xml version 1.0 encoding UTF-8 standalone no
record xmlns http:www.loc.govMARC21slim xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.loc.govstandardsmarcxmlschemaMARC21slim.xsd
leader nam Ka
controlfield tag 001 002001094
007 cr mnu|||uuuuu
008 090428s2008 flu s 000 0 eng d
datafield ind1 8 ind2 024
subfield code a E14-SFE0002572
Cripe, Michael M.
Can using one trainer solely to deliver prompts and feedback during role plays increase correct performance of parenting skills in a Behavioral Parent Training program?
h [electronic resource] /
by Michael M. Cripe.
[Tampa, Fla] :
b University of South Florida,
Title from PDF of title page.
Document formatted into pages; contains 33 pages.
Thesis (M.A.)--University of South Florida, 2008.
Includes bibliographical references.
Text (Electronic thesis) in PDF format.
ABSTRACT: Behavioral Parent Training refers to a broad range of instructional programs that teach parents and other caregivers ways to build and change behavioral repertoires of children. Most, if not all, such programs employ Behavioral Skills Training (modeling, prompting, role-play practice and feedback) to teach parenting skills. However, specific ways to use prompting during role plays have not been described in the behavioral parent training literature. The present study compared two methods of conducting role plays during parent training. A between group, pretest/posttest design was used to evaluate and compare the effect of using one versus two trainers on the role-play performance of parents and other caregivers involved in the child dependency system. Although both groups' posttest scores improved, there was no statistically significant difference between the one and two trainer groups. It was determined that foster and adoptive parents performed better on posttest measures than did biological parents and relative caregivers, regardless of group assignment.
Mode of access: World Wide Web.
System requirements: World Wide Web browser and PDF reader.
Advisor: Kimberly Crosland, Ph.D.
Behavioral Skills Training
x Applied Behavior Analysis
t USF Electronic Theses and Dissertations.
Can Using One Trainer Solely to Deliver Prompts and Feedback During Role Plays Increase Correct Performance of Parenting Skills in a Behavioral Parent Training Program? by Michael M. Cripe A thesis submitted in partial fulfillment of the requirements for the degree of Master of Arts in Applied Behavior Analysis College of Graduate School University of South Florida Major Professor: Kimberly Crosland, Ph.D. Raymond G. Miltenberger, Ph.D. Stacie A. Neff, M.S. Date of Approval: July 8, 2008 Keywords: rehearsal, Behavioral Skills Training, ch ild welfare, Behavior Analysis, Services Program, modeling Copyright 2008, Michael M. Cripe
Dedication I dedicate this text to my late mother, Ann E. Crip e, who believed in me when I did not believe in myself, and to Diana and Nuri, the b est family I could ever hope for.
Acknowledgments I would like to acknowledge David Geller, Terresa K enney, Bryon Neff and Stacie Neff for their support, personal as well as technic al. I also must acknowledge my fellow students who assisted with data collection and prov ided moral support, my advisor, Dr. Kim Crosland and Dr. Ray Miltenberger. Finally, I am grateful for Dr. Susan M. HavercampÂ’s assistance with the statistical analysi s of the data presented herein.
i Table of Contents List of Tables ii List of Figures iii Abstract iv Introduction 1 Method 7 Behavior Analysis Services Program 7 Participants and Setting 7 Group Selection and Assignment 9 Informed Consent 10 Training 10 In-class role play assessments 13 Independent Variables 13 One Trainer 13 Two Trainers 13 Pre and Posttests 14 Interobserver Reliability 14 Experimental Design and Analysis 15 Results 16 Discussion 19 References 23 Appendices 26 Appendix A: Tool Checklists 27 Appendix B: Role Play Scenarios 31
ii List of Tables Table 1. Caregivers in the two trainer group arran ged by type. 8 Table 2. Caregivers in the one trainer group arran ged by type. 9
iii List of Figures Figure 1. Mean Group Pre and Posttest Scores 16 Figure 2. Biological and Foster/Adoptive Parents Me an Scores 18
iv Can Using One Trainer Solely to Deliver Prompts and Feedback During Role Plays Increase Correct Performance of Parenting Skills in a Behavioral Parent Training Program? Michael M. Cripe ABSTRACT Behavioral Parent Training refers to a broad range of instructional programs that teach parents and other caregivers ways to build an d change behavioral repertoires of children. Most, if not all, such programs employ Be havioral Skills Training (modeling, prompting, role-play practice and feedback) to teac h parenting skills. However, specific ways to use prompting during role plays have not be en described in the behavioral parent training literature. The present study compared two methods of conducting role plays during parent training. A between group, pretest/po sttest design was used to evaluate and compare the effect of using one versus two trainers on the role-play performance of parents and other caregivers involved in the child dependency system. Although both groupsÂ’ posttest scores improved, there was no stat istically significant difference between the one and two trainer groups. It was determined t hat foster and adoptive parents performed better on posttest measures than did biol ogical parents and relative caregivers, regardless of group assignment.
