LeBlanc et al. (2016) A Proposed Model for Selecting Measurement Procedures for Problem Behavior PDF

Behav Analysis Practice (2016) 9:77–83 DOI 10.1007/s40617-015-0063-2 TECHNICAL ARTICLE A Proposed Model for Selecting Measurement Procedures for the Assessment and Treatment of Problem Behavior Linda A. LeBlanc 1 & Paige B. Raetz 1 & Tyra P. Sellers 1 & James E. Carr 2 Published online: 13 Oc...

Behav Analysis Practice (2016) 9:77–83 DOI 10.1007/s40617-015-0063-2 TECHNICAL ARTICLE A Proposed Model for Selecting Measurement Procedures for the Assessment and Treatment of Problem Behavior Linda A. LeBlanc 1 & Paige B. Raetz 1 & Tyra P. Sellers 1 & James E. Carr 2 Published online: 13 October 2015 # Association for Behavior Analysis International 2015 Abstract Practicing behavior analysts frequently assess and Numerous textbooks on ABA (e.g., Cooper et al. 2007; treat problem behavior as part of their ongoing job responsi- Mayer et al. 2012) and behavioral research methods bilities. Effective measurement of problem behavior is critical (e.g., Bailey and Burch 2002; Barlow et al. 2008; Johnston to success in these activities because some measures of prob- and Pennypacker 2008; Kazdin 2011) describe various lem behavior provide more accurate and complete information measurement procedures, often doing so in great detail. about the behavior than others. However, not every measure- Some of these textbooks provide guidance about matching ment procedure is appropriate for every problem behavior and specific measurement procedures with specific applied cir- therapeutic circumstance. We summarize the most commonly cumstances (e.g., Mayer et al.). However, to our knowledge, used measurement procedures, describe the contexts for none provide an integrated model for considering multiple which they are most appropriate, and propose a clinical aspects of an applied situation and the relative suitability decision-making model for selecting measurement produces of each primary measurement procedure. The recent given certain features of the behavior and constraints of the behavioral literature for practitioners suggests one possible therapeutic environment. solution. Clinical decision-making models provide a means to guide Keywords Clinical decision-making. Data collection. the selection of procedures when their optimal implementa- Measurement. Problem behavior tion depends on a match with specific environmental circum- stances. For example, Grow et al. (2009) and Geiger et al. The systematic measurement of behavior is foundational to (2010) described decision-making models for selecting be- the delivery of applied behavior-analytic services (Baer et al. tween multiple scientifically supported and function-based 1968; Sidman 1960). A practitioner’s choices about the pro- treatments for problem behavior maintained by attention and cedures used to track behavior over time are pivotal because escape, respectively. The models consist of a series of ques- direct observation data impact other important decisions. For tions that can be answered to lead a practitioner to recommen- example, proper measurement procedures allow one to exam- dations about interventions that are optimally matched to clin- ine the function of problem behavior and decide when to im- ical considerations (e.g., client safety, available resources, plement or change interventions. Indeed, there are no valid clinical goals). Fiske and Delmolino (2012) used a similar circumstances under which applied behavior analysis (ABA) approach to present a preliminary model for selecting between should be practiced without the collection of meaningful data. discontinuous measurement procedures for problem behavior. Their model includes two important questions (i.e., initial be- havior rate, terminal behavior goal) to consider when selecting * Linda A. LeBlanc between three discontinuous measurement procedures (i.e., [email protected] momentary time sampling, partial-interval recording, whole- interval recording), along with recommendations for design- 1 Trumpet Behavioral Health, 390 Union Blvd., Suite 300, ing the selected measurement procedure. Although a useful Lakewood, CO 80228, USA contribution to the literature, other measurement procedures 2 Behavior Analyst Certification Board, Littleton, CO, USA (e.g., event recording, permanent-product recording) and 78 Behav Analysis Practice (2016) 9:77–83 Requires constant vigilance; requires a timing device Requires constant vigilance; requires a timing device practical considerations should ultimately be included in a An indirect assessment of behavior; behavior must Requires a reliable and valid measurement device Generates an estimate of behavior; systematically (e.g., decibel meter) or rating system; requires Generates an estimate of behavior. Inappropriate for short-duration or low-frequency behavior; Requires constant vigilance; may be impractical reliably produce the product and be the only