Effective classroom management is "the most essential aspect of teaching," "the bedrock on which learning occurs." Teachers report that classroom management is "the number one problem that they struggle with." Definition of classroom management: "practices that a teacher follows to ensure that (1) students behave in ways that are not disruptive to the learning of others and (2) teachers are able to focus on academically relevant material."
Research: meta-analysis found that "classroom management had a large and significant effect (d = 0.80) on student behavior." Classroom management related to lower levels of bullying and higher levels of school connectedness (6). Longitudinal study: inner city first grade students randomly assigned to classroom in which Good Behavior Game implemented >> students tracked over time, at end of 1st grade had intervention group had significant decreased in aggressive behavior; in 6th grade the intervention students who were most aggressive in kindergarten were twice as likely to display aggressive behaviors, while the control group students who were most aggressive in kindergarten were 25 more likely to display aggressive behaviors (6).
Jacob Kounin (1970) found that both effective and ineffective teachers used "similar practices to address problem behaviors" but that the first group were better at using preventative strategies. Later researchers corroborated these findings.
There are five primary effective classroom management strategies (quoted mostly verbatim):
- Positively stated expectations made clear by establishing, teaching, and practicing a small set of classroom rules.
- Predictable routines for navigating different settings and situations within the school day must be developed and taught.
- Teachers use positive and specific praise to reinforce students for appropriate behavior.
- Inappropriate behavior should be responded to consistently with an appropriate level of consequence.
- Student engagement should be kept high by facilitating opportunities for active engagement and ensuring that the instructional content is interesting and meaningful.
* * * * *
Behaviorism
Attribution theory: "how we use information to come up with an explanation" for why someone acted as they did. Teachers tend to believe that students are in control of their behavior, meaning that most due not believe classroom variables cause the student's behavior. Most teachers believe that the most helpful intervention would be counseling, not changing their own classroom management.
Behaviorism: our behavior is determined by environmental factors. Principles of behaviorism:
Law of effect. We are more likely to engage in behaviors that generate pleasurable consequences.
Reinforcement. Reinforcement means rewarding behavior so it occurs more often. Positive Reinforcement: desirable consequence given, making behavior more likely to occur in future (e.g., teacher praises student for doing homework). Negative Reinforcement: undesirable consequence taken away, making behavior more likely to occur in future (e.g., teacher stops nagging student when she does homework).
Schedules of Reinforcement. Continuous Reinforcement Schedule: behavior is reinforced every time it occurs. Intermittent Reinforcement Schedule: behavior is reinforced intermittently -- (a) behavior can be reinforced at regular or irregular intervals (interval reinforcement systems) or (b) behavior can be reinforced based on the number of times it occurs (ratio reinforcement system). Interval reinforcement system: behavior reinforced after a certain interval; the interval may be fixed or variable. If the interval is variable, we don't necessarily know how long we need to do something to be reinforced. Reinforcing behavior at variable intervals tends to be more effective at producing behavioral changes because students cannot reliably predict when reinforcement will occur. For variable interval reinforcement schedules, teachers can use timer that buzz at random intervals to remind themselves to praise the students who are staying on task.
Punishment. A punishment is a consequence that decreases behavior. Positive Punishment: undesirable consequence given, making behavior less likely to occur in future (e.g., a child picks his nose, the teacher reprimands him). Negative Punishment: desirable consequence taken away, making behavior less likely to occur in future (e.g., student hits a peer and loses recess).
Extinction. Extinction "occurs when a behavior that was previously reinforced is no longer reinforced." Example: a student is reinforced when he tells jokes in class, and his peers laugh; students are told to no longer laugh when he tells jokes. Extinction burst: "an increase in a behavior that occurs when reinforcement stops" -- e.g., the student may start telling more jokes.
Matching Law. "We tend to engage in those behaviors that bring the greatest reinforcement."
Motivating Operations. What is rewarding one day might not be rewarding the next day. Satiation: when we have too much of something, it is no longer reinforcing -- e.g., a student who always gets praised for answering questions might come to no longer desire it. Deprivation: when we have too little of something, it is more reinforcing -- e.g., a student who has not received any attention from the teacher may find the smallest compliment very reinforcing.
* * * * *
Preventative Strategies
Preventing classroom problems requires addressing three areas: (a) Creating physical structures in the classroom that reduce the likelihood of behavior problems, (b) Teaching students expected behaviors, and (c) Engaging in effective teaching behaviors that that elicit desired behaviors.
Creating Physical Structures
Structuring desks in rows results in lower levels of off-task behavior during independent seatwork.
Teaching Expected Behaviors
Expectations must be positively stated. Classes should have approximately 4-6 rules.
Color Wheel System. Different expectations correspond with thee colors red, yellow, and green. Red: most restrictive, e.g., students expected to sit quietly with their eyes on the teacher. Green: least restrictive -- e.g., during free-time activities students should keep their hands and feet to themselves, respect others, and use inside voices. Yellow: between red and green; expectations during typical instructional periods. Tricolored wheel placed in front of class. Expectations taught during role plays. Teacher reviews expectations every day and cues students whenever color wheel is changed. Leads to more on-task behavior and less disruptive behavior.
Questions that may be used to establish expectations. How loud can students talk? How should students line up? What materials should students come to class with? What should students do to show they are paying attention? What should students do when they finish their assignments early? What should students do when they need help? What should students do when they need to use the restroom? When can students leave their seats? When can students work with classmates? Where can students move in the classroom?
Establishing routines. The most effective teachers spend most of their time during the beginning of the year establishing routines and teaching procedures. Questions to ask when considering which routines to teach: how long the routine should take, what is the end goal of the routine, what level of scaffolding needed to complete routine.
Teaching rules and routines. Four steps: telling, showing, doing, and generalizing. Doing = students practicing. Generalizing = students receive different opportunities to apply the skill to build fluency (that is, generalize). Never assume that students already know rules or routines. Never assume that practice just once is sufficient. Never assume that feedback is unnecessary.