1 Introduction Behavioral Parent Training (BPT) is a broad term us ed to describe various training programs that address the acquisition of s kills by parents or other caregivers which, when such skills are used, are believed to h ave some positive effect on the behavior of children (Graziano & Diament, 1992). BP T programs have taught parents to reinforce appropriate behaviors (OÂ’Dell et al., 198 2) withhold reinforcement for inappropriate behavior (Eyberg & Matarazzo, 1980) a nd, in some cases, punish inappropriate behavior (Eyberg & Robinson, 1982). P arents have been taught to use reinforcement for compliance with requests, while i gnoring non-compliance (Ducharme, Atkinson & Poulton, 2001) and to use reinforcement for compliance combined with a time-out procedure for a failure to comply within a specified latency period (Forehand & King, 1974; 1977). Some programs have involved teaching parents to use a variety of combinations of reinforcement, extinction, prompting, rule-stati ng and time-out depending upon various dimensions of the childÂ’s behavior, such as whether or not it is potentially physically harmful to self, others or property ( Van Camp, Borrero & Vollmer, 2003 ; Van Camp et al., in press ). Additionally, BPT programs have taught specific interaction skills (Eyberg & Matazarro, 1980; Van Camp et al., 2003; V an Camp et al. 2007, in press )
2 meant to improve relationships between child and pa rent by altering the parentÂ’s verbal and affective behavior. Common among many BPT training packages is the comb ination of didactic instruction (e.g. lecture) with modeling, role-play ing and the contingent delivery of feedback based upon performance (Ducharme et al., 2 001 ; Forehand & King 1977; Marcus, Swanson & Volmer 2001; Sandler, Van Dercar & Milhoan, 1982). Instruction that employs modeling, role-playing, and the contin gent delivery of feedback for correct and incorrect performance is known as Behavioral Sk ills Training (BST). It is typically used in cases where individuals must learn appropri ate responses to situations that, due to their nature, have to be staged (Poche, Brouwer & S wearingen, 1981). BST is often used in training individuals to react to potentially dan gerous situations such as the presence of a firearm (Himle, Miltenberger, Flessner & Gatherid ge, 2004) a potentially abusive person (Poche, Yoder & Miltenberger, 1988) and fire s (Jones, Kazdin & Haney 1981). However, a large number of BPT programs cite the us e of BST techniques in teaching parents to respond appropriately to child behavior in staged settings (Ducharme et al., 2001; Eyberg & Robinson, 1982; Forehand & K ing, 1977; Van Camp, Vollmer & Borrero, 2003; Wolfe et al., 1982). For example, Va n Camp et al. ( in press ) described a BPT program that presented parents with nine techni ques to use with children to reduce inappropriate behavior, teach appropriate alternati ve behavior and improve the quality of parent/child interactions. During class sessions, t he trainers demonstrated the techniques and then role-played with the parents using situati ons that were frequently encountered in the home environment. Correct responses were prompt ed as necessary during the role-
3 plays, with feedback being delivered as to which st eps of the relevant technique were completed and which ones were not. In the BPT literature, detailed descriptions of the ways in which modeling, roleplaying, prompting and delivery of feedback are use d are often not provided. For example, Ducharme et al. (2001) stated that a 45 mi nute session using, Â“modeling, rehearsal and performance feedback,Â” was conducted with two parents, but did not describe the procedure (p.860). Forehand and King ( 1977) noted that after a reinforcement procedure was discussed with a mother in training, therapist modeling of additional techniques and practicing in a role-play situation was Â“optional,Â” but did not describe the manner in which these should be conduc ted (p.99). Smagner and Sullivan (2005) cited the use of teaching methods including, Â“Â…role-plays followed by verbal feedback,Â” (p.433). Sandler et al. (1982) mentioned the use of role-plays where the trainer modeled desired parenting techniques while a participant acted the part of the child. Next, the trainer acted as the child while t he participant attempted to perform the skills that had just been demonstrated. However, th e authors provided no specific descriptions of prompting or delivery of feedback. It appears that in these studies, it was assumed that the types of models, prompts and feedb ack necessary for instruction in the BPT program were either obvious or widely known. Ho wever, the level of detail in describing the use of BST techniques in BPT program s may be insufficient for replicating (Baer, Wolf, & Risley, 1968). One goal of BPT is to establish stimulus control su fficient for a parentÂ’s use of various techniques upon a childÂ’s emitting a partic ular behavior or when taking part in a
4 certain activity such as feeding or toilet training For example, a childÂ’s picking up trash, or sharing toys with a sibling could, with training become a discriminative stimulus for the parentÂ’s contingent delivery of descriptive pra ise. BST procedures, such as roleplaying, prompting and contingent delivery of feedb ack are employed so that a parent may rehearse responding in the presence of stimuli similar to those that are encountered in-situ. Therefore, it is important that BST techni ques be properly employed by BPT trainers so that transfer of stimulus control from ex-situ to in-situ conditions may occur (Miltenberger, 2004). Generalization of a particular parenting skill from training to the home environment may be said to have taken place if it o ccurs in-situ when relevant stimuli are present (Stokes & Baer, 1977). Examples used in tra ining must, to the greatest extent possible, incorporate stimuli common to the environ ment in which parents will be expected to use skills being taught (Stokes & Baer, 1977). For example, if a parent in training has a child who cries and screams for ice cream, then a trainer acting as a child in a role-play designed to teach an appropriate parent al response should engage in similar behavior with similar dimensions (i.e., frequency, volume, tone of voice). In the BPT program described by Van Camp and collea gues, ( in press ) it was often necessary that one trainer conduct role-playi ng, acting as the child while also delivering corrective feedback and praise, with sma ll groups of parents. This entailed a trainer alternating between acting as the child and delivering prompts and feedback in role-plays. This is analogous to two boxers sparrin g while one delivers instructions to the other or two actors rehearsing a scene while one su pplies the other with prompts for