comprehensive model of problem-behavior measurement pro- for high-frequency or non-discrete behavior overestimates the occurrence of behavior; cedure selection. The purpose of the present article is to illustrate a clinical decision-making model for selecting between many problem- behavior measurement procedures in everyday practice. The requires a timing device. requires a timing device model includes a variety of measurement procedures and prac- source of the product tical considerations, including observability of behavior, per- constant vigilance Potential limitations sonnel resources and constraints, dimensions of behavior, and the nature of behavior (free or restricted operant), in addition to Fiske and Delmolino’s (2012) consideration about the ter- minal behavior goal. Direct measures of the temporal relation between the discriminative stimulus does not require constant vigilance; Various (e.g., wound size, number of holes Allows measurement of behavior that automated recording possible with multiple individuals or behaviors; Measurement Considerations and Decision-Making Does not require constant vigilance Allows concurrent measurement of some dimensions (e.g., volume) good correspondence to event Direct measures of the behavior; Frequency, rate, percentage of opportunities Direct measures of the behavior Total duration, mean duration, percentage Direct measures of the behavior recording compared to other occurs at inaccessible times discontinuous procedures Development of the Model (including frequency) The model described here was developed as part of a and the behavior clinical-practice standardization initiative within a large ABA human-service agency. The initiative was designed Strength(s) to synthesize best-practice guidelines and develop clinical decision-making tools in important areas of practice for delivering ABA services to individuals with special needs. The first three authors surveyed the published literature to Various (e.g., mean decibel, mean rating Features, strengths, and potential limitations of measurement procedures for problem behavior identify empirical articles, literature reviews, book chapters, Percentage of samples or intervals in Percentage of intervals in which the and textbooks on behavioral measurement. The group then which the behavior occurred synthesized the literature to develop five best-practice (for restricted operants) guidelines for collecting data on problem behavior. Implementation of the first of the guidelines—select an Resulting measure(s) behavior occurred optimal measurement system—was facilitated by the devel- Mean latency opment of the measurement summary in Table 1 and the per event) duration in wall) decision-making model in Fig. 1. Measurement Procedures to offset for each behavior as it occurs Record whether a behavior occurred at Record whether a behavior occurred at between the discriminative stimulus all during specific time intervals for Record the amount of time from onset Record the amount of time that passes Record each instance of behavior as it force, volume) for each instance of Record a dimension of intensity (e.g., The decision-making model includes seven different measure- Permanent-product recording Document the effects of a behavior defined observation period(s) and the onset of the behavior ment procedures as terminal points in the model. These pro- a given moment for defined cedures and their optimal circumstances are defined in the observation period(s) following section. on the environment Event Recording Event recording encompasses any proce- the behavior Description dure in which the frequency of each behavior is recorded occurs during an observation (aka frequency recording). Data from event recording are typically summarized as the frequency (count) of responses or response rate (frequency divided by Momentary time sampling time). For behaviors that have a limited opportunity to occur, Partial-interval recording Measurement procedure such as problem behavior that only occurs in response to a Duration recording Intensity recording task presentation, event-recording data are often summarized Latency recording Event recording as the percentage of opportunities in which behavior occurred (e.g., percentage of trials with problem behavior). The primary Table 1 strength of event recording is that it provides information about a specific and important behavioral dimension—its Behav Analysis Practice (2016) 9:77–83 79 Is the problem behavior observable? YES NO Is the problem behavior Does the problem behavior discrete and countable? produce a measurable, physical change in the environment? YES NO YES NO Are observer resources Are observer resources Permanent Start brainstorming with sufficient to count each sufficient to Product your supervisors about NO instance of problem continuously monitor Recording indirect measures, covert behavior? problem behavior? observation, or other behaviors to observe. YES YES NO