:01. .I 5-21 . 4 . 1 . 1. "Han. “mum... awe? , tin"... 5.. 12.11 .. 1 ugh . . 1 : -um... 1.1“" . "hubs”... Tl. 11\1:s.._. 21 1:1 :1. . 1 . , :1 . 1. . . 1 ...I u 1... .711 1 .1 1 .1 21.; 1 :9 151'117 . . . . .. 111...- .1111 tullervhlllrvla. 1 1.... !.k . . . . 1. .. .. . 1 . 1 . .1 11 1.1.? 0.5: I! ”if lHl“HIM”!!!UHUIHIIHIHHUWWHIIIHHNJIHWI ” 320 079 6334 This is to certify that the dissertation entitled ’fl’ié ifTF’LlffflCe 610 (L home VICr‘ict‘IiM‘h rerri” QC‘LCCLDLVOW pro mm OW (CHIS (Chiral CQAOL Pareh‘fi i1:f‘Q\/(OFS 9L CltmkhapL Presented by F COLUZC‘: LUC‘an. :Dmm K036!” has been accepted towards fulfillment of the requirements for DH D degree in HLth/ln FLD [03y LIBRARY Michigan State University PLACE IN RETURN BOX to remove this checkout from your record. TO AVOID FINES return on or before date due. MAY BE RECALLED with earlier due date if requested. DATE DUE DATE DUE DATE DUE AUG 1 8 “521292“ 0‘1 ’ MIDIEUO622001}, o 1100 WWW.“ THE INFLUENCE OF A HOME VISITATION, PARENT EDUCATION PROGRAM ON LOCUS OF CONTROL AND PARENTING BEHAVIORS OF LIMITED RESOURCE MOTHERS By Dawn Christine Koger A DISSERTATION Submitted to Michigan State University in partial fulfillment Of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Family and Child Ecology 1999 ABSTRACT THE INFLUENCE OF A HOME VISITATION PARENT EDUCATION PROGRAM ON LOCUS OF CONTROL AND PARENTING BEHAVIORS OF LIMITED RESOURCE WOMEN By Dawn Christine Koger In his model on the determinants of parenting behavior, Belsky (1984) suggests that parenting is multiply determined by characteristics of the child, contextual sources of stport and stress, and the psychological resources of the parent. He argues that the most critical component to parent-child interaction is parents’ psychological resources, and speculates that parents who are most capable Of responding to children in sensitive, nurturing and empathic ways are mature, psychologically healthy adults. While several personafity dimensions encompass the traits known collectively as psychological resources, one construct that lms been connected to parenting is locus Of control, or one’s belief about his or her ability to influence the outcomes of life. The purpose of this study is to examine the locus of control construct within the context of parent education and parenting behaviors. It is designed to determine if mothers’ locus of control orientations shift toward internality, and perceptions of parenting behaviors improve as a result of a home visitation, parent education program. In addition, the correlation between locus of control and parenting behaviors is explored. One hundred mothers with children three and younger, living in a large, Midwest, urban city participated in the study in 1999. Fifty of the mothers were enrolled in the experimental group through the Building Strong Families, home visitation parent education program, and fifty volunteered to participate in the non-equivalent comparison group after being recruited through their participation in the Women, Infants, and Children (WIC) supplemental food and nutrition education program Data was collected on a pretest-posttest basis using the following research instruments: The Adult Nowicki- Strickland Internal-External Control Scale (AN SIE), The Parenting Behavior Assessment (PBA) and the Family Record Form (FRF). Group differences were tested using t-tests, and correlations between key variables were tested through Pearson product moment correlation. Results indicate that mothers who complete the BSF program are likely to experience statistically Significant differences toward more internal orientations following the program, while women in the comarison group report no changes. These findings were significant at the p< .000 level. Also, mothers who complete the program experience significant increases (p< .000) in reports of positive parenting behaviors as well, when compared to a comparison group of mothers who did not receive any parent education treatment. Finally, correlations between locus of control orientation and parenting behaviors yielded inconsistent results. Correlations at pretest revealed no Significant relationship, however, analysis at posttest yielded a significant negative correlation. That is, internal locus of control scores were correlated with mother’s tendency tO identify self with positive parenting behaviors. In general, the results of this study suggest that locus of control orientations and parenting behaviors can be influenced by a home visitation, parent education program Belsky, J. (1984). The determinants of parenting: A process model. Child Development, 55, 83-96. Dedicated in loving memory Of my two Dads, Gerald Thomas Farmer and Douglas John Blanchard. One gave me the gift of life; the other the discipline to do something with it. I will be forever grateful to you both. ACKNOWLEDGEMENTS This accomplishment would not have been possible without the support of many wonderful people. First and foremost, I would like to recognize my advisory committee, Drs. Youatt, Bobbitt, Carolan & Whipple. Your contributions toward my graduate education have been tremendous. Not only have I learned the necessary knowledge and skills to perform well in the classroom, but also I have learned valuable information that I have applied in my personal and professional life. Each of you finds practical, yet challenging ways to stimulate learning and thinking. I am grateful for your wisdom, expertise and insight, and that you were willing to share them with me through this process. I would also like to acknowledge the contributions of the Building Strong Families stafl‘ in Wayne County who were willing to add a few more forms that interfered with their teaching time, in spite of the fact that paperwork is their least favorite part of the job! You are wonderfirl ladies and outstanding home visitors. You all Should be so proud to learn that what you do every day is making a difference in the lives of the women with whom you work. Also, I thank the Detroit Urban League Wellness Gateway WIC clinic staff (Janet, Regina, Angelique, Sharon, and Karen) for allowing me to come to your clinic and recruit clients for the comparison group. You willingly modified your routines and gave up some of your very precious space for me so that I could have access to the mothers I needed for this study. I really appreciated your flexibility and commitment to help. Additionally, I would like to acknowledge the families who so graciously agreed to participate in this study and were willing to share their personal thoughts and feelings. I hope the information you shared with me will allow us to better understand mothers and the role of parent education in strengthening families and communities. I would also like to thank my colleagues at Michigan State University Extension who have been a source Of encouragement and support through this process. I have truly appreciated the enthusiasm for my work and the high value placed upon graduate education. I am especially gratefirl to my colleague and friend Jodi Spicer, who contributed any way she could, especially in these last 2 months. She was my courier, my technical advisor, my equipment supplier, my editor, my copier, my friend, and often times my sanity. I would also like to express my graditude to my family for their love and encouragement. To my husband, Joe, thanks for supporting my decision to return to graduate school, for understanding the chaos wouldn’t last forever, and lightening the load when you could along the way. To my children, Makenna and Jaret, you inspire and motivate more than you will ever know. You both have such a zest for learning, and never give up when you’re faced with a challenging task. You reminded me to celebrate the “baby steps” along the way and delight in the process of learning. Finally, thanks to my Mom for being the mother that you are. For as long as I can remember, you have been my strength and support, believing in me, nurturing me. No matter how Old I become, I still feel you beside me, helping me along as I go. In case I haven’t said so recently, thanks! This one’s for you! TABLE OF CONTENTS LIST OF TABLES ............................................................... LIST OF FIGURES ............................................................. CHAPTER I — INTRODUCTION ............................................ Introduction ............................................................... Statement Of the Problem .............................................. Purpose of the Study .................................................... Significance of the Study ............................................... Theoretical Framework .................................................... Social learning theory ............................................. The process model of the determinants of parenting .......... Integration of Rotter’s and Belsky’s Theories ................. Operational map .................................................... Constructs and Definitions ........................................................ Research Questions ...................................................... Assumptions .............................................................. Hypotheses ............................................................... Limitations ............................................................... Background Information. .............................................. CHAPTER II - REVIEW OF LITERATURE ........................... Review of the Literature ................................................... vii PAGE xii xiii 1 10 13 19 23 28 28 29 29 32 34 34 Introduction to Locus Of Control ................................. Factors Related to Locus of Control Orientation ......... Socioeconomic status ................................ Ethnicity ............................................... Gender .................................................. Significance of Internal Orientation ......................... Internality .............................................. Extemality ............................................. Relationship Between Locus of Control and Other Related Constructs ............................................. Self esteem ............................................. Self efficacy ............................................ Shifts in Locus of Control Orientation ...................... Locus of Control and Parenting ................................ Parenting .......................................................... Competent parenting .................................... Relationship between competent parenting and child outcomes ........................................... Ethnicity and parenting ................................. Parent Education .............................................. Home Visitation Parent Education ............................. Summary of the Literature ....................................... Program Information ................................. . .................... Building Strong Families curriculum ......................... viii 34 35 35 36 38 4O 40 41 42 42 43 43 46 46 47 48 49 52 55 56 56 Building Strong Families program delivery ............. Program goals, objectives, and activities ................. Program values ............................................... Building Strong Families program evaluation ........... CHAPTER III — RESEARCH METHODOLOGY ................... Research Design ...................................................... Research Sample ...................................................... Research Instruments ................................................ The Adult Nowicki-Strickland Internal-External Control Scale (AN SIE) ..................................... Parenting Behavior Assessment (PBA) .................. Family Record Form ........................................ Data Collection Procedures ........................................... Data Analyses .......................................................... Ethical Considerations ................................................ CHAPTER IV - RESULTS ................................................ Descriptive Statistics ................................................. Demographic data ........................................... Pre-intervention analysis .................................... Inferential Statistics .................................................. Pre—intervention analysis of group means ................ H1: Post-intervention differences in locus of control orientation between group means ......................... ix 58 59 61 63 65 65 66 69 7O 72 73 74 78 79 82 82 82 89 89 89 91 H2: Post-intervention differences in parenting behaviors between group means ....................................... H3: Relationship between locus of control and parenting behaviors ..................................................... Intra-group analysis of locus of control ................... Intra-group analysis of parenting behaviors .............. Conclusion .............................................................. CHAPTER V - DISCUSSION,CONCLUSION, FUTURE RESEARCH, IMPLICATIONS ............................................ Discussion ............................................................. Hypothesis 1 ................................................... Hypothesis 2 .................................................. Hypothesis 3 .................................................. Intra-group analysis .......................................... Summary ............................................................... Conclusion ............................................................. Future Research ........................................................ Implications ............................................................. APPENDICES .................................................................. Appendix A—Experimental Group Consent Form ................ Appendix B—Experimental Group Enrollment Record ........... Appendix C—Comparison Group Consent Form .................. Appendix D—Comparison Group Enrollment Record ............ Appendix E—Adult Nowicki-Strickland Intemal-External Control Scale (AN SIE) ................................ 94 95 97 101 103 106 106 106 111 113 117 119 122 126 132 137 138 139 140 141 142 Appendix F—Parenting Behavior Assessment (PBA) ............ 145 Appendix G—Comparison Group Follow-Up Letter .............. 146 BIBLIOGRAPHY ................................................................. 147 xi TABLE 10 11 LIST OF TABLES TITLE Cell Counts and Percentages for Experimental and Comparison Group Demographics Means and Standard Deviations for Experimental And Comparison Group Demographics Testing Differences Between Group Means at Pretest on Key Demographic Variables Using Independent T-tests Testing Differences Between Group Means of AN SIE Scores at Pretest for Experimental and Comparison Groups Using Independent T-tests Testing Differences Between Group Means of PBA Scores at Pretest for Experimental and Comparison Groups Using Independent T—tests Pre- and Posttest Results for Locus of Control Scores for Experimental and Comparison Groups Using Paired Sample T-test Pre- and Posttest Results for Parenting Behavior Scores for Experimental and Comparison Groups Using Paired Sample T-tests Correlation Between Locus of Control and Parenting Behaviors at Pretest Correlation Between Locus of Control and Parenting Behaviors at Posttest Independent T-tests for Intra-group Differences Among Locus of Control Scores and Demographic Variables Independent T-tests for Intra-group Differences Among Parenting Behavior Scores and Demographic Variables xii PAGE 83 84 88 90 90 92 92 96 96 98 102 FIGURE LIST OF FIGURES TITLE Rotter’s Social Learning Theory: Predicting Behavior Social Learning Theory A process model of the determinants of parenting Integrated Framework of Rotter’s Social Learning and Belsky’s Process Model of the Determinants of Parenting Operational map of this study within the context of selected elements fiom Belsky’s and Rotter’s Theories Research Design Operational map of this study and demonstrated Results xiii PAGE 10 12 14 20 65 125 CHAPTER I INTRODUCTION Early reports of the success associated with home visitation programs lead to an influx of resources