Engaging in Effective Teaching Behaviors
Effectively communicating directions. (1) Get students' attention before giving a direction. (2) Don't give directions as questions (e.g., don't say, "Would you mind moving to another desk?"). (3) Don't give long directions. (4) Don't ask students to do things that are not behaviorally measurable (e.g., "Try harder"). (5) Don't create a power struggle by standing immediately over a student and waiting for him or her to comply. (6) Don't forget to check to make sure that the student actually followed the direction.
Making precision requests. This includes the student's name and a description of the required behavior, and it is delivered in a polite tone and allows wait time for the student to comply. The request should be made in as few words as possible in language that clearly states what you would like the student to do. Steps:
* * * * *
Preventative Strategies
Preventing classroom problems requires addressing three areas: (a) Creating physical structures in the classroom that reduce the likelihood of behavior problems, (b) Teaching students expected behaviors, and (c) Engaging in effective teaching behaviors that that elicit desired behaviors.
Creating Physical Structures
Structuring desks in rows results in lower levels of off-task behavior during independent seatwork.
Teaching Expected Behaviors
Expectations must be positively stated. Classes should have approximately 4-6 rules.
Color Wheel System. Different expectations correspond with thee colors red, yellow, and green. Red: most restrictive, e.g., students expected to sit quietly with their eyes on the teacher. Green: least restrictive -- e.g., during free-time activities students should keep their hands and feet to themselves, respect others, and use inside voices. Yellow: between red and green; expectations during typical instructional periods. Tricolored wheel placed in front of class. Expectations taught during role plays. Teacher reviews expectations every day and cues students whenever color wheel is changed. Leads to more on-task behavior and less disruptive behavior.
Questions that may be used to establish expectations. How loud can students talk? How should students line up? What materials should students come to class with? What should students do to show they are paying attention? What should students do when they finish their assignments early? What should students do when they need help? What should students do when they need to use the restroom? When can students leave their seats? When can students work with classmates? Where can students move in the classroom?
Establishing routines. The most effective teachers spend most of their time during the beginning of the year establishing routines and teaching procedures. Questions to ask when considering which routines to teach: how long the routine should take, what is the end goal of the routine, what level of scaffolding needed to complete routine.
Teaching rules and routines. Four steps: telling, showing, doing, and generalizing. Doing = students practicing. Generalizing = students receive different opportunities to apply the skill to build fluency (that is, generalize). Never assume that students already know rules or routines. Never assume that practice just once is sufficient. Never assume that feedback is unnecessary.
Engaging in Effective Teaching Behaviors
Effectively communicating directions. (1) Get students' attention before giving a direction. (2) Don't give directions as questions (e.g., don't say, "Would you mind moving to another desk?"). (3) Don't give long directions. (4) Don't ask students to do things that are not behaviorally measurable (e.g., "Try harder"). (5) Don't create a power struggle by standing immediately over a student and waiting for him or her to comply. (6) Don't forget to check to make sure that the student actually followed the direction.
Making precision requests. This includes the student's name and a description of the required behavior, and it is delivered in a polite tone and allows wait time for the student to comply. The request should be made in as few words as possible in language that clearly states what you would like the student to do. Steps:
- Make request using the word "please" >> Wait 5 seconds
- If student complies >> Deliver praise
- If student doesn't comply >> Make second request using the word "need" (e.g., "You need to give me that book") >> Wait 5 seconds
- If student complies >> Deliver praise
- If student doesn't comply >> Deliver reprimand
Make high-probability requests. Sometimes we can increase the chances of a student following a low-probability request by making a high probability request first. Example: "Please write a paragraph about your day" could be preceded by "Please pick up your pencil." It's hope that the principal of behavioral momentum will then take over. (The principal of behavioral momentum refers to the idea that once we start doing something, it is difficult to stop.) Begin by asking students to do something they have no problem doing (e.g., "Give me a high five"), and then ask them to do something they may be less likely to do ("Now tell me what happened on the playground"). Example: start with the Hokey Pokey.
Increasing opportunities to respond. Two alternatives to hand-raising: choral responding and response cards. It's also important to ensure a sufficient instructional pace, thus allowing students less time to act out. One challenge with choral responding is that it can be difficult to monitor individual responses. Alternatives to response cards: laptops using websites like polleverywhere.com or socrative.com; students can also respond to their partners.
Actively supervising the classroom. Two components. First, the teacher moves around the classroom, scans the room, and frequents those parts where behavior problems frequently occur. The less predictable the teacher's location, the less likely that students will be able to engage in problem behaviors. Second, the teacher should have frequent interactions with students to make sure they know that the teacher is around -- e.g., greeting students with a friendly tap on the shoulder, providing praise.
* * * * *
Providing Behavioral Feedback
Three types of verbal feedback: negative feedback (e.g., reprimands), positive feedback (e.g., praise), and planned ignoring.
Negative Verbal Feedback
The problem with yelling. Yelling scares students, and if they're scared, it's difficult to learn. Yelling can also trigger students who have been traumatized, as traumatic experiences are often precipitated by yelling. Also, some students don't find yelling to be punishing.
Effective reprimands. Key features:
- Brief
- Delivered immediately
- Delivered quietly and individually
- Made using a calm, firm, unemotional tone
- Paired with eye contact
- Paired with specific feedback regarding what the student is doing wrong and what the student should be doing instead
- Followed by teacher moving away from student while student modifies behavior
Positive Verbal Feedback
Effective praise must be specific and contingent. E.g., "Marcus, nice job taking the time to sound out the word!" "Karina, I know it can be hard to introduce yourself to someone new. I'm so proud of you for sitting next to a new friend today!"
When praise is public, it can also remind other students of classroom expectations. Some students find student praise, but not teacher praise, to be reinforcing. (Example: students asked to identify one peer who had done something nice for them, and these identified students were given a sticker.)
Self-monitoring can increase teacher praise -- e.g., teachers watch video of themselves and count number of times they praised students.
