" MESIE STATE UNIVERSITY LIBRARIES IIIIII II IIIIIII I IIII III III I I 3 129300 This is to certify that the thesis entitled Child Sexual Abuse Prevention from a Social Problem Solving Perspective presented by Jacqueline Secor Grober has been accepted towards fulfillment of the requirements for Master of Arts degree in Psychology /5D0Y3v12a/' V V M 'or professor [MnepAugust 8. 1991 0-7639 MS U is an Affirmative Action/Equal Opportunity Institution I \_ fi__ _ mommy “N Michigan State University —_ PLACE IN RETURN BOX to remove this checkout from your record. TO AVOID FINES return on or before date due. DATE DUE DATE DUE DATE DUE #— A A ~1— ‘u‘ ‘7‘ "I“: {2 1'1?!" I j L MSU Is An Affirmative ActIorVEquel Opportunity Institution encircmpma-p. 1 CHILD SEXUAL ABUSE PREVENTION FROM A SOCIAL PROBLEM SOLVING PERSPECTIVE By Jacqueline Secor Grober A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of MASTER OF ARTS Department of Psychology 1991 «I -. ’ 'l~"_ —_ ” .K I _<‘(‘_’/ /‘f'r' / 4;! - ABSTRACT CHILD SEXUAL ABUSE PREVENTION FROM A SOCIAL PROBLEM SOLVING PERSPECTIVE By Jacqueline Secor Grober Previous theory and research in social problem solving (SPS) suggests that situational factors affect the strategies children generate to solve con-on social dilemmas. However, researchers have not investigated how these situational variables influence children’s strategizing in "unsafe" situations in which they lust resist a request. This study investigated how the nature of the request (non-sexual vs. sexually abusive) and the age of the social partner (child vs. adult) affected preschoolers’ resistance strategies. Four vignettes that systematically varied the request and the age of the partner were administered to 61 preschoolers. The inpact of situational factors was assessed on six aspects of SPS--the quantity, content, effectiveness, planfulness, and flexibility of solutions. Results indicate that the nature of the request, but not the social partner’s age, affect the quantity and content of solutions. Results also indicate that girls were more effective and planful problem solvers than boys. To my husband, James S. Grober, whose steadfast understanding, encouragement, and kindness has helped me immensely. iii ACKNOWLEDGEMENTS I would like to give special thanks to Anne Bogat for the considerable time and effort she devoted to supervising this thesis. I would also like to thank all the preschool staff, parents, and children who participated in this study. iv TABLE OF CONTENTS Page LIST OF TABLES...........................................vii LIST OF FIGURES...........................................ix INTRODUCTION...............................................l Overview of Child Sexual Abuse Prevention ResearChOOOOOOOO0.00000000IOOOOOOOOOCOOOOOOOOOOOOOOOO.1 Integrative Review of the Social Problem Solving and Child Sexual Abuse Prevention Literatures.........3 The Validity of Social Problem Solving Assessment Techniques in Child Sexual Abuse Prevention ResearChOOOOOOOOOOOO0.00.0000IO...00.0.0000000000000040 Rational for the Present Study.......................44 StUdy Hypotheses .......... 0.000000.0.00.00.00.000000047 METHODOOOOOOOOOOOO0.000.000.00000.0000.000.000.0000000000053 Participants.... ......53 “easures. O O O 0 I O O O O I 0 O O O O O O 0 O O O O O O O O O O O O 0 O O O O O O O O 0 O O O O 53 Procedure. 0 O O O O O O O O O O O O O O O 0 O O O O O O O O O O O O O O O O O O 0 O O O O O O .56 scoring. 0 O O O O O O O O O O O O O O O O O I O O O O O O O O O O O O O O O O O O O O O O O O O O 57 RESULTSOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOO0.0.00.0000000000063 DISCUSSIONOOOOOO...O0..IOOOOOOOOOOOOOOOOOOO0......0.0.00.070 Study Results and Social Problem Solving Theory and Research..................................70 Study Results and Child Sexual Abuse Prevention EffortBOOOOOOOOOOOOOOOOOOOO0.0.0.0.000000084 APPENDICES.OOOOOOOOOOOOO0.000000000000000000000000.00.000.93 A: Preschool Alternative Solutions Test (PAST)............93 B: Preschool Alternative Solutions Test (PAST) Scoring ManuaIOOOOOOOOOOOOOOO0.00000000000000000000000000.00.0105 C: Letter of Invitation/Parental Consent Form............167 D: Tables. 0 O O O O O O O O O O O O I O O O O I O O O O O O O O O O O O O O O O O O O O O O O O O O O .169 E: Figures. 0 O O O O O O O O O O O O O I O O O O O O O 0 0 O O O O O O O O O O O 0 O O O O O O O O I .201 LISTOFREFERENCESOOOOOOOOOOOOOOOOOOOO0.0.0.0000000000000205 vi 10. 11. 12. 13. 14. 15. 16. LIST OF TABLES Components of SPS Assessed in CSA Prevention Studies ..... 169 Assessment of Strategy Selection Variables in CSA Prevention Research Studies..............................172 Recursive Aspects of the SPS Process Assessed in CSA Prevention stUdiGBOIOIQOOOOOI.0.0.0.0000...-00.0.00000000175 Number of Responses Coded Using Original Strategy Categories for Each Situation..................... ....... 178 Formation of 10 Strategy Categories from Original 29 Strategy Categories...................................180 Means and Standard Deviations for Five SPS Variables.....182 Repeated-Measures Analysis of Variance for Encoding ...... 183 Repeated-Measures Analysis of Variance for Quantity of Solutions ................... .... ............. 184 Means and Standard Deviations for Total Number of Solutions Across Situations ...... ........................185 Repeated-Measures Analysis of Variance for SOIUt‘ion Typ°OOOI0.0.00.00.00.00.0000IOIOIIII 000000000000 186 Means and Standard Deviations for Four Strategy Categories.......................... .......... ..187 Repeated-Measures Analysis of Variance for Non-confrontational Solutions............... ..... .. ...... 188 Repeated-Measures Analysis of Variance for Help-Seeking Solutions...................................189 Repeated-Measures Analysis of Variance for Assertive Solutions......................................190 Repeated-Measures Analysis of Variance for Ru‘e-B.8.d so‘utionQOOOOOOIIOOIOIIOOOIOOOOUIOOIOOOOO ..... ‘91 Repeated-Measures Analysis of Variance for Nonsense Solutions............................ ...... .....192 vii 17. 18' 19. 20. 21. 22. 23. 24. Repeated-Measures Analysis of Variance for Aggressive Solutions............................ ......... 193 Repeated-Measures Analysis of Variance for Finagle Solutions............................... ..... ....194 Repeated-Measures Analysis of Variance for Make a Deal--Restructure Situation to Meet Both Parties, N°.d8 so‘ution‘OIIIOIOIIOIIOIOOOOOIIII.0.0.0....195 Repeated—Measures Analysis of Variance for Passive so‘ution800000000IOOOOOIIIOIOIIIOICOO...000......196 Repeated-Measures Analysis of Variance for Compliance Solutions................................ ..... 197 Repeated-Measures Analysis of Variance for EffeCtivenesso0.0.00.0...0.00.00...IDIOOOIIOOOOII00.000.0198 Repeated-Measures Analysis of Variance for Planfulness.... ..................... .. ......... . ......... 199 Repeated-Measures Analysis of Variance for Problem Solving Flexibility..............................200 viii LIST OF FIGURES Rubin and Krasnor's (1988) Model of Social Problem Solving........... ..... ..........................201 Gender of Subject x Nature of Request Interaction for Non-confrontational Solutions............................202 Gender of Subject X Nature of Request Interaction for Help-seeking Solutions ....... ............................203 Partner Age x Nature of Request Interaction for “BID-seeking salutionS0.0000IOOOOOOOIOOIOOIO000......0.0.204 ix INTRODUCTION ve v'ew o ' d Sexua se eve tio ese c American society has become increasingly aware of the prevalence of child sexual abuse. Epidemiological and clinical studies, the news media, and grass-roots as well as state-funded child advocacy projects have described the pervasive nature and devastating effects of sexual victimization in childhood. In response to this recognition of the scope and negative emotional impact of child sexual abuse, a plethora of "personal safety" programs have been developed and implemented to prevent the sexual victimization of children (Finkelhor, 1984). Plummer (1986) estimated that by 1986, diverse plays, curricula, and coloring books had reached over one million children with sexual abuse prevention messages. Unfortunately, these child sexual abuse (CSA) prevention programs have evolved without a corresponding theoretical framework that can explain their effects or an assessment technology that can reliably and validly measure program outcome. The information and skills presented in most child sexual abuse prevention curricula were developed solely from an examination of the circumstances or ”preconditions" that surround child sexual victimization (e.g., children’s misconceptions or lack of knowledge 1 2 regarding sexual abuse, and the strategies used by perpetrators to identify victims and insure their cooperation) (Conte, 1986; Finkelhor, 1979, 1984; Plummer, 1986). Current CSA prevention programs provide factual information and behavioral skills which enable children to protect themselves in sexually abusive encounters. When program outcome is evaluated, curriculum effectiveness is usually determined by a self-report multiple choice questionnaire which assesses only abstract knowledge, pg; behavioral skills. This approach to CSA prevention fragments children’s performance in sexually abusive encounters into several seemingly unrelated pieces. First, it does not consider all of the cognitive and behavioral processes that must be used to effectively terminate a sexually exploitative encounter. Second, current CSA prevention research does not conceptualize these cognitive and behavioral components as working in an interdependent and integrated manner. Third, the current approach does not consider how situational factors affect the cognitive and behavioral processes involved in terminating a sexually abusive experience. Fourth, CSA prevention research has been plagued by criticisms of inadequate or inappropriate assessment techniques (Conte, 1987; Leventhal, 1987; Reppucci & Haugaard, 1989). As Finkelhor (1984) has suggested, empirical findings and theory from other areas of psychology need to be applied to CSA prevention efforts in order to 3 address these important criticisms. The social problem solving (SPS) literature provides the needed theoretical framework and assessment technology. In an effort to fruitfully integrate the social problem solving and child sexual abuse prevention literatures, this paper will, first, present a model of the social problem solving process and apply this model to children’s performance in both commonly occurring non-sexual social dilemmas and sexually abusive situations. Both theoretical and measurement issues will be examined for each component of the SPS model. Second, the validity of SPS assessment methods for predicting performance in both common non-sexual and sexually abusive situations is examined. Finally, a rationale for the present study is provided, the variables of interest in this study are specified, and predictions are made regarding children’s performance in common non-sexual and sexually abusive situations. Integrative Review of the Social Problem Solving (SP8) and ' Se us use CSA eve t' t es An Overview of the Social Egoblem Solving Literatgze Systematic research on how children solve every-day interpersonal problems began in the early 1970’s with the work of George Spivack and Myrna Shure (Spivack & Shure, 1985). Their research specified the cognitive skills young children employed to solve social problems and developed interventions that fostered the growth of these skills (Spivack, Platt, & Shure, 1976; Spivack & Shure, 1974). In 4 addition, Spivack and Shure (1974) pioneered the development of psychological measures which assess these interpersonal skills. Historically, social problem solving research has been concerned with relating SPS ability to teacher, peer, or self ratings of adjustment or sociometric status (Krasnor & Rubin, 1981). Much of the research in SPS was inspired by Shure and Spivack’s (1972) initial findings, which indicated that interpersonal cognitive problem solving skills distinguished normal from maladjusted preschoolers. Moreover, they later provided data that indicated training in SPS skills improved adjustment in maladjusted children (Spivack et al., 1976; Spivack & Shure, 1974), and that SPS intervention with normal preschoolers prevented development of problem behaviors up to two years later (Shure & Spivack, 1982). Thus most of the SPS literature, inspired by these exciting findings, has continued to investigate the relationship of SPS skill to social competence and/or psychological adjustment. However, two new areas of research in the social problem solving are of particular interest to child sexual abuse prevention efforts. First, the bulk of SPS interventions seek to enhance the ggngxgl interpersonal cognitive skills (e.g., identification of feelings, alternative thinking, consequential thinking) used to solve a myriad of every-day social problems (e.g., being teased, sharing objects, resolving differences of opinion with adults). However, SPS 5 training techniques and assessment methods have recently been applied to specific social problems children face (e.g., substance abuse) in an effort to enhance a specific type of social response (e.g., peer resistance). Caplan et a1. (1989) successfully employed SPS training techniques (as well as other substance use prevention techniques) to enhance adolescents’ ability to resist offers to use licit and illicit substances. Equally important, Caplan et al. (1989) adapted SPS assessment methodology to measure program effects on peer resistance. SPS training and assessment methods also have been utilized in interventions that teach specific prosocial behaviors to deviant populations (e.g., prosocial responses to aggressive or rejected children) (Bierman & Furman, 1984; Ladd, 1981; LaGreca & Santogrossi, 1980). These applications of SPS training techniques and assessment methods to specific social problems are most relevant to the child sexual abuse prevention literature. Second, SPS researchers have developed more rigorous and precise models that delineate the relationship between social cognition and social behavior (Dodge, 1986; Krasnor & Rubin, 1986; Rubin & Krasnor, 1981). These models of social problem solving can be applied to a wide variety of social decisions (Dolgin, 1986), including the strategizing children must engage in to protect themselves in a sexually abusive encounter. Def’ itio o Soc'a o v’ Social problem solving (sometimes conceptualized as 6 ”social competence") has been described as the complex interplay of cognitive and behavioral factors that people use to solve interpersonal conflicts (Dodge, 1986; Rubin & Krasnor, 1986; Spivack et al., 1976). Researchers have repeatedly found that social problem solving is not a unitary construct, but, instead, is a multifaceted phenomenon (Dodge, 1986; Krasnor & Rubin, 1981; Shure & Spivack, 1982; Spivack et a1. 1976). In the early 1970’s, D’Zurilla and Goldfried (1971) articulated a formal definition that outlined the essential elements involved in social problem solving: (a) identification of a situation as problematic, (b) generation of possible alternatives to solve the problem, (c) choosing the appropriate solution for the situation, and (d) implementation of the chosen strategy. Since that time, researchers have elaborated the cognitive and behavioral components of social problem solving ability. Spivack et a1. (1976) emphasized four cognitive skills necessary for effective social problem solving in preschoolers: (a) the ability to generate alternative solutions to a problem, (b) the ability to think of consequences of an interpersonal action, (c) the ability to reason step-by-step from the current situation to the goal, and (d) the ability to think of a problem in interpersonal terms. These social cognitive skills seem to reflect processes different from those of general intelligence (Shure, Spivack, & Gordon, 1972; Shure, Spivack, & Jaeger, 1971). Recent formulations of social problem solving have more precisely specified the cognitive and behavioral components of SPS performance and delineated the relationship among these components. Dodge (1986), for example, has proposed that cognitive processes such as sensory stimuli encoding, mental representation of the problem situation, and behavioral components such as motoric deployment of a chosen social problem solving strategy are essential aspects of social problem solving. A number of researchers have emphasized that social problem solving is a process comprised of many discrete but igtgzgependent cognitive and behavioral elements (Dodge, 1986; Rubin & Krasnor, 1986). Only a few models have been developed that integrate both the cognitive and behavioral components of social problem solving (Dodge, 1986; Rubin & Krasnor, 1986). Models that describe how social cognition relates to social behavior are necessary for a complete understanding of children’s functioning in interpersonal situations (Dodge, 1986). As argued above, multiple cognitive and behavioral skills determine a child’s performance in any given social situation. Thus the task becomes to dissect the complex phenomenon of social performance into researchable pieces. Models of the SPS process enable researchers to segment into discrete, yet interrelated, variables the complex interplay of social behavior and social cognition. By breaking down the SPS process we are Spivack, a Jaeger, 1971). Recent formulations of social problem solving have more precisely specified the cognitive and behavioral components of SPS performance and delineated the relationship among these components. Dodge (1986), for example, has proposed that cognitive processes such as sensory stimuli encoding, mental representation of the problem situation, and behavioral components such as motoric deployment of a chosen social problem solving strategy are essential aspects of social problem solving. A number of researchers have emphasized that social problem solving is a process comprised of many disczetg but intendgpendent cognitive and behavioral elements (Dodge, 1986; Rubin a Krasnor, 1986). Only a few models have been developed that integrate both the cognitive and behavioral components of social problem solving (Dodge, 1986; Rubin & Krasnor, 1986). Models that describe how social cognition relates to social behavior are necessary for a complete understanding of children’s functioning in interpersonal situations (Dodge, 1986). As argued above, multiple cognitive and behavioral skills determine a child’s performance in any given social situation. Thus the task becomes to dissect the complex phenomenon of social performance into researchable pieces. Models of the SPS process enable researchers to segment into discrete, yet interrelated, variables the complex interplay of social behavior and social cognition. By breaking down the SPS process we are 8 able to achieve, through detailed study of each component, a more sophisticated understanding of the SPS process as a whole. In addition, models of the SPS process enable researchers to make more accurate predictions about actual social performance. Since multiple processes are involved in social behavior, no single process can account for a given social outcome. As Dodge (1986) points out, even a cursory glance at the social competence promotion literature quickly reveals that the prediction of social behavior from a single social or cognitive factor has yielded mixed results. He has argued that the use of a model that specifies the relationship between multiple situational, cognitive, and behavioral variables will allow us not only to better understand children’s social performance but also to predict more accurately children’s social behavior. Rubin and flzasnor’s Model of SPS Processes Rubin and Krasnor (1986) present a model that delineates the role of both cognitive and behavioral variables in the social problem solving process (see Figure 1). The model partitions social problem solving into five principal components: (a) goal selection, (b) examination of the social context or "task environment," (c) assessment of information concerning strategies used in similar situations and other relevant knowledge, (d) strategy selection, and (e) strategy implementation. The remaining elements of the model delineate the recursive nature of children’s 9 strategizing in social dilemmas and indicate that the SPS process may proceed in either an automatic or a reflective fashion. In the following sections, each component of this model is examined in terms of children’s performance in both common non-sexual and sexually abusive situations. Both theoretical and measurement issues in the SPS literature and in the CSA prevention literature are considered for each stage in the SPS process. figgl Selection Theory. During this stage children formulate the criteria by which they will judge the success or failure of their problem solving attempts. In selecting a social goal, children define the end state they hope to achieve (Rubin & Krasnor, 1986). Coal selection is a critical component in SPS because children’s performance in subsequent stages is motivated and directed by their choice of a social goal. Equally important, accurate interpretation of children’s social behavior is not possible without a knowledge of their specific social goals (Krasnor, 1985). The relevance of goal selection in CSA prevention research is two-fold. First, as with the assessment of SPS in common non-sexual situations, one must take into account children’s goal selection to interpret accurately their behavior in a sexually abusive situation. When children do not defend themselves in sexually abusive situations, it may be because their goal is to comply with the perpetrator’s request 91 because their goal is to resist the proposition, 10 but they cannot generate any self-protective strategies that would achieve this end. Thus an accurate understanding of children’s performance in sexually abusive encounters depends upon a knowledge of their social goal in that encounter. Second, an understanding of children’s social goals in situations that require them to resist the request of an adult are pertinent to the development of effective CSA prevention programs. An examination of the types of social goals children choose in naturally-occurring non- sexual situations indicates that young children rarely choose resistance (of peer or adult requests) as a social goal (Krasnor, 1982). Hence, it may be important for CSA prevention programs to counteract children’s tendency £2; to resist by effectively communicating to children that it is o iat for them to choose self-protective resistance goals in sexually abusive situations. Measurement. The investigation of children’s social goals is a complex matter (Krasnor, 1985; Rubin s Krasnor, 1986); it is virtually impossible to be sure of another person’s internal motivation. On the one hand, if the experimenter sets the goal for the child (either in a hypothetical social dilemma or in a simulated social dilemma), the experimenter cannot be certain the child has actually assumed that goal. Instead, the child’s goal may be to provide a socially acceptable answer for the investigator! On the other hand, if the investigator attempts to infer children’s social goals from their natural 11 behavior, he/she will quickly become mired in philosophical questions regarding who is the best judge of an individual’s intention (Rubin & Krasnor, 1986). Social problem solving research has dealt with this methodological problem by specifying the social goal the child or adolescent is working to achieve (Caplan et al., 1989; Spivack & Shure, 1974; Weissberg, Gesten, Rapkin, et al., 1981). Essentially, by specifying the social goal, SPS assessment techniques control for the variety of social goals children can assume and thus researchers can focus their investigations on how well children reach specific social goals. The one drawback of this methodology is that it assumes the child is motivated to achieve the goal set by the experimenter (Krasnor & Rubin, 1981). In spite of the importance of this aspect of the SPS model for CSA prevention, none of the studies in child sexual abuse prevention literature have adequately investigated this variable (see Table 1). Nor do the majority of extant CSA prevention studies attempt to control this variable by explicitly stating the goal and then asking children how they would achieve it. Only one of the currently published studies in the CSA prevention literature controls for children’s goal selection. In their assessment of children’s assertiveness in sexually inappropriate situations, Hill and Jason (1987) instructed children to refuse to participate in a role-played sexually inappropriate and/or dangerous situation. 12 Encoding,of Social,Cue§ Theory. During this stage, children encode the stimulus features of the social context, or task environment. The type of setting (e.g., home, classroom, store) and its unique physical features (e.g., number of exits), as well as the physical characteristics of the target person, are important components of the task environment. Verbal or physical gestures that define the nature of the social interaction are additional aspects of the social context. Children also may encode more subtle aspects of the social situation such as internal cues indicating their emotional state and external cues indicating the target person’s emotional state (Dodge, 1986). The efficiency and accuracy with which a child perceives the social cues that define a situation is critical to successful performance at later stages of the SPS process, because it determines the appropriateness of the strategy the child considers and, eventually, enacts (Dodge, 1986). Similarly, in a sexually abusive encounter, children are more likely to deploy appropriate (i.e., self- protective) strategies if they accurately encode the relevant social cues indicating that it is a sexually abusive and not a non-threatening situation. Indeed, Hazzard, Webb, Kleemeier, and Pohl (in press) have recently argued that CSA prevention programs need to focus their efforts primarily on improving this component of the self- 13 protective process. These investigators point out that unless children accurately encode a situation as inappropriate, they will not enact gay self-protective strategies. Hence, child sexual abuse prevention programs often direct a considerable amount of the curriculum to increasing the accuracy with which children encode the cues of a sexually abusive situation. For example, CSA prevention programs often teach children to distinguish between different types of interpersonal touching (Conte, Rosen, & Saperstein, 1986; Gilbert, Berrick, LeProhn, & Nyman, 1989). In addition, most curricula instruct children that both familiar and unfamiliar persons might ask them to engage in inappropriate sexual touching (Wurtele, 1987). Two specific questions are suggested by an integration of empirical findings in the SPS literature with CSA prevention research. First, the SPS literature suggests that situational factors may affect the accuracy with which children perceive the social situation. Dodge and his associates have found that five-year-olds more accurately encode hostile social cues than prosocial ones (Dodge, Murphy, 5 Buchsbaum, 1984) Moreover, it seems that CSA prevention programs assume children have more difficulty encoding the highly unusual experience of inappropriate sexual touch than common non-sexual social dilemmas. However, this putative difference in the encoding of non- sexual and sexually abusive situations needs to be 14 empirically tested by a comparison of children’s ability to encode sexually abusive encounters with their ability to encode every-day non-sexual problem situations. Such a comparison could begin to answer the question of whether the sexual nature of the request influences the accuracy with which children encode the situation. Second, research in the SPS literature suggests that the accuracy of encoding is affected by numerous characteristics of the social partner involved in the dilemma (Dodge, 1986). For example, preschoolers and young children encode social cues with greater accuracy if the social partner is the same gender (Deutsch, 1975; Fleshbach & Roe, 1968) or the same age (Flapan, 1968). It may be that these variables affect the accuracy with which children perceive a sexually abusive encounter, too. If we can identify empirically the situational factors that confuse children, CSA prevention curricula can more effectively correct these specific misunderstandings and teach children to accurately label sexually abusive encounters. Measurement. This component of the SPS model has been investigated using both visual and auditory stimuli (e.g., film, videotape) (Dodge et al., 1984; Flapan, 1968) or using auditory stimuli alone (Dodge & Newman, 1981). Typically, the stimulus is presented to children individually and then they are asked to describe the nature of the social situation or to determine the intent or emotional state of the actors. 