5 missed lines. In any of these cases, the individual who receives prompts and feedback while rehearsing may learn to respond to stimuli ot her than those most relevant to the insitu environment. Bondy and Frost (2001) discussed prompting strategi es in a description of the Picture Exchange Communication System (PECS), a lan guage training program for children with autism. The basic assumption underpin ning the PECS training is that in verbal interactions between individuals, the verbal behavior of one (the speaker) is under the stimulus control of the other (the listener) (B ondy & Frost, 2001; Skinner, 1957). Thus, the PECS training uses two trainers to conduc t communication training with a child with autism. One acts solely as the Â“communication partner,Â” while the other delivers physical prompts in a manner so that no social inte raction takes place between the second trainer and the child. The authorsÂ’ stated purpose for this strategy was to have the child respond only to those stimuli that are being traine d and not to prompts from either trainer. An example of this procedure involves placing prefe rred items before the child with a communication partner seated facing her. When the c hild attempts to pick up an item, the second trainer physically prompts her from behind t o pick up a picture of the item and hand it to the communication partner. The child the n receives the item for a brief period of time. The childÂ’s selection and handing over of the picture is (ostensibly) reinforced by gaining access to the item; the Â“exchangeÂ” of pictu re for item constitutes a discrete instance of verbal behavior (Bondy & Frost, 2001). As Bondy and Frost (2001) suggest, training individ uals to respond to the ongoing behavior of others may be made more difficult when prompts that are not germane
6 to such interactions intervene. Thus, the authors c hose to have one individual act solely as the deliverer of physical prompts to a child who wa s learning to mand. In the case of using role-plays to approximate in-situ conditions, having an actor alternate between engaging in child-like behavior and delivering prom pts may interfere with the establishment of stimulus control with respect to t he stimuli relevant to the skill being taught (i.e. parenting skills). To evaluate this wi thin the context of parent training this study compared the use of two methods of conducting role-plays within a BST training environment. The use of one trainer who acted as th e child-model while also delivering prompts and feedback was compared with the use of t wo trainers, with one acting solely as the child model and the other delivering prompts and feedback. It was hypothesized that the performance of those individuals who were exposed to the use of two trainers, as described earlier, would improve over those who wer e only exposed to one trainer. If it were found that the use of two trainers for role-pl ays in a BPT program had a significant effect on the performance of participants, addition al resources for staffing such programs might be warranted. In addition, the results of thi s study may suggest further avenues for research into ways to program for skill generalizat ion in the area of behavioral parent training.
7 Method Behavior Analysis Services Program The behavioral parent training curriculum used in t he study was developed and promulgated by the Behavior Analysis Services Progr am. The Behavior Analysis Services Program (BASP) provides behavior analytic services to clients of the Florida child welfare system. It is funded by the Florida l egislature and administered by the University of South Florida and the University of F lorida. Teams of behavior analysts and senior behavior analysts (i.e., supervisory sta ff) are located throughout the state in service districts as defined by the Florida Departm ent of Children and Families. Local private, not for profit, community-based c are organizations deliver services in the areas of foster care, adoption and family reunification. The BASP works with these organizations to provide training to fos ter, adoptive and biological parents, as well as residential and care management staff. The services provided by the BASP all are behavior analytic in nature and include competencybased training, functional behavioral assessment and provision of technical assistance. Participants and Settings The participants were biological parents or alterna te caregivers (e.g., foster parents) involved with a community based care agenc y and referred to the BASP for training. Biological parents were referred for trai ning in order to meet the requirements of
8 a court-approved case plan. Some foster parents wer e referred by licensing agencies due to concerns about behavior management practices or because they were providing care for a child or children who engaged in challenging behaviors. Other foster and potential adoptive parents were referred for training prior t o having children placed in their homes. Finally, some foster, adoptive and other alternate caregivers received training due to having sought out help in providing care to childre n with challenging behavior (i.e. without any requirement from any agency or entity). Demographic and personal characteristics varied across participants. Tables 1 and 2 show demographic characteristics for participants and are arranged b y group assignment. Table 1 Caregivers in the One-Trainer Group (Group B) arran ged by type 0ne Trainer Group B Caregiver Type Gender Race Pretest Score (percent of correct steps) Posttest Score (percent of correct steps) 1 2 3 4 5 6 7 8 9 10 11 12 13 14 biological biological biological biological relative biological biological foster/adoptive foster/adoptive foster/adoptive foster/adoptive foster/adoptive foster/adoptive staff F M M F F F M F M F F M F F W W W W B W W W W W W W B B 48 48 34 18 24 24 40 22 41 54 26 61 43 36 56 60 45 58 45 60 74 79 75 88 53 98 76 70 Note .Â“BiologicalÂ” denotes a parent of a dependent child, Â“relativeÂ” denotes a blood relative of a dependent child. Â“Foster/adoptiveÂ” refers individua l who have or will be taking dependent children int o their homes. Â“StaffÂ” refers to individuals who regu larly interact with dependent children and their fa milies in a variety of settings.