Planned Ignoring
This means not responding to a particular behavior. Based on principle of extinction: i.e., a behavior that is no longer reinforced will stop. Teacher continues teaching the lesson as if nothing occurred. Planned ignoring of inappropriate only works if the teacher pairs it with the praising of appropriate behavior. Differential reinforcement of an alternative behavior (DRA): a student receives attention for doing what is desired and is ignored for the problem behavior. Keep in mind that when planned ignoring begins it will likely be accompanied by an extinction burst.
Noncontingent Attention
Noncontingent attention is one way to provide students with the attention they desire. When attention is given noncontingently, a teacher will reinforce students' behavior by providing attention frequently -- e.g., by greeting students at the door, engaging students who are working on assignments.
Conclusions
Rules only work when paired with verbal feedback.
* * * * *
Implementing Token Economies
Research shows that providing students with specific, contingent praise leads to improvements in appropriate behavior. Sometimes praise loses its reinforcing value, and some students don't find praise reinforcing. A token economy involves pairing positive verbal feedback with the delivery of tokens. The tokens can in turn be used to purchase different forms of positive reinforcement. Research shows that token economies result in stronger outcomes than praise alone; this could be because giving out tokens forcing the teacher to more closely monitor their own behavior. Point sheets are a type of token economy.
Four necessary conditions for any token system: (a) you should be able to administer the tokens effortlessly, (b) tokens should be administered quickly and contingently (if there's a delay, the student might not understand what they've done to receive the token), (c) the delivery of a token must be paired with verbal feedback.
Token economies only work if the rewards are reinforcing. You can give students a reward preference survey. Therefore, a token economy should incorporate a variety of possible rewards rather than just one. Four types of rewards:
* * * * *
Implementing Token Economies
Research shows that providing students with specific, contingent praise leads to improvements in appropriate behavior. Sometimes praise loses its reinforcing value, and some students don't find praise reinforcing. A token economy involves pairing positive verbal feedback with the delivery of tokens. The tokens can in turn be used to purchase different forms of positive reinforcement. Research shows that token economies result in stronger outcomes than praise alone; this could be because giving out tokens forcing the teacher to more closely monitor their own behavior. Point sheets are a type of token economy.
Four necessary conditions for any token system: (a) you should be able to administer the tokens effortlessly, (b) tokens should be administered quickly and contingently (if there's a delay, the student might not understand what they've done to receive the token), (c) the delivery of a token must be paired with verbal feedback.
Token economies only work if the rewards are reinforcing. You can give students a reward preference survey. Therefore, a token economy should incorporate a variety of possible rewards rather than just one. Four types of rewards:
- Edible reinforcers: candy, gum, small snack.
- Tangible reinforcers: small toys, books/comics, pencils.
- Activity reinforcers: five minutes early dismissal to lunch; ability to change desks for one day; free time to do art projects, play board games, etc.; homework pass; getting to watch video at end of the day.
- Social reinforcers: opportunity to visit the nurse, opportunity to talk to friend for 5 minutes, lunch with teacher.
Start using smaller-scale or more activity-based rewards such as stickers or extra free time. If these rewards don't work, then move to larger rewards such as tangibles or edibles. Research suggests it may be easier to to successfully fade out an intervention when behavior is being maintained by less power rewards.
How to introduce token economy: (a) provide rationale for why you're doing it, (b) clearly define target behaviors and provide a rationale for why these behaviors are being selected, (c) explicitly state why exhibiting these behaviors will enhance the functioning of individual students or the whole class, (d) the procedures for earning tokens must be made clear (i.e., what they need to do to earn tokens, when they will be able to earn tokens, and how often tokens will be delivered), (e) explain the rewards available, (f) explain the procedures for redeeming tokens.
Variations of the Token Economy
Response Cost. Reinforcers/tokens are taken away contingent on inappropriate behavior. E.g., all students begin day with same number of tokens or no tokens; as day progresses, students' number of tokens can go up or down; students must be given chance to earn back tokens. One study found that token economies to be more effective than response cost.
Level System. Like traditional token economy but expectations intensify as student moves up through different levels of the system. Usually contain four levels. Might look at percentages -- e.g., students must earn 75 percent of possible points for four consecutive weeks before they can move on. Students can also move down. The first level might include small rewards, while the highest level includes larger rewards. Preschool version: all students received shape that was placed in the middle level of a seven-level chart; if student exhibited appropriate behaviors, they received praise and moved up one level; two-four times per day teacher provided a group-based reward to students whose shapes were in the positive levels; students engaged in reward activity while classmates resumed working.
Common Problems
Problem #1: If there are too many students, teacher might have trouble recognizing desirable behavior. If students are routinely ignored for performing good behavior, they might feel that it doesn't matter. One solution: tootling.
Problem #2: There is too long of a delay between the administration of tokens and the receipt of a reward. Temporal discounting -- the phenomenon by which the value of a reward decreases as the amount of time until it is received increases.
Problem #3: If the rewards are priced incorrectly (i.e., rewards are too easy or too tough to obtain). It may be necessary to vary the threshold depending on individual students' abilities (e.g., could reward one student for talking quietly even if they're not silent).
Problem #4: If students have earned enough tickets to receive reward and therefore decide to misbehave. Solution: put expiration date on tokens.
Problem #5: If token economy is not implemented consistently across time.
Fading the Intervention
Many fade the token economy after its initial success. Many worry that students will think that they will come to think they should be rewarded for everything they do. Ways to fade. (1) Gradually reduce the frequency with which tokens are delivered. Continue to provide students with verbal feedback each time they exhibit the target behavior but provide tokens according to an intermittent schedule (e.g., every third or fifth time the behavior occurs) or in a random fashion. You can also reduce the amount of time during which the intervention is implemented (e.g., scale back from all day to select periods). (2) Increase the number of tokens needed to purchase rewards. (3) Increase the length of time that students have to wait to exchange their points for rewards.