Partial Momentary Can problem behavior Interval Time occur at any time? Recording Sampling Is one of the following NO behavior dimensions the primary concern or an YES important secondary Event measure for designing Recording treatment? (Percentage of Opportunities) YES NO Duration Latency Intensity Event Recording Duration Latency Intensity Recording* Recording* Recording* (e.g., force, volume) Fig. 1 A decision-making model for selecting measurement procedures for problem behavior. Note: an asterisk also generates a frequency count frequency of occurrence. However, event recording is only an observation or the total duration of all behaviors during appropriate for behaviors that have clear beginnings and end- an observation. The latter measure might also be expressed ings and do not occur so frequently that it is impossible to as the proportion of the observation in which behavior accurately record them. Event recording is also best suited occurred. Duration recording is a preferred measurement for behaviors that occur for comparable durations. In addition, procedure when information about how the long the behav- event recording requires the observer to constantly monitor ior occurs is the most relevant dimension of interest. In the behavior, which may not be feasible in many service addition, duration recording in which the duration of each environments. behavior is documented also generates a frequency mea- sure. Like event recording, however, duration recording re- Duration Recording Duration recording involves the mea- quires constant vigilance, which might limit its practicality. surement of the amount of time each behavior occurs dur- Duration recording also requires a timing device (e.g., stop- ing an observation. Duration-recording data are typically watch, clock app) that must be easily accessible yet summarized as the mean duration of each behavior during unobtrusive. 80 Behav Analysis Practice (2016) 9:77–83 Latency Recording Latency recording involves the measure- Partial-Interval Recording Partial-interval recording in- ment of the amount of time (usually in seconds) it takes each volves documenting whether the behavior occurs in each of behavior to occur following a specific environmental event a consecutive series of brief time periods. Partial-interval data (e.g., a discriminative stimulus) during an observation. are typically summarized as the percentage of intervals in Latency-recording data are typically summarized as the mean which behavior was scored. Because partial-interval recording latency to each behavior during an observation. Latency re- does not require constant vigilance (i.e., one no longer needs cording is a preferred measurement procedure when informa- to observe within an interval after a behavior has been scored), tion about a behavior’s latency is the dimension of interest. it is often used to measure high-rate behavior, as well as mul- For example, Call et al. (2009) evaluated the mean latency to tiple forms of behavior. However, partial-interval recording is problem behavior during different tasks as an index of each associated with several disadvantages. Unlike event, duration, task’s aversive properties (i.e., tasks associated with low la- latency, and intensity recording, which are continuous record- tencies were aversive). In addition, latency recording also gen- ing procedures, partial-interval recording is a discontinuous erates a frequency measure. However, like duration recording, measurement procedure because some behaviors are deliber- latency recording requires constant vigilance (for both the ately not recorded. Thus, partial-interval recording does not behavior and antecedent event) and a timing device. produce complete data about any behavioral dimension but instead generates an estimate of the frequency and duration Intensity Recording Intensity recording involves the mea- of behavior. In addition, partial-interval recording consistently surement of the intensity (or magnitude) of each behavior overestimates the level of behavior, and this overestimation is during an observation. Intensity might be recorded by the exaggerated by long intervals and observation periods (Wirth behavior’s force, loudness, or other relevant characteristic. et al. 2013). We refer the reader to Fiske and Delmolino Recording these characteristics can be done objectively (e.g., (2012) for a discussion about the determination of interval decibels) or subjectively (e.g., rating a behavior’s force on a 5- and observation-period sizes. point scale). How intensity-recording data are summarized after an observation depends on the specific measured charac- Momentary Time Sampling Momentary time sampling in- teristic, but the mean intensity is likely a relevant measure for volves documenting whether the behavior occurs at the end of many situations. Alternatively, the percentage of behaviors in each of a consecutive series of brief time periods. Momentary an