allocated to home service interventions, resulting in thousands of programs across the United States (Gomby, Culross & Behrman, 1999). Many agree that parents need effective social support networks and suggest that home visiting programs provide a valuable resource by functioning in that capacity (Baker et. al., 1999; Daro & Harding, 1999; Wagner & Clayton, 1999). However, Since home visitation describes a delivery strategy and not a program model, there is a great deal of disparity among individual programs and their primary activities. Clearly, these contextual differences lead to a host Of different outcomes, variable success rates and mixed findings. Yet there is still a substantial body of work that suggests tint home visitation programs can be an effective way to reach families with young children (Campbell, 1994; Olds et al., 1986; Hardy & Street, 1989). Olds et a1. (1986; 1997) demonstrated positive results of home visitation using outcomes related to child maltreatment, including child abuse and injury. Hardy and Street (1989) concluded that there was a significant reduction in children’s hospitalizations, a Sharp reduction in suspected child abuse and neglect, and a reduction in the substantiated incidents of child abuse and neglect in a home visited group compared to a control group who did not receive the intervention. Other investigators studying a home visitation program for families with newborns at risk for poor outcomes found that mothers in the experimental group reported numerous positive outcomes including improved parenting efficacy, decreased parenting stress, and increased use of non-punitive discipline techniques when compared to a control group not receiving services (Duggan et. al., 1999). While home visitation programs address multiple outcomes, parent education is Often a primary focus of the intervention. Parent education is aimed at the parent or parents, with the intent of providing support, information, skills, and/or referrals to community resources to improve the parent-child relationship and family system. Parent education programs report a range of goals and activities and Often look to the empirical evidence to guide conceptual and programmatic models in the development and implementation of programs. For example, existing research has substantiated that parents’ ability to nurture their children may be jeopardized for a variety of reasons. Living in a stressful or chaotic environment, having limited support networks, and lacking knowledge of normal child development may limit parents’ capacity to foster their child’s optimal development (Seitz, Rosenbaum & Apfel 1985). Therefore, one may conclude that stimulating, organized and safe home environments, supportive social networks, age appropriate developmental expectations, and positive parenting behaviors are among the significant predictors of competent parenting, and as such, a desired outcome of parent education programs. Some go further, however, and suggest that parenting is affected by several domains of parental functioning. These domains include not only those dimensions typically associated with parental functioning such as knowledge and performance, but also parental acceptance and capacity, as well (Pecora, Fraser, Nelson, McCroskey & Meezan, 1995). The first two domains are those most Often targeted by with parent education and more easily conceptualized. Parental knowledge refers to what parents know in terms of age appropriate expectations, normal child development, and appropriate discipline strategies, while performance is the actual parenting behaviors and strategies utilized. The second two domains are important, but less Often associated with parent education interventions. Parental acceptance ascribes the degree to which a parent has acknowledged the role and responsibilities of parenting and is often conceptualized in terms of the presence or absence of warmth and empathy. Finally, parental capacity is the parent’s capability to provide adequate care. It is an extremely complex construct and is influenced by characteristics such as parental depression, illness, developmental disabilities, low self esteem, lack of confidence, and lack of perceived control (Pecora et al., 1995). This concept of parents having the capacity to provide appropriate and responsive care is further supported by the process model Of the determinants of parenting (Belsky, 1984). Belsky’s model describes how parental psychological resources directly impact parental functioning. Like parental capacity, parental psychological resources can be a complex variable and represents numerous attributes. Belsky defines it as the product of the parent’s developmental history and personality. One aspect of parental psychological resources that has received attention in the literature is locus of control. Locus Of control refers to where individuals find the decision making factors that influence life (Swick & Graves, 1986). This concept originated in social learning theory and describes the degree to which individuals perceive reinforcement as contingent upon their own behavior. Locus of control orientations range fi'om internal to external (Rotter, 1966). Internal locus of control is the belief tlmt an individual can determine his/her own fate within limits. External locus of control is the belief that he/she is controlled by powerful forces outside of him/herself (Lefcourt, 1976). Many contextual variables have been linked to locus Of control orientation in the literature. In general, limited resource audiences tend to score as extemals, while individuals from middle and upper socioeconomic groups generally score as internals (Shaw & Uhl, 1971). There is also a substantial body of research which suggests that extemality is a firnction of gender. Women tend to score in the external direction while men typically score more internally (DeBrabander & Boone, 1990). These tendencies are especially Significant in relation to studies of limited resource women, and the influence that their inclination toward extemality might have in other areas of their lives, such as parenting. For exarrrple, investigators have explored the locus of control construct and its relationship to parenting. Many have documented that internality is associated with positive parenting behaviors and child outcomes. Specifically, studies have linked parental locus of control to child locus of control (Barling, 1982), child personality characteristics (Ollendick, 1979), parent child interaction (Clmndler, Wolf; Cook & Dugovics, 1980; Kleemeier, 1976), the incidence of child abuse (Ellis & Milner, 1981), parent perceptions ofchildren's problems (Harris & Nathan, 1973), and parents' ability to provide stimulating environments (Stevens, 1988). Statement of the Problem Studies support the notion that internal locus of control is related to positive behaviors (Lefcourt, 1982; Schaefer, 1983), that parental locus of control impacts the parenting process and children (Chandler et al., 1980; Stevens, 1988;), and that interventions have been successful in influencing locus of control orientation (Tait, 1976). Furthermore evidence documents that home visitation can be an effective delivery model (Campbell, 1994) and parent education programs are effective in creating positive changes in families (Brems, Baldwin & Baxter, 1993; Gross, Fogg & Tucker, 1995). Thus far, the majority of parent education programs has concentrated upon evaluating impact on narrow measures such as children’s cognitive development or parenting behaviors (Clewell, Brooks-Gunn, & Benasich, 1992). Recent research, however, has begun to seek out broader, non- traditional outcomes Of parent education. Evidence suggests tint parent education programs can transform mothers’ thinking about themselves, their own lives, and their personal resources (First & Way, 1995). Nonetheless, not enough is known about mothers’ changes in self as a result Of parent education. Some evidence suggests that a parent class may alter mothers’ perceptions of self (First & Way, 1995). Yet, little data exists which documents non- traditional outcomes Of parenting education, and few programs