15 Only fifteen out of twenty-six studies reviewed in the CSA literature assess whether CSA prevention programs effectively increase children’s ability to encode social cues (see Table 1). And none of these studies report findings that would shed light on whether gpggifig person or environmental variables make it more difficult for children to accurately encode a situation as sexually abusive. However, from the studies available, it seems that a wide age-range of children (i.e., four to eleven year-olds) are able to identify a wide variety of visual or verbal depictions of sexual abuse with great accuracy before a CSA intervention (Blumberg, Chadwick, Fogarty, Speth, a Chadwick, 1988; Miltenberger & Thiesse-Duffy, 1988; Swan, Press, & Briggs, 1985; Wurtele, 1990; Wurtele, Marrs, & Miller-Perrin, 1987). Clearly, the ceiling effects found in these studies limits the conclusions that can be drawn about children’s ability to encode sexually abusive situations (Reppucci a Haugaard, 1989). Thus, one of the basic assumptions of CSA prevention theory (i.e., that children have difficulty accurately encoding sexually abusive encounters) has not been supported by the limited research testing this notion. The ceiling effects reported in the literature suggests that either a) children are capable of encoding inappropriate situations without an intervention, or that b) the assessment techniques used in CSA prevention research do not adequately measure this skill. Victims of sexual abuse often report 16 that they did not recognize the sexual touching as inappropriate (Finkelhor, 1984). Hence it seems unlikely that the above findings represent an accurate assessment of children’s encoding. Rather, it seems that the techniques utilized by CSA prevention do not measure this skill adequately. CSA prevention researchers need to develop measures of encoding that are capable of determining a) whether or not children do, in fact, have difficulty recognizing sexually inappropriate situations, and b) how both person and situational factors affect the accuracy of children’s encoding in sexually abusive encounters. Accessing Strategy Inforggtion,Relevgnt to Current_nilgmma Theory. Rubin and Krasnor (1986) argue that accessing strategy information involves recall of behaviors previously deployed (or observed) in response to social dilemmas similar to the current one. In addition, children may remember other types of information relevant to the current situation (e.g., knowledge about the target person, available resources). The material recalled during this stage may have been taught didactically to children or may have been learned from trial-and-error social experience. Alternatively, children may formulate new untested strategies by modifying and/or combining old behavioral responses. The child’s ability to access gccgratg and rglgxgnr information is critical to the outcome of any interpersonal dilemma. In sexually inappropriate situations, children are 17 more likely to recognize and effectively terminate the exploitative encounter if they are able to access knowledge that helps them interpret the social situation and formulate relevant behavioral strategies. Child sexual abuse prevention programs devote considerable resources toward teaching children concepts and information that will aid them in successfully defending themselves in a sexually exploitative situation (Conte et al., 1986; Gilbert et al., 1989; Wurtele, 1987). CSA prevention programs provide children with general information as well as specific verbal and behavioral strategies. For example, many programs teach children abstract concepts such as "body ownership" (i.e., that children have the right to refuse touches they do not like) and that they are not at fault if an adult attempts or succeeds in touching them inappropriately (Gilbert et al., 1989). Most CSA programs also teach children behavioral strategies such as saying "no," giving a self-defense yell, leaving the sexually abusive situation, and telling an adult about the experience (Conte et al., 1986; Gilbert et al., 1989). Measuremen . A number of different assessment techniques have been used in SP8 and CSA prevention evaluations to measure this aspect of the SPS model. In CSA prevention research, both structured interviews which require children to generate responses (Borkin a Frank, 1985; Christian, Dwyer, Schumm, & Coulson, 1988; Conte, 18 Rosen, Saperstein, a Shermack 1985; Daro, Duerr, & LeProhn, 1987; Harvey, Forehand, Brown, & Holmes, 1988; Hazzard et al., in press; Hill a Jason, 1987; Kenning, Gallmeier, s Plemons, 1987; Saslawsky & Wurtele, 1986; Wurtele, 1990; Wurtele, Kast, Miller-Perrin, & Kondrick, 1989; Wurtele et al., 1987; Wurtele, Saslawsky, Marrs, & Britcher, 1986) and group or individually administered questionnaires which utilize a multiple-choice, true-false, yes-no, or Likert (e.g., never-always) response format have been employed to measure the knowledge component of the SPS process (Binder, Dale, & McNiel, 1987; Conte et al., 1985; Garbarino, 1987; Kolko, Moser, Litz, & Hughes, 1987; Nelson, 1981; Plummer, 1984; Ratto & Bogat, 1990; Saslawsky & Wurtele, 1986; Sigurdson, Strang, & Doig, 1987; Swan et al., 1985; Wolfe, MacPherson, Blount, & Wolfe, 1986; Wurtele, 1990; Wurtele et al., 1986; 1987; 1989). CSA prevention programs often include some measure of children’s knowledge. Twenty-two out of twenty-six of the studies reviewed measure children’s ability to access relevant factual knowledge about the nature of sexual abuse, characteristics of perpetrators, and specific actions children can take to defend themselves (see Table 1). All these studies report increases in children’s knowledge. The confidence with which these changes may be attributed to participation in CSA prevention programs varies with the experimental design employed by each study (see Wurtele, 1987, for a review). 19 However, the large number of studies that report data on this component of the SPS process and the substantial minority of studies that assess 9311 this aspect of the SPS process (Binder et al., 1987; Borkin & Frank, 1985; Conte et al., 1985; Kenning et al., 1987; Kolko et al., 1987; Nelson, 1981; Plummer, 1984; Ray & Dietzel, 1988; Wolfe et al., 1986) collectively suggest that children’s ability to access factual knowledge is a valid indicator of the effectiveness of CSA prevention programs. However, since none of the studies reviewed assessed both knowledge and actual behavioral performance (see Table 1), the strength of the relationship between these two variables in sexually abusive situations is unknown. Some lessons may be learned from the SPS literature in this regard. Dodge (1986) asserted that the research on the relationship between measures of SPS knowledge and 13 11x9 social competence demonstrates that assessment of only one component of SPS performance does not accurately predict in girg SPS behavioral performance because multiple factors affect SPS performance. Similarly, the ability to remember relevant factual information is only one of the components of successful strategizing in a sexually abusive encounter. Used alone, it may not provide a complete enough understanding of children’s strategizing to predict successful performance in a sexually abusive episode. Hence, the use of knowledge as a predictor of successful performance in sexually abusive encounters, without empirical support for the correlation between 20 knowledge and behavior, may be inaccurate. Strate Se ectio Theory. Once relevant knowledge and behavioral strategies have been accessed, the child must then select the most appropriate behavioral response. Strategy selection has often been viewed as the most crucial component of the SPS process (Dodge, 1986; D’Zurilla & Goldfried, 1971; Spivack et al., 1976); hence, considerable theory and research have been devoted to its delineation. This component of SPS requires the translation of more general information into one or more possible strategies that take into account the specific circumstances of the problem at hand (Rubin a Krasnor, 1986). In addition, strategy selection may involve a decision-making component, if more than one problem solving strategy was formulated. In this case, the child must then select the strategy that he/she perceives to be the most appropriate solution to the dilemma (Spivack et al., 1976; Spivack & Shure, 1982). Rubin and Krasnor (1981; 1983; 1986) suggest that strategy selection can be most meaningfully studied by examining how situational (i.e., contextual) and/or subject variables affect children’s strategy choice. A number of situational factors have been reported to influence children’s strategy selection in common non-sexual social dilemmas (e.g., object acquisition, friendship initiation). Characteristics of the other person involved in the social dilemma with the child (hereafter referred to 21 as the "social partner") have been found to affect the child’s strategy selection. Observational studies of young children indicate that the gender (Jacklin & Maccoby, 1978; Rubin a Krasnor, 1983), age (Holmberg, 1980; Krasnor, 1982; Langlois, Gottfried, Barnes, & Hendrics, 1978; Rubin a Krasnor, 1983; Whiting & Whiting, 1975), relative peer status (Putallaz & Gottman, 1981), and familiarity (Doyle, 1982; Gottman & Parkhurst, 1980) of the social partner affect strategy choice. Of these variables, the age of the social partner is the most frequently studied. In general, researchers have found that children generate more prosocial, non-aggressive strategies toward older targets (Holmberg, 1980; Krasnor, 1982; Langlois et al., 1978; Rubin & Krasnor, 1983; Whiting & Whiting, 1975). For example, children use more politeness markers (e.g., "please") when negotiating social problems involving older children or teachers (Krasnor, 1982; Rubin & Krasnor, 1983). The nature of the request (or behavior) upon which the social dilemma centers has also been found to influence children’s strategy choice (Krasnor, 1982; Krasnor 1983). (A social conflict can result from either a request to engage in unwanted activity or a threatening behavior. However, for the remainder of this study, this aspect of the situation will be referred to as simply "the nature of the request” regardless of the precise cause of the conflict.) In her observational study of preschoolers, Krasnor (1982) found a statistically significant relationship between the 22 problem solving strategy employed and the nature of the request. For example, she reports that children are most likely to use commands to stop another child from engaging in an activity and descriptive statements to gain attention. The gender of the problem solver (hereafter referred to as ”the gender of the subject") has also been found to affect the type of strategy used (for a review see Maccoby & Jacklin, 1974). Specifically, in common social dilemmas, boys tend to generate more aggressive strategies than girls (Barnett, Darcie, Holland, & Kobasigawa, 1981; Marsh, Serafica, a Barenboim, 1981). One study found that 3rd- grade girls exhibited better overall problem solving ability in common non-sexual social situations than did 3rd-grade boys (Reid, 1991). In sexually abusive encounters, children’s strategy selection may be affected by similar variables. For example, it could be that children are less likely to choose assertive self-protective responses with an adult who asks the child to engage in inappropriate touching than they would be with §_rhiTg who wants to engage in the same activity. However, CSA prevention research has not explicitly investigated the effect of the age of the social partner on children’s strategy choice in sexually abusive encounters. CSA prevention researchers frequently posit that the uncommon nature of a request to engage in sexual touching influences children’s self-protective ability in sexually 23 abusive situations (e.g., Conte et al, 1986; Finkelhor, 1979; Gilbert et al., 1989; Wurtele, 1987). Although the rationale behind this assertion has not been clearly explicated, it is generally thought that children may be confused by such a request. This confusion may result in an inability to enact self-protective behaviors. It may be that this deficit in self-protection can be fruitfully conceptualized as a deficit in strategy selection. SPS theory asserts that strategy selection requires both the application of general knowledge and the choice of an appropriate strategy. Thus, this view of children’s strategizing provides a hint of two specific ways in which problem solving could fail in sexually abusive encounters. Children may not be able to translate general knowledge into concrete behavior, or they may fail to select the most effective self-protective strategy. Perhaps just as the nature of the request influences children’s strategy choice in the common non-sexual situations investigated in the SPS literature, the nature of the request may also influence children’s strategy choice in sexually abusive encounters. However, the effect of a request to engage in unwanted sexual touching on children’s strategy choice has not been empirically studied. Although accurate statistics on the prevalence and incidence of child sexual abuse are difficult to obtain (Wyatt a Peters, 1986a; 1986b), the literature indicates that girls are more frequently victims of sexual abuse than 24 are boys (see Finkelhor, 1984 for a review; Sedlak, 1989). The SPS literature suggests that the gender of the subject influences children’s strategy selection in every-day social dilemmas. One important question that CSA prevention research could answer is whether the gender of the subject also affects the strategies chosen in uncommon, sexually abusive encounters. Thus, of the factors that have been found to influence children’s strategy choice in common non-sexual social dilemmas, two situational variables--the age of the social partner and the nature of the request--and one subject variable--the gender of the subject-—are of particular interest to CSA prevention efforts. Measurement. The nature of the solutions children select and the link between situational variables and strategy selection have been investigated in the SPS literature. Typically, SPS research has utilized what Cooney and Selman (1978) have categorized as "hypothetical- reflective" assessment techniques to explore this component of the SPS process. Hypothetical-reflective assessment techniques present children with a realistic hypothetical social dilemma as well as an appropriate social goal. Children are then asked to generate g; ngny golutions gs possible that will solve the hypothetical social dilemma and attain the goal. This method has been used successfully with preschoolers and kindergartners (Rubin a Krasnor, 1983), third- and fourth-graders (Weissberg, Gesten, 25 Carnrike, et al., 1981; Weissberg, Gesten, Rapkin, et al., 1981) and fifth- and sixth-graders (Herz, 1987; Spivack et al., 1976). An alternative approach employed with younger age groups has involved presenting a number of similar social situations that have the same social goal and asking children to provide only a single response to each situation (Spivack & Shure, 1974; Spivack et al., 1976). The strategies children suggest have been assessed along a number of dimensions. Both the quantity and quality of the solutions children generate are critical aspects of strategy selection since these features greatly influence the outcome of the social dilemma. If a child generates even a single adaptive solution, the dilemma will probably be resolved. On the other hand, if the child enacts one or more idiosyncratic or maladaptive solutions, the dilemma may remain unresolved or escalate. The quantity of strategies or "strategy repertoire" is often calculated by determining the number of relevant alternative solutions a child is able to generate for a specific type of social dilemma (Gesten et al., 1982; McKim et al., 1982; Spivack a Shure, 1974). "Strategy repertoire" is defined as the number of gijjgrgnr solutions generated to solve the same problem situation. This is simply a count of the different relevant solutions the child offers. Implicit in the use of the quantity of solutions as an indicator of SPS ability is the assumption that the more solutions a child is able to generate, the more solutions 26 the child will have available to deploy in an actual problem situation and hence, the more likely the child is to resolve successfully the interpersonal dilemma. However, as many researchers have pointed out, the relationship between strategy repertoire and successful social outcome may be more complex (Dodge, 1986; Krasnor & Rubin, 1981; Rubin & Krasnor, 1986). For example, a child who generates one or two effective solutions to a social dilemma may solve it more effectively than a child who generates ten solutions that are only marginally effective or idiosyncratic. Thus SPS researchers have argued that both the ggagtity and the quality dimensions of social strategies need to be assessed (Dodge, 1986; Kendall & Fischler, 1984; Krasnor & Rubin, 1981). The quality of children’s social strategizing can be evaluated along a number of different dimensions. First, the type of solutions children generate to various social dilemmas can be catalogued (Rubin & Krasnor, 1983; Spivack & Shure, 1974). In addition, solutions can be evaluated according to the "effectiveness" and the "planfulness" they exhibit (Caplan, Weissberg, Bersoff, Ezekowitz, 5 Wells, 1986). The quantification of a solution’s planfulness and effectiveness enables researchers to measure the extent to which a solution reflects the use of a number of the cognitive skills D’Zurilla and Goldfried (1971) and Spivack and Shure (1976) have defined as hallmarks of the SPS process. 27 In common non-sexual social dilemmas, effective solutions have been defined as strategies that are reasonably assured of helping the child achieve his/her goal, take into account the feelings and needs of the other party, and are not likely to cause negative consequences for the actor or for others (Caplan et al., 1986). Solution effectiveness, thus defined, reflects three of the cognitive skills D’Zurilla and Goldfried (1971) and Spivack et al. (1976) consider essential aspects of social problem solving. Effective SPS strategies are ones in which children (a) choose a solution appropriate to the specific situation at hand, (b) consider the consequences of the action, and (0) cast the problem in interpersonal terms. Planful solutions have been defined as responses that explicitly take into account situational factors and/or the means and probable end of implementing the proposed social strategy (Caplan et al., 1986). Planful solutions embody two defining characteristics of SPS cognition, namely, (a) the ability to think of consequences of an interpersonal action and (b) the ability to reason in a step-by-step manner from the current situation to the goal (Spivack et al., 1976). The use of both quantity and quality dimensions in the assessment of the strategies children select provides a detailed and complete understanding of the child’s strategy repertoire. Children’s strategizing in sexually abusive encounters can be assessed along similar dimensions. Responses can be 28 catalogued according to the type of self-protective strategy the child deploys. The effectiveness and planfulness of these responses can also be determined. However, the criteria for evaluating the effectiveness of a response in a sexually abusive situation must reflect the unique nature of that situation. In a sexually abusive episode, an effective response need not take into account the feelings or needs of the perpetrator. Indeed, such considerations may make a solution less effective since they indicate that the child is willing to consider the perpetrator’s needs. Hence, for situations in which a child is asked to resist the request of another, a more appropriate criteria for rating the effectiveness of a strategy are the maximization of the child’s short- and long-term safety, and the minimization of negative consequences for the child. The most common technique used to assess the strategies children choose to defend themselves in sexually abusive encounters is similar to the hypothetical-reflective method employed in the assessment of SPS skills. Twelve out of twenty-six studies employed an assessment instrument in which interviewers read children short scenarios depicting sexually abusive or dangerous situations and asked children how they would respond if faced with that situation (Christian et al., 1988; Daro et al., 1987; Harvey et al., 1988; Hazzard et al., in press; Hill & Jason, 1987; Miltenberger & Thiesse-Duffy, 1988; Ratto & Bogat, 1990; Saslawsky & Wurtele, 1986; Wurtele, 1990; Wurtele et al, 29 1986; 1987; 1989) (see Table 2). However, for a number of reasons, the results reported in these studies provide limited information on the strategies children choose to enact in sexually abusive situations. First, many of the assessment instruments used in these studies differ in important ways from the SPS hypothetical-reflective format (Christian et al., 1988; Ratto & Bogat, 1990; Hill & Jason, 1987; Saslawsky & Wurtele, 1986; Wurtele, 1990; Wurtele et al., 1986; 1987; 1989). A few studies have assessed children’s intentions to use specific self-protection strategies rather than their ability to generate a repertoire of such strategies (Binder et al., 1987; Kolko et al., 1987; Plummer, 1984). Alternatively, some investigators have assessed children’s strategies for coping with sexually abusive and dangerous situations by asking children to choose one or more responses to a hypothetical situation (Conte et al., 1985; Nelson, 1981; Swan et al., 1985). Assessing a child’s strategy repertoire using a format in which the correct response is described in the question does not test whether the child could generate the responses independently of the test instrument. This assessment procedure severely limits the generalizability of the findings. Equally important, multiple-choice questionnaire formats artificially limit the range of strategies children may select by providing all the possible responses. Second, the scoring systems employed rarely assess the 30 full range of children’s strategy repertoire. Yet in a sexually abusive situation a child may need to enact a number of self-protective strategies in order to repel the perpetrator. All of the studies reviewed in the CSA prevention literature employed assessment measures that arbitrarily restricted the number of solutions a child could generate. Children could generate a maximum of three responses to the same story (Hill & Jason, 1987; Miltenberger & Thiesse-Duffy, 1988; Ratto & Bogat, 1990; Saslawsky & Wurtele, 1986; Wurtele, 1990; Wurtele, et al., 1986; 1987; 1989). In addition, a number of researchers utilized a measure in which only one behavioral and one verbal response could be included in the final score, thus further restricting the child’s response repertoire (Ratto & Bogat, 1990; Saslawsky & Wurtele, 1986; Wurtele, 1990; Wurtele et al., 1986; 1987; 1989). In may of the extant studies either vague, inexplicit scoring criteria were employed (Ratto, 1988), or only the specific strategies taught in the CSA prevention curriculum were scored as viable solutions (Christian et al., 1988; Hazzard et al., in press; Miltenberger & Thiesse-Duffy, 1988; Saslawsky & Wurtele, 1986; Wurtele, 1990; Wurtele et al., 1986; 1987; 1989). For example, the only responses which can be scored in the systems employed by the last seven studies are: verbal refusal, leaving the situation, and telling an adult (Hazzard et al., in press; Miltenberger & Thiesse-Duffy, 1988; Wurtele, personal communication, 31 September 1, 1990). This type of scoring excludes many effective solutions and artificially restricts the range of alternative solutions that can be recorded. Third, results reported in some of these studies do not necessarily indicate changes in only strategy repertoire, because the scores reported reflect the quantity and quality of children’s solutions as well as responses to other types of questions (Ratto & Bogat, 1990; Saslawsky & Wurtele, 1986; Wurtele et al., 1986; 1987; 1989). A number of researchers report and analyze total scores that include, in addition to children’s strategies, children’s responses to questions unrelated to strategy repertoire (Ratto & Bogat, 1990; Saslawsky & Wurtele, 1986; Wurtele et al., 1986; 1987; 1989). For example, information regarding the appropriateness of sexually abusive situations and who the child would tell if he/she were abused are often included in the total score. Thus significant changes in total scores may be due to changes in dimensions of children’s performance not related to strategy repertoire. Recently, researchers have begun to recognize this problem. A number of studies now divide the total score into different "factors" (Hazzard et al., in press; Wurtele, 1990). For example, Wurtele (1990) now divides the total score on her measure of children’s self-protective abilities into a ”Discrimination” score that measures encoding and a "Prevention Skills" score that measures strategy quantity and quality. 32 Nevertheless, the frequent use of total scores as well as the format and scoring of extant measures limits the amount of information that can be gained regarding children’s strategy generation in sexually exploitative encounters. Few studies report information which adequately describes multiple dimensions of children’s strategy repertoire (see Table 2). In spite of the fact that the effectiveness of the solutions children choose in sexually abusive encounters is critical to successful termination of the experience, only six out of twenty-six studies in the CSA prevention literature assessed the effectiveness of the solutions children generated. Hill and Jason (1987), Miltenberger and Thiesse-Duffy (1988), Wurtele and her co-investigators (Saslawsky & Wurtele, 1986; Wurtele et al., 1986; 1987), and Ratto and Bogat (1990) score each response according to its effectiveness in protecting the child from sexual abuse. Responses that are scored as maximally effective, for example, include leaving the situation, and telling an adult. None of the published studies assessed the planfulness of children’s responses to sexually abusive encounters. The link between situational factors and strategy selection has also been assessed in the SPS literature. Krasnor and Rubin have designated this link the child’s "responsiveness to the task environment" (Krasnor & Rubin, 1981, p. 469; Rubin & Krasnor, 1986, p. 20-21). As Krasnor 33 and Rubin define it, the child’s responsiveness to the task environment indirectly measures a child’s encoding of important social cues by assessing how his/her strategies change under differing interpersonal conditions. This variable is operationally defined as the degree to which the child predictably modifies his/her social strategies across different social situations (Krasnor & Rubin, 1981). Krasnor and Rubin assume that the later stage of "strategy selection" is dependent upon the encoding process, hence they argue that a child’s sensitivity to key situational and task features will be reflected as variations in strategic behavior across different social contexts. Thus the child’s "responsiveness to the task environment" assesses the strength of the link between the situation and the strategy selection stage of the SPS process. In spite of the need to understand how children’s deployment of self-protective behavior varies under differing circumstances, none of the studies reviewed in the CSA prevention literature systematically varied the interpersonal or situational features of the sexually abusive situations they presented to children. Hence, the relationship between strategy selection and situational factors has not been adequately investigated (see Table 2). St a e In e ta '0 Theory. In this stage of the SPS process children use their physical and/or verbal skills to deploy the strategy selected in the previous stage. Thus, this model makes 34 clear the separation between children’s ability to Lnink of a solution and their ability to gnng; the solution in a relevant context. In a sexually exploitative encounter, strategy implementation may be critical to effective termination of the episode. If children generate potentially effective solutions but enact them in an ineffective manner, the perpetrator may not desist. For example, a child may think of an assertive verbal resistance strategy (such as saying "no") but may perform the verbalization in such an unassertive manner that the perpetrator is not deterred. Measungment. There have been relatively few attempts in the SPS literature to measure this component of the SPS model (Rubin & Krasnor, 1986). Studies that have assessed strategy implementation have utilized role-play methodology (Feldman, 1983; Herz, 1987; Nelson & Carson, 1988) which assessed deployment variables such as voice tone, eye contact, physical gestures, and body stance. Four of the twenty-six studies in the CSA prevention literature evaluated this component of the SPS model (see Table 1). Moreover, it should be noted that one of these studies (Miltenberger & Thiesse-Duffy, 1988) assessed only children’s ability to resist abduction and n2; to resist a request to participate in sexual touching. With that caveat in mind, Miltenberger and Thiesse-Duffy utilized role-play and in 21x9 follow-up assessment procedures. Both motoric (e.g., did child run away from perpetrator) and verbal 35 (e.g., did child say "no" to perpetrator’s request) dimensions of children’s performance were assessed. They found that all of the 24 children studied reached a criterion performance of two perfect role-plays after individual training by the experimenter. However, only the six to seven year-olds performed adequately in the two-month in xixn follow-up assessment. Daro et al. (1987), Hill and Jason (1987), and Kenning et al. (1987) evaluated children’s ability to respond assertively in a number of role-play situations. Assertiveness skills were assessed in both sexually inappropriate and abduction situations. Again both verbal and motoric dimensions of children’s responses were evaluated. Hill and Jason and Daro et al. report statistically significant improvements in children’s assertiveness after participation in a CSA prevention intervention. SPS Outcone and the Recursive Nature of the SPS Ezogess Iheozy. Rubin and Krasnor (1986) argue that children compare their perceptions of the social context after strategy implementation to the criteria previously established in the goal selection phase of the SPS process. Based on the extent to which the social context after strategy implementation matches the social context delineated in the goal stage, the strategy is judged a success, partial success, or failure. This judgement is a branch-point in the SPS process. If the strategy succeeds, 36 then the SPS process is terminated. If the strategy is judged to have only partially succeeded or to have failed, then a number of options are open to the child. The child may automatically or reflectively (a) stop formulating any type of social goals for that situation, (b) change social goals, (c) repeat the same strategy, or (d) modify the previous strategy while maintaining the same goal. Whether children proceed or stop problem solving may be affected by the importance of attaining the goal relative to the "cost" of reattempting the goal with the same or a different strategy or by the range of the child’s strategy repertoire (Rubin & Krasnor, 1986). Children are more likely to repeat a social strategy if the goal is important relative to the social cost of implementing the strategy or if they have access to relatively few strategies. In addition, Rubin and Krasnor (1986) assert that a child’s emotional response to problem solving failure may affect how the child proceeds. They hypothesize that failure to attain a social goal may be accompanied by strong negative affect, particularly if the target person or social goal was important to the child. These negative feelings may interfere with the cognitive processing necessary to formulate a new social strategy. Thus, the angry or anxious child is more likely to repeat a previous strategy since repetition requires less cognitive processing than does formulation of a new strategy. Alternatively, the child may terminate the problem solving process in order to end the 37 unpleasant emotional state that problem solving failure produces. Thus, Rubin and Krasnor argue that whether children reflectively persist with new or similar SPS strategies or automatically or reflectively terminate the problem solving process, depends heavily on the personal importance of the goals children attempt to attain, the repertoire of strategies they possess, and their emotional state. In a sexually abusive situation both persistence and response flexibility in the face of initial strategy failure are critical to effective termination of a sexually exploitative episode. Perpetrators often have a previously established relationship that allows repeated contact with children, and they often repeatedly approach their intended victims, and/or attempt to persuade their victims even when the children initially refuse to participate (Conte, Wolfe, & Smith, 1989; Finkelhor, 1974). Therefore, children who persist in enacting nlternative self-protective strategies when their first strategy fails, are more likely to terminate the abusive encounter successfully. However, the social context of a sexually exploitative encounter may mitigate against children’s persistence and flexibility in the face of initial strategy failure. It seems likely that the failure of children to reach their initial goal to fend- off the perpetrator would produce high levels of negative affect. This possibility combined with the fact that they are attempting a "high cost" behavior (i.e., to resist the 38 request of an adult who they may love or respect) may motivate them to terminate problem solving or may interfere with their efforts to continue formulating effective alternative strategies. If children continue to problem solve, they will lack multiple strategies for terminating such a novel situation and hence it is likely that they will repeat initial strategies rather than implement new ones. Measunement. Hypothetical-reflective assessment techniques reported in the SPS literature allow for the assessment of the flexibility of children’s strategizing in the face of initial strategy failure (Krasnor & Rubin, 1981). Response flexibility is assessed by telling children that their first strategy failed and asking them what they would do next (Krasnor & Rubin, 1981). None of the studies reviewed in the CSA prevention literature adequately assessed this aspect of SPS performance (see Table 3). The "What If Situation Test" (WIST), developed by Wurtele and her co-investigators (Saslawsky & Wurtele, 1986; Wurtele et al., 1986; 1987), queries children regarding what they would do if their initial strategy failed; however, the alternative responses children provide are not differentiated from the initial response. Similarly, Hill and Jason (1987) present scenarios in which children must respond to initial strategy failure, but specific analysis of their response in the face of initial strategy failure is not reported. Hence, no information has been reported in the literature regarding 39 the persistence or flexibility with which children respond to a failed problem solving attempt in a potentially sexually abusive encounter. Although the versions of the WIST utilized in extant studies (Ratto & Bogat, 1990; Saslawsky & Wurtele, 1986; Wurtele et al., 1986; 1987), as well as assessment measures used in other studies, (Miltenberger & Thiesse-Duffy, 1988; Hill & Jason, 1987) ask children for multiple responses to a sexually abusive situation, the sequencing of the questions and/or scoring systems employed restrict the type of responses that can be scored. For example, on the WIST the first two questions ask specifically for verbal responses and the third question requires children to generate a behavioral response. Hence, sequencing patterns are artificially imposed by the questions. Miltenberger and Thiesse-Duffy (1988) and Hill and Jason (1987) allow children to freely report the sequence of strategies they would deploy in sexually abusive and dangerous situations, however they do not report information regarding patterns in the sequence of strategies children indicate they would use. Hence, the natural sequencing of children’s responses to a potentially sexually abusive situation has not been adequately explored (see Table 3). Co c '0 Thus the SPS literature can provide CSA prevention research with the needed theoretical underpinnings. Rubin and Krasnor’s (1986) model of the SPS process delineates the 40 interplay of component processes involved in children’s performance social dilemmas. In addition, the SPS literature describes a long history of assessing children’s performance in social dilemmas. Hence, an examination of the assessment techniques presented in the SPS literature provides the technology needed to improve CSA prevention assessment methods. The Validity of Socia ob em Solv' sse s c ugs in Child SexunliAbuse Preventioanesenncn Introduction The goal of assessment in both SP8 and CSA prevention research is the same--to predict what children actually do in real-life situations from their performance on analogue tasks. Two central issues in CSA prevention assessment leave the validity of CSA prevention research techniques in doubt. First, the ethical problems associated with assessment of children’s self-protective skills in simulated situations (much less in actual sexually abusive encounters!) have made it virtually impossible to compare children’s descriptions of what they would do with their performance under more ecologically valid circumstances. Indeed, these ethical constraints disallow comparison of test performance to any type of criterion performance. Second, CSA assessment instruments lack an explicit rationale for the method(s) employed to elicit children’s social responses and for the scoring systems used to categorize these responses. 41 CSA assessment techniques can benefit from an integration of knowledge gained from the SPS assessment literature. Because SPS research examines children’s performance in common non-sexual social situations, a number of studies have been conducted comparing children’s in xign performance with their performance on tests utilizing hypothetical situations. Hence, questions of the validity of SPS assessment techniques have been empirically addressed. Also, the SPS literature documents a long history of experimentation with various methods of querying and scoring children’s social performance (see Krasnor & Rubin, 1981, for a review). Current SPS assessment techniques reflect the knowledge gained from a decade of research in this area. Hypothetical-Reflective Assessment Techniques nng In Vivo Eerfornnnce in Connon, Non-Sexunl_nnd Sexunllx Abusive Sitnntions A number of studies suggest that for preschoolers and young children, performance on hypothetical-reflective assessment techniques is a valid predictor of current in 1139 social problem solving performance as rated by teachers (McKim et al., 1982; Shure et al., 1971; Shure et al., 1972; Spivack & Shure, 1974; Spivack et al., 1976) and as observed in staged social problem solving tasks (Kendall & Fischler, 1984). Moreover, this relationship holds even when the effects of IQ and verbal fluency are statistically eliminated (HcKim et al., 1982; Shure et al., 1971; Shure et 42 al., 1973, cited in Spivack & Shure, 1974). Only a few studies with this age group have not found significant relationships between performance on hypothetical-reflective measures and teacher-rated social adjustment (Asher & Renshaw, 1981; Rickel & Burgio, 1982; Winer et al., 1982). Krasnor and Rubin (1981) have suggested that performance on hypothetical—reflective assessment measures may predict consciously-thought-out behavior better than it does automatic, unthinking, common, social behavior. Every- day social behavior may be enacted in a relatively unthinking manner (Langer, 1978) and may proceed according to well-learned social scripts (Schank & Abelson, 1977). A social script is a mental blueprint for the sequencing of events and behaviors in common situations (Schank & Abelson, 1977). If a child possesses a social script for a situation, then the child’s behavior will unfold automatically (Langer, 1978; Schank & Abelson, 1977). Yet as Cooney and Selman’s (1978) nomenclature suggests, "hypothetical-reflective" assessment techniques primarily measure prehensive or reflective responses. In contrast to scripted, unthinking social interactions, some social situations require more reflective cognitive processing. Situations that are nnggl or have inpnztnnt nonsegnenceg for the actor may require more reflective cognition (Flavell, 1981). Therefore, social performance on hypothetical-reflective measures may indeed reflect actual social strategy deployment in novel situations or 43 situations that have important consequences for the actor. Since a potentially sexually abusive encounter is both a novel situation and one that has important consequences for the child, children’s performance on hypothetical- reflective tests may be highly reflective of their in 1119 performance in an actual sexually abusive episode. In Vivo Perfornnncennnd the Scoring of Hypotneticnl- Reflective Techniques Dodge (1986) argues that social performance is not determined by a single cognitive factor (e.g., encoding or strategy repertoire), but by multiple cognitive processes that work together. Therefore, assessment of multiple aspects of the social problem solving process will increase the accuracy of our predictions regarding in yixn behavior (Richard & Dodge, 1982). If hypothetical-reflective techniques are utilized to assess multiple aspects of a child’s social strategizing, better correlations with in x139 performance may result. As Krasnor and Rubin (1981) have argued, hypothetical- reflective assessment techniques are capable of measuring many dimensions of social performance such as strategy repertoire, strategy effectiveness, responsiveness to social context, strategy persistence, and strategy flexibility. By assessing more than one component of the SPS process, Krasnor and Rubin obtain a more complete understanding of SPS performance. Since social performance in sexually abusive, as well 44 as non-sexual, situations is comprised of multiple components, testing a single component of the SPS process cannot adequately predict in 2139 performance in a sexually exploitative encounter. Hence, the adoption of SPS assessment techniques and scoring in CSA prevention assessment may increase the validity of CSA assessment. Bntionnlenfor theAPresent Study This study attempted to redress both the theoretical deficits and the assessment problems CSA prevention research currently faces by focusing on three specific shortcomings of extant child sexual abuse prevention research. First, current research does not explicate a clear and unified conception of how children think about and react to potentially sexually abusive situations. Without a model specifying the interplay of cognitive and behavioral factors that comprise children’s performance in abusive episodes, it is difficult to determine which cognitive processes and/or behavioral skills are critical to the successful deployment of self-protective behavior in an abusive encounter. In addition, the absence of such models hampers efforts to understand the function(s) of different CSA prevention educational components in enhancing children’s self- protective abilities. Second, CSA prevention program evaluation research has been plagued by criticisms of inadequate or inappropriate evaluation techniques (e.g., Conte, 1987; Leventhal, 1987; 45 Reppucci & Haugaard, 1989). Resolution of these criticisms may hinge partly upon assessing more of the cognitive and behavioral dimensions that comprise successful performance in sexually abusive situations. Assessment instruments in child sexual abuse prevention commonly measure only knowledge of relevant factual information and self- protective skills. However, a more accurate picture of the children’s functioning in sexually abusive encounters may be gained from measurement of additional variables that have been found to relate to competent problem solving performance in other social dilemmas. Research suggests that variables such as the encoding of social cues, quantity and type of strategies generated, their planfulness and effectiveness, and the ability to develop new strategies in the face of initial strategy failure are critical to problem solving success. Third, current child sexual abuse prevention research has not empirically determined which situational and subject variables influence successful deployment of self-protective skills in sexually abusive encounters. SPS research on children’s behavior in common non-sexual social dilemmas suggests that their problem solving is affected by a number of situational and subject variables. Because child sexual victimization can occur under a multitude of circumstances (Finkelhor, 1979), research needs to determine empirically which of these factors influence children’s perceptions and behavior. 46 The lack of theory defining the component processes involved in children’s strategizing in potentially abusive encounters, the lack of valid assessment techniques for measuring the multiple aspects of children’s performance in these encounters, and, lastly, the lack of information regarding which situational and subject variables influence deployment of self-protective skills has left the effectiveness of child sexual abuse prevention programs in question. This study applied a model of social problem solving to children’s performance in potentially sexually abusive encounters. Because a sexually abusive encounter is an interpersonal encounter, the use of a SPS model that describes how children think about and behave in social dilemmas is particularly appropriate. This model delineates multiple aspects of the social problem solving process and will thus be able to provide a more detailed view of children’s performance in sexually abusive situations. The application of a SPS model to the self-protection process also suggests the need for new assessment techniques. The use of sophisticated SPS assessment strategies that measure multiple variables may help answer some of the long-standing criticisms regarding the accuracy of the findings in child sexual abuse prevention research. An integration of SP8 and CSA prevention research also suggests that two situational factors--the age of the partner and the nature of the request--and one subject 47 variable-—the gender of the child--may influence children’s performance in sexually abusive dilemmas. Thus, this study will compare boys’ and girls’ strategizing in problem situations that vary systematically along these two situational dimensions. Examination of children’s SPS in non-sexual and sexual social dilemmas will enable researchers to better understand whether the situational factors that affect children’s SPS in common social dilemmas can be generalized to children’s problem solving in uncommon sexually abusive encounters. Study Hypotheses The SPS model previously presented was used as a heuristic device to guide the investigation of children’s performance in sexually abusive and non-sexual situations. This study was primarily concerned with how two situational variables--the age of the partner (child vs. adult) and the nature of the request (non-sexual vs. sexual)--and one subject variable-~the gender of the child (boys vs. girls)-- influence children’s SPS in unsafe situations. The effect of these three independent variables was examined on the following six aspects of SPS performance: a) encoding of social cues, b) quantity of solutions, c) quality (content) of solutions, c) strategy planfulness, d) strategy effectiveness, and e) problem solving flexibility in the face of initial strategy failure. 48 Encoding of Social Cuen I). II). It was expected that, regardless of the age of the partner (i.e., child vs. an adult), children would more accurately encode the non-sexual than the sexually abusive situations. It was predicted that, regardless of the nature of the request (i.e., non-sexual vs. sexually abusive), children would more accurately encode the two situations involving a child partner as compared the two situations involving an adult partner. To test these hypotheses, pretest responses to question #3 of the analyzed ( Eneschool Altennative SolutiongnTest (PAST) were see Appendix A for a copy of the PAST). For details on the scoring of the accuracy of encoding see pp. 4-5 in the P eschoo A e ative Solution§MI§§LUIEASTl Scoring Manual (in Appendix B). u ntit III). IV). '0 3 Children were expected to offer a greater number of solutions for the non-sexual as compared to the sexually abusive situations. In addition, children were expected to suggest more strategies for solving the social dilemmas involving a child partner as compared to the social dilemmas involving an adult partner. Quantity of solutions were calculated by summing all 49 the relevant qualitatively different responses to questions #4, #5, and #6 on the PAST (for a copy of the PAST see Appendix A). Quality (Content) of Solutiong V). It was hypothesized that children would offer more assertive, help-seeking, and non-confrontational solutions than any other types of solutions, regardless of nature of the request or the age of the social partner. VI). It was also predicted that children would offer more help-seeking and non-confrontational solutions to the two sexually abusive situations than to the two nonsexual situations. VII). It was predicted that more children would offer more aggressive solutions in the situations involving a child partner as compared to an adult partner. VIII). Boys were expected to offer a greater number of aggressive solutions than were girls, regardless of the nature of the request or age of the partner. The quality of the solutions given in response to questions #4, #5, and #6 on the PAST were scored according to nine content categories (for a copy of the PAST see Appendix A; for an explanation of procedures used to score solution content see pp. 6-18 of the £A§I_§nn;ing_flnnngl in Appendix B). 50 Strate P a f ness XII). It was expected that planfulness scores would be higher in the two non-sexual social dilemmas as compared to the sexually abusive dilemmas. XIII). It was hypothesized that planfulness scores would be higher in the situations involving a child partner as compared to the situations involving an adult. XIV). An interaction between the nature of the request and the age of the social partner was predicted. It was expected that planfulness scores would be highest in the non-sexual situation involving a child partner. The mean planfulness score for each subject for each story was calculated by summing the planfulness score for each response to questions #4, #5, and #6 for each story on the PAST and dividing by the number of responses per story (see Appendix A for a copy of the PAST; see pp. 32-36 in the EAST Scoring Manual, in Appendix B, for an explanation of the procedures used to score planfulness). Stnategy Effectivenesg IX). It was expected that effectiveness scores would be higher for the non-sexual social dilemmas as compared to the two sexually abusive dilemmas. X). It was hypothesized that effectiveness scores would be higher for the situations involving a child partner compared to the situations involving XI). 51 an adult partner. An interaction between the nature of the request and the age of the partner was predicted. It was expected that effectiveness scores would be highest in the situation involving a non-sexual request from a child partner. The mean effectiveness score for each subject for each story was calculated by summing the effectiveness score for each response to questions #4, #5, and #6 for each story on the PAST and dividing by the number of responses per story (see Appendix A for a copy of the PAST; see pp 22-31 of the PAST Scoring Ma ua in Appendix B for an explanation of procedures used to score effectiveness). Problen Solving Flexibility XV). It was predicted that flexibility scores would be higher in the non—sexual situations as compared to the sexually abusive situations. XVI). It was hypothesized that flexibility scores would be higher in the situations involving a child partner as compared to the situations involving an adult partner. XVII). An interaction between the nature of the request and the age of the partner was predicted. It was expected that flexibility scores would be highest in the non-sexual situation involving a child partner. Responses to question #5 on the PAST was assessed for 52 strategy flexibility (see a copy of the PAST in Appendix A; see pp. 37-39 of the PAST Scoring Manual for an explanation of procedures used to score response flexibility). METHOD Participants Participants were enrolled in one of six preschools located primarily in the Lansing area. One hundred thirty preschoolers received parental permission to participate. Because the data for this study was part of a larger study that utilized a Solomon four-group design, only half of these participants were randomly selected to receive a pretest. Complete prepoint data was available for 62 of the 64 pretested participants (32 boys, 30 girls). One participant withdrew from the study before completing pretesting, the other preschooler was ill during the testing period. The mean age of participants was 4.4 years (range 3.3 - 6.2 years). The mean age for boys was 4.5 years and for girls was 4.2 years. The difference between boys’ and girls’ ages was not statistically significant [2(1, 59) = 2.05]. Ninety percent of the subjects were Caucasian, 8% were African-American, and 2% were Asian-American. Mea§u1:e§ Data analyzed in this study was obtained using the Preschool Alternative Solutions Test (PAST). The PAST was administered as part of a larger battery of CSA prevention 53 54 evaluation measures (see Appendix A for a copy of this measure). The PAST is an individually-administered interview developed for this study. It is designed to assess both quantitative and qualitative features of social problem solving performance in both non-sexual and sexually abusive situations. It consists of four vignettes in a fixed order. In each vignette, the age of the perpetrator and the nature of the request are systematically varied. All characters in the vignettes are the same sex as the subject. A picture depicting the central action in the story accompanies each vignette. The first vignette describes a situation in which one preschooler asks a peer to cross a street that the target peer is not allowed to cross in order to purchase candy at a store. The second vignette describes an encounter in which a child asks a target peer to "touch private parts." In the third vignette, an adult asks a child to cross the street the target child is not allowed to cross. And in the fourth vignette, an adult asks a child to "touch private parts." After each vignette, children are reminded that the target child does not want to participate in the suggested activity. Children are then asked a series of questions to determine: a) whether they encoded the story accurately, b) what they think the target character could initially do or say if he/she didn’t want to participate, c) what they think 55 the target character could do if the previously suggested strategy fails, and d) all the different strategies the target character could employ to resist participating in the suggested activity. Children’s responses were initially coded into one of 29 content sub-codes (see the EAST Scoring Mannal, in Appendix B). These content sub—codes were grouped into nine major scoring categories. Initial inspection revealed that children provided few or no responses for some of the 29 sub-codes (see Table 4). For example, no "bribe" solutions and only four "manipulate own affect" responses were offered. Based upon these frequencies and analysis of solution content, sub-codes were re-grouped to form ten scoring categories. Of the initial nine major groups of content sub-codes, five remained unchanged, one (help- seeking) received an additional sub-code, two were collapsed (passive resistance and manipulate affect), and two new major categories were created from previous sub-codes (invoke a rule and restructure situation to meet both parties’ needs) (see Table 5). Thus, interrater reliability, data analyses, and discussion of solution content in this study are based upon the 10 categories that appear in Table 5. Interrater reliability was calculated separately for encoding, solution content, and planfulness. Sixteen percent of the protocols were scored for interrater reliability. Two scorers independently coded randomly 56 selected PAST protocols. Scorer reliability was calculated using the formula: scoring agreements / (scoring agreements + scoring disagreements). Interrater reliability was 99% for encoding, 87% for solution content (for 10 solution categories), and 89% for solution planfulness. Previous studies (Caplan et al., 1986) have obtained an interrater reliability (Kappa coefficient) from .91 - .94 for solution content and of .81 for solution planfulness. Procedure In the Spring of 1989, an information letter was sent to all Lansing area preschools inviting them to participate in a child sexual abuse prevention project to begin the following Fall. In the Fall of 1989, follow-up telephone contact was made with each preschool to determine their level of interest. A number of preschools asked to learn more about the project. A member of the research team met with each of these centers to provide more information about the sexual abuse prevention curriculum and the evaluation. After a preschool agreed to participate, letters were sent to the parents. The letter explained the interview and curriculum, invited the parents to attend a parent meeting, and included a consent form (see Appendix C). Only children with signed consent from their parent(s) were allowed to participate in the study. Child assent was obtained verbally prior to the interview. The design for this study was a Solomon four-group. 57 Children in each center were blocked for gender and then randomly assigned to either a delayed-treatment control or an experimental condition. Children from each of these two conditions were then randomly assigned to either a pretest- posttest or a posttest-only group. This study utilized only the pretest data. Hence, only procedures pertaining to the collection of pretest data are explained below. PAST interviews were conducted by trained undergraduates who received course credit for their work. Interviewers participated in at least ten hours of group training and two hours of individual training before interviewing preschoolers. In order to become more familiar to the children, all interviewers played with the children in the preschool for at least two hours (over a two-week period) prior to interviewing. All evaluation measures, including the PAST, were individually administered and interviewers were blind to children’s experimental condition and study hypotheses. Interviews were conducted in a quiet area of the preschool, away from other children, in 20- minute segments over two or three days. Each PAST vignette and question was read aloud to the child. Children’s responses were written verbatim on the protocol. Sconing Protocols were scored for encoding of the situation, quantity, quality, and flexibility of solutions. All scoring procedures for these variables are explained in 58 detail in the WWW Snnging_flnnnnl (see Appendix B for a copy), however, the criteria for scoring each variable is briefly summarized below. Ennnging. Responses to questions 01, #2, and 03 were scored for how accurately the child encoded the sex and age of the social partner in the vignette (question 01, 02) and the nature of the request, i.e., non-sexual vs. sexually abusive (question #3). Responses to questions 01, 02, and #3 were scored dichotomously, (accurate or inaccurate), (see the 2551 Scoring Mgnnnl for details). An accurate answer to question #3 was operationally defined as one in which the child indicated that the persons involved wanted to "touch private parts," in the case of the sexually abusive vignettes, or that the persons "wanted to cross the street and/or go get some candy," in the non-sexual vignettes. Questions 84, 85, and #6 asked children to generate strategies to solve the social dilemma presented in the vignette. Responses to these questions were evaluated along four different dimensions: a) quantity of solutions, i.e., strategy repertoire, b) content of solutions, c) planfulness of solutions and d) effectiveness of solutions. Both the criteria for the content categories and the criteria for planfulness are based on SPS scoring criteria developed by Caplan et al. (1986). A more detailed explanation of scoring criteria can be found in the £A§I_§nnning_finnnnl (see Appendix B). 59 t‘t o ut' . The quantity of solutions was defined as the total number of qualitatively different strategies a child generates to questions #4, #6, and #6 on the PAST. Snlntinn_ggntgnt. The content of each solution was determined by classifying each solution as falling into one or more of the following nine content categories: nnnlgngg (e.g., "he could blast off to outer space,” "he should protect his private parts"), nnngrtiyg (e.g., "she could say ‘no,’" "she could say 'I don’t have to’"), hgln;§ggking (e.g., ”she could go ask her mom if she could cross the street," "she could tell her mother that her friend wanted to touch private parts," "call the police"), nnn; gonfzontationnl (e.g., ”she could leave," "run, run, run, superfast"), nggnessive (e.g., "he could hit him," "he could call him names"), {inagle (e.g., "she could say, 'let’s play checkers instead,’" "she could say, ’I’ll give you $5 to stop asking me to cross the street’”), e- - e o nnnnnnin1;g_nnnnnng (e.g., ”My Mom said not to cross that street," "I’m not allowed to do that"), Inkg_g_ggnl;; restructure_situati2n_t2_meet_bgth_2artiesl_needs (9.8.. ”she should say ’I’ve got some candy at my house, let’s go there and eat it instead,’" "he should say 'touch your own private parts. 