observation that exceeded a certain threshold might also be time sampling data are typically summarized as the percentage relevant. Intensity recording is a preferred measurement pro- of intervals in which behavior was scored. Because momen- cedure when information about a behavior’s intensity is the tary time sampling does not require constant vigilance, it is dimension of interest. In addition, intensity recording also often used to measure multiple forms of behavior or the be- generates a frequency measure. However, intensity recording havior of multiple individuals. However, momentary time requires constant vigilance, as well as a reliable and valid sampling is associated with several disadvantages. measurement system. The latter requirements are especially Momentary time sampling is a discontinuous measurement important when intensity recording involves equipment or a procedure and thus does not produce complete data about subjective rating system. any behavioral dimension but instead generates an estimate of behavior. In addition, the type of error generated by mo- Permanent-Product Recording Permanent-product record- mentary time sampling is inconsistent (unlike the overestima- ing involves the measurement of behavior by its physical im- tion of partial-interval recording) and is exaggerated by long pact on the environment. Examples of permanent-product re- intervals and observation periods (Wirth et al. 2013). We refer cording include measuring tissue damage from self-injury the reader to Fiske and Delmolino (2012) for a discussion (e.g., Grace et al. 1996) and counting the number of food items about the determination of interval and observation-period missing as evidence of food stealing (e.g., Maglieri et al. sizes. 2000). How permanent-product recording data are summa- rized after an observation depends on the specific evidence Using the Model produced by the behavior, but the mean or frequency of these events is likely a relevant measure for many situations. The decision-making model consists of a series of questions Permanent-product recording is a preferred measurement pro- that a practitioner can ask and answer to select the optimal cedure when direct observation of the behavior is impossible measurement procedure for problem behavior given specific or impractical. However, the utility of this procedure is limited client and environment circumstances. This model is intended by its indirect nature (i.e., behavior is not directly observed) for use by behavior analysts with experience in the assessment and by the following required conditions: the behavior must and treatment of problem behavior and measurement. The reliably produce the product, and the product must not be model depicted in Fig. 1 may be most useful to those who frequently produced by any other behavior or event. are relatively new to practice and in need of a systematic guide Behav Analysis Practice (2016) 9:77–83 81 to selecting optimal measurement procedures for problem manner for off-task behavior to occur. Off-task behavior behavior. would likely be recorded using duration recording unless re- The questions in the model focus on several important con- sources prohibited constant vigilance, in which case a discon- siderations in the order that logically governs selection of a tinuous measurement procedure (e.g., momentary time sam- measure: (a) specific characteristics of the behavior, (b) person- pling) would be used. As another example, a demand must be nel resources and constraints, (c) important dimensions of be- presented in a context in which compliance could reasonably havior, and (d) the nature of behavior as a free or restricted occur in order for there to be any opportunity for noncompli- operant. That is, behaviors that occur covertly, or cannot be ant behavior to occur. Noncompliance would likely be scored directly observed, present unique measurement concerns that as the percentage of opportunities or demands resulting in often prevent direct observation, rendering all of the other ques- noncompliant behavior. It may also be the case that protective tions moot. Thus, this question is posed first in the manuscript equipment or the absence of some person or material elimi- and a Byes^ answer ends the process with options to use nates the possibility of the behavior occurring. For example, permanent-product recording or to brainstorm other strategies protective arm splints might be used to prevent the occurrence for gaining access to the behavior in real time (i.e., overtly). The of very severe hand or arm biting or intense head hitting. As next question refers to whether the behavior is discrete and another example, aggression towards a sibling can only occur countable, regardless of resources, in order to make the critical when the sibling is present. In these instances, it is best to decision between a continuous (e.g., event, duration, latency, collect a continuous measure (e.g., frequency, duration) but