9 Note. Â“BiologicalÂ” denotes a parent of a dependent child, Â“relativeÂ” denotes a blood relative of a dependent child. Â“Foster/adoptiveÂ” refers individu al who have or will be taking dependent children in to their homes. Â“StaffÂ” refers to individuals who reg ularly interact with dependent children and their f amilies in a variety of settings. The training took place in two settings. One was a large training room at a local university. The other was a training room on the gr ounds of a residential facility for dependent children. Group Selection and Assignment Two experimental groups were formed. Participants w ere selected from a list of referrals maintained by the BASP office. These refe rrals came from case managers, other agency staff and participants themselves (reasons f or individual referrals to the BASP Table 2 Caregivers in the Two-Trainer Group (Group A) arran ged by type Two Trainer Group A Caregiver Type Gender Race Pretest Score (percent of correct steps) Posttest Score (percent of correct steps) 1 2 3 4 5 6 7 8 9 10 11 12 biological biological relative biological relative biological relative foster/adoptive foster/adoptive foster/adoptive foster/adoptive staff F F F M F F F M M F F F W W W W B W B W W B W W 46 37 26 26 26 27 36 47 36 35 62 59 48 42 31 57 88 71 59 66 74 88 92 68
10 were mentioned earlier). Classes were drawn from th e referral pool in first-referred, firstserved order as mandated by contractual agreement w ith the local child welfare agency. Participants were assigned to either group followin g the administration of the course pre-test. In order to control for the possib ility that high or low scoring individuals could form the preponderance of either group, group assignment was made as follows. Pre-test scores were rank ordered from highest to l owest across all participants. Group assignment from highest to lowest score was made in this sequence: A B B A, repeating as necessary. Individuals whose pretest scores were 80 per cent or greater were excluded from the study. One potential participant was exclu ded from the study as a result of exceeding the upper limit for pretest scores. Informed Consent All participants, regardless of experimental group assignment, received the standard BASP Tools for Positive Behavior Change cu rriculum training. Informed consent was obtained to take part in the program pe r the standard practice employed by the BASP. The BASP program operates under a service provision agreement with the Institutional Review Board. An exemption for obtain ing specific informed consent from participants was granted for this study under exemp t category 45 CFR 46.101(b)(1) which covers the evaluation and/or comparison of ed ucational strategies, techniques or curricula. Training The Tools for Positive Behavior Change is a trainin g package that focuses on eight, task-analyzed techniques (or tools) for adul ts to use across a variety of routine
11 interactions with children and adolescents. Each te chnique is based on empirically validated behavioral procedures such as differentia l reinforcement, extinction, time-out and contingency management (Van Camp et al., in press ). For this study, four of the eight tools were used for evaluation of the independent variable. The exact same curriculum was used for bo th classes so that instruction in the four relevant techniques was no different in one cl ass than the other. Class session 1 consisted of an introduction and overview of the co urse and pre-tests. Class 2 covered basic behavioral principles such as the potential e ffects of consequences on the future likelihood of behavior. The remainder of the classe s included instruction of the tools that comprise the curriculum. As stated previously, the current study concentrate d on four of the eight tools; however, participants received instruction in the a dditional tools. The four tools of interest were covered in classes 3 through 5, in th e order that they appear herein (A complete set of task-analyses for each tool is atta ched in appendix 1). Tool 1; Stay Close was a cluster of adult behaviors is intended to enh ance interactions with children so that adult attention is established as a reinforcer. The use of open-ended questions, statements of empathy and concern, along with the maintenance of affect appropriate to the situation (e.g. facial expressions, tone of voice) were cover ed. Tool 2; Use Reinforcement was a series of steps that guides adults in the delivery of various stimuli (e.g., descriptive praise, items, privileges, pleasant touch) contingent upon a childÂ’s performance of desirable behavior. Tool 3; Pivot consisted of withholding attention while a child e ngaged in minor inappropriate behavior (Â“junk behaviorÂ”) that would not be harmful to anybody or
12 anything and could be ignored with little chance of harm occurring. When the inappropriate behavior ceased the adult then provid ed attention, as in a differential reinforcement procedure. Tool 4; Redirect/Use Reinforcement was a series of steps for interrupting behavior that may cause injury and thu s could not be ignored or for behavior that occurred because an acceptable alternative did not exist in the childÂ’s repertoire. In either case, on-going behavior would be interrupted and an appropriate alternative then prompted. Reinforcement would be delivered by the a dult contingent upon the childÂ’s performance of the alternative behavior. This skill is also essentially a differential reinforcement procedure (Van Camp et al., in press ). In each class, instruction began with a didactic pr esentation of the session materials with participants making comments, asking questions and being encouraged to respond. The tool or tools slated for that session were typically demonstrated early in the session, with some components broken out and covere d in more detail (e.g., Stay Close was demonstrated across a few scenarios that involv ed children of different ages with empathy statements relevant to each situation). Tra iners demonstrated the tools by roleplaying different scenarios that cover a variety of stimulus conditions likely to be encountered by adult caregivers (e.g., age of the c hild, the activity at hand). For example, the tool Redirect/Use Reinforcement was demonstrated with a young child running near a street, an older child climbing on a counter to get something or an older child trying to open a sealed compact disc with a sharp knife.