* * * * *
Group Contingencies
Advantages of group contingencies: (a) teachers like them because they're efficient (instead of keeping track of several individual point sheets), (b) it feels more fair to some teachers, as they're not only rewarding a small group of students. Group contingencies have been shown to be a powerful intervention.
Independent Group Contingencies. The same rules apply to all students, but the contingency is delivered to individual students based on their individual behavior. Example: token economies. Disadvantages of individual group contingencies: they can be time intensive and difficult to manage given the need to closely monitor the behavior of each student. Dependent or independent group contingencies are easier to manage.
One individual group contingency: the classroom password. At the beginning of class, the teacher announces a classroom password that will be said a certain number times throughout the period, as well as a prize (e.g., 25 pieces of candy). The students are then responsible for placing an X on a sheet of paper each time the teacher says the password. This encourages students to better attend to instruction. At the end of the period, the teacher collects the recording sheets and evaluates the accuracy of each student's estimate. Five students then randomly selected from amount those with correct answers to share the prize for the day.
Dependent Group Contingencies. This is where the behavior or an individual student or a small group of students determines the reward that all students will receive. Example: a group of students is misbehaving; the teacher chooses to reward the entire class when those particular students behave appropriately. The dependent group contingency that receives the highest rating from both teachers and students is one in which rewards are based on the behavior of a randomly selected student.
Example. Teacher provided social skills training focused on making positive verbal statements to peers, told students they could earn a grab bag reward each day for demonstrating target behavior. At the beginning of each day, the names of two students selected at random and announced publicly. If both students made at least four positive statements to peers during the day, the entire class would be able to choose a reward at the end of the day. One concern: by making the students known ahead of time, you're putting a high amount of pressure on them; this may result in aggression directed towards those students. One solution: keep the person or group a secret from the rest of the class. In one study, a randomly selected row of students was observed each time that a chime sounded at predetermined intervals of approximately 12 minutes. Teacher reported how they'd done at the end of the class period. If students were on-task during at least 75 percent of the observed intervals, the class received a reward.
Another example. Every 5 minutes the teacher scanned the class and noted which students were on-task at that moment. At the end of the period, one student's name was pulled from a jar but not shared with the class. If the student met the criteria (4/5 intervals on-task), then the whole class would receive a reward (e.g., homework pass).
Do not punish the whole class based on the performance of select students.
Interdependent Group Contingencies. Lots of research supporting these contingencies. Definition: all students in a group must meet the criteria for all students to receive the reward. Example: Teacher expects all students to enter classroom, remove their coats, hang up their bags, and be seated on the carpet within 90 seconds. If every student meets goal, the class receives 2 extra minutes of computer time. Because all students are working together, there is greater motivation for students to work together. These contingencies are easy for teachers because the decision regarding reinforcement is black and white (either everyone receives the reward or not).
Example. Teacher established three expectations: pay attention and finish your work; get the teacher's permission before speaking; and stay in your seat. Every 45 minutes, a student received a checkmark next to his/her name if he/she exhibited two or more prosocial behaviors during an interval. If every student had five of seven checkmarks on 4/5 days, the class was rewarded. Alternative: If the goal is to decrease an undesired behavior, a threshold could be established for each student's behavior (e.g., all students or the entire class must receive fewer than two reminders to sit quietly). Can also computer average: e.g., if the class lines up within 2 minutes on 80 percent of occasions.
The Good Behavior Game. Original: teams of students compete against one another; the team with the most points awarded a prize. Concern: members of the losing team may respond with anger or disappointment and problem behaviors motivated by jealous could occur. One variant: any team with more than 5 points at the day wins a reward. Another variant: two teams; teacher responded to misbehavior by identifying the student as well as describing the infraction; the student then received a point on the board; the team with the fewest points at the end of the day received a piece of candy; the team with the fewest points at the end of the week received a pizza or cupcake party.
More critiques of the Good Behavior Game: it fails to tell students what they should be doing; it sets up opportunities for students to argue; it encourages retaliation against a team member who caused team to lose points. Alternative: Caught Being Good Game: teachers assign points to teams whose students are following a specific expectation when a silent timer goes off; any team that has three or more points at the end of 10-minute period receives reward.
Mystery Motivator. Rewards made available only on certain days, and the reward is a mystery until revealed. By not revealing reward ahead of time, students' motivation and interest is piqued. Calendar: cover up all squares with a pieces of paper. If goal is met for the day, that particular calendar square is revealed. if there is an "M" on that day, a reward is randomly selected (can draw cards with different rewards on them from manila envelope). If there's not an "M," the teacher provides students with positive verbal feedback.
One can randomize different target behaviors (e.g., out-of-seat behavior, talking out), behavioral criteria (e.g., one to four behavioral occurrences), and available rewards (e.g., small tangibles, free time at the end of class).
Mystery Hero intervention. Goal to decrease the number of verbally disrespectful behaviors. At the start of class, teacher would randomly elect the name of one student (i.e., the Mystery Hero) and place this name in an envelope. In second envelope, teacher would place a randomly selected reward for the day. If at the end of the period, the Mystery Hero had reduce his or her number of inappropriate verbalizations from the previous day, the entire class would receive the Mystery Reward and the Mystery Hero would receive public verbal reinforcement. If at the end of the period the goal was not met, no public statement was made but the teacher would meet privately with the student to provide feedback and encouragement regarding future behavior.
The Timely Transitions Game. Teacher explains the explicit expectations for students during transitions and posts them. Next practice transitions with corrective feedback if necessary. Teacher then explains that the class can earn a reward if the amount of time that it takes to transition is equal to or less than a criterion that will be randomly drawn each day. Teacher times transitions to see how long they take. Class receives 1 point toward larger reward if goal met. Daily point can come in form of letters which spell out reward (e.g., P-A-R-T-Y).