have examined the effect that the intervention has upon locus of control orientation. Even fewer yet attempt to influence locus of control orientations toward internality, in spite of what is known regarding the impact of locus of control on parenting and child development. While maternal locus of control is a variable in many studies, few have explored the extent to which mothers’ locus of control orientation may be influenced as a result of a parent education program. Also, more information is needed to firrther explore the relationship between maternal locus of control orientation and parenting behaviors, and how this construct influences ways in which limited resource mothers interact with and respond to their children. Therefore, there is a need to better understand if a home visitation, parent education program can be an effective way to transform mothers’ thoughts of self and personal power, and influence parenting behaviors of limited resource mothers. Purpose of Study The pmpose of this study is to examine the locus of control construct within the context of parent education and parenting behaviors. Specifically, this study will determine if limited resource mothers’ locus of control orientations shift towards internality, and perceptions of parenting behaviors improve, as a result of a home visitation, parent education program Also the correlation between maternal locus of control and parenting behaviors will be explored. The materials used in the course Of the program were designed Specifically for the limited resource audience and include a personal development component for the mothers which is believed to influence the degree to which they perceive reinforcement as contingent upon their own behavior. Also, the program’s cmricula focuses upon providing mothers with knowledge and skills to understand and respond to their children in developmentally appropriate ways. Finally, the program utilizes a paraprofessioml, home visitation model tint is thought to influence and enhance the ways in which mothers perceive themselves and their sense of control and power. For these reasons, it is believed that locus of control orientation and parenting behaviors of limited resource mothers should be influenced by the intervention. Significance of the Study Findings of this study will be helpful in furthering the understanding of and changes in locus of control and its relationship to parenting. Also, the information gleaned will have direct implications for parent education program development, implementation, and evaluation for limited resource audiences. Thus far, the focus of most parent education programs has been on the dissemination of behavioral or technical information (First & Way, 1995). However, if findings suggest that a home visitation, parent education program is an experience tint transforms mothers’ thoughts about self and personal power, then progam planners should consider models which facilitate and encourage parental personal growth. Encouraging mothers to develop new ways of thinking could have significant impact on their ability to effectively parent, their relationship with their child and the child’s development, and ultimately upon improvements in other areas ofrmternal life comse as well. Theoretical Framework Two theoretical models are the foundation for this research, social learning theory and the process model of the determinants of parenting. This next section describes both of these theories independently and then integrated within the context of this study. Social lmmg’ theory The first model is Rotter’s social learning theory (1954) and provides a foundation for mxlerstanding personality and behavior. The basic assumption of social learning theory is tint persomlity is the result of the interaction between the individual and his environment, ' _ as Opposed to fixed internal traits with no capacity to be molded or shaped. Thus, personality represents learned behavior, meaning present behavior is shaped by past experiences. Rotter’s social learning theory depicts forn' elements that are utilized to predict behavior and describe persornlity. The first variable is behavior potential, or the likelihood for a given behavior to occur in a particular situation, in relation to a single reinforcement. The second major variable is the expectancy value, or the degree to which one believes tint a particular reinforcement will occur as a result of a specific behavior. Reinforcement is the third variable and describes the level Of preference for any particular reinforcement to occur if the possibilities Of all occurring were equal. The final variable is the situation to which the individual is responding. Rotter (1954) hypothesized that using these forn' variables, there is a formula which can predict behavior. That formula is, the behavior potential for a given situation and a particular reinforcement is a function of the expectancy value and reinforcement value. In other words, how one behaves in a given situation is a result Of the degree one expects a particular reinforcement to occur as a consequence of a Specific behavior, and the value for that particular reinforcement, within the context Of the environment and past experiences. Figure 1 is a visual representation Of the formula The elements of Rotter’s theory, which are particularly relevant to this study, are bolded in Figures]. Figure l: Rotter’s Social Learning Theory (1954) Predicting Behavior Personality Situation Behavior = Expectancy Reinforcement Potential Values Value Note. Adapted from Rotter, J. (1954). Social learning and clinical psychology. New York: Prentice Hall. The inclusion of the concept of expectancy values, or probability of reinforcement, is what sets Rotter’s theory aside from other social learning theories. Eventually Rotter (1966) refined this idea further and developed the locus of control construct which described generalized expectancy in terms of internal or exterrnl reinforcements. Rotter suggests that people will behave differently if they feel tint what happens to them is a result of their own behavior rather than controlled by chance, luck, fate, or powerful others. Figure 2 illustrates Rotter’s social learning theory. Elements particularly relevant to this study are bolded in Figure 2. Figure 2: Social learning theory Past \ { Personality W Behavior Experiences 1 I Expectancy [ Environment ] Values {Reinforcement 1 Values Note. Adapted from Rotter, J. (1954). Social learning and clinical psychology. New York: Prentice Hall. Rotter’s theory establishes a fiamework tint describes persornlity and behavior within the context of the environment and experience. In this study, social learning theory serves as a means to better understand how experiences influence the locus of control construct, and ways in which locus of control influences behavior. This model describes the range of human behavior, including belnvior within the context Of parenting. For example, according to Rotter’s theory, mothers perceptions of the influence Of their behavior on any given outcome is a result of their experiences and attitudes. Their behavior is formed by how they view the world and their past experiences. This model offers a useful fiarnework in determining how mothers make behavioral decisions tint Shape their lives and the lives of their children. The m model ofth_e determinants Ofmenting The second theoretical model for this study is the process model of the determinants of parenting (Belsky, 1984). It is based upon an integration of the profusion of literature 10 regarding children and parenting rather than an empirically demonstrated model, although it is a widely accepted theoretical fiamework. This model origirnted in the literature of child maltreatment and is an ecological model of parental functioning. Ecological theory emphasizes tint hunnn development is influenced by the context within which individuals live. For instance, children’s development is influenced by how parents and other significant people care for them, and how others care for them is influenced by the characteristics of, and the interactions among, families, social networks, neighborhoods, and