13 In class role-play assessments At the end of each class session, participants were broken into the two assigned groups. The trainer told the group that they would be practicing the use of the skill just covered in class by role-playing with a trainer. Th e trainer blindly selected a written scenario from several different ones in an envelope (see Appendix B). Each group member was instructed to act as the parent in the r ole-play scenario and attempt to use the skill that was taught in that class. Independent Variables One trainer. In this condition, one trainer acted as the child a s well as prompted participants during the in class role plays to comp lete steps that they missed per the checklist for the relevant skill. Thus, in order to deliver a prompt, she had to break the flow of the role-play, instruct the participant to complete a step and then re-assume the role of the child. For example, if a participant wa s to have provided a verbal prompt to the Â“child,Â” and did not, the trainer would stop th e role-play and say something like, Â“Remember to say, Â‘good job.Â’ Try again,Â” and then resume the role of the child. Two Trainers In this condition, one trainer acted as the child exclusively, while another prompted participants, when necessary, to c omplete steps per the relevant check list for that skill. The trainer who portrayed the child remained in character (i.e., continued to engage in Â“childÂ” behavior) while the second trainer delivered prompts as necessary for each participant to correctly perform each step. For example, if the participant was to get within armÂ’s reach of the Â“c hildÂ” and failed to do so, the prompter would say something like, Â“Get within armÂ’s reach,Â” while the Â“childÂ” continued to
14 behave in a way pertinent to the scenario (e.g., cr y and stomp his feet). If the participant moved closer to the child, the prompter delivered i mmediate positive feedback. If the participant did not respond to the first prompt, th en additional prompting sufficient to affect the correct response was issued. Pre and Posttests In order to evaluate the effect of the independent variable, each individual took part in pre and post course role-play assessments. Individuals were presented with four standardized scenarios where a BASP trainer portray ed a child. The role play scenarios were the same for both pre and posttests. The score rs were blind to the group assignment of the participants. The participants were told the age of the child and given details about the situation. They were asked to do whatever they would normally do in each situation, given that they were in the position of the primary caregiver. Checklists were used to record completion of the steps relevant to each ski ll being assessed (see appendix 1). Each step could be scored as having occurred or not occurred. Not applicable was scored if the opportunity to perform a step was not availa ble. This would apply if the trainer failed to perform one of the steps in the role play script. Scores were expressed as the percentage of steps correctly performed across all tools. No feedback was given to the participants as to their performance at any time du ring or immediately after the assessment. Interobserver Agreement Interobserver agreement was calculated for each ski ll assessment by comparing the responses of the two trained observers on each step of each tool. Observers were blind
15 to the group assignments of the participants. Agree ment on any given step was said to have occurred if both observers scored the step in the exact same way (i.e. each scored Â“yes,Â” Â“no,Â” or Â“not applicable) (Van Camp et al., in press ). Reliability scores for each pretest were added and divided by the number of pre tests to determine the overall rate of agreement across pre-tests. Posttest interobserver agreement was determined in the exact same manner Sixty-one percent of pre-course assessments and fif ty percent of post-course assessments across participants were scored indepen dently by a second trained observer in order to determine interobserver agreement. For pretests, average interobserver agreement was 64 percent ( r = 52 % to 94%). For posttests, agreement was assess ed on 50 percent of assessments. Average interobserver ag reement for posttests was 93 percent ( r = 87% to 100%). Experimental Design and Data Analysis A between group, pretest/posttest design was used t o evaluate the effectiveness of the independent variable. Scores on each of the fou r tools are expressed as a percentage of correct steps per checklist and were derived by dividing the number of correct steps by the number of steps possible for each tool and mult iplying the result by 100. Group means were derived by adding the percentage of corr ect steps of each individual and dividing the sum by the n of the group.