Common problems. (1) The criterion set is not appropriate, either too easy or too hard. First err on the easy side; if students find something new too difficult, they are apt to give up. Raise bar once students have experienced some success. Make sure teams are evenly composed. (2) Individual students may sabotage the group reward. If one student commits too many rule violations, exclude them from the reinforcer if his/her team wins. Another solution: place the student who is disrupting the intervention on their own independent group contingency; student allowed to earn his/her own reward and can rejoin team with improved behavior. Another solution: instead of deducting points for infractions, focus on reinforcing appropriate behavior. (3) Students retaliating against one another. Teach students to provide corrective feedback and encouragement, not yell at peer.
Randomization solves many problems -- e.g., some students don't find the reward equally motivating; some students might find the criterion too high (this also prevents students from no longer working once criterion has been met).
Fading the intervention. In the long term we want students to display these behaviors without external supports. At first reinforcement must occur frequently or even continuously. Creating behaviors that are sustained requires the use of variable reinforcement schedules whereby the behavior is reinforced less frequently -- e.g., having students work for larger rewards but providing those rewards less frequently helps students begin to regulate their own behavior. Also: instead of reinforcing behavior every 5 minutes, start reinforcing every 6 or 7 minutes.
* * * * *
Self-Management
Some argue that the use of teacher-directed interventions means that student behavior remains externally, not internally, managed. Self-management = those interventions in which students are expected to play a more active role in the modification of their behavior. The basis of self-management is self-monitoring. Self-monitoring = systematically observing and recording instances of a predetermined target behavior.
Why is self-management effective? One explanation: Principle of reactivity: a behavior changes when somebody (others or oneself) watches it. Another explanation: Self-reinforcement or punishment plays a role. Self-management has been shown to be effective.
Self-management has been shown to (a) increase on-task behavior, (b) reduce disruptive behaviors (e.g., inappropriate verbalizations, rule-following behavior), (c) teach ASD kids appropriate play behaviors.
Ways to self-record. (a) If you want to know how frequently a behavior occurs. Give student blank form and ask them to keep a tally of how frequently behavior occurs (Form 7.1). Easy for students to learn. (b) If you want to know whether the behavior occurs. Use a checklist (Form 7.2). (c) When the behavior cannot be easily counted or checked off. Interval recording: observing and recording one or more target behavior at predetermined periods of time -- e.g., student marks sheet to indicate he's paying attention when tone sounds (Form 7.3). (d) Both state and event behaviors can be assessed using a rating scale format. This involves making an overall judgment at the end of a block of time or day. E.g., can use 5-point scale. (Form 7.4). (e) Directly record the permanent products. E.g., a student records the number of math problems completed by the end of class or number of pages read of a story.
Should student ratings be externally evaluated? An advantage of being evaluated is that the student can use the feedback to make conscious and deliberate changes. Two types of evaluations: comparison of ratings to a predetermined goal (e.g., goal is to call out no more than 7 times), comparison of ratings to an external standard (student and teacher both compare ratings, then teacher meets with student to compare recording sheets).
It's not clear whether self-management needs to be paired with extrinsic rewards to be effective.
Introducing the intervention. See Coach Card 7.3. Teachers should explain what behaviors students should demonstrate and how exhibiting those behaviors will help them to succeed.
Classwide self-management. Example: Middle schoolers expected to do SLANT: students taught to self-assess their SLANT by holding their thumbs up, down, or sideways each time that a timer buzzed. Can also have students monitor their behavior in groups: e.g., team chart placed at center of each table (p. 140).
Self-management with interdependent group contingency -- e.g., students taught to self-evaluate the degree to which they were prepared for class; students received points based on absolute performance and accuracy. With group contingency -- e.g., class divided into groups; every 10 minutes students instructed to record whether they met behavior; at end of class teacher drew a percentage criterion from bag (e.g., 75%); teams who met that received reward. Peer-assisted self-management (p. 141).
Common Problems. #1: Students not rating accurate. Problem #2: Self-monitoring is too cognitively demanding. Problem #3: Students too focused on self-monitoring.
Fading the intervention. Standard fading techniques: gradually reduce frequency with which rewards are earned, increase the threshold for earning rewards. Specific fading techniques for self-management: reduce the number of times student is expected to self-monitor (e.g., move from every 10 minutes to every 20 minutes); instead of asking student to individually assess whether each specific behavior occurred, he might start to provide one global assessment; move from teacher-determined to student-determined rating periods; teacher can move to spot checks, that is, monitoring student's behavior to once every two days.
* * * * *
Collecting Classwide Data
Must first operationally define target behavior (desirable and undesirable target behaviors); this means (a) listening examples of what target behavior is and is not, (b) passing the stranger test (complete stranger could read definition and know what to observe), (c) passing the dead man test (i.e., behavior could not be demonstrated by a dead man; e.g., sitting on rug without bothering others would not pass this test because a dead man could refrain from bothering others while on rug; change behavior to sits quietly while focusing on teacher).
All behaviors are either event or state behaviors. Event behaviors have a clear beginning and end (can be easily counted). State behaviors vary in duration and are difficult to count. Recording event behaviors: (a) frequency (Figure 8.1), (b) scatterplot can be used to better understand when problem behavior is occurring and to identify any patterns (Figure 8.2), (c) duration of behavior, (d) latency (how much time elapses before a behavior is initiated). Recording state behaviors are best recorded using interval procedures; this involves observing and recording target behavior at predetermined periods of time (Figure 8.3, Figure 8.4, Planned Activity Check). Direct Behavior Rating (DBR) can be used for individual students or whole class.
Permanent Product Data = naturally occurring sources of data that already exist. Examples: tests, completion of assignments, submission of homework. Self-management examples: students rate their behaviors and turn them in. Example of token economy: add a step by keeping a central log of the number of tokens earned each day.
Baseline data is essential: needed to verify that there is in fact a problem, needed to set appropriate goals for intervention. How long should baseline data be collected? There's no magic number, but the general rule is that it should be collected until you can discern a predictable pattern of behavior. Fewer data points will be needed to derive a general sense of baseline performance when estimates of behavior are highly consistent.