1’. coins to watch TV’"). nsssixs_rssistangs£ Inninnlnte_n££gnt (e.g., "she could cry," "he could ignore him and keep watching T.V.”), nnnplinngg (e.g., "he could give-in and touch private parts," "he could take off his 60 pants"). So t'o 've . The effectiveness of each solution was rated on a 0 - 3 scale as very effective, effective, marginally effective, or ineffective. A very effective solution is one that maximizes the child’s short- and long-term safety and minimizes other negative consequences to the child. An ineffective solution does not achieve any of those goals. Since the effectiveness of a solution is situation-specific, the content codes (listed above) receive different effectiveness scores depending primarily upon the situation to which they are given (see the EAST Scozing Manual in Appendix B for details). However, a solution could be given an effectiveness score greater than zero only if it was judged to be a realistic "do-able" solution that the child could, in fact, carry out. This judgement was made for each solution individually. Effectiveness ratings were derived from the results of four adult judges’ ratings of the effectiveness of each of sub- content code for each situation. finlnninn_£lgn1nlng§§. The planfulness of each solution was rated on a scale of 0 -2 as very planful, planful, or not planful. Solution planfulness is defined as the extent to which the solution reflects a formal plan for executing the proposed strategy. The following are considered indicators of planfulness: a) the explicit consideration of possible obstacles or contingent events that might change the child’s deployment strategy, b) the articulation of 61 intermediate steps or prerequisite conditions/behaviors needed to carry-out a strategy, and/or c) the use of situational factors (including time) to implement a plan. Solutions are scored on a 0 -2 scale as very planful, planful, or not planful. For example, "she should go wait in another room until her parents come home, and then she should tell” constitutes a very planful response since clearly states the intermediate steps which would be taken and exhibits a consideration of time (for a more detailed explanation of planfulness see the EAST Scoping finnnnl in Appendix B). o e Solv' e ' ' . In addition, children’s responses to question #5 were rated for strategy flexibility on a 0 - 2 scale as flexible persistence, inflexible persistence, and diffidence. Procedures for scoring flexibility in the face of initial strategy failure closely follow those developed by Krasnor and Rubin (1983). The flexibility rating for each subject is determined by comparing his/her response to question #4 on the PAST to his/her response to question #5. Question #4 asks children for their initial solution to the social dilemma presented to them in the vignette. Question #5 informs them that the initial solution did not work and asks them what they would do next. A flexible and persistent solution is one which offers a qualitatively different solution in response to strategy failure. An inflexible persistent solution is one that repeats the 62 previous solution. A diffident solution is one in which the child suggest capitulating to the request of the perpetrator, or indicates he/she cannot generate a solution (e.g., says ”I don’t know what she would do") or provides an unscorable response (see the £A§I_§§9;ing_flnnngl in appendix B for details). RESULTS This study investigated how two situational variables (the age of the partner and the nature of the request) and one subject variable (gender of the child) influenced six different aspects of the SPS process (encoding, quantity of solutions, quality of solutions, solution effectiveness, solution planfulness, and solution flexibility). These dependent variables were analyzed in a series of 2 X 2 X 2 (Age of Partner, Nature of Request, Gender of Subject) repeated-measures, fixed-model ANOVAs. The age of the partner and the nature of the request were considered within subjects factors. The gender of the subject was treated as a between subjects factor. All analyses follow this format except where specified. Encod'n o Soc'a It was hypothesized that children would more accurately encode the non-sexual compared to the sexually abusive situations (hypothesis I) and more accurately encode the stories involving a child partner as compared the stories involving an adult partner (hypothesis II). Results indicated that children’s ability to detect social cues was affected by the age of the partner [2(1, 59) 8 6.26, p<.05], but not by the nature of the request [2(1, 59) = .65], or the gender of the subject [3(1, 59) = .66]. No significant 63 64 interaction effects were obtained. Contrary to hypothesis 11, children achieved higher encoding scores in situations involving an adult social partner rather than situations involving a child partner. See Tables 6 and 7. ua t o S 'o It was hypothesized that children would offer more solutions to the non-sexual as compared to the sexually abusive situations (hypothesis III) and that they would suggest more solutions to the social dilemmas involving a child partner rather than those involving an adult (hypothesis IV). Results indicated that the nature of the request did influence the quantity of solutions generated [£(1, 60) = 11.16, p<.001]. As predicted, children offered significantly more solutions to non-sexual situations (fl = 2.29) than they did to sexually abusive encounters (M = 1.90). However, results indicated that neither the age of the partner [£(1, 60) = .03] nor the gender of the subject [E(1, 60) = .30] affected the number of solutions children suggested. See Tables 6 and 8. WW It was predicted that children would produce more assertive, help-seeking, and non-confrontational solutions than any other type of solution (hypothesis V). Because the relationship between solution types had been stated a priori, a planned comparison was conducted to test this hypothesis. The results indicated that the mean for assertive, help-seeking, and non-confrontational solutions 65 ngnningg (5 8 .89) was significantly greater than the mean for all other solution types combined (5 a .32) [2(1, 61) = 69.80, p<.0001]. Follow-up 2 X 10 repeated-measures ANOVA (Gender of Subject X Solution Type) and post-hoe Scheffe tests were computed. The gender of the subject did not significantly influence children’s strategy choice [2(1, 60) = .20], nor did the interaction between gender of subject and strategy type reach statistical significance [£(9, 540) 8 1.44]. However, a significant overall effect for solution type was found [E(9, 540) = 31.81, p < .0001]. Scheffe post-hoc analyses comparing individual types of strategies to all other strategies combined revealed that children offer significantly more assertive (M = 2.77), non-confrontational (M = 1.69), and nonsense (M = 1.50) solutions than all other solution types (5 = .81; H = .94; n = .96, respectively). Additionally, when assertive, non-confrontational, and help- seeking solutions were compared to each other, significant differences were observed between all three means (2 < .001 for all comparisons). See Tables 9 - 10. Hypotheses VI through VII investigated the impact of subject and situational factors on children’s strategy selection, particularly, on their use of help-seeking, non- confrontational and aggressive solutions. It was hypothesized that children would offer more help-seeking and non-confrontational solutions to the sexually abusive than to the non-sexual situations (hypothesis VI). The nature of 66 the request significantly affected the number of non- confrontational [2(1, 60) = 6.19, n<.05] and help-seeking [E (1, 60) = 8.20, p<.05] solutions suggested. In sexually abusive situations, children offered 39;; non- confrontational (M = .48) but {£331 help-seeking (M 8 .21) solutions than in the non-sexual social situations (M = .37, M =.32, respectively). Thus, hypothesis VI was partly supported. A main effect for gender was also found for non- confrontational solutions [2(1, 60) = 5.72, p<.05]. Overall, boys offered significantly fewer non- confrontational responses (M = .32) than did girls (M = .53). No significant main effects for the age of the partner were found for help-seeking or non-confrontational solutions. See Tables 12 - 13. The main effects for non-confrontational and help- seeking solutions are further explained by a number of significant interactions. For non-confrontational solutions the Nature of the Request X Gender of the Subject interaction reached significance [2(1, 60) = 8.10, p<.05]. Boys suggested about the same number of non-confrontational solutions to sexually abusive and non-sexual situations (M = .31, M 8 .33, respectively). Girls, however, offered more of these strategies to sexually abusive than to non-sexual dilemmas (M = .65, M = .43, respectively). See Figure 2 and Tables 11 - 12. Two interactions approached significance for help- seeking solutions. In the Nature of the Request X Gender of 67 the Subject interaction [E(1, 60) 8 3.74, p<.10], girls chose help-seeking solutions with nearly equal frequency in the sexually abusive and non—sexual situations (M 8 .27, M 8 .30, respectively). Boys, however, offered fewer help- seeking solutions in the sexually abusive (M 8 .16) than in the non-sexual social dilemmas (M 8 .33) [E(1, 60) 8 3.74, n<.10]. See Figure 3 and Tables 11 and 13. The Age of Social Partner X Social Request interaction also approached significance [2(1, 60) 8 3.03, p<.10]. In situations involving an adult social partner, children suggested help-seeking about as often in sexually abusive and non-sexual situations (M 8 .23, M 8 .26, respectively). However, when the social partner was a child, children chose help-seeking solutions less frequently in sexually abusive (M = .19) than in non-sexual social dilemmas (M 8 .37). See Figure 4 and Tables 11 and 13. In addition, exploratory ANOVAs were performed on the seven remaining solution categories. No main effects for the age of the partner or the gender of the subject were obtained for these solution categories, nor did any interactions reach significance. See Tables 16 - 21. However, the nature of the request did significantly affect the number of assertive [£(1, 60) 8 5.21, p<.05] and rule- based [2(1, 60) 8 21.61, p<.001] solutions children generated. Children offered fewer assertive (M 8 .65) and rule-based (M 8 .09) solutions in sexually abusive than in non-sexual social dilemmas (M 8 .74, M 8 .45, respectively). 68 See Tables 11, 14, and 15. It was predicted that children would offer a greater number of aggressive solutions to social dilemmas involving a child than an adult partner (hypothesis VII), and that boys would offer a greater number of aggressive strategies than girls (hypothesis VIII). No significant effects for any independent variable were found for aggressive solutions. See Table 17. St ct've es Hypotheses IX through XI predicted that children would generate more effective solutions to non—sexual situations (hypothesis IX) and situations involving a child partner (hypothesis X). It was also predicted that children would suggest the most effective solutions to non-sexual situations involving a child social partner (hypothesis XI). No significant main effects for the age of the partner [E(1, 54) 8 1.46] or the nature of the request [E(1, 54) 8 .04] were found. However, the main effect for the gender of the subject approached significance [2(1, 54) 8 2.73, p<.10]. Girls produced more effective strategies than did boys (M 8 2.06, M 8 1.83, respectively). No interactions between these factors reached significance. See Tables 6 and 22. WW Hypotheses XII through XIV predicted that children would obtain higher planfulness scores in non-sexual situations (hypothesis XII) and situations involving a child partner (hypothesis XIII). It was also predicted that 69 children would generate the most planful solutions in non- sexual situations involving a child social partner (hypothesis XIV). No significant main effects for the age of the partner [E(1, 54) 8 2.14] or the nature of the request [E(1, 54) 8 1.26] were found. However, a significant main effect for the gender of the subject was obtained [£(1, 54) 8 6.02, p<.05]. Again, girls produced more planful solutions than did boys (M 8 .29, M 8 .15, respectively). None of the interactions between these factors reached significance. See Tables 6 and 23. Strategy Flexibility Hypotheses XV - XVII predicted that after their first strategy failed, children would produce more flexible solutions in situations involving non-sexual requests (hypothesis XV) or a child partner (hypothesis XVI). It was also predicted that children would attain the highest flexibility rating in the non-sexual situation involving a child partner (hypothesis XVII). No significant main effects for the age of the partner, the nature of the request, or the gender of the subject were obtained. In addition, none of the interactions between factors reached significance. See Table 24. DISCUSSION The results of this study will be discussed in terms of their implications for SP8 theory and research as well as CSA prevention efforts. In each of the following two sections, this study’s findings will be linked to the relevant SPS variables-~encoding, solution quantity, quality, planfulness, effectiveness, and flexibility. Study Results and Social Problem Solving Theory and Research Current SPS research and theory focuses on children’s social performance in commonly occurring dilemmas. By examining SPS in uncommon (i.e., sexually abusive) situations, this study extends our current understanding of children’s social performance. Results regarding five SPS variables have particular relevance to the SPS literature: solution quantity, quality, planfulness, effectiveness and flexibility. We: Contrary to findings reported in the SPS literature (e.g., Dodge et al. 1984), the nature of the request (non- sexual vs. sexually abusive) did not affect the accuracy of preschooler’s encoding in this study. However, the encoding skills measured in the current study differed from previous SPS research (Dodge et al, 1984; Flapan, 1968). Children 70 71 in Dodge et al.’s (1984) and Flapan’s (1968) investigations were asked to determine the intent of the provocateur (e.g., prosocial, hostile). However, this study investigated children’s encoding of more salient aspects of the social situation (e.g. gender and age of social partners; non- sexual or sexually abusive nature of the situation). Perhaps situational factors influence the accurate detection of more subtle social cues while not affecting the apprehension of more obvious ones. The fact that encoding scores were uniformly high for all stories (M 8 2.79; range 1 - 3) indicates that these social cues were salient for most children. See Table 6. t' t 0 Although children generated significantly more solutions to non-sexual than to sexually abusive situations (M 8 2.31; M 8 2.00, respectively), a comparison of the mean number of solutions generated in this study to those in other SPS studies suggests that this gtatigticnllx gignifinnng difference may be of little actual importance. The number of solutions generated to both sexually abusive and non-sexual situations in this study are well within the range of mean scores (2.00 - 2.40) generated in other studies of young children that utilize similar hypothetical- reflective assessment techniques (Getz, Goldman, & Corsini, 1984; Rickel & Burgio, 1982; Rubin a Krasnor, 1983; Spivack & Shure, 1982). Given that the quantity of solutions generated in this study was similar to the quantity 72 generated in other studies, the next section explores how the quality (or content) of children’s solutions offered in this study compares to those reported in other studies. u t Con So t S te Se ect'o ' U sa e tuatio Comparison of the types of solutions children offered in this study to the types of solutions children offered in other studies reveals both similarities and differences in children’s problem solving. The frequency of assertive, non-confrontational, nonsense, and aggressive solutions in this study is comparable to the frequencies reported in extant SPS research; the frequency with which help-seeking solutions were generated differs. In this study, children suggested assertive responses (e.g., "He should say 'No, I won’t!’”) more frequently than all other problem solving strategies. This finding is consistent with both observational (Krasnor, 1982) and hypothetical-reflective (Getz et al., 1984; Rubin & Krasnor, 1983) assessment of preschoolers’ strategy choice in common SPS situations. It is not surprising that a frequent problem solving strategy in common situations is also the strategy of choice in social dilemmas requiring resistance. In this study, non-confrontational solutions were the second most frequently suggested solution. They were offered significantly less frequently than assertive strategies and more frequently than all other types of strategies. Because SPS research usually examines common 73 situations, such as object acquisition, non-confrontational strategies are not appropriate. Only one study could be found in the extant literature that examined non- confrontational responses (Strayer & Strayer, 1976). In this study non-confrontational strategies were the second most frequently employed response to peer aggression (e.g., hit, push). Results of this observational study of preschooler’s behavior support the validity of this study’s findings regarding non-confrontational solutions. The mean number of nonsense solutions (e.g., "He should sting him with a ray gun.") in the current study was uniformly high across situations. SPS investigators rarely include information regarding the number of nonsense or irrelevant solutions. However, in two studies of third- graders’ SPS skills, the number of irrelevant responses obtained is comparable to the number obtained in this study (McKim et al., 1982; Weissberg et al., 1981). The frequency with which nonsense solutions were offered suggests that children may often generate unrealistic or illogical, and therefore ineffective, solutions to social dilemmas. The high number of unrealistic solutions could also be an artifact of the interviewing techniques used (similar techniques were used in the McKim et al., 1982 and Weissberg et al., 1981 studies). After children provided a solution, they were asked to "think of something else" the child in the story could try. This technique may have led some children to produce unrealistic solutions in a effort to 74 comply with task demands. Contrary to SPS research assessing preschoolers in common situations (Rubin & Krasnor, 1983), in this study, boys and girls did not differ in the number of aggressive strategies they suggested. Nor did the age of the partner or nature of the request affect the number of aggressive solutions offered. The number of aggressive strategies offered in this study is similar to the number suggested in Rubin and Xrasnor’s (1983) study. However, in the Rubin and Krasnor study, aggressive responses were the second most frequently offered response, while in this study, children suggested almost all solution types more frequently than aggressive strategies. See Table 9. Existing research on preschooler’s SPS has found that children rarely seek help from others to solve common social dilemmas (Getz et al., 1984; Rubin & Krasnor, 1983; Strayer & Strayer, 1976). Contrary to this finding, help-seeking responses (e.g., "She should tell her Mom.") were the fourth most frequently suggested response in this study. The increased frequency with which help-seeking solutions were suggested could be due to the greater dangerousness of the "unsafe situations" utilized in this study (i.e., crossing a busy street the child is not allowed to cross, unwanted sexual touching). In these uncommon and threatening situations, the child may view the judgment and protection of an adult as more useful that in object acquisition situations. 75 Clearly the above comparisons suggest that both similarities and differences exist between children’s problem solving in common and in more dangerous, uncommon situations. Existing SPS research examines children’s performance only in common peer situations. Future research needs to complement the existing literature on SPS by further investigating children’s SPS in more threatening, uncommon social dilemmas. Systematic comparison within the same study of children’s SPS in common (e.g., object acquisition) and uncommon (e.g., sexually abusive) situations would allow for more confident conclusions regarding differences in children’ SPS across situations. Only systematic investigation of children’s SPS in a wide variety of circumstances will yield an accurate understanding of their social skill. The Inflngnce of Situational Engtnng 9n Stnntggz Selegtion Contrary to research on SPS in common problem situations (Rubin a Xrasnor, 1983; Holmberg, 1980), the age of the partner did not affect preschoolers’ strategy choice in this study. One explanation for such a discrepancy can be found in Rubin and Krasnor’s (1983) explanation of the mutual influence of situational factors upon one another and upon and children’s strategy selection. They propose that children’s sensitivity to the characteristics of the partner depends upon other situational factors (including the nature of the request) that define the problem situation. That is, children’s strategy choice, rather than being statically 76 controlled by a predictable set of factors, is determined by the interplay of situational elements that increase and recede in importance depending upon the nature of the problem. Thus, it may be that the age of the partner influences children’s strategy choice in common problems such as object acquisition but is less important in resistance situations such as those assessed in this study. The age of partner X nature of request interaction for help-seeking solutions provides some support for this notion. In this interaction, children chose help-seeking with about equal frequency in sexually abusive situations, regardless of the age of the partner. However, in non- sexual situations, children chose help-seeking much more frequently when with children than when with adults. In other words, it seems that the age of the social partner did not influence strategy choice in sexually abusive situations, but preschoolers did consider this characteristic of their social partner in non-sexual situations. These findings, as well as Rubin and Xrasnor’s suggestion, again underscore the importance of further investigating children’s SPS in a wide variety of common and uncommon dilemmas. Researchers cannot generalize about the impact of particular person characteristics on children’s behavior until their influence has been systematically investigated in a number of different social dilemmas. Krasnor’s (1982) observational study, recent SPS theory 77 (Dodge, 1986; Rubin & Xrasnor, 1981; 1986), as well as the above cross-study comparisons of children’s strategy preference suggest that the nature of the request influences the strategy chosen. The results of this study further support this suggestion and extend its application to children’s performance in different types of situations. Children offered fewer help-seeking, assertive, and rule- based responses in sexually abusive than in non-sexual social dilemmas. Children’s suggestion of these responses in non-sexual situations indicates they can perform competently in some resistance situations. What accounts for children’s failure to suggest these competent self- protective strategies in sexually abusive encounters? Although SPS research and theory provides evidence tnnt children vary their SPS according to situational factors, this literature does not supply an explain of En: this variation in performance occurs. Script theory furnishes one possible explanation. Script theory argues that individuals possess "mental blue prints" for the usual sequence of actions in common events (Schank & Abelson, 1977). These scripts help people understand and respond to every-day events (Langer, 1978). Familiar contextual cues trigger the enactment of a specific sequence of social behaviors. Both the cues and the behaviors comprise the script for that particular situation (Abelson, 1981). Children acquire scripts through both observation and participation in events (Nelson, 1981). 78 Developmental psychologists have found that children as young as three years of age possess scripts for common social interactions (Nelson, 1978). Preschool-age children are preoccupied with the concrete details of a script and do not easily generalize scripted behavior to new situations (Nelson, 1981). Thus, preschoolers display competent social behavior in some situations and are unable to perform in other similar encounters (Nelson, 1981). For example, when a child encounters a dilemma such as being asked to cross a street he/she is not allowed to cross, situational cues trigger a script for that event which includes a sequence of behavior the child has used to terminate similar incidents. The script enables the child to understand the event and enact resistance strategies. Because a request to engage in sexual touching is uncommon, children do not possess a script to help them understand this event. Moreover, script theory argues that the change in the request may alter the situational cues sufficiently so that the preschoolers’ script for non-sexual situations is not triggered. Therefore, they fail to enact, in sexually abusive encounters, those frequently-performed resistance strategies (e.g., assertive, help-seeking strategies) contained in scripts to common non-sexual events. This study’s finding that children frequently offered rule-based resistance strategies to non-sexual situations but rarely offered them to sexually abusive encounters is 79 consistent with this interpretation of the data. Rule-based responses such as "He should say 'No, we might get hurt, my parents said never cross this street’" reflect a familiarity with the social event. The low frequency of rule-based responses in sexually abusive encounters suggests that it is a novel situation for preschoolers. Hence, scripts utilized in common resistance situations are not triggered by a request to engage in sexual touching. In this study, children offered more non- confrontational solutions to sexually abusive than non- sexual situations. This finding is not inconsistent with the script theory described above. Internal (emotional) as well as external cues can trigger a script (Abelson, 1981). Hence, children may possess a script that dictates that they run away if a situation becomes too anxiety-provoking, threatening, or novel. mm For girls and boys, mean planfulness scores in this study were low (M 8 .15, range 0 - 2). The complex cognitive processes necessary to generate planful strategies (e.g., the ability to consider a complex series of events and to consider the perspective of others) may not be well developed in preschoolers. Shure and Spivack, who have studied planfulness over a wide age range of children, conclude that preschoolers exhibit only rudimentary planfulness, while older children and adults display a wide range of planfulness in their problem solving (Shure, 1982). 80 Fischler and Kendall’s (1988) study supports this conclusion. They report that 6-7 year-olds were significantly less planful than 8-9 or 10-11 year-olds while differences between the 8-9 year-olds and 10-11 year-olds were not significant. The low frequency with which finagle and manipulate affect solutions were suggested in this study (see Table 9) and other studies of preschooler’s SPS (Rubin a Krasnor, 1983; Getz et al., 1984) may also reflect this age group’s lack of skill at social role-taking, reciprocity, and sequential thinking. Consistent with this interpretation, Getz et al. (1984) found one-step, simple strategies (e.g., grabbing a toy) were higher among 3 and 4 year-olds than among 5 year-olds. Five year-olds, on the other hand, suggested more complex strategies such as bargaining. In the current study, preschoolers also most frequently offered simple, one-step solutions (i.e., assertive, non- confrontational, help-seeking). In this study, girls provided significantly more planful solutions than did boys. Fischler and Kendall (1988) report similar findings in 6 to 11 year-olds. On the other hand, Getz et al. (1984) and Shure and Spivack (1982) report no differences in planfulness between male and female preschoolers. The effect of gender on planfulness is difficult to determine since assessment of this cognitive ability has not always been considered (Gesten et al., 1982; Shure et al., 1972; Shure & Spivack 1972) or reported as a 81 separate score (e.g., Weissberg, Gesten, Rapkin, et al., 1981). Although the issue of gender differences in social cognition has not received much attention in the developmental literature (Cohn, 1991), the few studies that have addressed this issue have found that girls possess better role-taking skills, greater empathy, and are more socially appropriate than are boys (for a review see Shantz, 1983). Planfulness is defined as consideration of the steps necessary to implement a solution. It requires the ability to consider other’s emotional states and the appropriate use of social convention (Spivack & Shure, 1982). For example, a solution such as "A friend be mad, so I would say no politely, or say my Mom was calling me, and run home superfast" reflects role-taking, empathy, and sensitivity to social convention. Thus, if these skills are considered components of planfulness, this study’s findings are consistent with the literature on children’s social cognition. The planfulness of a problem solving attempt can determine its success or failure. In both sexual and non- sexual unsafe situations, this skill becomes particularly critical because problem solving failure (i.e., the solution generated does not stop the provocateur’s threatening behavior) may have greater adverse consequences for the child. Yet researchers in SP8 or CSA prevention do not always consider planfulness in assessing preschooler’s 82 social performance; nor do they consider how subject variables (e.g., gender, age) influence the ability to develop planful solutions. Further study of the development of planfulness and more careful consideration of possible age and gender differences in planfulness is essential to an accurate understanding of children’s ability to cope with both every-day and unusual social situations. t te c ’v As was the case with planfulness, boys’ lower effectiveness scores may be interpreted as indicating in a general way that boys may be less capable than girls of generating adequate solutions to unsafe situations (both non-sexual and sexually abusive). Strategy effectiveness has not usually been assessed or reported as a separate score in the CSA prevention or SPS literatures. However, this study’s finding is consistent with the one study in which strategy effectiveness was reported as a separate score (Reid, 1991). In Reid’s study, 3rd-grade girls provided more effective solutions to common non-sexual dilemmas than did 3rd-grade boys. This study’s findings regarding the influence of subject gender on strategy effectiveness suggest that girls may be better problem solvers than boys. Two important questions remain to be answered. First, this study examined SPS performance in only unsafe dilemmas. Are girls more effective problem solvers in other types of social situations? Reid’s (1991) finding that girls achieved 83 significantly higher scores than boys in both peer aggression and friendship initiation situations suggests that this might be the case. Second, the age-range of participants in this study was restricted. Do these putative gender differences in SPS effectiveness continue throughout development, or do they disappear at a certain age? Future research needs to investigate both the generalizability and the developmental course of these putative gender differences in problem solving effectiveness. 