communities (Olds et al., 1999). Figure 3 shows Belsky’s process model of the determinants of parenting. In the model, Belsky suggests that parenting is multiply determined through the interaction of forces within the parent, forces within the child, and forces within the social context in which the parent-child relationship is embedded. According to his model, forces fi'om within the parent include the elements Of the parent’s developmental history and personality. Child characteristics such as developmental stage, temperament, behavior, as well as the “fit” between those characteristics and the parent, describe examples of the forces within the child. Those forces fiom within the social context represented in the model are the parents’ marital relationship, work, and social networks. The model assumes that parent’s personality is influenced by marital relations, work and social networks, that the child along with the parent’s developmental history, marital relations, work, and social network influence persornlity, which in turn influences parenting, and ultimately affects child development. The particular elements ofBelsky’s model tint are relevant to this study are bolded in Figure 3. ll sea .3 age case 25 .388 ass: "weenie afiaacaoe 2: $3: a 520m eon usage—95: :33 230 been A 3:03:29»:— - mac—.2...— baa—EEO.— gutfloflaau D 250 mecca—om :33: 43.502 32% 1 H wfifioaa .«o $555.83 2: me .038 3805 < ”m Gama 12 Belsky’s model on the determinants of parenting is reflective of the ecological theory, which emphasizes that hunnn development is a result of the interaction between the individual and the environment. Belsky’s model has multiple layers, each nested within a larger system, which impacts hunnn development. Belsky acknowledges the inrportance Of individual characteristics and contributions to development as evidenced by his inclusion of elements such as the temperament of the child and the developmental history and personality of the parent. Consistent with ecological theory, the model explicitly states, however, that child development does not occur in isolation, and tint the family is also a critical conrponerrt of hunnn development. Yet, while family is a significant setting in which hunnn development occurs, ecological theory emphasizes that development occurs in multiple settings simultaneously (Bronfenbrenner, 1986). Belsky’s model reflects the belief tint not only is development affected by the various systems and multiple environments in which children live, but also the multiple settings in which their parents live, and then the interaction among them. This assumption is reflected in the inclusion of multiple domains, such as the parent and child, and multiple settings such as social networks, nnrital relationship, school, and work. Mion ofRotter’s and Bels_ky’s Theories Figm'e 4 is the conceptual fiarnework created to depict this study and represents an integration ofRotter’s social learning theory and Belsky’s process model of the determinants of parenting. The following section will describe the elements selected fiom both models which are meaningful within the context of this research. 13 .3 8o: as 280 a .eoeoana mo 35:00 Eons mama?» E85 new momoawwoonxo EQEOO .Gwa: .p. .uofiom Bob was .waumw .lildlllallmm ass 2269 :6 .388 saga < Maegan a assesses. 2: .918: a seam 82m 582 “I.-IIIIIIIII-III.-ICU-III...-.IIIIIIUIICIIUIIIIIICC-III.-II.‘.III..IIII...IIIIII." 3882mm m m fig 5...... m m n 3325 83> manoeoaxm " . . . . . @235 m 3332“ EoEooaoEoM 5580qu 6.3 m Al. oozes—EH m L " 1m. m water—a.— __ baa—SES— _ .985: _ m Beam u . w _ m 13:25.29»:— ‘ ... beam cacao; m m 'I.‘ ICU-IUIIIUIIIIIUII. .I-IUIIIIIIUCII .II IIUIIIII.II-III-II...ICU-IICIIIUUICII‘ 33.32226 €252 38m 95:22— ue 8.35.582. 2: «e .258 23...:— a..a_n_om a... been. 9553. .593 9.333— wnflflufifi {9393.51899an < ”v 0.5»...— 14 Belsky’s model is reflected in the five elements Of social network, developmental history, persornlity, parenting, and child characteristics, and describe forces fi‘om within the parent, within the child, and within the social context which contribute to, or influence, parenting. The first two elements “Developmental History” and “Personali ” are the constructs tint Belsky refers to as parental psychological resources. Together, these elements describe the forces fiom within the parent tint contribute to parental fimctioning. One’s history and developmental processes contribute significantly to the adult he or she becomes, shaping persornlity and psychological well being. Belsky suggests that it is mature, psychological healthy adults who are the most capable of providing sensitive, nmtming, developmentally appropriate parenting. While numerous characteristics and attributes nnke up what is referred to as personality, one particular personality dimension that Belsky identifies as critical to parenting is the individual’s locus Of control orientation. In fact, Belsky (1984) states that direct support for the personality--parenting connection can be found in data which links internal locus of control to levels of Observed warmth, acceptance and helpfulness and low levels of disapproval when interacting with young children (Mondell & Tyler, 1981). In support of Belsky’s theoretical model, other investigators have examined parenting within the context of locus of contrOL as well. For example, in a review of the literature, Swick & Graves (1986) argue that the evidence suggests tint personal psychological characteristics are critical components Of effective parenting and suggest that locus of control is one Of the significant personath attributes. They assert that it is a “core skill” which influences nnny dimensions of parents’ and children’s lives, and contributes to healthy families. Scinefer, Hunter & Edgerton (1983) also propose that locus of control be identified as 15 one set of the maternal variables that describes and influences the child’s psychosocial environment. When researching maternal locus of control and parenting behaviors, the authors found internal locus of control to be associated with having progressive child rearing beliefs, providing educational experiences, talking with children and encouraging curiosity. Other parenting behaviors associated with parental internal locus of control are warmth, protectiveness, consistency with discipline and encouragement of independence. Conversely, studies have found parents who are extremely externally oriented have a negative impact upon their child’s development (Graves, 1986) and children raised in powerless settings model those behaviors in their own lives (Kempe & Kempe, 1978). The next element of Belsky’s model selected as relevant for this study is “Social Network”. Social network is defined as the social context in which the parent-child rehtionship is embedded. Particularly, Belsky is interested in the functions and sources of social support, and ways in which the network can contribute to or detract from growth promoting parenting. In fact, there is a compelling body of literature tint demonstrates the impact tint overall support influences parental psychological well-being and support is positively correlated to parental fimctioning (Belsky, 1984). Bronfenbrenner (1986) suggests that the informal and formal support systems of both the parent and the child influence a parent’s child rearing behavior. For example, women who had social networks to call on following the birth of their babies had more confidence in their ability to perform well as mothers than women who did not (Cutrona & Troutnnn, 1986). In addition, researchers have found that more support networks appear to promote feelings of well being (Melson, Ladd & Hsu, 1993) and higher levels of social support were 16 associated with higher parental satisfaction and higher maternal esteem (Koeske & Koeske, 1990). The fourth domain ofBelsky’s model is “Child Characteristics”. This component describes individual characteristics, including temperament and persornlity, from within the child. Belsky argues (1984) that the plethora of knowledge which describes the child’s influence on parents and parental functioning provides a foundation for this