16 Results Figure 1 shows the pre and posttest scores for both groups. Group A received the two-trainer treatment during role-plays. The mea n pretest score for the two trainer group (group A) was 40 percent, SD =13.2. The mean posttest score for the two trainer group was 65.3, SD=19.0. For the one trainer group (group B), the mean prete st score was 37.2 percent, SD=13.0. The mean posttest score was 66 percent SD =15.7. Figure 1. Mean pre and posttest scores for two tra iner (group A) and one trainer (group B) groups. 0 10 20 30 40 50 60 70 80 90 100 Group Group A Group B Pretest Posttest Percent Correct Steps
17 A 2 x 2 mixed analysis of variance with one between group factor (group assignment Â– one vs. two trainers) and one within g roup factor (time of assessment Â– pre vs. posttest) was used to analyze the results. Over all, there was a main effect for time, F (1, 48) = 41.38, p<.001 but no main effect for group, F (1, 48) =.021, ns. There was no statistically significant time X treatment group in teraction effect (F (1,48)=.27, ns). Thus, although each groupÂ’s performance on the assessment s improved over time, the use of one versus two trainers during BST did not appear t o make any appreciable difference in the scores. It was observed that foster and adoptive parentsÂ’ s cores appeared to be higher on posttests than the scores of biological parents and relative caregivers. A second 2 x 2 analysis of variance was conducted to assess the ef fects of time and caregiver type. There was a main effect for type of caregiver (i.e. foster/adoptive versus biological), F (1, 48) = 24.6, p<.001 with the foster and adoptive parent group perform ing better overall and a main effect for time (see above). However, t here was not a significant time X caregiver type interaction effect (F (1,48)=1.79, ns) indicating the two caregiver groups had similar gains between pretest and post-test.
18 Figure 2. Mean pre and posttest scores by caregiv er type. 0 10 20 30 40 50 60 70 80 90 100 Pretest/Posttest Foster & Adoptive A Biological & Relative Pretest Posttest
19 Discussion The present study looked at the effects of using an additional individual to provide prompts and feedback during role plays in a BPT class. A dearth of explicit explanations of such procedures in the BPT literatu re gave rise to this study (Ducharme et al, 2001). Given the outcome, that using an additio nal trainer did not produce a statistically significant difference between groups exposed to one and two trainer conditions, it may be concluded that behavioral par ent training using curricula similar to that employed in this research can be conducted wit h fewer trainers. However, due to the small number of participants and the potentially co nfounding variable of type of caregiver discussed below, these results should be considered with caution. A number of limitations of the present study should be considered. First, the onetrainer group, group B, had a higher proportion ado ptive and potential adoptive parents who had also been exposed to state-mandated trainin g that included a brief overview of the structure and key concepts of the BASP curricul um, but did not provide instruction of any of the four skills examined in the present stud y. This prior exposure could have had an effect on performance in that certain concepts ( e.g. the importance of using positive reinforcement) had been covered. Second, regardless whether they were trained in either the one or two trainer group, foster and adoptive p arents overall performed better on pre and posttests than did biological parents. Others h ave speculated that contextual variables such as socioeconomic disadvantage and family dysfu nction might negatively impact outcomes of behavioral parent training (Assemany & McIntosh 2002). All of the
20 biological parents and relative caregivers in this study had been referred to parent training due to some difficulties sufficient to warrant the removal of children from their homes (e.g., substance abuse, excessive use of corporal p unishment, neglect of a child). Future research in the use of instructional techniques in behavioral parent training program might take into account participant variables that could have an impact on outcomes. A potential avenue for research could involve the use of role play scenarios derived from each participantÂ’s experience so that training exem plars more are more closely related to conditions that a participant is likely to encounte r. For example, if parent trainers were able to observe families prior to training, role pl ay scripts could be developed that contain common and relevant exemplars. Although this does n ot directly address the hard variables such as socioeconomic status or substance abuse history, it would provide individuals with practice in conditions most simila r to those they have encountered in the past and perhaps have a positive effect on future p erformance of skills. Another limitation of the present study involves ho w prompting was delivered in cases where a participant did not emit a correct re sponse upon receiving a verbal and/or gestural prompt. In such cases, the trainer repeate d verbal and gestural prompts until the participant engaged in the relevant response. In ot her settings, such as when teaching a child to perform a skill, an instructor might use a touch prompt in order to get the child to engage in the required behavior, thus providing an opportunity to provide feedback on the childÂ’s performance more quickly. With adults in th e current setting, using any physical prompts would not have been appropriate.