5! le El .!.].l The nonsignificant findings for strategy flexibility may also be attributable to the limited cognitive sophistication of preschoolers. Flexibility scores were low (M 8 1.26) indicating that when told their initial strategy failed, preschoolers usually repeated the same strategy again. Rubin and Xrasnor’s (1983) finding that kindergartners achieved significantly higher flexibility scores than preschoolers supports the notion that this component of problem solving may not be well-developed in preschoolers. Preschooler’s lack of flexibility in resistance encounters could place them at greater risk for abuse than older children. Moreover, the lack of research in the SPS literature on this variable makes interpretation difficult. Clearly, more SPS research needs to examine how children’s flexibility in solving social problems changes as they grow older. Equally important, future research needs 84 to explore the impact of situational factors (e.g., common vs. uncommon social dilemmas; safe vs. unsafe request) on children’s problem solving flexibility. Study Results and Child Sexual Abuse Prevention Efforts A number of results from this study are particularly relevant to CSA prevention efforts. First, the adequacy of CSA prevention’s assessment of children’s encoding will be discussed. Next, the implications of this study’s results regarding factors affecting children’s strategy choice will be examined in terms of CSA prevention curriculum development. cod' Soc' e Integration of this study’s findings with SPS and CSA prevention research and with CSA victims’ reports suggests that encoding can be accurately assessed only if CSA prevention research employs more sophisticated assessment techniques that present more ecologically valid stimuli. This study’s finding that children accurately encoded social cues in potentially sexually abusive encounters is consistent with results of other CSA prevention research that has assessed similar social cues (e.g., nature of the request) (Blumberg et al., 1988; Miltenberger a Thiesse- Duffy, 1988; Swan et al., 1985; Wurtele, 1990; Wurtele et al., 1987). However, these findings are inconsistent with CSA victims’ retrospective reports. Victims of child sexual abuse often indicate that they misconstrued sexual touch as 85 common, appropriate touch (Finkelhor, 1979), and ignored feelings of emotional discomfort with the activity (Finkelhor, 1986). This discrepancy suggests that more sophisticated presentation of stimuli and more ecologically valid scenarios may be needed in CSA assessment measures. This study employed a method of stimuli presentation similar to the methods used in published CSA prevention research: children are presented with simple verbal descriptions of the social situation accompanied by a picture. These descriptions are capable of presenting only the most salient social cues and may omit more subtle cues that confuse children in actual sexually exploitative encounters. Perhaps CSA prevention research could utilize filmed interactions such as those used to assess encoding in SPS research (e.g., Dodge et al., 1984; Flapan, 1968). Filmed interactions may be capable of portraying more subtle social cues such as voice intonation and multiple facial expressions that indicate the complex array of feelings, motivations, and reactions that both perpetrator and victim experience. Equally important, currently used vignettes describe very simple relationships between perpetrator and victim. For example, these vignettes describe only the abusive encounter. They rarely include background information that would make the perpetrator’s relationship to the victim salient for the subject. However, research indicates that 86 perpetrators are often well-known to their victims and may, in fact, have a well-developed relationship with them (Finkelhor, 1984; DeYoung, 1982). To more accurately assess children’s encoding, CSA prevention assessment measures may need to portray a more well-developed relationship between the perpetrator and victim. Moreover, the vignettes employed rarely embed the actual request to "touch private parts" in a more subtle and complex sequence of touching. Instead, in current assessment measures, the perpetrator directly asks the child to "touch private parts." Research on the methods perpetrators use to engage children in sexual touching suggest that this depiction of an abusive encounter is not ecologically valid (Conte et al., 1989; DeYoung, 1982). In order to obtain more accurate assessment of encoding, vignettes may need to depict the more ambiguous non-sexual touching that often precedes "touching private parts" in abusive encounters (DeYoung, 1982). For example, vignettes could describe the perpetrator’s overtures to engage in sexual touching as starting with a hug or back rub and then proceeding to less ambiguous touch (e.g., "Would you give me a hug?" followed by more explicit statements such as "Would you like to touch private parts?”). Some CSA prevention researchers have suggested that perpetrators’ descriptions of how they lured children into sexually abusive relationships should guide the development of assessment techniques and CSA prevention curricula (Conte et al., 87 1989). A recent study employed videotaped vignettes in order to measure children’s encoding of sexually abusive encounters (Hazzard et al., in press). However, the 30- second vignettes did not permit the portrayal of a more complex relationship between perpetrator and victim, nor did they depict the escalation of sexual touching. More widespread use of filmed presentation of sexually abusive interactions and the use of information provided by perpetrators to create more realistic scenarios may yield more valid measurement of encoding in future CSA prevention research. Consistent with previous research, this study found that the age of the social partner significantly influenced preschooler’s apprehension of social cues. However, contrary to findings reported in the SPS literature on young children (Flapan, 1968), preschoolers in this study perceived social cues in interactions involving a peer lgnn accurately than social cues in situations involving an adult. The significant effect found for age of social partner on encoding is confounded by the order of vignette presentation. Story order was fixed. Vignettes involving a child partner were presented first to participants; stories involving an adult were presented last. Any misperceptions by the child were corrected by the interviewer before proceeding to the next story. Thus, children’s more 88 accurate encoding of the adult stories may reflect their having ”learned” the correct answers over the course of the interview. Thus, future studies of children’s SPS in potentially sexually abusive encounters need to assess the encoding of a wider variety of social cues, present more ecologically valid social stimuli, and randomize the order of story presentation. acto s at at Se t'o 'n Se u Abus'v c te The finding that children offered more non- confrontational but fewer assertive, help-seeking, and rule- based solutions to sexually abusive as compared to non- sexual dilemmas suggests that children’s repertoire of self- protective strategies may be limited in sexually abusive encounters. This finding indicates that CSA prevention efforts to teach children a wide range of effective strategies to defend against unwanted sexual touching are, indeed, needed. Typical CSA prevention curricula teach children three strategies for self-protection--saying no, telling an adult, and running away (for a review see Conte et al., 1986; Gilbert et al., 1989). These programs can be conceptualized as providing children with a script that enables them to generate multiple, effective problem solving strategies to sexually abusive situations. The curricula sensitize children to both internal (emotional) and environmental cues that indicate potential sexual abuse as 89 well as help children repeatedly rehearse self-protective behaviors that can be enacted in response to these cues. Both SPS and CSA prevention research suggest that the fewer effective problem solving strategies a child can generate, the less likely it is that he/she will be able to effectively resolve the social dilemma. The interaction effects for non-confrontational and help-seeking solutions are also pertinent to CSA prevention efforts. The interactions involving gender of subject suggest that boys are less likely than girls to suggest non- confrontational or help-seeking solutions in sexually abusive situations as compared to non-sexual situations. Therefore, once they are asked to engage in sexual touching, boys may be at greater risk for actual sexual abuse because they utilize fewer effective strategies to stop the unwanted touching. Thus CSA prevention curricula may need to devote particular attention to material that encourages boys to use help-seeking and non-confrontational responses to sexual abuse. Two CSA prevention programs currently available, the ”Grossmont Sexual Abuse Prevention Curriculum" (Behana, Gamble, & Stevens, 1985) and ”Talking About Touching" (Committee for Children, 1986), were examined to determine the amount of material devoted to depicting boys using self- protective skills in potentially abusive encounters. These two curricula include nearly the same number of sexually abusive vignettes involving boys as those involving girls. 90 However, these programs might be more effective in helping boys to develop a more diverse repertoire of self-protective skills if they devoted proportionally more time to exercises that teach boys help-seeking and non-confrontational responses (e.g., through role-plays). Future research needs to investigate whether CSA prevention programs are differentially effective with boys and girls. The interaction involving the age of the social partner revealed that in both sexual and non-sexual situations with an adult, children suggested help-seeking responses with about the same frequency; however, children chose help- seeking more frequently in non-sexual situations involving a child partner. Help-seeking responses (e.g., "He should tell his Mom") assume that the adult solicited will view the incident as inappropriate and harmful to the child. The nearly equal and relatively low rate of help-seeking with adult social partners, regardless of the nature of the situation, suggest that children may view nny activity suggested by an adult as appropriate. They seem to assume that if an adult suggested the activity, other adults (e.g., Mom or Dad) would also approve; hence, they do not seek help from others. CSA prevention researchers have suggested that children’s inability to discriminate between appropriate and inappropriate requests by adults may be fostered by parental admonishments to "listen to adults" and indiscriminately obey authority figures such as babysitters and teachers (McGrath, Speth, & Bogat, 1989; Wurtele, 1987). 91 CSA prevention programs often include a lesson that helps children make judgements regarding the appropriateness of an event, independent of whether an adult requested they engage in it (for a review see Conte et al., 1986; Gilbert et al, 1989). These programs also often teach children that certain adults actually may request that a child engage in harmful activities (for a review see Conte et al., 1986; Gilbert et al., 1989; Wurtele, 1987). Thus, this study’s finding suggests that these components of CSA prevention programs are, indeed, necessary if children are to learn to enact self-protective behavior. CSA prevention programs need to continue to include lessons that teach children to judge the appropriateness of a request independent of the age or social status of the person making the request. Conclusion This study’s results are important for both SP8 and CSA prevention research. The data supports and challenges previous findings in each domain. Recent theories of SPS (Rubin & Krasnor, 1986) were supported by this study’s finding that children varied their strategy choice according to the nature of the request. However, comparison of this study’s results to studies of SPS performance in common dilemmas suggests that researchers must investigate a wider range of situations if they are to gain an accurate understanding of children’s social competence. This study’s findings regarding the influence of the nature of the request on children’s strategy choice support APPENDICES APPENDIX A Preschool Alternative Solutions Test (PAST) 92 previous CSA prevention data that indicates that children are often unable to defend themselves in potentially sexually abusive encounters. However, it challenges CSA prevention researchers to consider more carefully the importance of the child’s gender when developing programs to teach children self-protective skills. Moreover, this study suggests that conceptualization and assessment of children’s performance in sexually abusive encounters according to SPS variables (e.g., encoding, strategy choice, planfulness) yields important information regarding children’s competence in protecting themselves from sexual abuse. CSA prevention is a new area of research; in order to more effectively prevent sexual victimization, it must develop a theoretical framework capable of explaining children’s performance in sexually abusive encounters. This study suggests that the literature on children’s SPS is a viable place to start. 9 3 Code# Interviewer Initials (FEMALE VERSION) Preschool Alternative Solutions Test (PAST) mm I'm ' totell uastoryabouttwolitfle ‘laBeckyandherfriend. Theywereboth thneeyearsod. Oneygay theywenplayingoutsi andB sfriemdeaid.‘1have60cente. Let’scrouthesbeetandgeteomecanh!" ButBeckyisn’t owedtoaossthatstnetwith anyonehutherparents. SOSHEDOESN'I‘WANTTODOTHAT. mam ForQueetioneI-J,beeuntomarhtheflntanswertluchildgivee~eventfitie wrong.) . 1. Are Becky and her friend both three years old? (Please CIRCLE one) Yes Mdonotnadchoiceeunlesschilddoeenotnspond.) No Don’t know 2. Is Becky's friend a boy or a girl? (Please CIRCLE one) Girl Boy Don’t know 3. What does Becky’s friend want her to do? m: If child has answered all 3 questions correctly, go directly to question #4. If child has answered any of the 3 qua-done incorrectly, repeat the story for the child and comet the child's nu'sperceptione. It in ertremely important that the child understand these three facts of the story before proceeding to question #4.] 94 4. What could Becky Mun when her friend asks her to cross that street she’s not allowed to cross? W Record one response and then proceed immediately to question #6.] MJZ; If child does not respond or Jaye. 'IDK' ask, "What could Becky say to her Marhwcouldeheacttfherfi-iemdaehedhertocrmaetreeteheenot allaeedtom'? 5. 0.3., (name of child). so you said Becky could (repeat child’s "spouse to #4). But suppose tried that and her friend kept on eayin , “Awe come on, , come on! Let's Just gathecandyetoreforaminntel IHEEw toouldBeckydoorsayifshex-eally 'twanttocroestheetreet. 95 on (name of child) you’re doing a great job thinking so hard! So far you said Becky could (repeat child’s responses). Now I want you to tell me ALL THE DIFFERENT THINGSBeekycouldsayordoifehe didn’twanttocrossthe streettogotothe candy Wlfachlldbegfluto aflerthneenumemtionastopchildandsay. '0L(nmofehad),thafsmwwmthw1ht8«byeboulddo “fen. wmmthr'eepnoiousm 90'6de I’dlibeyouto m0, gy-kah" mot-do. What’s sometime; else W Uchfldsays 'IDK' rompt saying, '01 (child’s ncmeJJet’stldnbmlly hardaboutwhataeckycoulddogaaarap staysndrepeatquesttoanl m Number responses.) WW AFTER EVERY RESPONSE PROM BY SAYHVG, 'OL what else mums-yr” 'OL, whatelseeouldfleehysaya'dom 96 W mwmmRGETIThemmatminmhuoqshauubethemmezendcrastlmchild WW I’m go' to tell you a story about two little girls, Molly and her friend. They are both three years old. One day when th were watchin' g T.V. gather, the friend said to Molly, 'let’s touch private parts.“ BUT M LLY DIDN'T WANT DO THAT. (301%: FmeioMI-abemtomarhtheflntmthechildgives—evenifitis wrong 1. Are Molly and her friend both three years old? (Please CIRCLEone) Yes Wanamadchoicesmlesschilddoesnotnspond.) No Don’t know 2. IsMolly’sfi'iemdaboyoragirl? (Please CIRCLE one) Girl Boy Don’t know 8. What does Molly's friend want her to do? W°Ifchildhasansweredall3questions con-ectly,gotodirectlytoquestion #4. If childhasansueredanyofthesquesflonsineorreetlynepeatthestoryfortluchild and coma the child’s misperaptions. It is atromely important that the child understand thesetlu'eefactsofthestorybefireproceedingtoquestion #4.] 97 4. Remember Molly doesn’t want to touch private parts. So what could she do or say when her friend asks her to do that. W11; Record one response and then proceed immediately to question #6.] Wlfchilddoesnatrupond 'IDK'uh’WhatoouldHollysaytoher Mwhwcouldfldbutifzrafltfl'endakedbernbuchpfimemm 5. 0.1L ,(natne chil,d)soyousaidMoll could(repeatchild’sresponseto#4). Butsuppose Mollytn tried that and her friend kept sayying," “Awe come on, Molly, come onl Let's just touch private t”partsforalittlewhile." mnwhatcoulduollydoorsayifshereany didn’t wtan totouchprivateparts. 98 6. 0.1:. (name of child) you’re doing a great job thinking so hard! So far you said Molly could (repeat childs responses). Now I want you to tell me ALL THE DIFFERENT THINGS Molly could say or do if she didn’t want to touch private parts with her fi'iend. mu.- If a child begins to perseverate, after three enumerations. stop child and say, "0.1. (name of child), that's hind of like what you thought Molly should do before. Gsplainhowthethmpreoiommpomarealihe.) I’dliheyoutotellme something else Holly could do orsayJ? W F child says 'IDK.’ prompt by saying‘ , 'O.K. (child’s name). let's thin): really hardaboutwhatllolly coulddoifaiaraphrasestoryandrepeatquestion “J." W Number responses] NW AFTER EVERY RESPONSE PROMPT BY SAYING. '0; what else eomldlfollytryI"OR "0.x, whatelseeoaldlollysayordofm‘l 99 WE I’m goin to tell you a story about a little girl named Suzie and sogrownup lady she know. One they were walking outside and the grownup said. “Suzie, I have some money. So let’s cross e street and get some candy!” But Suzie isn’t allowed to cross that street With anyone but her parents. SO SHE DOESN'T WANT TO DO THAT. 1m: For-Questions 1-3, besunetomarh theflntauswerthechildgives-evenifit is WWW.) 1. Is the grownup in this story older than Suzie? (Please CIRCLE one) Yes [flQIEdonotreaddwicesunlesschilddoesnotnspondJ No ’ Don’t know 2. Isthegrownupinthe storyaladyoraman? (Please CIRCLE one) Man Lady Don’t know 3. What does the grownup want Suzie to do? m- Ifchild has answered all 3 uestions correctly, go directly to question #4. If child has answe any of the 8 questions incorrectly, repeat the story [hr the child and mrrect the child's misperceptions. It is extremely important that the child understand these three facts of the story betbre proceeding to question #4.] 100 4. 3h“ :07qu Suzie Mm when the lady asks her to cross that street she’s not allowed cross. MAL; Record one response and then proceed immediately to question #6.] Wlfchild does not respondorsays. 'DK‘ash. "Whateould Susie say to the :dyorrhyowcouldeheactiftheladyaehedhertooroesastmtshe’snotallowed ems 5. 0.x, (name of child). so you said Suzie could (repeat child’s response to #4). But suppose Suzie tried that and the grownup kept on saying, ”Awe come on, Suzie, come on! Let’s {fiu‘sitn go to the candy store for a minute!” 1m what could Suzie do or say if she really ’t want to cross the street. lOl 0.K. (name of child) you’re doing a great job thinking so hard! So far you said Suzie could (repeatochild's responses). Now I want you to tell me ALL THE DIFFERENT THINGS Sumo could say or do if she didn’t want to cross the street to go to the candy W If a child begins to perseverate, after three enumerations, stop child and say. '01. (name of child), that’s hind of like what you thought Suzie should do before. Explain how the three previous responses are alike.) I’d like you to thinh of all rrgm things Suzie could say or do. What’s something else Susie could W F child says 'HJK' prompt by saying. '03. (child’s name). let’s think really hard about what Suzie could do if @araphrase story and repeat question #6).' ME: Number responses.) (”W AFTER EVERY RESPONSE PROM BY SAYHVG. '0‘. what else couldSusietryr'OR '01.. whatelsecouldSues‘esayordoW‘] W This is a story about a little girl named Linda and a grownup lady she knows. One day they were reading a book together and the grownup said to Linda, "Linda, let's touch private parts.’ BUT LINDA DIDN’T WANT TO DO THAT. %° For Questions 1-3. be sure to mark the first answer the child gives~eoen if it is 1. Is the grownup in this story older than Linda? (Please CIRCLE one) Yes W'donotreadchoicesunlesschilddoesnotnspond.) No Don’t know 2. Isthegrownupinthe storyaladyoraman? (Please CIRCLE one) Man Lady Don’t know 3. What does the grownup want to do? m Ifcltildhasartsweredall3questionsconectly,goWytoquestion #4.!fchild hasansw anyofthe3questionsincorrealy,repeatthestozforthechildandcorrect the child's misperceptions. It is. atremely important that the ild understand these three factsofthestorybeforeprooeedmgtoquestionfld 103 ‘- Remember, Linda doesn't want to touch private parts. So what could she m when the lady asks her to do that? W’ Record one response and then proceed immediately to question #5.] Wlfchilddoesnotnspondorsays. 'HJK'M'WhateouldLindasaytothe ladyorhoeseouldsheactifthisladyashedhertotouchprioateparbr‘l :- 53 i dehild). so said Linda could (repeat child’s response to #4). But suppose thatand theyolfikept saying. 'Awe comeomLinda. come on! Let’sjust tepartsfora' whfla'THENwhatcouldIJndadoorsayit‘shereally totonchprivateparts. ii? iii 6. 104 0.): (name of child) you’re doing a great job thinking so hard! So far you said Linda could (repeat child’s responses). Now I want you to tell me ALL THE DIFFERENT THING-S Lmda could say or do if she didn’t want to touch private parts with the grownup. mu; If a child begins to perseverate. after three enumerations. stop child and say. "0-!- (um of child), We hind oflihe what you thought Linda should do before. (Es-plain how thethreepnsoious responsesarealihe.) I’dliheyouto W Ofdaflwmthingsundacouldmordo. What’s somethingelse Linda could do” W 1? child my! 'mK'prompt by saying, '01 (child's name). let's thinh really ha-daboutwhatLindacoulddoipraruphmsestayandrepeotquestioniwf W’ Number responses] PW- Am EVERY RESPONSE PROM BY W0. '0‘ what also ooaldUndahyVOR "01., whotelseeouldLindaeuyordo?"“] APPENDIX B Preschool Alternative Solutions Test (PAST) Scoring Manual THE PRESCHOOL ALTERNATIVE SOLUTIONS TEST (PAST) SCORING MANUAL Jacqueline S. Grober Michigan State University 1990 105 TABLE OF CONTENTS IntrOduction.I.0........OOOCIOOIOOIOOOO00.0.000000000000000107 SECTION I: Overview of scoring procedure..................108 SECTION II: Scoring accuracy of encoding...................110 SECTION III: Scoring solution content......................112 IrrelevantOOOOO...OOOOOOOIOOOOOOOOOOOOII‘ AssertiveOOOOOOOOCOOOOOOO00.000.000.00115 Help-seekingi..00....0.0.0000000000000117 Nonconfrontational....................117 AgngSSiveeeeeeeeeeeeeeeeeeeeeeeeeeeee118 FinagleOOOOOOOOOOOOOOOOIIOOOOOOOOOOOOOIZO Passive resistance....................123 Manipulate affect.....................124 Conpliance............................124 Deter-ining multiple content acorea...126 Part A: Multiple content responses....126 Part 8: Responses containing similar strategies...........127 SECTION IV: Scoring solution effectiveness.................128 SECTION V: Scoring solution planfulness....................138 SECTION VI: Scoring solution flexibility and persistence..143 SECTION VII: Counting Alternative Solutions Per Reaponse...146 Blank scoring for-.0.000000000000000000000.0.0.00.000000000147 Blank PAST protocol (Female Version).......................151 PAST pictures (Fe-ale Version).............................163 106 107 INTRODUCIION The Preschool Alternative Solutions Test (PAST) presents four realistic stories to children which describe a social dilemma that the children are then asked to solve. The PAST measures children’s ability to accurately encode both benign and sexually abusive social dileamas and to generate alternative solutions to them. Four stories are presented in the PAST. In each story, the age of the actors and the type of situation is systematically varied. Children’s responses to each story are recorded verbatim. These responses are then scored along the dimensions listed below: accurate encoding of story solution content solution effectiveness solution planfulness solution flexibility and persistence (for question #5 ONLY) “I“MNF‘ vvvvv . . O C . The remainder of this manual provides, first, an overview of the scoring procedures to be used, and then, a detailed explanation of how to score each of the variables listed above. 108 OVERV W S ORING PROCEDUR PAST scoring can be divided into two phases. Each phase quantifies aidifferent aspect of children’s social problems solving ability. The first three questions (#1 - 3) on the PAST examine the accuracy with which children have encoded the social dilemma presented in the story. The next three questions (#4 - 6) assess the child’s ability to offer alternative solutions to resolve this dilemma. The first three questions of each PAST story are scored dichotomously as accurate or inaccurate. The goal of scoring encoding is simply to determine if the child understood each of the important elements in the vignette. Scoring of the second three questions on the PAST is more complex. Here, children generate solutions to the social dilemma presented in the story. The goal in scoring these responses is to accurately translate into numbers the essence of the solutions the child has proposed. Each response to questions #4 through #6 is scored along a number of different dimensions: content, effectiveness, and planfulness. These scores are a numerical short-hand description of the verbal responses the child actually gave. Thus, these scores should capture as fully as possible the nature of that particular solution. Please note that while questions #4 - #6 are scored for solution content, effectiveness, and planfulness, question #5 is scored, in addition, for flexibility and persistence. However, only question five is scored for flexibility and persistence. Coders should score one story at a time. They should first score the first three questions. Then, the first response to question #4 should be scored along all three dimensions (content, effectiveness, and planfulness), then the coder should move to the next response, and so on, until all responses to story 1 are scored. The coder should then repeat this process for stories 2 through 4. Below is a list of the questions the coder should ask him/herself when coding each response. 109 $3ttittttttttlttttttttlit¥$$¥$¥$3333¥¥33333$¥ttt##3333338 W o ’ se Po 3 ' - : 1. Is the answer accurate (=1) or inaccurate (=0)? WW: 1. Is the solution irrelevant? 2. What content categories best describe this solution? 3. How effective is this solution? 4. How planful is this solution? 5. (For question 85 only) How flexible & persistent is this solution? ti$.13#ttlttltitt‘ttttttttttttttttttttttlitit'ttflifittiitt 110 ECT ON I : SCORING CCURAC OF ENCOD NC On the PAST, questions #1 - #3 for each story are scored for how accurately the child understood the major elements of that particular story. One question deals with each major element in the story. Please note that coders MUST BE AWARE OF WHICH STORY THEY ARE CODING AND WHAT GENDER THE SUBJECT 18 in order to accurately score responses to questions 1 - 3!! The answer for question 01 is always the same, regardless of the story or gender of the subject. However, the correct answer to question #2 depends upon the gender of the subject. Thus, the sample answers listed below differ for male and female subjects. Also, the correct response to question 83 differs depending upon the story; The answer to question #3 for stories 1 a 3 is the same, and the answer for question #3 for stories 2 l 4 is the same, see the sample answers below. If a child answers a question correctly a "1” should be placed in the column on the score sheet corresponding to that question. If the child answers incorrectly or says "I don’t know," a "0" should be placed in the column corresponding to that question. Examples of correct and incorrect answers are provided below to help guide scoring. Each box provides answers for one PAST question. £A§I_9!£§IIQE_11 QORRECI INCORREQT ST R O : ANSWERS AN§EEB§ 1). Are and his friend Don”tknow both three years old? Yes No OR Is the grownup in this story older than ? 