elements’ inclusion in his model. However, he also supports others who suggest that perhaps the relationship between child characteristics and parental functioning is not a result of direct shaping on the child’s part, but rather an issue of the “goodness of fit” between parent and child (Lerner & Lerner, 1985). The fifth and final element of Belsky’s model represented in this study is “Parenting”. In the article describing his model, Belsky seems to use the terms parenting and parental firnctioning interchangeably. He suggests that parenting refers to an individual’s childrearing attitudes, strategies and behaviors. He proposes that those parents who are able to provide sensitive and responsive care to their children facilitate optimal child development. Many other researchers have established the linkage that supports that competent parenting enhances child development as well. For example, Maccoby & Martin (1983) concluded that warm, involved, maturing parents had children who were competent, responsible, independent, confident, achievement oriented, and able to control aggression. In addition, parental sensitivity to task, level of aggression towards child, and extent to which independence is encouraged inve been linked to secm‘e attachments and children’s problem solving skills (Easterbrook & Goldberg, 1984). 17 Belsky’s model provides a comprehensive, ecological explanation of the influences on parenting and parenting behaviors. However, to give meaning to the ways in which individuals behave in general, and to the locus of control construct, social learning theory must be included in the theoretical fiamework, as well. Rotter (1954) suggests that past experiences, or developmental histories, influence and shape future behaviors. In addition, Rotter indicates that one’s personality is influential in determining behavior. He asserts that behavior is defined in terms of one’s ideas regarding anticipated degree of success and eventually labeled this construct locus of control. The relevant elements of Rotter’s social learning theory which are integrated into the conceptual model are developmental history, personality, expectancy value, reinforcement value, and behavior. Personality, expectancy value and reinforcement value are interrelated concepts. That is, the values one has regarding the probability tint a specific reinforcement will occur and the value one places upon tint reinforcement, reflect personality characteristics. To identify this interconnectedness, these elements are represented in the model as intersecting ovals. Overlap, or terms that have consistent meanings and themes between the two theoretical models, is indicated on the integrated conceptual framework by a heavy broken line grouping common elements. Clearly there are parallels between the two models that are apparent upon examination of the integrated conceptual framework. Both researchers suggest that developmental-histories shape personality, and personality influences behavior. Furthermore, both describe the contextual factors of the environment as influential in terms of personality development and predicting behavior. Firnlly, both authors argue that locus of control is a significant personality construct which contributes to and 18 influences the ways in which individuals behave. While Rotter’s work has focused upon the construct of locus of control in terms of general behavior, Belsky has addressed the concept in terms of a more narrow aspect of human behavior, parenting. These parallels between the two theoretical models validate the integration and introduce a composite conceptual model for this study. The integrated conceptual framework is characterized by nesting Belsky’s model of the determinants of parenting within the larger context of social learning theory. Rotter’s work provides the foundation for ways in which individuals behave and Belsky’s suggest ways in which this process is applied within the context of parenting. (Mmtional map Figure 5 is the operational map for this study and describes the key variables within the context of the integrated conceptual model. The “Developmental History” element, includes the key variables gender, maternal age, ethnicity, educational level, income, number Of children, and first time parent status. Each describe part of a mother’s past experiences, or a segment of her developmental history, and are critical components which influence experiences and developmental processes. Both Belsky and Rotter consider the impact that variables such as these influence decision making, personality, and behavior. Each of the above terms represents a control variable in this study. The next element within the conceptual model is “Personality”. In general, the term personality represents numerous individual attributes. However, both Belsky and Rotter agree tint locus of control orientation is a specific personality dimension and focus upon the role tint it plays in determining behavior. Furthermore, others have identified 19 A 8:39: A . I . I . I 338332 335:0:qu $852K: beam A 323352 320803 305093 beam 3an beam Bantam ll. U cacao..— ., _ _ fiesta; 3.580 owe FEED . m0—Hm—fifih0ém0 Aug—EU 56—330 c3852 . Saute: “amnemmoQ £53m eoefleoto 75:8 E88 2:: RE . fiesta; EoESQoQ .«o 9.62 $8502 .. w .8250 . 80333 wfifipam . a , £2930 {2:02 . a .. senses: . . | mmm 9.5502 . imp—m:— . 1: HZHZmOAEM—n Score; #636335 88w?“ All, 5:838 “:83 _I GomusmmTr Show I .l a II - I . I .. MMOEHZ 300m Becca. wanted e5 mac—Eon 89a 3380? 3828 .«o 6888 05 Eng? .33» £5 mo 38 .eeomafloao m 95$..— 20 locus of control as an irrmortant personality dimension in relation to parenting (Swick & Graves, 1986) and it has been linked to positive outcomes for both parents and children (Barling, 1982; Stevens, 1988). Because of the empirically demonstrated significance of locus of control, it was selected as the key construct addressing the “Personality” dimension of the operational map, and serves as the dependent variable of the study. Another concept that is consistent in both models is referred to as “Behavior” on the operational map and describes specific parenting behaviors. Parenting behaviors reflects the child rearing strategies utilized by parents and is the second dependent variable for this study. This variable assesses parental perceptions regarding the frequency and consistency of positive parenting behaviors and relates the behaviors to locus of control and parent education. Thus, one of the primary purposes of the study is to explore the extent to which participation in a parent education program influences clmnges in mothers’ locus of control orientation , parenting behaviors and the relationship between those behaviors and locus of control. While Rotter’s model suggests that personality is the interaction of the individual and his meaningful environment, he does not explicitly state what those environmental components are. While the element “Social Network” represented on the operational map might be considered part of what he thought to be “meaningful environment”, it was never specifically defined. Therefore, social network is attributed to Belsky’s work only. In this study the independent variable parent education represents the element of Belsky’s model known as “Social Network”. Although the parent education intervention intends to influence ways in which parents respond to and interact with their children and change behavior, another significant priority is to improve the social network of the mother. 