21 Difficulty in delivering effective prompts during r ole plays was also encountered when the relevant skill required a participant to d emonstrate particular tone of voice or facial expression. When working with adults in this setting, a prompt to smile, or change vocal inflection might be met with resistance and, if the behavior is not emitted upon a verbal prompt, it likely will not occur at all duri ng the role play. In fact, one of the individuals in the two trainer group stated that he r tone of voice was Â“the way (I) talk,Â” and that she would not change it. Interobserver agreement data was very low for the 6 1 percent of pretests sampled. In the future, additional training should be conduc ted with observers immediately prior to conducting pretests. Due to constraints of schedule and the number and availability of trained observers who were blind to participantsÂ’ g roup assignments, different observers were used for posttest than were used for pretests. Future efforts at collecting interobserver agreement data in this venue might be nefit from employing the same observers throughout. Future research might involve developing specific B PT curricula for targeted groups (e.g., biological parents). As others have n oted, it may be the case that other, molar contingencies such as economic and social str essors have an overall deleterious effect on the outcomes of parent training. Therefor e, programs that provide such supports as are relevant (e.g., assistance with transportati on) might be more likely to be associated with improved outcomes on curricular measures. Anot her potential avenue for research might involve the use of numerous and varied exempl ars of parent-child interactions that could be easily imitated by learners, prior to enga ging in actual skills training. Finally,
22 adding explicit instructions for how to use behavio ral skills training methodology to behavioral parenting training curricula could be us eful for instructors and participants. The present study did not demonstrate the superior effectiveness of using one trainer to act solely as the child during role play s in a behavioral parent training class. One potential confounding variable, the overall gre ater improvement on posttests for foster and adoptive parents, may have played a role in the findings. Future research in this area should control for the variable of caregi ver type.
23 References Assemany, A.E. & McIntosh, D.E. (2002). Negative o utcomes of behavioral parent training programs. Psychology in the Schools, 39 (2), 209-219. Baer, D.M, Wolf, M.M & Risley, T.R. (1968). Some c urrent dimensions of applied behavior analysis. Journal of Applied Behavior Analysis, 91-97. Bondy, A. & Frost, L. (2001). The picture exchange communication program. Behavior Modification, 25 (5), 725-744. Ducharme, J.M., Atkinson, L., & Poulton, L. (2001). Errorless compliance training with physically abusive mothers: a single-case approach. Child Abuse & Neglect 25, (6), 855-868. Eyeberg, S.M. & Matazarro, R.M. (1980). Training p arents as therapists: A comparison between individual parent-child interaction trainin g and group didactic training. Journal of Clinical Psychology, 36 (2), 492-499. Eyeberg, S.M. & Robinson, E.A. (1982). Parent-chil d interaction training: effects on family functioning. Journal of Clinical Child Psychology, 11, 130-137. Forehand, R. & King, E.H. (1974, 1977). Noncomplia nt children: effects of parent training on behavior and attitude change. Behavior Modification, 1 (1), 93-108. Graziano, A. M. & Diament, D. M., (1992). Parent b ehavioral training: an examination of the paradigm. Behavior Modification 16 (1), 3-38.
24 Jones, R.T., Kazdin, A.E. & Haney, J.I. (1981). So cial validation and training of emergency fire safety skills for potential injury p revention and life saving. Journal of Applied Behavior Analysis, 14 (3), 249-260. Marcus, B.A., Swanson, V. & Vollmer, T.R. (2001). Effects of parent training on parent and child behavior using procedures based on functi onal analyses. Behavioral Interventions, 16 (2), 87-104. Miltenberger, R. G. (2004). Behavior Modification: Principles and Procedures, 3rd edition. Belmont, CA: Thomson/Wadsworth. OÂ’Dell, S.L., OÂ’Quin, J., Alford, B.A., OÂ’Briant, A .L., Bradlyn, A.S., & Giebenhain, J.E, (1982). Predicting the acquisition of parenting sk ills via four training methods. Behavior Therapy, 13, 194-208. Poche, C., Brouwer, R., & Swearingen, M. (1981). Te aching self-protection to young children. Journal of Applied Behavior Analysis, 14 (2), 169-175. Sandler, J., Van Dercar, C. & Milhoan, Mariann, (19 77). Training child abusers in the use of positive reinforcement practices. Behavior Research and Therap y, 16, 169-175. Skinner, B.F., (1957). Verbal Behavior. Cambridge, MA: Prentice-Hall. Smagner, J.P. Smagner, J.P. & Sullivan, M.H. (2005). Investigati ng the effectiveness of behavioral parent training with involuntary clients in child w elfare settings. Research in Social Work Practice, 15 (6), 431-439. Stokes, T.F. & Baer, D.M. (1977). an implicit tech nology of generalization. Journal of Applied Behavior Analysis, 10 (2), 349-367.
25 Van Camp, C. M., Borrero, J. C. & Vollmer, T. R. (2 003). The family safety/applied behavior analysis initiative: an introduction and o verview. The Behavior Analyst Today, 3 (4), 389-404. Van Camp, C.M., Vollmer, T.R., Goh, H. Whitehouse, C.M., Reyes, J. & Montgomery, J.L. (In press). Behavioral parent training in chi ld welfare: evaluations of skill acquisition.