How long should intervention last? RTI Action Network recommends 6-12 weeks. Others recommend that intervention should be implemented long enough to be able to say with confidence that the intervention was or was not effective.
Evaluating Data. Method #1: Visual Analysis: (a) can examine whether there is a change in the degree or variability of the data (i.e., how much the values change from day to day), (b) can examine whether there is a change in the trend (slope) of the data, (c) can examine whether there is a change in the mean level of the behavior since baseline (or median to exclude outliers). Must consider which of three methods is best for situation; won't generally use all three.
Method #2: Quantifying Effect Size: (a) Percentage of Nonoverlapping Data (PND), (b) Percentage of All Nonoverlapping Data (p. 177).
Ways to fade effective intervention: limit intervention to certain times, deliver reinforcement less frequently. Continue to collect outcome data, even if only every week as opposed to every day.
If intervention hasn't worked. It could be that intervention wasn't implemented with fidelity. It could be that the intervention wasn't implemented with sufficient intensity. It could be that intervention simply wasn't effective.
* * * * *
Assessing Intervention Fidelity
To improve intervention fidelity, break intervention down into smaller parts.
Ways to measure treatment fidelity: (a) Direct observation (produces accurate estimates; limitations: requires additional manpower and also there can be a reactivity effect in which the teacher changes behavior when being observed. (b) Self-report (problem with this is that teachers tend to overestimate their fidelity to a treatment. (c) Permanent products.
Along with assessing whether an intervention is being implemented, we should assess how it's implemented. Observer can rate the quality of delivery on a scale of 1-5. Should also consider whether students responded appropriately to the intervention (e.g., were they bored, engaged?) (can also use qualitative assessment for this).
What level of fidelity is sufficient? Many would say that we should shoot for 80 percent. That said, there are elements of most interventions that are more important than others.
Strategies for Promoting Treatment Fidelity. Prior to Intervention: intervention must be acceptable by consumers; intervention must be feasible (e.g., not too time-consuming); intervention must be well understood. During Implementation. Teachers might not immediate feedback, coaching, modeling, practicing, ongoing support. Teachers might need cueing (e.g., a visual reminder). After Implementation. Drift can happen. Therefore should have periodic checkups.
* * * * *
Individualizing Intervention
Classwide interventions will help 80-85 percent of students. To reach these students, teacher can consider intensifying elements of intervention: (a) increase dosage (e.g., increase the amount of time each day an intervention is implemented; keep intervention dosage for class intact but increase intervention dosage for certain students), (b) alter intervention delivery (e.g., enhance explicit instruction, "I do, we do, you do," (c) increase the frequency and explicitness of feedback, (d) increase the power of the reinforcement (can create reinforcement survey or use forced-choice reinforcement menu, p. 204).
Incorporating additional intervention components to intensify supports: (a) add a self-monitoring component, (b) introduce an external interventionist, (c) incorporate a home-based component.
* * * * *
Group Contingencies
Advantages of group contingencies: (a) teachers like them because they're efficient (instead of keeping track of several individual point sheets), (b) it feels more fair to some teachers, as they're not only rewarding a small group of students. Group contingencies have been shown to be a powerful intervention.
Independent Group Contingencies. The same rules apply to all students, but the contingency is delivered to individual students based on their individual behavior. Example: token economies. Disadvantages of individual group contingencies: they can be time intensive and difficult to manage given the need to closely monitor the behavior of each student. Dependent or independent group contingencies are easier to manage.
One individual group contingency: the classroom password. At the beginning of class, the teacher announces a classroom password that will be said a certain number times throughout the period, as well as a prize (e.g., 25 pieces of candy). The students are then responsible for placing an X on a sheet of paper each time the teacher says the password. This encourages students to better attend to instruction. At the end of the period, the teacher collects the recording sheets and evaluates the accuracy of each student's estimate. Five students then randomly selected from amount those with correct answers to share the prize for the day.
Dependent Group Contingencies. This is where the behavior or an individual student or a small group of students determines the reward that all students will receive. Example: a group of students is misbehaving; the teacher chooses to reward the entire class when those particular students behave appropriately. The dependent group contingency that receives the highest rating from both teachers and students is one in which rewards are based on the behavior of a randomly selected student.
Example. Teacher provided social skills training focused on making positive verbal statements to peers, told students they could earn a grab bag reward each day for demonstrating target behavior. At the beginning of each day, the names of two students selected at random and announced publicly. If both students made at least four positive statements to peers during the day, the entire class would be able to choose a reward at the end of the day. One concern: by making the students known ahead of time, you're putting a high amount of pressure on them; this may result in aggression directed towards those students. One solution: keep the person or group a secret from the rest of the class. In one study, a randomly selected row of students was observed each time that a chime sounded at predetermined intervals of approximately 12 minutes. Teacher reported how they'd done at the end of the class period. If students were on-task during at least 75 percent of the observed intervals, the class received a reward.
Another example. Every 5 minutes the teacher scanned the class and noted which students were on-task at that moment. At the end of the period, one student's name was pulled from a jar but not shared with the class. If the student met the criteria (4/5 intervals on-task), then the whole class would receive a reward (e.g., homework pass).
Do not punish the whole class based on the performance of select students.
Interdependent Group Contingencies. Lots of research supporting these contingencies. Definition: all students in a group must meet the criteria for all students to receive the reward. Example: Teacher expects all students to enter classroom, remove their coats, hang up their bags, and be seated on the carpet within 90 seconds. If every student meets goal, the class receives 2 extra minutes of computer time. Because all students are working together, there is greater motivation for students to work together. These contingencies are easy for teachers because the decision regarding reinforcement is black and white (either everyone receives the reward or not).