2] Paraprofessional parenting instructors are intended to serve as a source of support and enhance parental functioning. This influence on the mother’s social support network is believed to effect parental behavior directly, when the parenting instructor provides feedback and support regarding a mother’s role and her parenting behaviors, and indirectly, as the instructor provides emotional support and encouragement to the mother herself. The final element included on the operational map is “Child Characteristics”. The variable selected to represent this piece of the model is child’s age. While many other child characteristics influence parenting and parental functioning, age is certainly one factor that affects the ways in which mothers respond to and interact with their children. There are vast differences in the information needed and caregiving patterns associated with parents of children newborn through three years of age, the target population for this particular parenting edumtion program For this variable, the age of the target child is categorized into two groups, 0-18 months and 19-36 months This distinction is in recognition of the role that children and their developmental stage play in influencing parental functioning. The arrows among the elements of the operational map indicate influence and directionality. Thus, according the operational map, developmental history and social network influence personality, personality influences social network, and child characteristics, social network and personality influence parenting. The thick lines indicate the relationships that this research intends to test, and the arrows identify the hypothesized direction of the relationship between the variables. The thick, intact line suggests a relationship that has already been demonstrated in previous research, yet the relatiomhip will be evaluated in this study as well. In these instances, results are useful to either confirm or refute previous findings. The broken lines on the operational map represent anticipated 22 relationships between variables, or those that have not yet been consistemly documented through previous work. Based upon the operational map, there are three hypotheses that are being tested. They are the influence of a parent education program on parenting behaviors, the relationship between locus of control and parenting behaviors, and the influence of a parent education program on locus of control. In conclusion, the highly regarded work of Rotter (1954) and Belsky (1984) provides a strong theoretieal model that supports the integration of human development and belnvior in the environmental context. Moreover, Belsky’s model clearly supports the idea that the individual, family, neighborhood, community and societal systems, along with the interaction among them all, determine the course of human development. However, while Belsky suggests that parenting is multiply determined through forces from within the parent, forces within the child, and within the social context in which the parent child relationship is embedded, hearguesthatthedomainsarenotequalintheir influence. Ofthethreeareas, he identifies parental psychological resources as the most critical component to optimal parental functioning. For this reason, this study focuses on the locus of control construct within the context of parenting education and parenting behaviors. Constructs and Definitions This section presents the major variables and concepts that are represented in this research, along with their conceptual and operational definitions. 23 Independent variable: Parent Education Program Conceptualization: Educational intervention aimed at the parenting process, intended to improve parenting competence. 9mm All mothers who are enrolled in and complete the Building Strong Families home visitation program within a 16 week period. Dependent variable: Locus of Control Orientation Conc_§ptualization: Locus of control, which ranges fiom internal to external, refers to the degree to which an individual perceives reinforcement as contingent upon his or her own behavior. Internal locus of control is the individual's belief that he or she is an actor and can determine his/her fate within limits. External locus of control is the person's belief that he or she is controlled by forces outside of him or herself (Lefcourt, 1976). erationalization: The locus of control construct is measured by the participant's score on the Adult Nowicki-Strickland Internal-Extemal Control Scale (AN SIE), a self report questionnaire which measures participant's attitudes regarding the extent to which they are able to control their own fate within limits. 24 Dependent variable: Parenting Behaviors Conceptualization: Childrearing strategies and behaviors utilized by mothers to interact with, respond to, and influence children and their behaviors. gmtionalization: Parenting behaviors is measured by the mother’s score on the Parenting Behaviors Assessment (PBA), a self report questionnaire which assesses parental perceptions of parenting behaviors in relation to young children. Other key terms: Home Visitation Program Conceptualization: A program delivery strategy that sends staff into the homes of families with young children to encourage changes in the knowledge, attitudes or behaviors of parents by providing social support, practical assistance, and/or parent education. gmfionalization: Home visitation program is measured by the Building Strong Families parenting instructor indicating that the participant received individual lessons on her Family Record Form. Limited Resource Conceptualization: Limited resource describes mothers with income levels 185% or less of the poverty level. 25 erationalization: Limited resource is measured by the monthly income identified on the Family Record Form, which should not exceed one twelflh of 185% of the annual poverty level. Maternal Age Conc_eptualization: The period of time in which a mother has been alive. erationalization: The number a mother reports on the Family Record Form as her current age in years. Maternal Education Cogc__eptualization: The number of years that a mother has participated in formal education through a licensed institution. gmtionalization: The number a mother reports on the Family Record Form as the last full year of formal education that she has completed. Ethnicity Conceptualization: The ethnic group with which a mother identifies. Qrgrationalization: The ethnic group that the mother selects on the Family Record Form 26 Number of Children Conceptualization: The total number of children to whom a woman has given birth and/or provides care. erationalization: The number of children is measured by adding together all children who are listed under family members on the Family Record Form. First Time Parent Status Con§_eptualization: Mother’s first parenting experience. erationalization: First time parent status is determined by those parents who list only one child on the Family Record Form under household members. The responses are coded into two discrete categories: Yes (One child) or No (More than one child). Age of Child Concmiization: The period of time in which a child has been alive. erationalization: Tire number of months in which the targeted child has been alive as recorded on the Parenting Behavior Assessment. Building Strong Families (BSD—Experimental group A home visitation, parent education program for limited resource mothers of children 0-3 years of age. 27 Women, Infants and Children (W IC)—Non-equivalent comparison group A federally funded program for limited resource mothers who are pregnant or parenting young children that provides supplemental food packages and nutrition education at community based clinics. Research Questions The following questions are posed regarding this research. 1. To what extent does locus of control change in mothers who participate in a home visitation, parent education program? 2. To what extent do parenting behaviors clmnge in mothers who participate in a home visitation, parent education program? 3. What is the relationship between locus of control and parenting behaviors for limited resource mothers? Assumptions The following assumptions are made and accepted as the truths that guide the development of this research. 1. Parental psychological resources influence parenting (Belsky, 1984). 2. Locus of control orientation is a characteristic of parental psychological resources (Belsky, 1984). 3. Locus of control orientation can change (Rotter, 1966; Tait, 1976). 4. The Adult Nowicki-Strickland Internal-External Control Scale (AN SIE) provides a valid measure of the locus of control construct (Nowicki & Duke, 1983). 28 5. The Parenting Behavior Assessment (PBA) provides a valid measure of mothers’ perceptions of parenting behaviors. 