19 and 20 because Emma responded again to the timer in the classroom before the teacher had a chance to deliver the intervention.
Jenny's (Figure 1, bottom panel) response latency, during baseline, was variable and averaged 62.27 s (range, 24.16 s to 117.21 s). Data were not collected during days six through nine because Jenny was absent from school. During the individual beta command phase, Jenny's level of responding decreased to an average of 19.82 s (range, 6.62 s to 60 s). During the individual alpha command phase, Jenny's response latency further decreased and stabilized near the 5 s level (M = 5.40 s). Data were not collected on days 21 through 23 because Jenny was absent (days 21 and 23), and she attended a class field trip (day 22).
To determine if the effects of the individual beta and alpha commands generalized to non-targeted commands that occurred directly following the clean-up command, response latency was collected for the come to the carpet and sit/"crisscross applesauce" commands which were only delivered in the group beta command format. Results of these observations are depicted in Figure 2 and summarized as the mean response latency in Table 1.
For Jake (Figure 2, top panel), response latency for the come to the carpet command (open squares) had an increasing trend for four of the five days (M = 42.76 s) when the clean-up command was in baseline. At the same time, response latency for the sit/"crisscross applesauce" command (closed diamonds) demonstrated an upward trend (M = 59.46 s). When Jake received an individual beta command for the clean-up command, response latency for both the come to the carpet command and the sit/"crisscross applesauce" command demonstrated initial decreases in level with a subsequent increasing trend that occurred at similar levels to data collected on days one through five. When Jake received an individual alpha command for the cleanup command, response latency for the come to the carpet command and sit/"crisscross applesauce" command occurred at similar levels to previous days, but then decreased to an average of 0 s for the come to the carpet command and to 0.82 s for the sit/"crisscross applesauce" command on days 14 and 15.
For Emma (Figure 2, middle panel), when she received the typical group beta command for the clean-up command (baseline), response latency for the come to the carpet command was relatively stable (with the exception of session six), whereas the response latency for the sit/"crisscross applesauce" command was variable with averages of 27.75 s (range, 0 s to 120 s) and 62.76 s (range, 0 s to 120 s), respectively. When Emma received an individual beta command for the clean-up command, response latency for both the come to the carpet command and the sit/"crisscross applesauce" command immediately decreased and remained stable. When Emma received an individual alpha command for the clean-up command, response latency for both the come to the carpet command and the sit/"crisscross applesauce" command remained at low and stable levels.
For Jenny (Figure 2, bottom panel), response latency for the come to the carpet command and the sit/"crisscross applesauce" command was variable with a decreasing trend when the clean-up command was provided in a group beta format. When Jenny received an individual beta command for the clean-up command, response latency for the come to the carpet command was at low levels and remained stable until days 17 through 19, in which an upward trend occurred. Similar results were obtained with the sit/"crisscross applesauce" command. When Jenny received an individual alpha command for the clean-up command, response latency for the come to the carpet command and the sit/"crisscross applesauce" command demonstrated a low and stable level of responding.
Discussion
The results of this study indicate that response latency during transitions decreased when individual beta and individual alpha commands were directed specifically toward pre-kindergarten and kindergarten students at the same time as typical classroom transition procedures. Response latency decreased for both typically developing students and a student with Autism Spectrum Disorder. Although beta commands are generally non-specific in nature, low latency levels were observed in the individual beta command intervention condition, suggesting that simply addressing a student by name may serve as a common stimulus to initiate task engagement (or decrease response latency). Furthermore, this study found that individual alpha commands resulted in reductions in response latency across most participants compared to baseline, which converge with previous findings supporting the utilization of alpha commands to increase compliance (Forehand, 1977; Houlihan et al, 1994; Williams & Forehand, 1984).
With regard to instructional time frequently lost in preschool classrooms during transition times, the command interventions utilized in this study demonstrate implications for maximizing instructional time in the classroom by decreasing response latency during transition periods. For instance, if Jake were to continue in the baseline phase responding to group beta commands without intervention and were to have the same average per day (114.90 seconds) for a school year consisting of 180 days, instructional time lost in that single transition period would account for over 5.5 hours of time during the school each year. If the most specific intervention used in this study (individual alpha commands) was delivered to Jake, transitioning during this period in the morning would only use approximately 19 minutes of time during the school each year. Thus, with this intervention, over 300 minutes of time in school could be gained and allotted for instructional time rather than time spent transitioning from one activity to another. Additionally, by gaining instructional time over the course of a school year, it's possible that the number of opportunities to respond in the classroom would increase, allowing for maximizing active student responding, which has been shown to produce higher levels of academic achievement (Heward, 1994; Jez & Wassmer, 2015; Patall et al., 2010).