and intensity recording) and a discontinuous (e.g., time sam- use a denominator that reflects the restricted observation win- pling, interval recording) measurement procedure. Some be- dow in which the behavior can occur in the conversion to a haviors may be initiated as a single instance but occur for a rate or percentage duration measure. long duration (e.g., vocal stereotypy, hand mouthing, off-task If the behavior can occur at any time, consider all dimen- behavior), making it difficult for a continuous measurement sions of the response and select the ones that are most criti- procedure to capture and provide the most useful information cally important to fully capture the important features of the about the behavior. When it is possible to capture information behavior and the potential change in the behavior that may on each event, continuous measurement procedures are strong- occur due to intervention. This should lead you to select the ly preferred because they eliminate the error associated with most appropriate type of continuous measurement procedure estimate-based discontinuous measurement (see prior descrip- or a combination of those procedures. The most commonly tions of momentary time sampling and partial-interval record- used continuous recording procedure is event recording in ing). Some discontinuous measurement procedures (i.e., which a frequency (count) is often converted to a rate measure whole-interval recording) tend to systematically underestimate by dividing it by the duration of the observation window. The the actual level of behavior and thus are inappropriate for mea- inter-response time (IRT) can also be calculated from event suring problem behavior. For this reason, whole-interval re- recording to inform intervention planning (e.g., the start inter- cording is not recommended in our model. val for NCR). When other dimensions of the behavior (e.g., Personnel and resource constraints figure into the model duration, intensity) are not particularly important or are rela- next. Note that even if the behavior itself is discrete and ob- tively equal across instances of the behavior, event recording servable and thus, amenable to continuous measurement, the is the recommended procedure. Note that is not particularly available resources may constrain the use of such procedures. useful to have a frequency count without making note of the If a person cannot remain constantly vigilant for all instances observation time in order to subsequently convert it to a rate. of behavior (e.g., a teacher in a classroom with 25 students), Even when it seems like an observation window will always then a discontinuous measurement procedure in the form of be the same (e.g., the school day), certain factors may change momentary time sampling would likely be a more practical the duration of the observation at some point (e.g., the child and preferred option. If there are sufficient resources to collect leaves school ill, an unexpected field trip occurs). data using a continuous measurement system, then the next When some other dimension of behavior can vary greatly, questions lead a practitioner to select the optimal measure- capture a measure of that dimension (e.g., duration, intensity) ment procedure given the nature of the behavior as a free or and derive the frequency count and rate from the number of restricted operant and the important dimensions of the observed instances. That is, each event is scored for intensity, behavior. so the number of scored events is equal to the frequency count. Many behaviors can occur at any time and in any setting Tantrums are often scored using a duration measure but may (i.e., a free operant) while others can only occur under certain also include an intensity rating or a frequency count of a spe- circumstances (i.e., a restricted operant). That is, with a re- cific intense problem behavior that is then divided by the stricted operant behavior, there must be some condition in duration of the tantrum (e.g., rate of aggression). Over the place for the behavior to occur. For example, some instruc- course of an intervention, one might observe a change in the tional event must be presented and occur in an ongoing number of tantrums, the average duration of each tantrum, or 82 Behav Analysis Practice (2016) 9:77–83 the rate of aggression during tantrums. Other important and Conclusion useful measurement procedures, such as latency recording, not only produce information about the specific temporal di- Practicing behavior analysts frequently assess and treat prob- mension, a frequency measure can be obtained from the data lem behavior as part of their ongoing employment responsi- and in many instances a rate measure as well. bilities. Effective measurement of problem behavior is critical to success in these activities because some measures of prob- lem behavior provide more accurate and complete information Case Example about the behavior than