111 fitttt‘tttlItttttttttttfit##tttttilitttitttttttttttitt8##tilttttttt W 9933321 139988391 Lfllflfil_fififléhfi_figfifl Aflfiflfiflfi 6351335 2). Is ’s friend a boy or Girl, Lady DonLtknow a girl? Boy, Man LBLHEl_flAL£_£QBfl 2). Is 's friend a boy or Boy, Man DonLtknow a girl? Girl,Lady titttttOtIt.itfitttttttttifitit3tit.#tttttt‘ittfitfittfittttttttt$8¥¥*$$ ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++ EA§I_Q!E§IIQE_13 QQEEEQI CORR STORY 1 EVSTQRY 3 AE§EEB§ AE§EEB§ 3). What does ___‘s friend want him/her to do? get some candy go with him buy some candy get something cross the street IDK OR What doe the grownup " " want ___ to do? 9953221 IEQQBEEQI §IQEX_2_L_§IQBI_1 Aflfiflfififi Aflfiflfiflfi 3). What does ___’s friend touch p.p., I.D.K., want him/her to do? touch his/her touch body p.p. parts, touch each other, take his/her clothes off OR What does the grownup want to do? +++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++ 112 SEC ION : U ION ON On the PAST, questions 04 - #6 ask children to generate solutions to the presented social dilemma. These solutions are scored, first, according to their content. All solutions can be classified as falling into gng_21_gg;g of the major content categories and a content subcategory. Solution content is recorded on the scoring sheet by writing the corresponding content code number into the assigned space. Include the umber before and after the decimal place. SCORING OF CONTENT SHOULD NOT BE IMPRESSIONISTIC. INSTEAD, CODERS SHOULD CAREFULLY FOLLOW THE GUIDELINES LISTED BELOW. Example ‘1 If a child said "I would say 'no’" this response falls under the major content category of ”assertive” (see pp. 9) and under the subcategory of "assertion of the subject’s intention not to comply" (2.2), hence a "2.2" should be placed on the scoring sheet under the first column labeled content codes. The remaining columns under the content codes can be left blank since no other content codes describe the child’s response. X8!!! e If a child said, "I would say no and hit him," this response can be categorized by two different content codes. First, the 'I would say no’ part of this solution should be coded as in example '1, by putting a "2.2" in the fist column. Then, the coder must score the 'and hit him’ part of the child’s response. This part of the response can be categorized under the aggressive major content category (see pp. 12) and under the "direct physical harm to person or possessions (5.1) subcategory, so a "5.1" should be placed in the second column under content codes. The remaining columns under content codes can be left blank sine no other content categories describe the child’s response. 1). 2). 3). 4). 5). 6). 7). 8). 9). 113 OVERVIEW OE THE CONTENT CODES W 1.1 Nonsense solutions or irrelevant comments 1.2 Vague solutions that were not clarified by the interviewer W 2.1 Assertion of subject’s rights 2.2 Assertion of subject’s intention not to comply 2.3 Other-directed commands or requests for cessation of behavior 2.4 Nonviolent verbal threats or physical gestures that do not escalate the situation 1 Telling another person 2 Solicitation of aid 3 Solicitation of third-party advice 3. 3 3 4.1 Subject leaves the scene AW 5.1 Direct physical harm to person or possessions 5.2 Third-party physical assault 5 3 Verbal threats or physical gestures which escalate the situation W Bribe Make a deal--solicitation of 3rd party advice Make a deal--distraction Make a deal--restructure situation to meet both parties' needs Stalling Invoke a rule or authority’s command Make excuses Distraction comma: mama: canon» ubqu-o 1 Ignore 2 Passive or Slight Resistance, Equivocation 7 7. 8.1 Manipulate own affect 8.2 Manipulate perpetrator’s affect Mam-121.911.2111 9.1 Active compliance 9.2 Change goal 9.3 Meets own needs later 9.4 Delayed compliance 114 We” (1) Note: "I don’t know” is scored as irrelevant if it is offered as the ONLY response to question 04 (This should be a highly unusual response, since most children can think of at least one solution to the presented social dilemma!) When "I don’t know” is given in response to question 85, it is also scored as irrelevant and is then scored along the flexible-persistent dimension (see section III). Irrelevant responses are responses which do not in any perceivable way relate to the situation at hand. Nonsensical answers, responses which simply repeat the goal, irrelevant comments, irrelevant feelings, or responses in which the interviewer did not obtain sufficient detail to accurately classify the response elsewhere are also categorized as irrelevant. 1.1) Nonsense solutions, irrelevant comments, "I don’t know" responses ”He could go fishing" ”He should sting him with a bee" "I would blast off to outer space" "Throw the house at him"* "Tell my dog, Santa Claus" (any nonhuman or fictional being} "My friend wouldn’t do that" ”Stay home” ”I don’t know" ”I’m not sure" ”Can’t say” "Beats me" ”Hope he stops”¥' *Egtg: Most aggressive solutions should be scored under category 5. However, if the solution requires the performance of an inhuman feat (e.g. lifting a house and throwing it), then category 1.1 should be used. **Ng£g: See note to category 2.3 1.2) Vague solutions that were not clarified by the interviewer "He should stop his friend” "He should protect his private parts" "She shouldn’t cross the street" "He shouldn’t go across it [the street)" 115 Mile (2) Assertive strategies are nonaggressive statements, questions, or physical gestures used to assert or defend the subject’s rights; his/her intention not to comply; request cessation of the other party’s behavior; express nonviolent threats or ‘warnings; or express disagreement with the other party’s perceptions.or'motives. Examples of different types of assertive solutions are listed below. 2.1) Assertion of subject’s rights "Tell him he’s not the boss of my p.p." ”Say ’you can’t make me’" "Say 'I don’t have to cross the street’" (321;: this category should be used when it is apparent that the child has internalized the right. In other words, there should be no mention of an external rule or authority figure (e.g., parents) which adds to the legitimacy of the rule. If a reference to an external rule or authority figure is made score under 6.6) 2.2) Assertion of subject’s intention not to comply C 9" "Say no ”Say 'no way, Jose’" "Say 'I don’t want to [touch p.p./ cross street]’" ”Say ’I don’t like to [touch p.p. / cross street]’" "Say 'No, I won’t touch p.p.’" ”Say ’Bye-bye’" "Say ’I want to go home’" ”Say ’I want to go to bed’" Note: if a child includes a simple one-word "no" as part of a larger response which is coded under a different category (especially 6.1-6.8), a 2.2 need not be scored in addition since a 2.2 is assumed in those cases. However, a 2.2 should be scored in addition to another content code if the assertive (2.2) part of the response is elaborate (e.g. ”no way, Jose", "no I don’t want to"), even if the remainder of the response if scored elsewhere. 2.3) Other-directed commands or requests for cessation of behavior "Tell her to stop that" "Tell her to quit bugging me" 116 "Tell her to leave me alone" "Tell the grownup to go home" ”Say 'I don’t want you to cross that street’" "Say 'stop!’" "Say 'you go across the street’" ”Say 'I wish you would stop’" Note: Compare this response to the last response under category 1.1. If the child specifically indicates he/she would gay "I wish you would stop" this is coded as assertive (2.3) since the child is verbalizing his/her request for the behavior to stop; but if the child only silently wishes or hopes it will stop, this is simply an irrelevant feeling and should be coded 1.1. 2.4) Stated nonviolent threats or gestures that do no; escalate the situation i.e., threats or gestures that probably will go; cause a verbal or physical fight "Say ’I’ll tell.’"* ”If you do that, I’ll tell"‘ ”Say ’I’m going to tell my Mom and Dad’"* "Say 'I’m going to call the police’"* "Say 'You’re driving me crazy’" "I don’t like you anymore" "She could say, 'If you do that, I won’t like you anymore’" ”I won’t be your friend anymore" ”I don’t want to be your friend anymore" ”If you do, I’ll leave"’ "She should say, 'If you make me do that, I’ll cry’"# I'No§,g: There are a number of responses which if the child W the described action are coded under this category. If the child actuallz_tgkg§ the described action, the response is categorized elsewhere. For example, if a child says "I’m going to call the police" this is coded as an assertive nonviolent threat. However, if the child says "Call the police" and the interviewer confirms that the child is proposing to take this action, then this response is categorized as falling into another category (help-seeking). Similarly, if a child threatens "I’m going to tell" this is coded as assertive, but if the child indicated he/she would actually take this action by saying "I’d go tell,” then this should be coded as help-seeking. Similarly again, if the child threatens "if you make me, I’ll cry" this should be coded assertive, but if the child says "I’d cry" this would be coded manipulate affect. 117 Masking (3) Solutions in which the subject involves someone else in solving the problem are coded as "help-seeking." This can take the form of telling and adult or peer after the episode has ended, soliciting immediate aid from and adult or peer, or asking an for advice on how to cope with the problem when it arises. 3.1) Telling another person "She should tell" "Tell" "Tell my [Mom, Dad, sister, friend, teacher, etc. {any living human being)]” "Call the police" 3.2) Solicitation of aid "She should get help" ”She should scream for help" "Ask her Mom to help her cross the street" "Ask another adult to help them cross the street" 3.3) Solicitation of third-party advice ”She should ask her Mom if she can cross the street" "She should ask her Mom for permission" "Ask [my mom, sister, friend] if she should” ”Ask her teacher what to do” Wining (4) Strategies are coded as ”nonconfrontational” if the subject meets his/her own needs immediately and avoids a confrontation, argument, or physical fight. N953: This means that the child must intend to leave along. Thus the proposed exit should not imply or state that the other child/adult would accompany the target child. Compare to category 7.1 (used when child does not leave the scene) and 6.8 (often used when both actors leave the scene together to engage in a different activity). 118 4.1) Subject leaves the scene "Hide” ”She should leave" ”Walk away” ”Go home" ”Go to her room” ”Run away" ”Go play with someone else" ”Go somewhere else" ”Go watch a different T.V." ”Go climb a tree" (it should be assumed that tree climbing is a solitary activity unless the child indicates he/she would include the other actor) ”Go outside and sleep in a tent" (it should be assumed that sleeping is a solitary activity unless child indicates he/she would include other actor) "Go outside and play" (if interviewer clarified child would go alone) Amalie (5) Responses which express physically hostile infringements on another’s person or possessions; verbal insults; threats to do physical harm; or the use of intimidation (e.g., threat of violence) constitute aggressive solutions. 5.1) Direct physical harm to person or possessions "Hit him" "Throw her out the window" "Throw numbchucks at him” ”Smash his T.V.” "Run him over” ”I would shoot him” Note: Aggressive solutions must be at least marginally realistic. Thus the last response is coded as aggressive, but "I would shoot him with a 135;: gun" is considered unrealistic since no such technology exists it should be scored 1.1. 5.2) Third-party physical assault ”Get my friends to beat him up" 119 5.3) Verbal aggression which does escalate the situation i.e., could cause a verbal or physical fight "Call him names" ”Say ‘you’re a doo-doo brain’" "Say 'I’m going to hurt you if you try it’" "I’m going to crush you head" "Stomp my feet" ”Stick out my tongue at the babysitter [i.e. grownup] Yell at her "0.x. I’ll do Karate on you" 120 Einnls (6) Strategies in which the subject pegists papticipating by persuading, enticing, or tricking the other actor into a giffgppgt activity, or persuades, entices, or tricks the other actor to cease his/her efforts to get the target child to participate in the forbidden act. 6.1) Bribe "Say 'if you watch T.V. I’ll give you a cookie’" "Say 'I’ll give you 55 not to do that’” 6.2 - 6.4) Make a deal solutions Note: In all of the "make a deal" categories (6.2 - 6.4) the child seems to be entering into a verbal negotiation with his/her partner. Often the word "let’s" indicates he/she is proposing an alternative to the suggestion to which the other actor may agree or object. Other solutions by their very nature require the cooperation of the other actor, and thus implicitly ask for his/her assent/dissent. The key to "make a deal" solutions is that the child seems to invite negotiation with the other actor. (In contrast to 6.1 and 6.8 in which the child is more directive.) 6.2) Make a deal--solicitation of 3rd party aid "She could say 'Let’s go ask my Mom if I can cross the street with you.’” "He could say, 'Why can’t we go ask our parents if we can do that?’” 6.3) Make a deal--distraction ”Say 'let’s play checkers instead’" "Say 'lets’ play with our stuff’" ”Let’s play a game about monkey bars" "How about we go outside and get popsicles together" 6.4) Make a deal--restructure situation to meet both parties’ needs "His friend should go across and bring the candy back" 121 "Say 'You go to the store and get the candy. I’ll stay here’" "Say 'Let’s go to my house and get some candy’" "Say ‘Let’s go to another store" ”Say 'I have some candy here; let’s just eat that’" [for p.p. story] "She could say 'go touch p.p. yourself’ and then she could leave and watch TV at her own house" [for p.p. story] Say "Let’s touch tummies instead" 6.5) Stalling "She could say 'Let’s do that later’" ”Say ’Let’s do that tomorrow’" ”Say 'I can’t talk about it now’" Note: A code of 6.5 differs from a code of 9.4 in that the child never agrees to comply at any time with the request. To be given a score of 6.5 the response 125L923: indicate the child would comply at any present or future time. 6.5 responses emphasize a delay or stalling for time and, most important, leave gogplgtelz unanswepeg the question of future compliance. (If a child indicates he/she will comply immediately, the response is scored 9.1. If child indicates he/she will comply in the future, the response is scored 9.4, delayed compliance). Npte: A code of 6.5 differs from a code of 7.2 in that 7.2 responses imply or state that the child may comply at a future time. Thus the child explicitly states at least a possible receptiveness to participation. However, 6.5 responses leave the question of compliance com ete unanswer 6.6) Invoke a rule or authority’s command "Say 'My parents said I don’t have to touch p.p.’" "I can’t. My Mom and Dad won’t let me without their permission" "I’m not allowed to do that" "My parents said never touch p.p" "Say ’I should go get my parents to go with me’" "Say 'Candy is bad for my teeth’" "Say ’Candy is bad for you’" "Say 'That’s bad’" "Say 'That’s wrong’" "Say '1 can’t touch p.p.’" "Say 'I don’t want to get hit by a car’" "He could get hit by a car" Note: In many of these examples, the response only implies or alludes to a rule or authority’s command. The child sometimes does 122 not articulate the specific rule or the specific authority from which the rule came. Nevertheless, these responses should be coded 6.7) Make excuses "Because he doesn’t like candy" "He could say he doesn’t have time" "I’d say ’I’m tired’" "Say 'I can’t; I’m going to my friend’s house.’" "Say 'I can’t right now, I have to eat lunch’" Note: this category should only be used if the child is obviously making an excuse. 6.8) Distraction "They could go outside and play" "Go to the kitchen to get something to eat" "Ask for a drink of water" "They could play with the cat" "Play with playdough" "She could drew her [other child] a picture" ”Turn off TV and put in new tape 'Land Before Time’" "Eat peanut butter sandwiches together" "Get some cigarettes and smoke them" "They could just make paper or something--better than touch private parts" "They could try going to the pet store to buy a fish instead" Note: The 6.8 category differs from 6.3 in that 6.8 responses are more directive. The child does not indicate he/she would negotiate with the other actor, but, instead, would simply begin to the described action. uptgg The 6.8 category is different from 4.1 in that the activities either imply the participation of both actors or the child explicitly states both actors would be involved. Thus 6.5 differs from 4.1 in that both children leave the scene together. Ngte: A 6.8 code differs from 7.1 in that the activity mentioned in 7.1 responses is generally’ considered not to require the interaction of two persons whereas the activities articulated in 6.5 responses suggest or require the participation of two persons. 123 We: (7) Solutions in which the subject does not comply, or actively resist are coded as "passive resistance." 7.1) Ignore "he could just keep watching T.V." "he could ignore him" Ngte: Activities in this category do not require that the actors interact in any way. 7.2) Passive, Slight resistance, Equivocation "Ralph could say nothing" "I wouldn’t do anything" "Say a prayer to God that they will stop" "He could just stand there and not do anything" "He could just sit in the chair and wait" [p.p. story] "She could scoot over from her friend" "He could fall down on the ground" "Andy could wait for his Mom and Dad to come home" [cross the street story] "She should let go of her friend’s hand" "Say 'O.K., maybe’" "Say, ’I don’t know’" ""EYbe :0 "Maybe tomorrow" 124 W (8) Strategies that center upon the manipulation of affect are ones in which the subject states that he/she would either change his/her own emotional state or attempt to change the emotional state of the perpetrator. In addition, solutions in which the subject attempts to deter the perpetrator by threatening to become emotionally upset fall under this category. 8.1) Manipulate own affect "Act scared” "Act mad" "Act sad" "Act frustrated" "Cry" "I’d cry" "I’d be real mad" "She could give them mean looks" 8.2) Manipulate perpetrator’s emotions "I’d make him feel bad" (Q: what do u mean? "make him (ppl_bgg that he asked") We: (9) Solutions in this category are not resistance strategies. Instead, they are behaviors or statements which indicate the child would give up the goal of resistance and comply with the perpetrator’s request. 9.1) Active compliance "Say ’Yes’ "I would do it" "I would take off my pants" "I would take off all my clothes and touch p.p." "I would say 'O.K.’" "we can touch p.p./cross the street" "I’ll touch p.p./cross the street" "Let’s touch p.p./cross the street" "Hang on to her hand and cross the street" 125 "Have her friend carry her across the street" "Jump across the street" "Touch p.p. / cross the street, but keep it a secret 9.2) Change goal "I would give-in" 9.3) Meets own needs late: "I would touch p.p. for a while and then watch T.V." "I’d say O.K. just this one time" Note: Only responses in which the child touches p.p. fiplt and meets his/her own needs lptgg should be scored here. Responses in which the child tries to gglgz touching private parts by suggesting another activity figst yet indicates he/she would, ultimately comply, both distraction (6.3 or 6.8) and delayed compliance (9.4) should be scored (see below). 9.4) Delayed compliance "Say 'I’ll do it later’" (Note: child indicates he/she will definitely participate at a future time.) "Let’s watch T.V. until this show is over, then touch p.p." (Note: scored both 6.3 and 9.4) 126 DETEREININQ EULIIELE CONTENT SCOEEE As was mentioned in the introduction to content scoring, a child’s response may be coded under more than one content category 11 go :e than,one type of problee selvigg gtrategz we; 9 ffer eg 19 2 single [espopee. Multiple content codes are needed when a child describes gualitetjvelz giztepent problem solving strategies in a single response. Part A below explains how to score qualitatively different strategies given in the same response. However, sometimes a child describes a gepietien of a content strategy that has already been scored for that response. Part B explains how to distinguish these types of solutions (which do get contain qualitatively different strategies) from ones that d_q contain true alternative strategies. Examples of responses are discussed below. It should be noted that these responses are merely egegpleg and are not exhaustive of all of the possible combinations of multiple content codes. PART A: EULIIELE CONTENT RE§PON§E§ ’ e ons Co t t 0 I’d say, ’first let’s watch T.V., 6.3 9.4 then we can touch p.p.’ She could say 'I’ll do that later, 9.4 6.1 if you watch TV with me now’ Say 'I can’t now, maybe later’ 6.7 7.2 9.4 He could say 'No, I don’t want to touch 2.2 6.5 private parts, let’s do that later Say 'no and ride my bike and tell’ 2.2 3.1 Say 'no’ and run away 2.2 4.1 I’d say 'no way Jose’ and hit him 2.2 5.1 Say 'you go across the street. I’m not allowed’ 2.3 6.3 Say ’yell stop’ and hope he stops 2.3 1.1 Say ’I don’t want to’ and cry 2.2 8.1 Say 'no’ and run away 2.2 4.1 He could leave and go tell his mother 4.1 3.1 127 R : S ONS S CONT IN SI A TR G 8 Sometimes a child will, in_e_eingle_1e§penee, offer a number of solutions which are gexietipne of a major content category. In this case, each variation should be coded according to the subcategory which best describes it. We: mm Say 'I don’t want to do that. I’m going to tell.’ 2.2 2.4 Say ’You can cross the street by yourself. I’m going to tell my Mom and Dad.” 2.3 2.4 128 SECTION IV: RATING EFFECTIVENESS After coding a solution’s content, it should be rated in terms of effectiveness. A response’s effectiveness is determined by examining both the solution’s content code and the particular solution the child provided. Both the original response and the assigned content code should be considered. Each solution is rated on a ‘4-point effectiveness scale. Ratings of solution effectiveness should be entered in the "EFF." column of the scoring sheet. Effectiveness is rated according to the criteria delineated below. The effectiveness of a given strategy is highly dependant upon the situation in which it is enacted. For example, physical aggression against a peer is more likely to repel the other actor and less likely to result in severe negative consequences (retaliation) than physical aggression directed toward an adult. Similarly, cognitively oriented arguments which may persuade a peer, are less likely to persuade a more cognitively sophisticated adult. Hence, when four adult judges used the criteria listed below to independently rate the effectiveness of over fifty solutions, a clear pattern emerged. A solution’s effectiveness could be loosely determined by the situation to which it was given and the content code it was assigned. Idiosyncratic solutions were sometimes awarded slightly lower scores than the majority of solutions in a content group because they were not feasible or accrued unusual negative consequences. Therefore, each particular solution then must also be evaluated along two other: dimensions: a) could this solution realistically be carried out) by' a ‘preschooler, and b) ‘would negative consequences for the child be created by this solution. A score of "O" is given if the solution could not be enacted by a preschooler. And one point is subtracted from the effectiveness score if the solution was likely to produce negative consequences for the child (down to a minimum score of 0). V V N S flighly Ettegtive (3) These strategies virtually guarantee the child’s short-term safety and have a high probability of insuring the child’s long-term safety. No negative consequences will result from enacting these strategies. 129 Ve 've These strategies have a high probability of maximizing the child’s short-term safety but do not necessarily protect the child’s long term safely. In addition, no negative consequences will result from enacting these strategies. Alternatively, in a small number of cases, the child’s short- and long-term safety is virtually guaranteed but enacting the strategy may cause immediate negative consequences for the child. (e.g. "I would go tell my friends what he wanted me to do and we’d beat him [peer] up. Then I’d tell Mom and ask her to whoop him too.") Effective (1) These solutions protect the child’s short-term safety. However, the child’s long-term safety is not considered. Negative consequences will probably also result. Alternatively, in some rare cases, short-term safety is ignored in favor of protecting long-term safety. Negative consequences are accrued in order to protect the child’s short- or long-term safety. (E.g., "I’d say no, but if he [grownup] kept up, I’d say 0.x. I’ll do it [touch p.p.] after we read the story. Then I’d run away when he got up.") Ineffective (01 These do not protect the child’s short-term or long-term safety. They may create negative consequences for the child. And/or these solutions are impossible for a child to enact. .‘3 91 V N 130 0° N _0 '0‘ '10. °._ R STORY 1: Two Peers Crossing the Street £2£§£Il¥£fl§§§ QQDE 030900 NNN heth UNNN NNNN CONTENT CODES 1W 1.1 Nonsense solutions or irrelevant comments 1.2 Vague solutions that were not clarified by the interviewer > eseztive soletions 2.1 Assertion of subject’s rights 2.2 Assertion of subject’s intention not to comply 2.3 Other-directed commands or requests for cessation of behavior 2.4 Nonviolent verbal threats or physical gestures that do not escalate the situation Wanting 3.1 Telling another person 3.2 Solicitation of aid 3.3 Solicitation of third-party advice No con out 'o o 4.1 Subject leaves the scene ! 'v o] l' 5.1 Direct physical harm to person or possessions 5.2 Third-party physical assault 5.3 Verbal threats or physical gestures which escalate the situation W1 6.1 Bribe 6.2 Make a deal--solicitation of 3rd party advice 6.3 Make a deal-~distraction 6.4 Make a deal--restructure situation to meet both parties’ needs 6.5 Stalling 6.6 Invoke a rule or authority’s command 6.7 Make excuses 6.8 Distraction HH HM HOOD 131 W 7.1 Ignore 7.2 Passive or Slight Resistance, Equivocation 8.1 Manipulate own affect 8.2 Manipulate perpetrator’s affect We 9.1 Active compliance 9.2 Change goal 9.3 Meets own needs later 9.4 Delayed compliance V’Jh’ 132 9° . " _ 191 ‘2 '1 1‘1 STORY 2: Two Peers Touching Private Parts EEFECTIVENESS CODE 09000) HNH NNNN NNNN CONTENT CODES Win: 1.1 Nonsense solutions or irrelevant comments 1.2 Vague solutions that were not clarified by the interviewer ser 'v o 'o 2.1 Assertion of subject’s rights 2.2 Assertion of subject’s intention not to comply 2.3 Other-directed commands or requests for cessation of behavior 2.4 Nonviolent verbal threats or physical gestures that do not escalate the situation e - ee t' 3.1 Telling another person 3.2 Solicitation of aid 3.3 Solicitation of third-party advice H I o ! l' l 1 !' 4.1 Subject leaves the scene Wishes: 5.1 Direct physical harm to person or possessions 5.2 Third-party physical assault 5.3 Verbal threats or physical gestures which escalate the situation W Bribe Make a deal--solicitation of 3rd party advice Make a deal--distraction Make a deal--restructure situation to meet both parties’ needs Stalling Invoke a rule or authority’s command Make excuses Distraction 05050305 03030303 QOUI abQNt-fi pump-e H000 133 W 7.1 Ignore 7.2 Passive or Slight Resistance, Equivocation 8.1 Manipulate own affect 8.2 Manipulate perpetrator’s affect sag-mm“ 9.1 Active compliance 9.2 Change goal 9.3 Meets own needs later 9.4 Delayed compliance 134 m e e '- m \ I .k .‘ .1 .s 7 s _‘.l STORY 3: A child and an Adult Crossing the Street EEEEQIIEEEE§§ 9925 0.30009 NNF" OHO uranop HHHH CONTENT CODES Wen: 1.1 Nonsense solutions or irrelevant comments 1.2 Vague solutions that were not clarified by the interviewer W 2.1 Assertion of subject’s rights 2.2 Assertion of subject’s intention not to comply 2.3 Other-directed commands or requests for cessation of behavior 2.4 Nonviolent verbal threats or physical gestures that do not escalate the situation - e ' so ‘0 3.1 Telling another person 3.2 Solicitation of aid 3.3 Solicitation of third-party advice c o t o 4.1 Subject leaves the scene W 5.1 Direct physical harm to person or possessions 5.2 Third-party physical assault 5.3 Verbal threats or physical gestures which escalate the situation W 1 Bribe 2 Make a deal--solicitation of 3rd party advice .3 Make a deal--distraction .4 Make a deal--restructure situation to meet both parties’ needs 5 Stalling .6 Invoke a rule or authority’s command 7 Make excuses 8 Distraction 03QO @0490) HH HN H000 135 W 7.1 Ignore 7.2 Passive or Slight Resistance, Equivocation o 8.1 Manipulate own affect 8.2 Manipulate perpetrator’s affect Wen: 9.1 Active compliance 9.2 Change goal 9.3 Meets own needs later 9.4 Delayed compliance ' a,- 136 01 C, ' .3 1°; 17.. ; ‘1’; STORY 4: A child and an Adult Touching Private Parts EFEECTIVENESS QQDE NNH owo (out...) N Nuawau “pip-AH ONTEN ODES Whine: 1.1 Nonsense solutions or irrelevant comments 1.2 Vague solutions that were not clarified by the interviewer e 'v o 2.1 Assertion of subject’s rights 2.2 Assertion of subject’s intention not to comply 2.3 Other-directed commands or requests for cessation of behavior 2.4 Nonviolent verbal threats or physical gestures that do not escalate the situation - e o s 3.1 Telling another person 3.2 Solicitation of aid 3.3 Solicitation of third-party advice 32W; 4.1 Subject leaves the scene 5.1 Direct physical harm to person or possessions 5.2 Third-party physical assault 5.3 Verbal threats or physical gestures which escalate the situation Einnlmlnnm 6.1 Bribe 6.2 Make a deal--solicitation of 3rd party advice 6.3 Make a deal--distraction 6.4 Make a deal-~restructure situation to meet both parties’ needs 6.5 Stalling 6.6 Invoke a rule or authority’s command 6.7 Make excuses 6.8 Distraction H000 137 7.1 Ignore 7.2 Passive or Slight Resistance, Equivocation 8.1 Manipulate own affect 8.2 Manipulate perpetrator’s affect We: 9.1 Active compliance 9.2 Change goal 9.3 Meets own needs later 9.4 Delayed compliance 138 SECTION V; PLANFULNESS After scoring a solution for content and effectiveness, it should be scored for planfulness. WWW 'b't a majes- In other words, "planfulness" is simply how carefully the child has thought about what he/she would need to do in order to successfully carry out his/her chosen strategy. Scorers should realize that planfulness requires cognitive sophistication. Since preschoolers are only beginning to develop this type of complex cognitive functioning, many of their answers will not receive a high planfulness score. Planfulness should not be confused with effectiveness. In rare instances, a solution may be very planful but ineffective. For example, "Ralph should wait until a car is coming and then push his friend into the street and say "you get some candy, stupid!" This solution is planful (since it considers timing and uses objects/situational factors to help the child carry out the plan). However, it would be rated as ineffective (since it may create severe negative consequences for Ralph, e.g., guilt over hurting his friend, criminal record, and threatens his safety, e.g., Ralph’s friend might retaliate if Ralph’s plan fails etc.). Solutions are scored on a 3-point scale as "Very Planful" (=2), "Planful" (=1), and "Not Planful" (=0). A rating of planfulness should be entered in the "PLAN." column of the scoring sheet for each solution. Listed below are criteria that indicate the presence of a plan. If any one indicator is present, a score of at least 1 should be given. The higher score of 2 should be awarded to solutions that exhibit either multiple indicators of planfulness or articulate a complex and/or detailed consideration of a single criterion. 