Additionally, the results of this study demonstrated that decreased response latency may generalize to additional behaviors (i.e., group beta commands) when only one specific, targeted command is delivered to students at the beginning of transition periods. Due to the chained schedule of command sequence utilized in this study (clean-up to come to the carpet to sit/"crisscross applesauce"), the data suggest that targeting the first command within a chained schedule may result in generalization to the other commands within the sequence. However, it remains unknown whether or not generalization would occur for distinct transitions that occur at different times during the school day, such as cleaning up and lining up 1 hr later for music time. The results of generalization to non-targeted commands in the present study further demonstrate the potential for gained instructional time in the classroom, with teachers using less time to deliver commands by only providing one initial command specifically directed at an individual student. This allows for minimal use of commands that possess higher specificity, while still producing maximum effects (i.e., the behavior produced by the use of one command generalizing to other vague classroom commands).
The intervention used in this study is also user-friendly, as it simply involves reading a statement off of a cue card while talking to a student. Given the easy implementation of the present study's intervention and the little time needed to train teachers on how to utilize the intervention in their classrooms, it's possible that teachers may be more likely to use the intervention with integrity, because as suggested by Martens, Peterson, Witt, and Cirone (1986), one of the greatest factors influencing teacher use and implementation integrity of school-based interventions is the ease of use. Taken together, these results extend and support the literature of instructional time in the classroom by investigating how the reduction of response latency provides implications for quicker transitions to instructional or academic activities in the classroom.
Limitations
Although the present study lends support to the authors' hypothesis, the study does have several limitations. One limitation for the study was that concurrent commands were delivered in the classroom. For example, in all phases of the study, the teachers in the classroom used an audible timer and songs to signal when it was time for all students in the classroom to clean up. These common classroom discriminative stimuli could have caused confusion for the participants in the study (however, data were not collected if participants responded to the timer and not the teacher's command). Additionally, the presence of all of these stimuli (class-wide procedures and individual commands) across phases may have confounded the effects of the commands delivered on response latency across participants because they interacted with each other. Thus, a component analysis of each of these stimuli would be beneficial to further analyze the sole effects of individual beta and individual alpha commands on response latency.
A second limitation to this study included the downward trend in data during baseline for Jake's response to the group beta cleanup command. Although there is a clear distinction between level and variability between the baseline and treatment phases, this downward trend limits the interpretation of the data for Jake and must be taken into consideration. This downward trend makes it difficult to definitively rule out maturation effects and history effects.
Third, the authors did not control for sequence effects. Because each intervention phase was conducted sequentially and consecutively across participants, the lower latencies to comply in subsequent intervention phases of the study may be due to exposures of similar commands in previous phases (e.g., individual beta and individual alpha commands both use the discriminative stimulus of addressing the student by name). For example, when examining all participant data, both intervention phases of individual beta commands and individual alpha commands for the clean-up command demonstrated lower levels of latency when compared to baseline.
In addition to sequence effects, the authors did not control for the quality of attention provided by the teachers to the participants. As suggested by Allday et al. (2011) and Hamlet et al. (1984), additional treatment variables such as direct teacher attention (e.g., greeting students with positive statements) or obtaining student eye contact prior to the delivery of commands may increase student compliance to commands. Thus, it is possible that, across intervention phases of the present study, teachers provided participants with different qualities of attention (e.g., closer proximity to the student, different tone of voice). This possibility cannot be ruled out as a confounding factor because teachers who participated in the present study were not blind to the changes in intervention conditions. For example, the proximity of the teacher to the student while delivering the command, whether the teacher was farther away from a student or close to the student, may have affected response latencies. For example, during baseline/ normal classroom procedures, teachers rarely approached individual students, resulting in addressing the class as a whole, whereas during intervention phases, the teacher approached the student participants and directly said their names, which may have impacted the participants' behavior.