27 Appendix A-Role Play Checklists
31 Appendix B-Role Play Scenarios Stay Close General instructions o Begin role play sitting or standing far enough away so the caregiver has to move towards you o Make emotional comments like, Â“this sucks, I had a crappy/shitty day, itÂ’s stupid, etcÂ”, and engage in minor junk behavior; make these types of comments intermittently o Stop these comments immediately once an empathy sta tement is made o Avoid eye contact until the caregiver makes empathy statement o If caregiver asks questions, answer them without ta lking too much o Respond to any problem solving with more junk behav ior o If the caregiver doesnÂ’t ask why you are upset, com plain about your issue so that the role play continues. o Remember that you want to talk to your caregiver. Scoring tips o Watch the caregiverÂ’s body language. Arms folded, h ands on hips, standing over the top of the child and looking at things other th an the child are not appropriate. Wait to see if they change. o Getting close and relaxed body language must occur by the halfway point for it to be scored as Â“yes.Â” o If an appropriate touch occurs, even at the very en d, it is scored as a Â“yes.Â” o End the role play when you have the information nee ded. Role Plays o Friend has moved away (official pre/posttest) o Someone made fun of my bike o Just found out friend has cancer o Someone at school is bullying me o This boy took my lunch from me on the bus this morn ing. o My favorite teacher is sick and we will have a new teacher for the rest of the year. o My teacher asked me to read in front of the class t oday and I messed up and the class laughed at me. o This girl at school is spreading rumors about me th at are not true. o This boy at school told me my caregivers donÂ’t love me anymore. o My best friend and I got into a fight and she is no t talking to me anymore. o An older student called me stupid and ugly today. o My mother missed our visit today. o My father was supposed to call yesterday and he did nÂ’t. Tool: Use Reinforcement General instructions o Begin role play sitting or standing far enough away so the caregiver has to move towards you o You will be engaging in appropriate behavior
32 Role Plays o Come home and immediately do your homework (officia l pre/posttest) o YouÂ’re making your bed o YouÂ’re setting the table for dinner o You made an A on your science project o YouÂ’re folding your clothes o YouÂ’re doing you homework o YouÂ’re helping your younger sibling with his homewo rk. o Your reading teacher said that you did a great job reading in front of the class today. o You made the baseball team. o You finished your homework assignment two days befo re it is due. o YouÂ’re loading the dishwasher. o YouÂ’re mowing the lawn. o YouÂ’re taking out the trash. Pivot General instructions o If doing individual child scenario, start w/ junk b ehavior, then eventually stop and engage in the appropriate behavior You roll your eyes, slam your hand on the table, an d then slowly get up. Â• Walk very slowly, shuffling your feet, engage in the requested task. Say something like: Â“How come I always have to (do the damned task)?Â” Emit some more junk, but pause occasionally, allowi ng the caregiver time to speak. Once task is done, slam the door, pick up your maga zine, and say, Â“There, are you happy now?Â” o If doing two child scenario, one child immediately engage in the appropriate behavior (for at least 15s or until the caregiver p raises you), while the second child engages in junk behavior, then eventually sto ps and engages in appropriate behavior Role Plays o Ask one child to take out the garbage (official pre /posttest) o Ask one child to go make their bed o Ask both children to sit for dinner (one plays with food) o Ask one child to do their homework o Asking two children to do the dishes (one plays wit h water) o Ask two children to put on their coat to go get rea dy to go to the store (one flops down on the chair and complains) o Ask one child to clean dishes off table (child comp lains) o Ask two children to help you bring the groceries in from the car (one stomps around the kitchen and refuses) o Ask one child to pick up their toys in the living r oom o Ask two children to turn the TV off and get ready f or bed (one child refuses to turn off TV)
33 Redirect-Use Reinforcement General instructions o You will engage in a semi-dangerous problem behavio r, something that will need to be redirected o Generally the caregiver will walk in or suddenly no tice the child engaging in this behavior o If you are redirected, give a brief bit of whining or crying and briefly resist by pulling against the caregiver, falling to the floor and stomping feet, saying Â“I can do itÂ”, etc., but not for more than three to five s econds. o If there is no intervention, continue to engage in the behavior. o If the caregiver redirects you to an alternate beha vior, engage in the alternate behavior If there is no redirection, eventually go engage in an appropriate behavior so the caregiver has a chance to provide praise Role Plays o You see your 3 YO throw small plastic toy in bassin et with your 2 month old (official pre/posttest) o You walk into the kitchen and notice that your 9 YO has a kitchen knife in his hand and he is trying to open a new CD o You are grocery shopping with your 6 YO, he starts to tip cereal boxes off the shelf o Your 3 YO keeps taking off her arm tubes when she g oes into the pool; you want he to wear them before she can go in o You see your 12 YO sneaking into the pantry and gra bbing some cookies before dinner o You are at the gas station; 9 YO grabs random candy bars and is asking you if he can have them. o Your 6 YO is about to chase your dog across the str eet o Your 7 YO is digging in her momÂ’s makeup bag o 4 YO drawing on the wall with crayons o 14 YO isnÂ’t supposed to be talking on the phone, bu t you see her reach for it