139 N S C N These responses indicate the subject has considered a possible obstacle or contingent event that might change the deployment of the strategy. Examplee: gm 14'; Eegponse W "He could go get his parents to see if he can [cross the street], and then go back. And if they don’t [let him], he can stay home." 2 "She could say no, no! But if the grownup didn’t stop, then she could give-in for a little while and later when he wasn’t looking, she could run home and tell her Mom everything." 2 "She could say 'go touch p.p. yourself’ and then she could leave and watch TV at her own house." 2 "If you want candy let’s go to another candy store on this side of the street." 2 "I’d say stop! And if he didn’t, I’d leave." 1 "Say you go across the street and get candy. I’ll wait here." 1 "Let’s get candy at my house instead." 1 Note: the above solutions were awarded only a "1" since the plan considered was less complex, and the description of contingencies and how the child would deal with them was less detailed. "Say 'I can’t I’m going to a friend’s house." 0 "Say maybe, maybe not." 0 140 . I {a 0 e' H“; t- : : :1 e; ,7”, -1 ;_, e c These responses describe steps or prerequisite behaviors that would be taken in order to carry out the proposed problem solving strategy. This type of response may simply be a crude listing of behavior (awarded a 1 for planfulness), or a more complex and/or detailed description of problem solving steps (awarded a 2 for planfulness). ' e se W "I would say no, then run outside to my neighbor’s house, and wait there for my Mom and Dad to come home." 2 "No or else I’ll ride my bike home and tell my Mom and you’ll get in big trouble." 2 "no because my parents have to be with me or be watching me if we are going to cross the street. So let’s go get my Mom first." 2 "She should go to her Mom’s work and tell her Mom." 1 "Not answer and walk home." 1 "Say ’no’ and them just walk away." 1 "Go home and tell my parents." 1 "Say 'no’ go and tell." 1 "Tell my parents" "Leave" "Just walk away" 0000 "Go home" 141 Ql__IIflIEQ_QE_BE§EQN§E These responses include a description of the time at which the solutions, or steps in the solution, will be deployed. SEW ’ 08 21W "I would say no and wait in my room until my parents came home. Then I would tell them what happened. 2 "She could say ’I want to go to bed right now.’ And then wait for her Mom--to tell her everything." 2 "She should wait until she wasn’t looking and then run away real fast." 2 "Oh, no, not now, maybe tomorrow." 1 "Tell Dad when he gets home" 1 "Just walk away slowly and calmly." 1 "Run away real quick." 1 "Dial 911 super fast." 1 "Tell" 0 "Run" 0 "Walk away" 0 142 Dx__QQE§IDEBAIIQN_QE_§IIUAIIQNAL_EAQIQB§ In these responses a specific aspect of the setting or situation (i.e. an object or person that is likely to be present) is used to carry-out the strategy. The more complex or detailed use of situational factors, the higher the planfulness score. However, solutions which articulate the use of any specific aspect of the setting should be coded as at least 1. Use of the object/person must be explicitly stated; it cannot be implied. _il—EJLsCh' d’s Res 0 s W "She should tell some other older grownup to help her." 2 "She should go to her room and lock the door." 2 "She call her Mom to come get her, then wait outside where there are lots of people until her Mom comes." "Ride home on his bike." 1 "Turn off TV and put in new tape Lagd Befope Time," 1 "Hide under a table" "Say let’s play a game about monkey bars." 1 "Play with the cat." 1 "Say 'Oh these are nice plants, let’s look at these for a while." 1 "Slam the door so loud it scares her." 1 "Hide" 0 "Bite" 0 "Shoot him with a laser gun." O 143 SE N V : U ON EX BI RSIS NC 0 uestio Obsta le ue t'o 's so e o 'b' 't end persistence. Ratings on this dimension indicate the extent to which the child persists in generating alternative (i.e., flexible) resistance strategies in the face of his/her initial strategy failure. Ratings of solution flexibility/persistence should be entered in the "FLEX." eolupg and on only the "Question #5" pew. flete: "I don’t know" responses to question. #5 are coded as irrelevant (see pp. 8) In order to score flexibility and persistence, the coder must CAREFULLY compare the MAJOR content scoring for question #4 with the content scoring for question #5. ONLY THE MAJOR CONTENT CATEGORIES ARE USED FOR SCORING FLEXIBILITY. DO NOT PAY ATTENTION TO THE MINOR CONTENT CATEGORIES (i.e., the number to the right of the decimal point)! 2 = FLEXIBLE and PERSISTENT The child offers a response to question 45 which contains at least one MAJOR content code that was not present in his/her response to question 04. The response to question #5 must not be coded as irrelevant, compliant, or as passive resistance to receive this score. E] 1.1.! Re nse Content Code §core 4). I d say no 2.2 5). I’d yell "no" and run way. 2.2 4.1 2 4). I’d say no 2.2 5). I’d run away 4.1 2 4). I’d say no and go tell 2.2 3.1 5). I’d hit him 5 1 2 4). I’d shoot him with a laser gun. 1.1 5). I’d run and tell my Mom. 4.1 3.1 2 (Note: In order for a response to receive a "2" for flexibility, the content codes that appeared in question #4 d9 not have to be 144 duplicated in question 45. Instead, one new MAJOR content code which did not appear in question #4 must be present in question 05.) 1 = PERSISTENT The child offers a relevant to question #5 in which one or more of the content codes given for question #4 are duplicated and no new content codes are added. The response to question 45 must not be coded as irrelevant or compliant to receive this score. W Response Contegt Code Scepe 4). I'd say no 2.2 5). I’d yell "no" 2.2 1 4). I’d say no and run away 2.2 4.1 5). I’d leave 4.1 1 4). I’d say no and go tell 2.2 3.1 5). I’d scream for help 3.2 1 0 = LACK OF FLEXIBILITY AND PERSISTENCE A "0" should be given for flexibility if the child offers solutions to question 45 which are coded as any of the following: a. irrelevant (scores of 1.1 - 1.2) b. "I don’t know" ( " " " ) c. compliant (scores of 9.1 - 9.4) d. passive resistance (score of 7.2 only ) 2131111211112 Re 0 se Cont t de fieepe 4). I’d say no 2.2 5). I’d give in. 9.1 0 4). I’d say no and run away 2.2 4.1 5). I don’t know 1.2 O 145 4). I’d say no and go tell 2.2 4.1 3.1 5). I’d sting him with a bee 1.1 4). I’d say no 2.2 5). I’d say ’maybe’ 7.2 146 ON V : N NC 0 A U B R OF E V RNA VE fiIBAIEQlE§ EQB_EAQH_BE§EQE§E For each response, the number of relevant alternative strategies must be counted. Simply examine the MAJOR content codes for that response each MAJOR content category counts for ONE alternative solution except for the irrelevant and the compliant content categories. Enter the total number in the appropriate column of the score sheet. For example: We: 9211mm Cede 9_f_Al_t_s I would say no 2.2 1 I would say no and say I’m going to tell if you don’t stop. 2.2 3.1 2 I would say no and run away 2.2 4.1 2 147 SCORING SHEET PRESCHOOOL ALTERNATIVE SOLUTIONS TEST (PA§D Subjea I Scorer Initials Date CHEQK 95:: Pretest(-1) Posttest(-2) Foiiowup(=3) §TQRY 1 (.1) (peer/cmss street) Mm: Question 1: Question 2: Quesiton 3: Problem-sowing: NTENT DES EFF P N Question 4: Question 5: ' Question 6: TOTAL ALTS 148 5m; (-2) (peer/touch p.p.) Liming: Question 1: Question 2: Quesiton 3: Pmplem-geivigg: QQNTENT QOD§§ Question 4: Question 5: Question 6: TOTAL ALTS m m m 149 STORY 3 (-3) (adult/cross street) 53m: Question 1: Question 2: Quesiton 3: Problem-sowing: Question 4: Question 5: Question 6: TOTAL ALTS 150 51064.4 {-4) (adult/much p.p.) Mg: Question 1: Question 2: Quesiton 3: ngglemeelvigg: ONTENT ODES Question 4: Question 5: Question 6: TOTAL ALTS l 5 l Coded! Interviewer Initials ("FEMALE VERSION) Preschool Alternative Solutions Test (FAST) $19.31 ONE I‘m ' to tell you a story about two little ' ls, Becky and her friend. They were both three years aid. One day they were playing outsi and Becky’s friend said, 'I have 50 cents. Let's nose the street and get some canb!" But Becky isn’t allowed to cross that street with anyone but her parents. SO SHE DOESN'T WANT TO DO THAT. {55%; ForQueetz'one 1-3.beeuntomarhtheflntmwerthechtldgiv¢s-¢venifitis 1. Are Becky and her fi'iend both three years old? (Please CIRCLE one) Yes mandreadchoicesunkuchilddocsnotmpond.) No Don’t know 2. Is Becky's friend a boy or a girl? (Please CIRCLE one) Girl Boy Don’t know 3. What does Becky’s friend want her to do? m: If child has answered all 3 questions correctly. go directly to question #4. If child has answered any of the 3 questions incorrectly. repeat the story for the child and comet the child's misperceptions. It is attemdy important that the child understand these three facts of the story before proceeding to question #4.] 152 " What 90““ Becky Mm when her friend asks her to cross that street she’s not snowed to cross? mu; Rccord one response and then proceed immediately to question #6.] W'Ifchild doesnot respond area a. 'IDK'oeh. "Whotcould Becky my to her fl'ienderhaeeouldeheactifha-fi-iezudaehedhertoermaetreetehe’e not allowable-m? 5. 0.x,(nameofchfldxeoyooeaid8eckycouldfiepeatchfld'sreeponsetofl). But-oppose Beckyta'ied that andhern-ien' dkepton sayingAwe come on, Becky, come on! Let'ejuet go oanchretoreforaminnte! We! tcoquBeckydooreayifehex-eslly 153 0.3. (name of child) you’re doing a great job thinking so hard! So far you said Becky could (repeat child's responses). Now I want you to tell me ALL THE DIFFERENT THINGSBeckycouldeayordoifehe didn'twanttocroeetheetreettogotothe candy store. Wlfachddbegimwpcrwwutc.efivthrummmtiom.uopehildandmy. "OJ. (molehfld),thafehindoflihewhatyouthough38¢chyehoulddo ”fore. wmmmmpnoiomm emclihe.) I’dliheyouto mo, if'mmhbn mot-do. We something else W H'chfldeoye'mz' rompt eaying,’O.K(child’enamd.lct’ethii-1hmlly Moboutmhathchycoulddogmp staymdnpeatquattonfififl W Number responses] WWWWWPROWBYWG, "0.1., what else eonldBeehytuPOB’DLtehateleeeouldBeehyemya-dor‘”) 154 mm WDONTFORGETI ”amalgam cachetotyshouldbetheumundczumchfld I'm ' to tell you a story about two little girls, Molly and her friend. They are both three year-a old. One day when th were watchin T.V. author, the friend aaid to Molly, 'let'a touch private pate.“ BUT M LLY DIDN'T ANT DO THAT. (N013): FmeationelQbemtowhtheflntmthechildgivea—evenifitie wrong. 1. Are Molly and her fiend both three years old? (Please CIRCLE one) Yea Mdonatnadchoicamlaachilddoeemmnd.) No Don’t know 2. laMolly'afi'iendahayoragirl? CPleaae CIRCLE one) Girl Boy Don't know 3. What doea Molly's friend want her to do? W‘Ifchiflhnaamwcredallsqueetioueconectly.gotodinalytoquestian#4. If chddhammyofthesqwflomincmdy,npeathecmrfluchfld audcomctthechild’amiepcrocptzbne. Itieextnmelyimportantthatthe ' understand theaedmefaaaoftheuolybcfirepmceedingmquextionflJ 155 4. Remember Molly doesn’t want to touch private parta. So what could she do or say when her friend asks her to do that. my; Record one response and then proceed immediately to question #6.] W'Ifcfilddoesnotnxpondor 'IDK'nsh'WlsatoouldHoaysaytoher Mos-Meowuolbnctifher askedhertotouohpdoaeepartsr'l 5- 03-. (name a! child). so you said Molly could (repeat child’s respome to 94). But suppose Molly tried that and her friend kept saying. “Awe come on, Molly. come on! Let’s just touchprivetepartaforalittlewhile.‘ mnwhatcouldhlollydoorsayifshereally didn’t want to touch private pate. 156 8. QR (name of child) you’re doing a great job thinking so hard! So far you said Molly could (repent child’s responses). Now I want you to tell me ALL THE DIFFERENT THINGS Molly could say or do if she didn’t want to touch private parts with her fi'iend. W If a child begins to aerate, after three enumeration. stop child and say. "0.1 (name of child). that’s ind of lite what you thought Molly should do before. Whosethethmpreotouamponsesmallhe.) I’dlibeyoutotellrrse somethtn; else Molly could do orsayJ? W 0' child we 'IDK.‘ ro t som' . '03. (child's name). let's thin): really Moboutwhctfloflywulddoi‘f’gopfiumwnpmqmstion #6)} W‘ Number responses.) MW AFTER EVERY RESPONSE PROM BY SAYING. '0; what else could Holly try!" OR "0.1. wisest else could Holly say or dorwl 157 W I’m going to tell you a story about a little girl named Suzie and a grownup lady she knew. One day they were walking outside and the grownup said, ”Suzie, l have some money. So let's cross the street and get some candy!“ But Suzie isn’t allowed to cross that street with anyone but her parents. SO SHE DOESN'T WANT TO DO THAT. Lfléfiz‘; For Questions 1-3, besure tornarh theflntanatoerthechildgives—cven ifit is 1. la the grownup in this story older than Suzie? (Please CIRCLE one) Yes ME'donotreadchoicesunlesschilddoesnotrespoW No I Don't know 2. lsthegrownupinthe storyaladyoraman? (Please CIRCLE one) Man Lady Don’t know 3. What does the grownup want Suzie to do? [131912; If child has answered all 3 questions correctly. go directly to question N. If child has answered any of the 8 questions ineomctly, repeat the story [or the child and correct the child's misperceptions. It is eturernely important that the child understand these three facts of the story befitre proceeding to question #4.] 158 4. ghat could Suzie Mm when the lady asks her to cross that street she’s not allowed cross. MAL: Record one response and then proceed immediately to question #6.] W: Ifchild does not nepondorsays. 'lDK'ash. ”Whatcould Suzie say to the :dyorrhyosocouldsheactiftheladyashedhertoeroeeastreetshe’snotallowed cross 5. 0.11. (name of child). so you said Suzie could (repeat child's response to #4). But suppose Suzie tried that and the grownup kept on saying, “Awe come on, Suzie, come on! Let's just go to the candy store for a minute!“ THEE what could Suzie do or say if she really didn’t want to cross the street. 159 0.}: (name of child) you're doing a great job thinking so hard! So far you said Suzie could (repeatehild’s responses). Now I want you to tell me ALL THE DIFFERENT $68 Suzie could say or do if she didn’t want to cross the street to go to the candy W If a child begins to perseverate. after three enumerations. stop child and say. "0.1:. (name of child). that’s kind of lihe what you thought Suzie should do before. maplain how the threepreoious responses an alike.) I’d lthe you to thinh ofall $1.?“ things Susse could say or do. What’s something else Suzie could W 11’ child says 'mz' prompt by saying. '03. (child’s name). let’s thinh really hard about what Suzie could do if(paraphrase story and repeat question #6).' O (N01113: Number narrow.) [mm AFTER EVERY RESPONSE PROM BY SAYHVG. 'OL who! else could Susie try?" on "01.. what else could Suzie say or do?""”] 160 mm This is a story about a little girl named Linda and a grownup lady she knows. One day they were reading a hook together and the grownup said to Linda, “Linda. let's touch private parts.“ BUT LINDA DIDN'T WANT TO DO THAT. %' For Questions L3. be sure to mark the first answer the child gives-even if it is 1. Is the grownup in this story older than Linda? (Please CIRCLE one) Yes WdonMnadchoicesunlesschilddoesnotrsspondJ No Don‘t know 2. Isthegrownupinthe storyaladyoraman? (Please CIRCLE one) Man Lady Don’t know 3. What does the grownup want to do? mIfchildhasansweredall3questionscorrectly.godirectlytoquestion#4.Ifchild hasansweredanyofthe3questionsincorrectly,repeatthesto forthechildandcorrect the child's misperceptions. It is extremely important that the «Zild understand these three factsofthestorybeforeproceedingtoquesttbnul 4. 161 Remember, Linda doesn't want to touch private parts. So what could she dam when the lady asks her to do that? W‘ Record one response and then proceed immediately to question #6.] Wlfchilddoesnotnspondorsays. 'meash, "Whatoouldldndasaytothe ladyorhoseeouldsheactifthisladyashedhertotouchprioatepartsr'l 0.3.. (name d child), you said Linda could (repeat child's response to 64). But suppose so triedthstandthe hcptsaying. ”Awe comeomlsinda. come on! Let'sjust privatepartsfora' ewhile.‘ THENwhatcouldLindadoorsayifshereally E i a 6. 162 on (name of child) you're doing a great job thinking so hard! So far you said Linda could (repeat child’s responses). Now I want you to tell me ALL THE DIFFERENT THINGS Linda could say or do if she didn’t want to touch private parts with the grownup. W If a child begins to perseverate. after three enumerations. stop child and say. "0.1. (name of child), that's hind of lihe what you thought Linda should do hour thethrveepreoious responsesansalihe.) l’dliheyouto before. (Explain . thinh ofall the dim things Linda could say or do. What’s somethsng else Linda could dot? W 3’ child says may prompt asying. '01 (child's name). let's thinh really hardaboutwhatLindacoulddoifwamp nayartdrepeotquestiontwf W“ Number responses.) WW AFTER EVERY RESPONSE PROM BY SAYHVG. '01.. what else oouldLirsdatryPOR "0.8., whateleeeouldLindasayordo?""] 163 164 165 166 ../\.l M)... «xmurfim \ svnvcflllll - wV. r /.p «4......6 .. .. I . e . s\\... e. awrauerf. . ..g at, ,, a. .4... e 1_ _, ,. Met-33?: .\ \\\.\ \xx..... PARENT MEETING FOR THIS PROGRAM WILL TAKE PLACE ON THURSDAY, FEBRUARY 8, 7:00 PM AT LESLIE NURSERY Dear Parents, Leslie Nursery is concerned about all types of personal safety that involve children. He routinely teach children safety rules about fire and crossing the street. Last year we offered a sexual abuse prevention program for our children, and because it was so successful, we are again offering it this year. We believe it is important to teach children to recognize threatening situations, should they arise, and to seek help. This is a personal safety program (The Grossmont College Child Sexual Abuse Prevention Program) that teaches children a "NO, Go, Tell” message; 11 is not a sex education pzogzgg. It does not provide any explicit information about human sexuality. The program starts with an optional parent meeting, Parents get to learn more about the problem of sexual abuse and how to talk about it with their children. There is also a parent—child workbook that can be used to reinforce the message of the program at home. The program for the children takes palace in the classroom and involves five 20 minute presentations that include a puppet show, picture stories, and activities. We are presenting this program in cooperations with Dr. Marianne McGrath and Dr. Anne Bogat, or one of their trained assistants. Children will be asked if they want to answer some questions about personal safety. They’ll be told that there are no right or wrong answers; we’d just like to know what they think. The questions ask children to identify appropriate and inappropriate touch and to state what they would say and do in various pretend situations. Half of the children who have permission to participate will receive the program first, the other half will receive it several weeks later. This procedure makes it possible to conduct the program in smaller groups and also to determine what children already know about the topic, how much of what they know is correct, and how much they learn from the program. 0 ow ow o w c pagticipa g, please complete the attached 29:; and :gfigzn 1; o ' ’ to c e NU . You may keep this page. If you have questions about the program, please feel free to discuss these with Adrienne Bigg before completing the form, or to leave a message for Dr. McGrath or Dr. Bogat at 353-8690 and they’ll return your call and answer your questions. Remember, 0 c d a permission. And, if you decide to let your child participate, there will be a parent meeting before the sexual abuse prevention program begins. 167 Parent Permission Form Sexual Abuse Prevention Program 1. I understand the evaluation process of the Grossmont College Child Sexual Abuse Prevention Program. I have read the description of the project and I understand what my child’s participation will involve. 2. I understand my child’s participation is optional and that my child must also give his/her verbal permission. 3. I understand that either myself or my child may discontinue my child’s participation in the evaluation at any time without penalty and that my child will still have the opportunity to participate in the Grossmont Child Sexual Abuse Prevention Program. 4. I understand that all results of the study will be kept in strict confidence and all responses of my son or daughter will remain anonymous except if my child reports possible sexual abuse. If such a report occurs, Dr. McGrath or Dr. Bogat will notify both myself and the preschool teacher immediately. Within these restrictions, the results of the study will be made available to me at my request. 5. I understand that participation in the study does not guarantee additional benefits to my child or to me. I hereby give my approval to allow my child to be interviewed about his/her knowledge of personal safety. I understand that this information will be used as part of a research project being conducted by Dr. Marianne McGrath and Dr. Anne Bogat at Michigan State University. NAME OF CHILD BIRTHDATE SIGNATURE OF PARENT/GUARDIAN DATE ------ > Please complete a separate form for each child in your falily who you want to participate. lttttt ------ > RETURN THIS FORM TO YOUR CHILD’S TEACHER AS SOON AS POSSIBLE. THANKS! 168 APPENDIX C Letter of Invitation/Parental Consent Form APPENDIX D Tables 169 Table l C I 525 ! I . E5! 2 v !i 5! I. Study Authors Components of SPS Goal Encoding of Access to Strategy Selection Social Context Knowledge Implementation Binder -- -- Y -- et al., (1987) Blumberg -- Y -- -- et al., (1988) Borkin -- -- Y -- a Frank, (1986) Conte -- -- Y -- et al., (1985) Christian -- Y Y -- et al., (1988) Daro et al., -- Y Y Y (1987) Garbarino, -- Y Y -- (1987) Harvey -- Y Y -- et al., (1988) Hazzard -- Y Y -- et al., (in press) Table l (cont’d) 170 Study Authors Components of SPS Goal Selection Encoding of Access to Strategy Social Context Knowledge Implementation Hill a C Jason, (1987) Kenning -- et al., (1987) Kolko -- et al., (1987) Miltenberger -- & Thiesse- Duffy, (1988) Nelson, -- (1981) Plummer, -- (1984) Ratto a -- Bogat, (1990) Bay a -- Dietzel, (1988) Saslawsky s -- Wurtele, (1986) Sigurdson -- et al., (1987) -- Y Y 171 Table 1 (cont’d) Study Authors Components of SPS Goal Encoding of Access to Strategy Selection Social Context Knowledge Implementation Swan -- Y Y -- et al., (1985) Wall, -- -- -- -- (1983) Wolf -- -- Y -- et al., (1986) -- y y -- Wurtele (1990) Wurtele -- Y Y -- et al., (1986) Wurtele -- Y y -- et al., (1987) Wurtele -- Y Y -- et al., (1989) Nate: Dashes (--) indicate variable was not assessed. Y indicates that variable was assessed. C indicates that variable was controlled. Table 2 .""-o 0 Studies I . ’ 172 Study Authors Strategy Selection Variables Strategy Quantity Strategy Quality Responsiveness to Social Context Binder -- et al., (1987) Blumberg -- et al., (1988) Borkin -- & Frank, (1986) Christian Y et al., (1988) Daro et al., Y (1987) Conte -- et al., (1985) Garbarino, -- (1987) Harvey Y et al., (1988) Table 2 (cont’d) 173 Study Authors Strategy Selection Variables Strategy Quantity Strategy Quality Responsiveness to Social Context Hazzard -- et al., (in press) Hill a -- Jason, (1987) Kenning ~- et al., (1987) Kolko -- et al., (1987) Miltenberger -— a Thiesse- Duffy, (1988) Nelson, -- (1981) Plummer, -— (1984) Ratto a -- Bogat, (1990) Ray a -- Dietzel, (1988) Saslawsky a -- Wurtele, (1986) 174 Table 2 (cont'd) Study Authors Strategy Selection Variables Strategy Strategy Responsiveness to Quantity Quality Social Context Sigurdson -- -- -- et al., (1987) Swan -- -- -- et al., (1985) v.11. -- -- -- (1983) Wolf -- -- -- et al., (1986) Wurtele -- Y -- (1990) Wurtele -- Y -- et al., (1986) Wurtele -- Y ~- et al., (1987) Wurtele -- Y -- et al., (1989) Ngtg: Dashes (--) indicate variable was not assessed. Y indicates that variable was assessed. C indicates that variable was controlled. 175 Study Authors Variables Related Recursive Aspects of SPS Persistence After Strategy Failure Flexibility After Strategy Failure Strategy Sequencing Binder et al., (1987) Blumberg et al., (1988) Borkin a Frank, (1986) Christian et al., (1988) Daro et al., (1987) Conte et al., (1985) Garbarino, (1987) Harvey et al., (1988) Table 3 (cont’d) 176 Study Authors Variables Related Recursive Aspects of SPS Persistence After Flexibility After Strategy Strategy Failure Strategy Failure Sequencing Hazzard -- -- -- et al., (in press) Hill & Jason, (1987) Kenning et al., (1987) Kolko et al., (1987) Miltenberger & Thiesse- Duffy, (1988) Nelson, (1981) Plummer, (1984) Ratto & Bogat, (1990) Ray a Dietzel, (1988) Saslawsky a Wurtele, (1986) 177 Table 3 (cont’d) Study Authors Variables Related Recursive Aspects of SPS Persistence After Flexibility After Strategy Strategy Failure Strategy Failure Sequencing Sigurdson -- -- -- et al., (1987) Swan -- -- -- et al., (1985) v.11. —- -- -- (1933) Wolf -- -- -- et al., (1986) Wurtele -- -- -- (1990) Wurtele -- -- -- et al., (1986) Wurtele -- -- -- et al., (1987) Wurtele -- -- -- et al., (1989) flgtg: Dashes (--) indicate variable was not assessed. Y indicates that variable was assessed. C indicates that variable was controlled. Table 4 178 Number of Responses Coded Using Original Strategy Categories for Each Situation Strategy Situational Factors Category Benign Request Sexual Request Child Adult Child Adult Nonsense Irrelevant 21 (11.5%) 21 (12.1%) 20 (12.6%) 19 (11.9%) Vague 4 ( 2.2%) 5 ( 2.9%) 2 ( 1.3%) 5 ( 3.1%) Assertive Subject’s Rights 0 ( 0.0%) 0 ( 0.0%) 2 ( 1.3%) 0 ( 0.0%) Subject’s Intent 43 (23.5%) 39 (22.5%) 38 (23.9%) 39 (24.5%) Command 7 ( 3.8%) 6 ( 3.5%) 6 ( 3.8%) 2 ( 3.1%) Nonviolent Threat 4 ( 2.2%) 2 ( 1.2%) 7 ( 4.4%) 2 ( 1.3%) Help-seeking Telling Another 15 ( 8.2%) 10 ( 5.8%) 11 ( 6.9%) 13 ( 8.2%) Solicit Aid 9 ( 4.9%) 7 ( 4.0%) 1 ( 0.6%) 2 ( 1.3%) Solicit Advice 2 ( 1.1%) 0 ( 0.0%) 0 ( 0.0%) 0 ( 0.0%) Non- confrontational 22 (12.0%) 24 (13.9%) 29 (18.2%) 30 (18.9%) Aggressive Direct 2 ( 1.1%) 2 ( 1.2%) 0 ( 0.0%) 5 ( 3.1%) 3rd Party 0 ( 0.0%) 0 ( 0.0%) 0 ( 0.0%) 0 ( 0.0%) Violent Verbal Threats 2 ( 1.1%) 1 ( 0.6%) 3 ( 1.9%) 2 ( 1.3%) Finagle Bribe 0 ( 0.0%) 0 ( 0.0%) 0 ( 0.0%) 0 ( 0.0%) Deal--Solicit Aid 0 ( 0.0%) 1 ( 0.6%) 0 ( 0.0%) 0 ( 0.0%) Deal--Distract 1 ( 0.5%) 1 ( 0.6%) 4 ( 2.5%) 2 ( 1.3%) Restructure Situation 4 ( 2.2%) 8 ( 4.6%) 0 ( 0.0%) l ( 0.6%) Table 4 (cont’d) 179 Strategy Category Situational Factors Benign Request Sexual Request Child Adult Child Adult Finagle (cont’d) Stalling 1 ( 0.5%) 0 ( 0.0%) 0 ( 0.0%) 1 ( 0.6%) Invoke a Rule 20 (10.9%) 16 ( 9.2%) 5 ( 3.1%) 6 ( 3.8%) Make Excuse 3 ( 1.6%) 3 ( 1.7%) 0 ( 0.0%) 1 ( 0.6%) Distraction 0 ( 0.0%) 3 ( 1.7%) 8 ( 5.0%) 3 ( 1.9%) Passive Ignore 1 0.5%) 1 ( 0.6%) 6 ( 3.8%) 4 ( 2.5%) Equivocate 5 2.7%) 7 ( 4.0%) 5 ( 3.1%) 7 ( 4.4%) Manipulate Affect Own Affect 2 1.1%) 1 ( 0.6%) 0 ( 0.0%) 1 ( 0.6%) Perpetrator’s Affect 0 0.0%) 0 ( 0.0%) 0 ( 0.0%) 0 ( 0.0%) Compliance Active 15 8.2%) 14 ( 8.1%) 11 ( 6.9%) 11 ( 6.9%) Change Goal 0 0.0%) 1 ( 0.6%) 1 ( 0.6%) 2 ( 1.3%) Meet Needs Later 0 0.0%) 0 ( 0.0%) 0 ( 0.0%) 0 ( 0.0%) Delayed Compliance 0 0.0%) 0 ( 0.0%) 0 ( 0.0%) 0 ( 0.0%) fiotg: The percentage of the total number of responses for each situation appears in parentheses. 180 Table 5 Formation of 10 Strategy Categories from Original 29 Strategy Categories 1). 2). 3). 4). 5). 6). 7). 8). 