Example. Teacher established three expectations: pay attention and finish your work; get the teacher's permission before speaking; and stay in your seat. Every 45 minutes, a student received a checkmark next to his/her name if he/she exhibited two or more prosocial behaviors during an interval. If every student had five of seven checkmarks on 4/5 days, the class was rewarded. Alternative: If the goal is to decrease an undesired behavior, a threshold could be established for each student's behavior (e.g., all students or the entire class must receive fewer than two reminders to sit quietly). Can also computer average: e.g., if the class lines up within 2 minutes on 80 percent of occasions.
The Good Behavior Game. Original: teams of students compete against one another; the team with the most points awarded a prize. Concern: members of the losing team may respond with anger or disappointment and problem behaviors motivated by jealous could occur. One variant: any team with more than 5 points at the day wins a reward. Another variant: two teams; teacher responded to misbehavior by identifying the student as well as describing the infraction; the student then received a point on the board; the team with the fewest points at the end of the day received a piece of candy; the team with the fewest points at the end of the week received a pizza or cupcake party.
More critiques of the Good Behavior Game: it fails to tell students what they should be doing; it sets up opportunities for students to argue; it encourages retaliation against a team member who caused team to lose points. Alternative: Caught Being Good Game: teachers assign points to teams whose students are following a specific expectation when a silent timer goes off; any team that has three or more points at the end of 10-minute period receives reward.
Mystery Motivator. Rewards made available only on certain days, and the reward is a mystery until revealed. By not revealing reward ahead of time, students' motivation and interest is piqued. Calendar: cover up all squares with a pieces of paper. If goal is met for the day, that particular calendar square is revealed. if there is an "M" on that day, a reward is randomly selected (can draw cards with different rewards on them from manila envelope). If there's not an "M," the teacher provides students with positive verbal feedback.
One can randomize different target behaviors (e.g., out-of-seat behavior, talking out), behavioral criteria (e.g., one to four behavioral occurrences), and available rewards (e.g., small tangibles, free time at the end of class).
Mystery Hero intervention. Goal to decrease the number of verbally disrespectful behaviors. At the start of class, teacher would randomly elect the name of one student (i.e., the Mystery Hero) and place this name in an envelope. In second envelope, teacher would place a randomly selected reward for the day. If at the end of the period, the Mystery Hero had reduce his or her number of inappropriate verbalizations from the previous day, the entire class would receive the Mystery Reward and the Mystery Hero would receive public verbal reinforcement. If at the end of the period the goal was not met, no public statement was made but the teacher would meet privately with the student to provide feedback and encouragement regarding future behavior.
The Timely Transitions Game. Teacher explains the explicit expectations for students during transitions and posts them. Next practice transitions with corrective feedback if necessary. Teacher then explains that the class can earn a reward if the amount of time that it takes to transition is equal to or less than a criterion that will be randomly drawn each day. Teacher times transitions to see how long they take. Class receives 1 point toward larger reward if goal met. Daily point can come in form of letters which spell out reward (e.g., P-A-R-T-Y).
Common problems. (1) The criterion set is not appropriate, either too easy or too hard. First err on the easy side; if students find something new too difficult, they are apt to give up. Raise bar once students have experienced some success. Make sure teams are evenly composed. (2) Individual students may sabotage the group reward. If one student commits too many rule violations, exclude them from the reinforcer if his/her team wins. Another solution: place the student who is disrupting the intervention on their own independent group contingency; student allowed to earn his/her own reward and can rejoin team with improved behavior. Another solution: instead of deducting points for infractions, focus on reinforcing appropriate behavior. (3) Students retaliating against one another. Teach students to provide corrective feedback and encouragement, not yell at peer.
Randomization solves many problems -- e.g., some students don't find the reward equally motivating; some students might find the criterion too high (this also prevents students from no longer working once criterion has been met).
Fading the intervention. In the long term we want students to display these behaviors without external supports. At first reinforcement must occur frequently or even continuously. Creating behaviors that are sustained requires the use of variable reinforcement schedules whereby the behavior is reinforced less frequently -- e.g., having students work for larger rewards but providing those rewards less frequently helps students begin to regulate their own behavior. Also: instead of reinforcing behavior every 5 minutes, start reinforcing every 6 or 7 minutes.
* * * * *
Self-Management
Some argue that the use of teacher-directed interventions means that student behavior remains externally, not internally, managed. Self-management = those interventions in which students are expected to play a more active role in the modification of their behavior. The basis of self-management is self-monitoring. Self-monitoring = systematically observing and recording instances of a predetermined target behavior.
Why is self-management effective? One explanation: Principle of reactivity: a behavior changes when somebody (others or oneself) watches it. Another explanation: Self-reinforcement or punishment plays a role. Self-management has been shown to be effective.
Self-management has been shown to (a) increase on-task behavior, (b) reduce disruptive behaviors (e.g., inappropriate verbalizations, rule-following behavior), (c) teach ASD kids appropriate play behaviors.
Ways to self-record. (a) If you want to know how frequently a behavior occurs. Give student blank form and ask them to keep a tally of how frequently behavior occurs (Form 7.1). Easy for students to learn. (b) If you want to know whether the behavior occurs. Use a checklist (Form 7.2). (c) When the behavior cannot be easily counted or checked off. Interval recording: observing and recording one or more target behavior at predetermined periods of time -- e.g., student marks sheet to indicate he's paying attention when tone sounds (Form 7.3). (d) Both state and event behaviors can be assessed using a rating scale format. This involves making an overall judgment at the end of a block of time or day. E.g., can use 5-point scale. (Form 7.4). (e) Directly record the permanent products. E.g., a student records the number of math problems completed by the end of class or number of pages read of a story.
Should student ratings be externally evaluated? An advantage of being evaluated is that the student can use the feedback to make conscious and deliberate changes. Two types of evaluations: comparison of ratings to a predetermined goal (e.g., goal is to call out no more than 7 times), comparison of ratings to an external standard (student and teacher both compare ratings, then teacher meets with student to compare recording sheets).
It's not clear whether self-management needs to be paired with extrinsic rewards to be effective.