Further potential mechanisms of behavior change that may account for the lower response latencies in intervention phases include the novelty of the new commands delivered. As mentioned previously, the teacher participants of the study infrequently delivered commands in their classroom such as the ones used in the present study. The novelty of the commands may have confounded the observed treatment effects.
Lastly, another limitation to the current study is the lack of data collection for the length of time to complete the issued commands. These data could be helpful in evaluating the interventions' effectiveness.
Future Directions
In the future, the present study can be applied to several clinical applications. One such clinical application includes using this command intervention with parents and their children who have difficulty initiating task engagement in a timely manner after a command is delivered. This antecedent behavioral management approach could be used in conjunction with other treatment approaches to help maximize treatment effects. Due to the ease of implementation and limited number of resources needed to implement the intervention outlined in the present study, the intervention could serve as a primer or introductory strategy for parents to use at the beginning of a parent training program. Simple behavioral management strategies could increase parent-practitioner rapport and buy-in to behavior analytic strategies to address common behavior problems.
From a research standpoint, replicating this study may also benefit from the collection of maintenance or follow-up data to analyze the long-term effects of the specificity of commands and generalization of these commands to non-targeted activities. During further research efforts, it may also be beneficial to evaluate the effects of the intervention on the length of time it takes a student to complete the task, after compliance to a command is initiated. Taken together, these data may be useful in analyzing the utilization of instructional time when examined in conjunction with response latency. If such connections are well studied, this line of research could further inform the link between instructional time in the classroom and student achievement.
Shelby M. Wolf
Meredith A. Weber
Gary Duhon
Oklahoma State University
Kelly M. Schieltz
University of Missouri
References
Allday, R. A., Bush, M., Ticknor, N., & Walker, L. (2011). Using teacher greetings to increase speed to task engagement. Journal of Applied Behavior Analysis, 44, 393-396.
Berliner, D. C. (1990). What's all the fuss about instructional time? In M. Ben-Peretz, & R. Bromme (Eds.), The nature of time in schools. New York, NY: Teachers College Press.
Carroll, J. B. (1963). A model of school learning. Teachers College Record, 64, 723-733.
Connell, M. C., Carta, J. J., & Baer, D.M. (1993). Programming generalization of in-class transition skills: Teaching preschoolers with developmental delays to self-assess and recruit contingent teacher praise. Journal of Applied Behavior Analysis, 26, 345-352.
Forehand, R. L. (1977). Child noncompliance to parent commands: Behavior analysis and treatment. In M. Hersen, R. M. Eisler, & P. M. Miller (Eds.), Progress in behavior modification: Volume 5. New York: Academic Press.
Forehand, R. L. & McMahon, R. J. (1981). Helping the noncompliant child. New York: Guilford Press.
Goetz, E. M., Ayala, J. M., Hatfield, V. L., Marshall, A. M., & Etzel, B. C. (1983). Training independence in preschoolers with an auditory stimulus management technique. Education and Treatment of Children, 6,251-261.
Hamlet, C., Axelrod, S., & Kuerschner, S. (1984). Eye contact as an antecedent to compliant behavior, journal of Applied Behavior Analysis, 11,553-557.
Heward, W. L. (1994). Three low-tech strategies for increasing the frequency of active student response during group instruction. In R. Gardner III, D. M. Sainato, J. O. Cooper, T. E. Heron, W. L. Heward, J. W. Eshleman, & T. A. Grossi (Eds.), Behavior analysis in education: Focus on measurable superior instruction (pp. 283-320). Pacific Grove, CA: Brooks/Cole.
Hine, J. F., Ardoin, S. P., & Foster, T. E. (2015). Decreasing transition times in elementary school classrooms: Using computer-assisted instruction to automate intervention components. Journal of Applied Behavior Analysis, 48, 495-510.
Hollowood, T. M., Salisbury, C. L., Rainforth, B., & Palombaro, M. M. (1994). Use of instructional time in classrooms serving students with and without severe disabilities. Exceptional Children, 61,242-253.
Houlihan, D., Vincent, J., Ellison, P. J., & Jones, R. N. (1994). Assessing childhood noncompliance: Subtle differences in one-step commands and their effects on response topography. Child and Family Behavior Therapy, 16, 9-20.