others. However, not every measure- ment procedure is appropriate for every problem behavior. In Joey is a 7-year-old student in a classroom with 22 other addition, the resources available on an ongoing basis in natu- students. The teacher and paraprofessional aide are willing ral environments may not always support the most labor- to collect data for the behavior analyst who is consulting on intensive measurement procedures. One concern is that be- the case. Joey is often off-task and disruptive during indepen- havior analysts who encounter barriers to complicated or op- dent seatwork, and he says rude things to other students in the timal data collection systems might fail to collect data alto- class, as well as the adults. The behavior analyst used the gether if they do not have a system for selecting the most clinical decision-making model to select measurement proce- useful procedures given their constraints. Another concern is dures for the teacher and aide to use. The teacher and aide that one might select a measurement procedure that does not indicated that the rude statements occur multiple times per provide sufficient information about the behavior to allow a day and are discrete, noticeable, and often reported by other meaningful evaluation of the effects of a given intervention. students. As long as the teacher and aide do not have to record Other clinical decision-making models have been devel- exactly what Joey says when he misbehaves, they can collect oped to guide behavior analysts in selecting among options data on each event that occurs almost immediately after it for intervention for problem behavior when multiple function- occurs. The behavior is not covert (i.e., it is observable), there based treatments have evidence to support their potential ef- are sufficient resources for continuous measurement, the be- fectiveness (e.g., Geiger et al. 2010; Grow et al. 2009). The havior could occur at any time, and there is no other critically use of this type of model may introduce a comprehensive and important dimension of behavior besides occurrence. Thus, thoughtful framework for decision-making when choices the model leads the behavior analyst to use event recording might otherwise be guided by familiarity with only a few of and calculate a frequency per school hour. the options. Similarly, selections of measurement procedures During independent seatwork, both the teacher and the aide for problem behavior might also be determined by recent use are roaming the classroom providing assistance to various or prior history of a procedure not working well in a different students and cannot remain vigilant to continuously score context. Fiske and Delmolino (2012) provided an example of Joey’s attending or inappropriate behavior. They often have how to consider the pros and cons for a more limited set of difficulty telling whether he is actively engaged in his work or procedures (i.e., discontinuous measurement procedures on- is looking down towards his desk while thinking about other ly), and the current model expands this idea to a broader range things. They are both in the room and can observe periodically of potential measurement procedures for problem behavior. but not continuously, as long as they do not have to be right This article may provide multiple benefits to the applied next to him when the observation occurs. There is at least one user. First, the table and text describing each measure provides component of his behavior that is somewhat covert as they a convenient and succinct summary of the strengths and con- cannot always tell whether he is actively engaged in the work siderations for the most frequently used measurement proce- at any given moment from across the room. Thus, there could dures for problem behavior. Second, the selection model may be a permanent product measure from each observation such guide practitioners through the most commonly encountered as the percentage of the assigned work completed. This is not barriers to effective data collection: (a) inability to observe the a direct measure of on-task behavior, but it can be a useful behavior, (b) lack of resources for continuous data collection, supplement to other measures. Disruptive behaviors may be and (c) a mismatch between the properties of the behavior discrete and countable, but other off-task behaviors are not, itself and the procedure. When there are no specific barriers and the environment does not have sufficient resources for and there are no other specific dimensions of behavior that continuous observation. Thus, the permanent product measure require special attention, the resulting selection is a frequency could be supplemented with momentary time sampling with count that can be converted into a rate measure. When barriers either the teacher or aide periodically (e.g., every 3 or 5 min) exist or it is important to capture other dimensions of behavior, recording whether Joey appears