Introducing the intervention. See Coach Card 7.3. Teachers should explain what behaviors students should demonstrate and how exhibiting those behaviors will help them to succeed.
Classwide self-management. Example: Middle schoolers expected to do SLANT: students taught to self-assess their SLANT by holding their thumbs up, down, or sideways each time that a timer buzzed. Can also have students monitor their behavior in groups: e.g., team chart placed at center of each table (p. 140).
Self-management with interdependent group contingency -- e.g., students taught to self-evaluate the degree to which they were prepared for class; students received points based on absolute performance and accuracy. With group contingency -- e.g., class divided into groups; every 10 minutes students instructed to record whether they met behavior; at end of class teacher drew a percentage criterion from bag (e.g., 75%); teams who met that received reward. Peer-assisted self-management (p. 141).
Common Problems. #1: Students not rating accurate. Problem #2: Self-monitoring is too cognitively demanding. Problem #3: Students too focused on self-monitoring.
Fading the intervention. Standard fading techniques: gradually reduce frequency with which rewards are earned, increase the threshold for earning rewards. Specific fading techniques for self-management: reduce the number of times student is expected to self-monitor (e.g., move from every 10 minutes to every 20 minutes); instead of asking student to individually assess whether each specific behavior occurred, he might start to provide one global assessment; move from teacher-determined to student-determined rating periods; teacher can move to spot checks, that is, monitoring student's behavior to once every two days.
* * * * *
Collecting Classwide Data
Must first operationally define target behavior (desirable and undesirable target behaviors); this means (a) listening examples of what target behavior is and is not, (b) passing the stranger test (complete stranger could read definition and know what to observe), (c) passing the dead man test (i.e., behavior could not be demonstrated by a dead man; e.g., sitting on rug without bothering others would not pass this test because a dead man could refrain from bothering others while on rug; change behavior to sits quietly while focusing on teacher).
All behaviors are either event or state behaviors. Event behaviors have a clear beginning and end (can be easily counted). State behaviors vary in duration and are difficult to count. Recording event behaviors: (a) frequency (Figure 8.1), (b) scatterplot can be used to better understand when problem behavior is occurring and to identify any patterns (Figure 8.2), (c) duration of behavior, (d) latency (how much time elapses before a behavior is initiated). Recording state behaviors are best recorded using interval procedures; this involves observing and recording target behavior at predetermined periods of time (Figure 8.3, Figure 8.4, Planned Activity Check). Direct Behavior Rating (DBR) can be used for individual students or whole class.
Permanent Product Data = naturally occurring sources of data that already exist. Examples: tests, completion of assignments, submission of homework. Self-management examples: students rate their behaviors and turn them in. Example of token economy: add a step by keeping a central log of the number of tokens earned each day.
Baseline data is essential: needed to verify that there is in fact a problem, needed to set appropriate goals for intervention. How long should baseline data be collected? There's no magic number, but the general rule is that it should be collected until you can discern a predictable pattern of behavior. Fewer data points will be needed to derive a general sense of baseline performance when estimates of behavior are highly consistent.
How long should intervention last? RTI Action Network recommends 6-12 weeks. Others recommend that intervention should be implemented long enough to be able to say with confidence that the intervention was or was not effective.
Evaluating Data. Method #1: Visual Analysis: (a) can examine whether there is a change in the degree or variability of the data (i.e., how much the values change from day to day), (b) can examine whether there is a change in the trend (slope) of the data, (c) can examine whether there is a change in the mean level of the behavior since baseline (or median to exclude outliers). Must consider which of three methods is best for situation; won't generally use all three.
Method #2: Quantifying Effect Size: (a) Percentage of Nonoverlapping Data (PND), (b) Percentage of All Nonoverlapping Data (p. 177).
Ways to fade effective intervention: limit intervention to certain times, deliver reinforcement less frequently. Continue to collect outcome data, even if only every week as opposed to every day.
If intervention hasn't worked. It could be that intervention wasn't implemented with fidelity. It could be that the intervention wasn't implemented with sufficient intensity. It could be that intervention simply wasn't effective.
* * * * *
Assessing Intervention Fidelity
To improve intervention fidelity, break intervention down into smaller parts.
Ways to measure treatment fidelity: (a) Direct observation (produces accurate estimates; limitations: requires additional manpower and also there can be a reactivity effect in which the teacher changes behavior when being observed. (b) Self-report (problem with this is that teachers tend to overestimate their fidelity to a treatment. (c) Permanent products.
Along with assessing whether an intervention is being implemented, we should assess how it's implemented. Observer can rate the quality of delivery on a scale of 1-5. Should also consider whether students responded appropriately to the intervention (e.g., were they bored, engaged?) (can also use qualitative assessment for this).
What level of fidelity is sufficient? Many would say that we should shoot for 80 percent. That said, there are elements of most interventions that are more important than others.
Strategies for Promoting Treatment Fidelity. Prior to Intervention: intervention must be acceptable by consumers; intervention must be feasible (e.g., not too time-consuming); intervention must be well understood. During Implementation. Teachers might not immediate feedback, coaching, modeling, practicing, ongoing support. Teachers might need cueing (e.g., a visual reminder). After Implementation. Drift can happen. Therefore should have periodic checkups.
* * * * *
Individualizing Intervention
Classwide interventions will help 80-85 percent of students. To reach these students, teacher can consider intensifying elements of intervention: (a) increase dosage (e.g., increase the amount of time each day an intervention is implemented; keep intervention dosage for class intact but increase intervention dosage for certain students), (b) alter intervention delivery (e.g., enhance explicit instruction, "I do, we do, you do," (c) increase the frequency and explicitness of feedback, (d) increase the power of the reinforcement (can create reinforcement survey or use forced-choice reinforcement menu, p. 204).
Incorporating additional intervention components to intensify supports: (a) add a self-monitoring component, (b) introduce an external interventionist, (c) incorporate a home-based component.
No comments:
Post a Comment