Jez, S. J. & Wassmer, R. W. (2015). The impact of learning time on academic achievement. Education and Urban Society, 47, 284-306.
Martella, R. C. & Marchand-Martella, N. E. (2015). Improving classroom behavior through effective instruction: An illustrative program example using SRA FLEX Literacy. Education and Treatment of Children, 38,241-272.
Martens, B. K., Peterson, R. L., Witt, J. C., & Cirone, S. (1986). Teacher perceptions of school-based interventions. Exceptional Children, 53, 213-223.
Olive, M. L. (2004). Transitioning children between activities: Effective strategies for decreasing challenging behavior. Beyond Behavior, 14,11-16.
Patall, E. A., Cooper, H., & Allen, A. B. (2010). Extending the school day or school year: A systematic review of research (1985-2009). Review of Educational Research, 80, 401-436.
Pennsylvania State Education Association. (2010). The power of a great education: PSEA's 20/20 vision for the future--maximize instructional time. Retrieved from https://www.psea.org/uploadedFiles/LegislationAndPolitics/Vision/Vision_MaximizeInstructionalTime.pdf
Register, D. & Humpal, M. (2007). Using musical transitions in early childhood classrooms: Three case examples. Music Therapy Perspectives, 25, 25-31.
Starkweather-Lund, A. R. (2001). Training teachers to give effective commands: Effects on student compliance, academic engagement, and academic responding. Retrieved from Bell and Howell Information and Learning Company. (UMI Microform 3016747).
Starkweather-Matheson, A. & Shriver, M. D. (2005). Training teachers to give effective commands: Effects on student compliance and academic behaviors. School Psychology Review, 34, 202-219.
Williams, C. A. & Forehand, R. (1984). An examination of predictor variables for child compliance and noncompliance. Journal of Abnormal Psychology, 12,491-504.
Wurtele, S. K., & Drabman, R. S. (1984). "Beat the buzzer" for classroom dawdling: A one-year trial. Behavior Therapy, 15, 403-409.
Yarbrough, J. L., Skinner, C. H., Lee, Y. J., & Lemmons, C. (2004). Decreasing transition times in a second grade classroom. Journal of Applied School Psychology, 20,85-107.
Address correspondence to: Shelby M. Wolf at Behavioral Therapy Solutions of Minnesota, 700 Commerce Avenue, Suite 260, Woodbury, MN 55125. E-mail: swolf@btsofmn.com
Caption: Figure 1. Response latency (in seconds) for the clean-up command for Jake (top panel), Emma (middle panel), and Jenny (bottom panel).
Caption: Figure 2. Response latency for the come to the carpet command (open squares) and sit/"crisscross applesauce" command (closed diamond) for Jake (top panel), Emma (middle panel), and Jenny (bottom panel).
Table 1
Mean Response Latency for All Commands
Baseline Condition-
Group Beta
Come
Clean to the
Up Carpet Sit
Jake 114.90 42.76 59.36
Emma 67.83 27.75 62.76
Jenny 62.27 14.45 15.23
Individual Beta Individual Alpha
Condition (for Condition (for
Clean-Up Command) Clean-Up Command)
Come Come
Clean to the Clean to the
Up Carpet Sit Up Carpet Sit
Jake 31.68 8.11 61.56 6.28 18.85 32.66
Emma 13.03 3.45 5.38 11.14 0 0.37
Jenny 19.82 13.51 23.09 5.40 8.10 0
Note. Mean response latency is recorded in seconds.
 Printer friendly
 Cite/link
 Email
 Feedback
| |
| Author: | Wolf, Shelby M.; Duhon, Gary; Schieltz, Kelly M. |
|---|---|
| Publication: | Education & Treatment of Children |
| Article Type: | Report |
| Date: | Feb 1, 2019 |
| Words: | 7386 |
| Previous Article: | Teaching Unison Responding During Small-Group Direct Instruction to Students with Autism Spectrum Disorder Who Exhibit Interfering Behaviors. |
| Next Article: | Effects of a Guided-Notes Intervention Program on the Quiz and Note-Taking Greek History Performance of High School Students with Learning... |
| Topics: | |