9). ‘Nonsense Solutions Irrelevant Vague *Assertive Solutions Assertion of Subject’s Rights Assertion of Intention Not to Comply Other-Directed Command to Stop Nonviolent Threat Help-Seeking Solutions Telling Another Person Solicitation of Aid Solicitation of 3rd Party Advice Make a Deal-~solicitation of 3rd party advice I"don-confrontational Solutions Subject Leaves the Scene I'Aggressive Solutions Direct Physical Harm to Person or Possessions 3rd Party Assault Verbal Threats that Escalate Situation Finagle Solutions Bribe Make Excuses Make a Deal-~distraction Distraction Stalling Invoke a Rule or Authority’s Command Solutions Make a Deal--Restructure Situation to Meet Both Parties’ Needs Passive Resistance/Manipulate Affect Solutions Ignore Passive, Equivocation, Slight Resistance Manipulate Own Affect Manipulate Perpetrator’s Affect 181 Table 5 (cont’d) 10). *Compliance Solutions Active Compliance Change Goal Meets Own Needs Later Delayed Compliance flgtg: 4 Indicates that category is the same as original scoring category. Table 6 Means and Standard Deviations 182 for Five SPS Variables Strategy Situational Factors Category Benign Request Sexual Request Child Adult Child Adult Quantity of Solutions Boys 2.34 (.94) 2.25 (.99) 1.88 (.87) 1.94 (.95) Girls 2.40 (.99) 2.23 (.98) 2.03 (.99) 2.17 (.95) Total 2.37 (.99) 2.24 (.99) 1.95 (.93) 2.05 (.95) Encoding Boys 2.71 (.53) 2.74 (.63) 2.77 (.50) 2.94 (.25) Girls 2.83 (.46) 2.93 (.25) 2.77 (.57) 2.90 (.31) Total 2.77 (.50) 2.84 (.49) 2.77 (.53) 2.92 (.27) Strategy Effectiveness Boys 1.90 (.71) 1.87 (.63) 1.84 (.58) 1.71 (.72) Girls 2.05 (.60) 1.98 (.77) 2.16 (.69) 2.03 (.66) Total 1.97 (.70) 1.92 (.70) 1.99 (.65) 1.87 ( 71) Strategy Planfulness Boys 0.11 (.19) 0.17 (.24) 0.14 (.26) 0.18 (.29) Girls 0.21 (.26) 0.31 (.45) 0.31 (.36) 0.32 (.36) Total 0.15 (.23) 0.24 (.36) 0.22 (.32) 0.25 (.33) Strategy Flexibility Boys 1.05 (.92) 1.33 (.86) 1.24 (.83) 1.14 (.85) Girls 1.21 (.98) 1.26 (.93) 1.37 (.76) 1.47 (.61) Total 1.13 (.94) 1.30 (.88) 1.30 (.79) 1.30 (.76) Note: Means followed by Standard Deviations in parentheses. Quantity score range 8 0 - 5 183 Table 7 Repeated-Measures Analysis of Variance for Encoding Source of Variation SS DF M8 2 2 Gender of Subject .28 1 .28 .66 .421 Age of Partner .69 1 .69 6.26 .015 Nature of Request .10 1 .10 .65 .424 Gender of Subject X Nature of Request .49 1 .49 3.03 .105 Gender of Subject X Age of Partner .01 1 .01 .05 .816 Age of Partner X Nature of Request .10 1 .10 .65 .423 Gender of Subject X Nature of Request X Age of Partner .03 1 .03 .23 .636 Within Subjects Error Age of Partner 6.55 59 .11 Nature of Request 8.66 59 .15 Between Subjects Error Gender of Subject 25.39 59 .66 Table 8 Repeated-Measures Analysis of Variance for 184 Quantity of Solutions Source of Variation 88 DF MS I 2 Gender of Subject .71 1 .71 .30 .586 Age of Partner .02 1 .02 .03 .860 Nature of Request 5.71 1 5.71 11.16 .001 Gender of Subject X Nature of Request .47 1 .47 .92 .342 Gender of Subject X Age of Partner .00 1 .00 .00 .995 Age of Partner X Nature of Request .81 1 .81 1.55 .218 Gender of Subject X Nature of Request X Age of Partner .08 1 .08 .15 .696 Within Subjects Error Age of Partner 30.98 60 .52 Request of Request 30.71 60 .51 Between Subjects Error Gender of Subject 141.47 60 2.36 185 Table 9 Means and Standard Deviations for Total Number of Solutions Across Situations Strategy Category Boys Girls Total Assertive X 2.94 2.60 2.77 SD 1.37 1.45 1.41 Noneconfrontational X 1.28 2.13 1.69 SD 1.35 1.46 1.46 Nonsense X 1.53 1.47 1.50 SD 1.44 1.17 1.30 Help-Seeking X 0.97 1.13 1.05 SD 1.18 1 53 1.35 Compliance X 0.91 0.83 0.87 SD 1.25 1.12 1.18 Rule-Based X 0.88 0.63 0.76 SD 1.10 0.77 0.95 Passive X 0.66 0.50 0.58 SD 1.00 0.82 0.92 Finagle X 0.31 0.53 0.42 SD 0.69 0.86 0.78 Aggressive X 0.31 0.20 0.26 SD 0.82 0.55 0.70 Restructure X 0.16 0.27 0.21 SD 0.45 0.52 0.48 Table 10 186 Repeated-Measures Analysis of Variance for Solution Type Source of Variation SS DF MS 2 2 Gender of Subject .20 1 .20 .20 .658 Type of Solution 349.46 9 38.83 31.81 .0001 Gender of Subject 15.79 9 1.75 1.44 .169 X Type of Solution Within Subjects Error Type of Solution 659.24 540 1.22 Between Subjects Error Gender of Subject 61.82 60 1.03 Table 11 Means and Standard Deviations for Four Strategy Categories 187 Strategy Situational Factors Category Benign Request Sexual Request Child Adult Child Adult Non-confrontational Strategies Boys .344 (.48) .312 (.47) .312 (.47) .312 (.47) Girls .400 (.49) .467 (.50) .663 (.49) .667 (.48) Total .355 (.48) .387 (.49) .468 (.50) .484 (.50) Help-Seeking Strategies Boys .375 (.49) .281 (.46) .125 (.34) .187 (.40) Girls .367 (.49) .233 (.43) .267 (.45) .267 (.45) Total .371 (.49) .258 (.44) .194 (.53) .226 (.42) Assertive Strategies Boys .781 (.42) .750 (.44) .719 (.46) .687 (.47) Girls .800 (.41) .633 (.49) .600 (.50) .567 (.50) Total .790 (.41) .694 (.47) .661 (.48) .629 (.49) Rule-Based Strategies Boys .312 (.47) .281 (.46) .156 (.37) .125 (.34) Girls .333 (.48) .233 (.43) .000 (.00) .067 (.25) Total .323 (.47) .258 (.44) .081 (.28) .097 (.30) 188 Table 12 Repeated-Measures Analysis of Variance for Non-confrontational Solutions Source of Variation 88 DF MS 2 2 Gender of Subject 2.81 1 2.81 5.72 .020 Age of Partner .04 1 .04 .25 .618 Nature of Request .73 1 .73 6.19 .016 Gender of Subject X Nature of Request .96 1 .96 8.10 .006 Gender of Subject X Age of Partner .10 1 .10 .65 .422 Age of Subject X Nature of Request .00 1 .00 .03 .863 Gender of Subject X Nature of Request X Age of Partner .04 1 .04 .23 .634 Within Subjects Error Age of Partner 9.61 60 .16 Nature of Request 7.11 60 .12 Between Subjects Error Gender of Subject 29.48 60 .49 189 Table 13 Repeated-Measures Analysis of Variance for Help-Seeking Solutions Source of Variation SS DF MS I 2 Gender of Subject .10 1 .10 .23 .635 Age of Partner .10 1 .10 .83 .367 Nature of Request .65 1 .65 8.20 .006 Gender of Subject X Nature of Request .30 1 .30 3.74 .058 Gender of Subject X Age of Partner .04 1 .04 .32 .575 Age of Partner X Nature of Request .32 1 .32 3.03 .087 Gender of Subject X Nature of Request X Age of Partner .00 1 .00 .02 .891 Within Subjects Error Age of Partner 7.61 60 .13 Nature of Request 4.77 60 .08 Between Subjects Error Gender of Subject 27.61 60 .46 190 Table 14 Repeated-Measures Analysis of Variance for Assertive Solutions Source of Variation SS DF MS 2 2 Gender of Subject .44 1 .44 .89 .350 Age of Partner .27 l .27 1.96 .167 Nature of Request .59 1 .59 5.21 .026 Gender of Subject X Nature of Request .08 1 .08 .68 .412 Gender of Subject X Age of Partner .07 1 .07 .54 .466 Age of Partner X Nature of Request .07 1 .07 .65 .424 Gender of Subject X Nature of Request X Age of Partner .07 1 .07 .65 .424 Within Subjects Error Age of Partner 8.17 60 .14 Nature of Request 6.84 60 .11 Between Subjects Error Gender of Subject 29.77 60 .50 Table 15 Repeated-Measures Analysis of Variance for Rule-Based Solutions 191 Source of Variation SS DF MS E 2 Gender of Subject .23 1 .23 1.00 .322 Age of Partner .04 1 .04 .28 .601 Nature of Request 2.56 1 2.56 21.61 .000 Gender of Subject X Nature of Request .14 1 .14 1.15 .288 Gender of Subject X Age of Partner .00 1 .00 .03 .873 Age of Partner X Nature of Request .11 1 .11 .99 .325 Gender of Subject X Nature of Request X Age of Partner .11 1 .11 .99 .325 Within Subjects Error Age of Partner 7.17 60 .13 Nature of Request 7.09 60 .12 Between Subjects Error Gender of Subject 13.62 60 .23 192 Table 16 Repeated-Measures Analysis of Variance for Nonsense Solutions Source of Variation SS DF MS E 2 Gender of Subject .02 1 .02 .04 .847 Age of Partner .01 1 .01 .03 .860 Nature of Request .00 1 .00 .01 .914 Gender of Subject X Nature of Request .32 1 .32 1.87 .177 Gender of Subject X Age of Partner .20 1 .20 1.04 .313 Age of Partner X Nature of Request .04 1 .04 .25 .618 Gender of Subject X Nature of Request X Age of Partner .10 1 .10 .65 .422 Within Subjects Error Age of Partner 11.56 60 .19 Nature of Request 10.42 60 .17 Between Subjects Error Gender of Subject 25.86 60 .43 Table 17 Repeated-Measures Analysis of Variance for Aggressive Solutions 193 Source of Variation 88 DF MS I 2 Gender of Subject .05 1 .05 .40 .531 Age of Partner .06 1 .06 .83 .366 Nature of Request .06 1 .06 2.70 .106 Gender of Subject X Nature of Request .00 1 .00 .00 .958 Gender of Subject X Age of Partner .06 1 .06 .83 .366 Age of Partner X Nature of Request .06 1 .06 2.64 .109 Gender of Subject X Nature of Request X Age of Partner .06 1 .06 2.64 .109 Within Subjects Error Age of Partner 4.37 60 .07 Nature of Request 1.44 60 .02 Between Subjects Error Gender of Subject 7.42 60 .12 194 Table 18 Repeated-Measures Analysis of Variance for Finagle Solutions Source of Variation SS DF MS E 2 Gender of Subject .19 1 .19 1.25 .269 Age of Partner .02 1 .02 .24 .625 Nature of Request .15 1 .15 2.51 .118 Gender of Subject X Nature of Request .15 1 .15 2.51 .118 Gender of Subject X Age of Partner .15 1 .15 1.84 .180 Age of Partner X Nature of Request .14 1 .14 1.76 .189 Gender of Subject X Nature of Request X Age of Partner .01 1 .01 .16 .688 Within Subjects Error Age of Partner 4.84 60 .08 Nature of Request 3.70 60 .06 Between Subjects Error Gender of Subject 9.09 60 .15 195 Table 19 Repeated-Measures Analysis of Variance for Make a Deal -- Restructure Situation to Meet Both Parties’ Needs Solutions Source of Variation SS DF MS E 2 Gender of Subject .05 1 .05 .80 .373 Age of Partner .10 1 .10 2.02 .160 Nature of Request .49 l .49 10.70 .002. Gender of Subject X Nature of Request .01 1 .01 .16 .689 Gender of Subject X Age of Partner .04 1 .04 .78 .381 Age of Partner X Nature of Request .04 1 .04 1.01 .319 Gender of Subject X Nature of Request X Age of Partner .00 1 .00 .13 .718 Within Subjects Error Age of Partner 3.11 60 .05 Nature of Request 2.75 60 .05 Between Subjects Error Gender of Subject 3.52 60 .06 ‘ Ngtg: Since no solutions of this type were offered for one cell, ANOVA computations are not valid. 196 Table 20 Repeated-Measures Analysis of Variance for Passive Solutions Source of Variation SS DF MS I 2 Gender of Subject .09 1 .09 .45 .506 Age of Partner .07 1 .07 1.02 .316 Nature of Request .06 l .06 .45 .505 Gender of Subject X Nature of Request .14 1 .14 1.07 .305 Gender of Subject X Age of Partner .02 1 .02 .28 .598 Age of Partner X Nature of Request .00 1 .00 .00 .969 Gender of Subject X Nature of Request X Age of Partner .15 1 .15 1.49 .227 Within Subjects Error Age of Partner 3.92 60 .07 Nature of Request 7.80 60 .13 Between Subjects Error Gender of Subject 12.68 60 .12 197 Table 21 Repeated-Measures Analysis of Variance for Compliance Solutions Source of Variation 88 DF MS 2 2 Gender of Subject .02 l .02 .06 .810 Age of Partner .00 1 .00 .00 .976 Nature of Request .07 1 .07 1.02 .316 Gender of Subject X Nature of Request .02 1 .02 .28 .598 Gender of Subject X Age of Partner .15 1 .15 .88 .351 Age of Partner X Nature of Request .01 1 .01 .12 .728 Gender of Subject X Nature of Request X Age of Partner .06 1 .06 .54 .464 Within Subjects Error Age of Partner 9.85 60 .16 Nature of Request 3.92 60 .07 Between Subjects Error Gender of Subject 21.22 60 .35 Table 22 198 Repeated-Measures Analysis of Variance for Effectiveness Source of Variation SS DF MS E 2 Gender of Subject 2.87 1 2.87 2.73 .104 Age of Partner .46 1 .46 1.46 .232 Nature of Request .01 1 .01 .04 .849 Gender of Subject X Nature of Request .49 1 .49 2.33 .133 Gender of Subject X Age of Partner .01 1 .01 .02 .884 Age of Partner X Nature of Request .09 1 .09 .29 .595 Gender of Subject X Nature of Request X Age of Partner .01 1 .01 .03 .863 Within Subjects Error Age of Partner 16.97 54 .31 Nature of Request 11.45 54 .21 Between Subjects Error Gender of Subject 56.68 54 1.05 199 Table 23 Repeated—Measures Analysis of Variance for Planfulness Source of Variation 88 DF MS 2 2 Gender of Subject 1.04 1 1.04 6.02 .017 Age of Partner .16 1 .16 2.41 .126 Nature of Request .09 1 .09 1.26 .266 Gender of Subject X Nature of Request .01 1 .01 .19 .667 Gender of Subject X Age of Partner .00 1 .00 .01 .926 Age of Partner X Nature of Request .05 1 .05 .66 .420 Gender of Subject X Nature of Request X Age of Partner .02 1 .02 .21 .646 Within Subjects Error Age of Partner 3.54 54 .07 Nature of Request 3.70 54 .07 Between Subjects Error Gender of Subject 9.37 54 .17 200 Table 24 Repeated-Measures Analysis of Variance for Problem Solving Flexibility Source of Variation 88 DF MS 2 2 Gender of Subject .77 l .77 .55 .461 Age of Partner .30 1 .30 .74 .394 Nature of Request .34 1 .34 .80 .377 Gender of Subject X Nature of Request .34 1 .34 .80 .377 Gender of Subject X Age of Partner .00 1 .00 .01 .936 Age of Partner X Nature of Request .27 1 .27 .39 .534 Gender of Subject X Nature of Request X Age of Partner .47 1 .47 .69 .413 Within Subjects Error Age of Partner 15.44 38 .41 Nature of Request 16.11 38 .42 Between Subjects Error Gender of Subject 52.48 38 1.38 APPENDIX E Figures 201 [ SELECTS GOAL EXAMINES THE "TASK ENVIRONMENT" i.e.,TARGET CHARACTERISTICS, SOCIAL CONTEXT l I ACCESSES INFORMATION CONCERNING STRATEGIES USED IN EARLIER I AND/OR SIMILAR SITUATIONS TO REACH GOAL SUCCESSFULLY [ STRATEGY SELECTION [— g _ [_ STRATEGY IMPLEMENTATION J I I _ ‘ OUTCOME 1 L I ¥ A._ [ _ [SUCCESS—J [ FAILURE ‘ g I l I l u REPEAT STOP MODIFY a). automatic a). automatic reflective revision b). reflective b). reflective ELQUB£_1: Rubin & Krasnor’s (1986) model of social problem solving 202 0.7 - ems a 0.6 - C .2 E o 0.5 " a) ‘5 h 0.4 " 0 .0 E :z— :1 BOYS 2 0.3 ~ 2 G é’ 0.2 ~ 0.1 Benign Sexually Abusive Nature of Request 5mm Gender of Subject X Nature of Request Interaction for Non-confrontational Solutions 0.4 - a c .2 é o 0.3 " a) ‘5 ems b o a E a z 0.2- c a o 5 BOYS 0.1 Non-sexual Sexually Abusive Nature of Request ENABLE—3 Gender of Subject X Nature of Request Interaction for Help-Seeking Solutions 204 0.4 - n c .2 2.5 o 0.3 J 0) "5 h- B E ADULT 2 0.2 ‘ CHILD :: a O 2 0.1 Non-sexual Sexually Abusive Nature of Request Eigu re 4 Partner Age X Nature of Request Interaction for Help-Seeking Solutions L I ST OF REFERENCES LIST OF REFERENCES Abelson, R. P. (1981). Psychological status of the scrip concept- Asszigan_£azghslgsiat. 36. 515-729- Asher. S. R., & Renshaw, P. (1981). Children without friends: Social knowledge and social skill. In S. R. Asher & J. M. Gottman (Eds.), v children’s leendshlpg (pp. 273-296). New York: Cambridge. Barnett, R., Darcie, 6., Holland, C., & Kobasigawa, A. (1981). Children’s cognitions about effective helping. Developmehtal Psychplogy, l_, 262-277. Behana, N., Gamble, N., & Stevens, C. (1985). up, g9, te l: h' ’s ' e a out h' e prevention. El Cajon, Ca: Grossmont College Sexual Abuse Prevention Program. Bierman, K. L., & Furman, W. (1984). The effects of social skills training and peer involvement on the social adjustment of preadolescent. thlg_ngxglgp;gnt, 55. 151-162. Binder, R. L., & McNiel, D. E. (1987). Evaluation of a school-based sexual abuse prevention program: Cognitive and emotional effects. Child Abpgg ppd Neglect, 11, 497-506. Blumberg, E. J., Chadwick, M. W., Fogarty, L. A., Speth, T. W-. & Chadwick. D- L- (1988). Ihe.§220_lgushzfiad ou Co 0 e t e u v i ' ' Unanticipgted Epsitivg Congequencgp. Unpublished manuscript. (Available from Elaine J. Blumberg, Center for Child Protection, Children’s Hospital and Health Center, 8001 Frost Street, San Diego, Ca 92123). Borkin, J., 6 Frank, L. (1986). Sexual abuse prevention for preschoolers: A pilot program. Child_!elfazg, 65, 75.82 s Caplan M. 2., Weissberg, R. P., Bersoff, D. M., Ezekowitz, R., & Walla. M. L. (1986). The_aiddls_sshssl altsrnstixe_a2luti2na_tsst_LA§Il_asgzins_asnusl- Unpublished manuscript, Yale University, New Haven, CT. 205 206 Caplan M. 2., Weissberg, R. P., Grober, J. S., Sivo, P. J., Grady. K.. & Jacoby. c. (1989). Sgsial_sgsnstanse o ot'o t ate ev u s ' Outcome f ndin s wit nne - pdolescentg. Manuscript submitted for publication, Yale University, New Haven, CT. Christian, R., Dwyer, S., Schumn, W. R., & Coulson, L. A. (1988). Prevention of sexual abuse for preschoolers: Evaluation of a pilot prostaao Bsxshslssissl_fisngrts. __6_Z’ 387-3960 Cohn, L. D. (1991). Sex differences in the course of personality development: A meta-analysis. WM!!! .19—9.: 252'2660 Committee for Children. (1983). i bou o ' : A personal safety curpiculup. (Available from the Committee for Children, P.O. Box 15190, Seattle, WA 98115). Conte, J. R. (1987). Ethical issues in evaluation of prevention programs. Child A u e d Ne ct, ll, 171-172. Conte, J. R., Rosen, C., Saperstein, L., & Shermack, R. (1985). An evaluation of a program to prevent the sexual victimization of young children. thld_Apggg Conte, J. R., Rosen, C., & Saperstein, L. (1986). An analysis of programs to prevent the sexual victimization of children. lpgphhl_pt_£pigppy Erevgption, 6, 141-155. Conte, J. R., Wolf, S., 8 Smith, T. (1989). What sexual offenders tell us about prevention strategies. Child Ahu§s_snd_flsslsst. 13. 293-301. Cooney, E., & Selman, R. (1978). Children’s use of social conceptions: Towards a dynamic model of social cognition. In W. Damon (Ed.), §pp1§l_§pgh1tjph (pp. 23- 44). San Francisco: Jossey-Bass. Daro, D., Duerr, J., a LeProhn, N. (1987, July). Child assaul reve t u t 0 ° W w s w' ppeschpplepg. Paper presented at the Third National Family Violence Research Conference, Durham, NH. 207 Deutsch, F. (1975). Effects of sex of subject and story character on preschoolers’ perceptions of affective responses and intrapersonal behavior in story sequences. ngelopmentgl Esyghplpgx, 11, 112-113. DeYoung, M. (1982). he ex v ti za o 1d . Jefferson, NC: McFarland and Company. Dodge, K. A. (1986). A social information processing model of social competence in children. In M. Perlmutter (Edo). Whales! (Vol. 18., pp. 77-125). Hillsdale, NJ: Lawrence Erlbaum Associates. Dodge, K. A., Murphy, R. M., & Buchsbaum, K. (1984). The assessment of intention-cue discrimination in children: Implications for developmental psychopathology. Chilg Development, §§, 163-173. Dodge, K. A., & Newman, J. P. (1981). Biased decision making processes in aggressive boys. u a o Abnormpl Psychology, CC, 375-379. Dolgin, K. G. (1986). Needed steps for social competence: Strength and present limitations of Dodge’s model. In M. Perlmutter (Ed.), The Minnesota gygpogia op child psyghology (Vol. 18., pp. 127-135). Hillsdale, NJ: Lawrence Erlbaum Associates. Doyle, A. B. (1982). Friends, acquaintances, and strangers: The influence of familiarity and ethnolinguistic background on social interaction. In K. H. Rubin & H. S. Ross (Eds.), Beep [glhtlgng and"§2£1£l_lklll§_ifl philghppd. New York: Springer-Verlag. D’Zurilla, T. J., & Goldfried, M. R. (1971). Problem-solving and behavior modification. lpgphhi_pt W. 16. 107-126. Feldaan. E. (1983. April) WW i__philgpgn. Paper presented at the Biennial Meeting of the Society for Research in Child Development, Detroit, MI. Feshbach, N. D., & Roe, K. (1968). Empathy in six and seven year olda- W. .12. 133-145. Finkelhor, D. (1979). S u v' tim e c ' d . New York: MacMillan. 208 Finkelhor, D. (1984). Child sexual sbusg. New York: MacMillan. Finkelhor, D. (1986). Prevention education in perspective. In M. Nelson, & K. Clark (Eds.), Ihs egucator’s gpidg totpreventing child sexual sbuse (pp.133-149). Santa Cruz, CA: Network Publications. Flapan, D. (1968). Childpeh’s updspstahdihg of social iptspsgtiph. New York: Teachers College Press. Flavell, J. H. (1981). Monitoring social cognitive enterprises: Something else that may develop in the area of social cognition. In J. H. Flavell & L. Ross (Eds.), Socisl cggsitive development: Epohtigps spd possiblesfutures (pp. 272-287). New York: Cambridge University Press. Fischler, G. L., & Kendall, P. C. (1988). Social cognitive problem solving and childhood adjustment: Qualitative and topological analyses. o n' 'v n Research, i2, 133-153. Garbarino, J. (1987). Children’s response to a sexual abuse prevention program: A study of the Spigsppsh comic. Child Abuse and Neglect, 11, 143-148. Gesten, E. L., Rains, M. H., Rapkin, B. D., Weissberg, R. P., de Apodaca, R. F., Cowen, E. L., & Bowen, R. (1982). Training children in social problem-solving competencies: A first and second look. Apspipsh Journsl of_commuplty Eszgholpgy, 19, 95-115. Getz, J. A., Goldman, J. A., & Corsini, D. A. (1984). Interpersonal problem solving in preschool children: A comparison of assessment procedures using two- dimensional versus three-dimensional stimuli. lppppsl 2I_A22lisd_Dexelgnaental_£sxshslgsx. 5. 293-304. Gilbert L. J., Berrick, L., LaProhn, T. E., & Nyman, C. C. (1989). Er2tsctins_z2uns_2hildren_fr2s_sexual_ehusei_ ngs ppgsghool tpsinihg wpph? Lexington, MA: Lexington Books: D.C. Heath 6 Co. Gottman, J. & Parkhurst, J. (1980). A developmental theory of friendship and acquaintanceship processes. In W. Collins (Ed-l. uinnsssts_§zsnsaiua_in_ghild Development, (Vol. 13, pp. 197-253). Hillsdale, NJ: Lawrence Erlbaum Associates. 209 Harvey, P., Forehand, R., Brown, C., & Holmes, T. (1988). The prevention as sexual abuse: Examination of the effectiveness of a program with kindergarten-age children. Dghsviop Ihepspx, 13, 429-435. Haugaard, J. J., & Reppucci, N. D. (1988). Ihs_sszgsl use ' ' c v ' know ed n 'nt v ' ’ . San Francisco: Jossey-Bass. Hazzard, A., Webb, C., Kleemeire, C., Angert, L, Pohl, J. (in press). Child sexual abuse prevention: Evaluation and one-year follow- -up- WW. Herz, S. (1987). o arison soc ob o v’ skills in video-tsped role-plays snd is hypotheticsl- psilective.interviews. Unpublished manuscript, Yale University, New Haven, CT. Hill J. L., & Jason L. A. (1987). An evaluation of a school-based child sexual abuse primary prevention program. Esychotherapz Bulletih, 22, 36-38. Holmberg, M. (1980). The development of social interchange patterns from 12 to 42 months. Child Devglopmsht, ii, 448-4560 Jacklin, C., & Maccoby, E. (1978). Social behavior at thirty- -three months in same- sex and mixed sex dyads. Child Development, 12, 557- 569. Kendall, P. C., & Fischler, G. L. (1984). Behavioral and adjustment correlates of problem-solving: Validational analyses of interpersonal cognitive problem-solving measures. Child Development, §§, 879-892. Kenning, M., Gallmeier, T., s Jackson, T. L. (1987, July). v 'o ' d e b v . a o wo ' . Paper presented at the Third National Family Violence Research Conference, Durham, NH. Krasnor, L. R. (1982). An observational study of social problem solving in young children. In K. H. Rubin & H. 8. Ross (Eds.), eer ela d oc k’ ' phildhppd (pp. 113-132). New York: Springer-Verlag. Krasnor, L. R. (1983). An observational case study of failure in social problem solving. J r a o 'e Devglopgental Psycholpgz, 1, 1-9. 210 Observational assessment of social 1 J. Krasnor, L. R. (1985). In B. H. Schneider, K. aH. Rubin, : e problem solving. ee New York: E. Ledingham (Eds.), in ssssssgght and inte ezv vsntign (PP- 57 74) Springer-Verlag. R., s Rubin, K. H. (1981). Assessment of social In T. Merluzzi, C. Krasnor, L. problem-solving in young children. Glass. 5 H. Geneat (Eda-l. Cogniti_s_asaesamsnt (pp- 452-476). New York: Guilford. J., Moser, J. T., Litz, J., & Huges, J. (1987). Kolko, D. Promoting awareness and prevention of child sexual victimization using the red flag/green flag program: An evaluation with follow-up. Jou a Violence, 2, 11- 35. Ladd, G. W. (1981). Effectiveness of a social learning method for enhancing children’s social interaction and peer acceptance. Child_nexelonaent. 52. 171-178. A M & Santogrossi, D. A. (1980). Social skills A behavioral LaGreca, . ., training with elementary school students: group approach. gpurhal pt Cghsglting shd Clihigsl Esychology, 18, 220-227. Langer, E. (1978). Rethinking the role of thought in social In J. Harvey, W. Ickes, & R. Kidd ' (Vol. interactions. (Eds.), w ec t 'b a , 2). Hillsdale, NJ: Lawrence Erlbaum Associates. Langlois, J., Gottfried, N., & Seay, B. (1973). The influences of sex of peer on the social behavior of preschool children. passionsental_£sz§hglssx. 6. 93- 98. Programs to prevent sexual abuse: Leventhal, J. M. (1987). What outcomes should be measured? thld.Ahuse_snd NM! 1.1.: 169-171. Marsh, D., Serafica, F., & Barenboim, C. (1981). Interrelationships among perspective taking, interpersonal problem solving, and interpersonal functioning. l2uLnal_sf_fisnetis_£szshglssz. 136. 37- 48. (1989, April). McGrath, M. P., Speth, T., a Bogat, G. A. d ob : o ce 0 e t o lit The effect oi sexual abuse ppeventiph pppgpsns. Paper presented at the Biennial Meeting of the Society for Missouri. Research in Child Development, Kansas City, 211 McKim, B. J., Weissberg, R. P., Cowen, E. L., Gesten, E. L., a Rapkin, B. D. (1982). A comparison of the problem- solving ability and adjustment of suburban and urban third-grade children. an o Miltenberger, R. G., & Thiesse-Duffy, E. (1988). Evaluation of home-based programs for teaching personal safety skills to children. e e v Analzsis. 21. 81-87. Nelson. D. E. (1981). An_exsluation_21_ths_student outcpsss shd s io a c e c 0 "You’re in Charge" ppogpsh. (Available from David Nelson, Utah State Office of Education, Salt Lake City, UT, 84111). Nelson, K. (1978). How young children organize knowledge of the world in and out of language. In R. Siegel (Ed.), Children’s thinking: Whit dgvglpps? Hillsdale, N.J.: Lawrence Erlbaum Associates. Nelson, K. (1981). Social cognition in a script framework. In J. H. Flavell & R. D. Ross (Eds.), oc' 'tive developsent. Frontier§_2ndwpossihle futurgs (PD. 97- 118). New York: Cambridge University Press. Nelson C., & Carson P. (1988). Evaluation of a social problem-solving skills program for third- and fourth- grade students. e 'c n o Co Esychology, l_, 79-99. Plummer, C. A. (1984, August). r v ' e a u e: Wh ' -s 00 o ac . Report prepared for the Second National Conference for Family Violence Research, Durham, NH. Plummer, C. A. (1986). Prevention education in perspective. In N. Nelson. & K. Clark (Eds-). Ihs_edusstsnls_suids to ppgvghtihg ghild ssxusl 59233 (pp.1-6). Santa Cruz, CA: Network Publications. Putallaz, M., & Gottman, J. (1981). An interactional model of children’s entry into peer groups Child Devslopment, fl, 986-9940 Ratto, R. (1988). v s o c n v b e ' . Unpublished master’s thesis, Michigan State University, East Lansing. 212 Ratto R., & Bogat G. A. (1990). An evaluation of a preschool curriculum to prevent the sexual abuse of children. W. 1.6. 289- 297. Ray. J. a Dietzel. H. (1988). WWW ppsxghtiph. (Available from Rape Crisis Network, N. 1226 Howard St., Spokane, WA 99201). Reid, G. J. (1991, April). Soc 1 o e v n childpen: Thssrols ofmsslf-etficacz shd stiegt. Paper presented at the Biennial Meeting of the Society for Research in Child Development, Seattle, WA. Reppucci N. D., & Haugaard J. J. (1989). Prevention of child sexual abuse: Myth or reality. Alepipan Ps cholo 'st, 11, 1266-1275. Rickel A. U., & Burgio, J. C. (1982). Assessing social competencies in lower income preschool children. Apepigan Jodphsl 91 Cpphhhity Esychology, 19, 635-645. Rubin, K H., & Krasnor, L. R. (1983). Age and gender differences in the development of representative social problem solving skill. Jo r e v o enta Ps cholo , 5, 463-475. Rubin, K. H., & Krasnor, L. R. (1986). Social-cognitive and social behavioral perspectives on problems solving. In M. Perlmutter (Ed.), Ihg hingesots szhppsia pg dhiid_psyghdidgx (Vol. 18., pp. 1-67). Hillsdale, NJ: Lawrence Erlbaum Associates. Saslawsky, D. A., & Wurtele, S. K. (1986). Education children about sexual abuse: Implications for pediatric intervention and possible prevention. WM. 1.1. 235-245. Schank. Ru 4 Abelson. 8- (1977). W shd_dhdppstshdihg. Hillsdale, NJ: Lawrence Erlbaum Associates. Sedlak. A. J. (1989. April). WM hhdsg_hhd_ngglgdt. Paper presented at the Biennial Meeting of the Society for Research in Child Development, Kansas City, Missouri. Shantz, C. U. (1983). Social cognition. In P. H. Mussen (Ed.), ndb o t d 'o Vo . 3° nghitive ngslppgght, (pp. 495-555). New York: John Wiley. 213 Shure, M. B. (1982). Interpersonal problem solving: A cog in the wheel of social cognition. In F. Serafica (Ed.), c'a o o s c v cohtext (Pp. 133-166). New York: Guilford Press. Shure, M. 8., & Spivack, G. (1972). Means-ends thinking, adjustment and social class among elementary school-aged children. lsurnal_2£_92naultins_snd Clihigsl Esychology, C8, 348-353. Shure, M. B., 1 Spivack, G. (1982). Interpersonal problem solving in young children: A cognitive approach to prevention. hpericsn,Journsl of Codhdhitx_£sydhdldgx, 19, 341-356. Shure, M. 8., Spivack, C., 1 Gordon, R. (1972). Problem- solving thinking: A preventive mental health program for preschool children. Resdihg World, ll, 259-273. Shure, M. 8., Spivack, G., & Jaeger, M. A. (1971). Problem- solving thinking and adjustment among disadvantaged preschool children. Child_Dg_glppg_ht,12, 1791- 1803. Sigurdson, E., Strang, M., Doig, T. (1987). What do children know about preventing sexual assault? How can their awareness be increased? Cahadiah dpuphal of Psychistpy, 12, 551-557. Spivack, G., Platt, J. J., 1 Shure, M. B. (1976). The problem-solvihg approash t9 adjustpght. San Francisco: Jossey-Bass. Spivack. 6.. & Shure. M. B. (1974). Social_sdiustaent_sf young children. San Francisco: Jossey-Bass. Spivack, G., & Shure, M. B. (1982). Interpersonal cognitive problem solving and clinical theory. In B. Lahey & A. E. Kazdin (Eds.). Adxance§_in_shild psyphplpgy (Vol. 5, pp. 323-327). New York: Plenum Press. Spivack, C., and Shure, M. B. (1985) ICPS and beyond: Centripetal and centrifugal forces. ,Algpipsh_dppphdl 21.92aasnitz_£szchglssx. 13. 226-243- Swan, H. L., Press, A. R., L Briggs, S. L. (1985). Child sexual abuse prevention: Does it work? Child_flsl1s;s, C1, 395-405. Urbain, E., & Kendall, P. (1980). Review of social-cognitive problem-solving interactions with children. Eszch_losicsl_fiulletin. 63. 109-143- 214 Wall, H. R. (1983). Ch' ass u v i ot o ev t' . (Available from Harvey Wall, Office of Research and Development, Mt. Diablo Unified School District, 1936 Carlotta Dr., Concord, CA 94519). Weiner, J. 1., Hilpert, P. L., Gesten, E. L., Cowen, E. L., & Schubin, W. E. (1982). The evaluation of a kindergarten social problem-solving program. lpdphsl of Brigapz Epevehtiph, 2, 205-216. Weissberg, R. P., Gesten, E. L., Carnrike, C. L., Toro, P. A., Rapkin, B. D., Davidson, E., & Cowen, E. (1981). Social problem-solving skills training: A competence- building intervention with second- to fourth-grade children. Ame a J u of o n't 3 ch 411-423. Weissberg, R. P., Gesten, E. L., Rapkin, B. D., Cowen, E. L., Davidson, E., de Apodaca, R. F., & McKim, B. J. (1981). Evaluation of a social-problem-solving training program for suburban and inner-city third-grade children. Jo n Co nd ClinicslcPsgchology, d_, 251-261. Whiting, B., & Whiting, J. (1975). ’ re o si cultures. A psycho-cultural analysis. Cambridge, MA: Harvard University Press. Wolfe, D. A., MacPherson, T., Blount, R., a Wolfe, V. V. (1986). Evaluation of a brief intervention for educating school children in awareness of physical and sexual abuse. Child hbgse ahd nglggt, 19, 85-92. Wurtele, S. K. (1987). School- based sexual abuse prevention programs: A review. Child_Ahus__an__fleslect. 11. 483-495. Wurtele, S. K. (1990). Teaching personal safety skills to four-year-old children: A behavioral approach. Behaxiorsl_1heranz. 21. 25-32- Wurtele, S. K., Kast, L. C., Miller-Perrin, C. L., 1 Kondrick, P. A. (1989). Comparison of programs for teaching personal safety skills to preschoolers. __uL__l_2I_Q2nault1na_and__linisal_£szch_lssx.51. 505- 511. 215 Wurtele, S. K., Marrs, S. R., & Miller-Perrin, C. L. (1987). Practice makes perfect? The role of participant modeling in sexual abuse prevention programs. ' Earsholosx. 55. 599-602. Wurtele, S. K., Saslawsky, D. A., Marrs, S. R., & Britcher, J. C. (1986). Teaching personal safety skills for potential prevention of sexual abuse: A comparison of treatments. Journgl of Consdltihg_shd_Clihiddl Esychology, Dd, 688-692. Wyatt, G. E., 1 Peters, S. D. (1986a). Issues in the definition of child sexual abuse in prevalence research. Child Ahuss i Ngglggt, 19, 231-240. Wyatt, G. E., 1 Peters, S. D. (1986b). Methodological considerations in research on the prevalence of child sexual abuse. Child Abuse i nglsst, 19, 241-251.