to be writing or reading or the resulting selections are the measurement procedures best otherwise working on his task and whether any disruptive suited to those circumstances. Additionally, if a practitioner behavior occurs at the observation point. These data would has already unsuccessfully attempted to use one measurement then be converted to a percentage-of-observations measure. procedure that may have been optimal from a technical Behav Analysis Practice (2016) 9:77–83 83 perspective, this model might assist them in selecting the best References of the remaining possible procedures that is better suited for their constraints. Baer, D. M., Wolf, M. M., & Risley, T. (1968). Current dimensions of We hope that practicing behavior analysts will find the applied behavior analysis. Journal of Applied Behavior Analysis, 1, 91–97. article useful when they are selecting measures for a problem Bailey, J. S., & Burch, M. R. (2002). Research methods in applied be- behavior or environmental constraint that they have not ad- havior analysis. Thousand Oaks: Sage. dressed before or for refining their existing experience and Barlow, D. H., Nock, M. K., & Hersen, H. (2008). Single case experi- expertise in selecting measurement procedures for problem mental designs: strategies for studying behavior change (3rd ed.). Boston: Allyn and Bacon. behavior. We also hope that this article and the others like it Call, N. A., Pabico, R. S., & Lomas, J. E. (2009). Use of latency to (e.g., Geiger et al. 2010; Grow et al. 2009) might assist prac- problem behavior to evaluate demands for inclusion in functional titioners in using a more systematic approach in all of their analyses. Journal of Applied Behavior Analysis, 42, 723–728. critical clinically related decisions and processes (e.g., mea- Cooper, J. O., Heron, T. E., & Heward, W. L. (2007). Applied behavior analysis (2nd ed.). Upper Saddle River: Pearson. surement and data collection, treatment planning, curricular Fiske, K., & Delmolino, L. (2012). Use of discontinuous methods of data assessment). collection in behavioral intervention: guidelines for practitioners. Finally, the model presented herein has not been empirical- Behavior Analysis in Practice, 5(2), 77–81. ly tested to determine if measurement procedures selected Geiger, K. A., Carr, J. E., & LeBlanc, L. A. (2010). Function-based treatments for escape-maintained problem behavior: a treatment se- based on the decision-making model produce more compre- lection model for practicing behavior analysts. Behavior Analysis in hensive or sensitive evaluations of the effects of interventions Practice, 3(1), 22–32. than measures selected without the use of the model. Thus, Grace, N. C., Thompson, R., & Fisher, W. W. (1996). The treatment of there is no empirical demonstration of any differential effects covert self-injury through contingencies on response products. Journal of Applied Behavior Analysis, 29, 239–242. of this model on the ease or utility of practitioners’ measure- Grow, L. L., Carr, J. E., & LeBlanc, L. A. (2009). Treatments for ment efforts. However, because the model is based on a well- attention-maintained problem behavior: empirical support and clin- developed literature on measurement, it might be prudent to ical recommendations. Journal of Evidence-Based Practices for consider it a potentially useful starting point until validation Schools, 10, 70–92. data can be generated. Towards that end, research efforts Johnston, J. M., & Pennypacker, H. S. (2008). Strategies and tactics of behavioral research (3rd ed.). New York: Routledge. might include assessments of the model’s social validity Kazdin, A. E. (2011). Single-case research designs: methods for clinical (i.e., are the selected measures easier to use and do they result and applied settings (2nd ed.). New York: Oxford University Press. in more generated data) and empirical comparisons between Maglieri, K. A., DeLeon, I. G., Rodriguez-Catter, V., & Sevin, B. M. model-generated and default measurement procedures in (2000). Treatment of covert food stealing in an individual with Prader-Willi syndrome. Journal of Applied Behavior Analysis, 33, terms of improved data-based decision-making. 615–618. Mayer, G. R., Sulzer-Azaroff, B., & Wallace, M. (2012). Behavior anal- ysis for lasting change (2nd ed.). Cornwall-on-Hudson: Sloan. Author Note This article does not represent an official position of the Sidman, M. (1960). Tactics of scientific research: evaluating experimen- Behavior Analyst Certification Board. Paige Raetz is now at the South- tal data in psychology. New York: Basic Books. west Autism Research and Resource Center. Tyra Sellers is now at Utah Wirth, O., Slaven, J., & Taylor, M. A. (2013). Interval sampling methods State University. This model was developed as part of the Clinical Stan- and measurement error: a computer simulation. Journal of Applied dards initiative at Trumpet Behavioral Health. Behavior Analysis, 47, 83–100.