6. Participants respond to self report questions honestly and accurately. Hypotheses The following three hypotheses are formulated for this research. Each of the hypotheses describes an expected outcome regarding the study’s results. H1: Limited resource mothers who complete a home visitation, parent education program will show an increase in internal locus of control orientation when compared to the comparison group. H2: Limited resource mothers who complete a home visitation, parent education program will show an increase in perceptions of positive parenting behaviors when compared to the comparison group. H3: As mothers’ locus of control scores decrease, parenting behavior scores will increase. Limitations Although the best possible research design was developed, some limitations of this study exist. First of all, only those mothers over the age of 18 years, participating in the Building Strong Families program from February 1999 through October 1999, in a large, urban community were included in the sample. Therefore, the findings are 29 generalizable only to the mothers in that particular sample rather than the BSF program as a whole, since the demographic characteristics of those in the sample is quite different than those participating across the state. In addition, while findings may present implications for program planning and implementation in general, these findings are true only for mothers participating in a specific parent education intervention, and are not generalizable to all parents enrolled in parenting education. Furthermore, while mothers participating in the WIC non-equivalent comparison group are similar to the experimental group in terms of demographics such as age, race, educational level, family composition, and monthly income, there may be other subtle differences between groups, which may affect findings. For example, BSF is a home visitation program, and WIC is a clinic-based service. There may be differences in mothers who receive services inside their homes and those who receive services outside of their homes. In addition, there may be differences between mothers who enroll in educational programs (BSF) and those who do not (WIC). Another limitation of this research is its inability to address the long term effectiveness of the BSF intervention. Many investigators have documented short term changes in parenting knowledge and/or behaviors, although there appears to be little evidence of long term changes. For those mothers participating in the BSF program, the post test is administered during the last visit, and there is no follow up contact to determine if changes are sustained over time. Testing procedures may present another limitation as well. It is possible that pretesting subjects may increase participants’ sensitivity to the locus of control construct and/or their parenting behaviors and alter their perceptions of their own power to 30 influence their lives or parenting skills. Also, data are self reports, and only measure mothers’ perceptions of parenting behaviors and locus of control constructs. Another limitation of the study is experimental mortality, or the number of subjects who drop out of the experimental and comparison groups prior to the conrpletion of the experiment. Because it was not feasible to contact those who chose to drop out of the program, it is impossible to discern if there are differences in those who completed the research study and those who did not. In addition, the reasons of those who decided to drop out of the experimental group may be different than the reasons of those who dropped out of the comparison group. These differences may present a threat to the validity of the study. Multiple treatment interface is another potential limitation of this research. The majority of mothers who participate in the Building Strong Families and WIC programs also participate in other services ranging from adult education, job training, therapy, health care, case management, and so on. Some of these interventions have preceded the BSF or WIC program in time, while others run concurrently. It may be difficult to attribute clmnges in locus of control orientation to BSF alone when many other influences have affected participants’ attitudes and behaviors as well. Another limitation of this study is small sample size. According to a power amlysis by Pecora et al., (1995), the recommended sample size for an alpha level of .05 and a moderate effect level of .80 is 50 per group. While this study meets these requirements with 50 subjects per group, the small sample is not large enough to detect smaller effects. 31 Finally, while the design for this research was the strongest possible considering program and staff resources, it was a quasi-experimental design. The results gleaned certainly have implications for program evaluation and future directions of the BSF program However, a more rigorous scientific design which includes a control group through random assignment, would permit researchers to attribute changes within the experimental group to the intervention with certainty. It is this type of research that is needed to influence program design and policy regarding home visitation, and family support at the state and national levels. Background Information This research is conducted within the context of a major university’s land grant extension system. Its purpose is to deliver research-based information and services in communities through a variety of educational strategies, technologies, and collaborations. The Extension programs and activities of this particular Midwest university are focused in three primary areas: Agriculture and Natural Resources, Economic and Community Development, and Children, Youth and Family Programs. Children, Youth and Family programs offer educational opportunities for individuals across the life cycle. The programs are designed to help people improve the quality of their own lives and the communities in which they live. Of noteworthy emphasis is the area of family strengths, which integrates community and university resources to help families succeed. One specific CYF program that was designed to build family capacity is Building Strong Families (BSF). 32 chili! mm with . EN additi duel with i this p BSF i appro Building Strong Families is a parent education program designed for parents with children 0-3 years of age. The program was developed in 1988 to provide an additional resource to home visitation nutrition instructors working with limited resource families with children. In the mid-eighties, staff began reporting that participants enrolled in EFNEP, a nutrition education program, were asking for more information on several additional topics, the most critical being parenting. A parenting curriculum was developed and was used as an additional resource for the EFNEP instructors working with families with young children. However, it soon became apparent that the need for this program exceeded staff resources so additional funding was secured in 1991 to pilot BSF in 4 counties. The program has since expanded and is currently offered in approximately half of the state’s 83 counties. 33 CHAPTER II REVIEW OF LITERATURE This chapter of the study has two segments. The first portion presents a review of the empirical literature that supports the need for the study, identifies and explains key variables of the research, and establishes the relevance of the concepts in relation to the theoretical framework. The second halfof the chapter describes the educational program that provides the context in which this research is conducted. Review of the Literature This section presents a review of the empirical literature to identify and explain key variables of the research They are, locus of control, parenting, mothers, limited resource, parent education and home visitation. These constructs are reviewed and discussed in relation to the theoretical fiamework and serve as a rationale for this study. Introduction to Locus of Control Theconceptoflocusofcontrolemerged fi'omsociallearningtheoryoverthree decades ago fi'om the work of Julian Rotter. The construct describes the degree to which individuals believe reinforcement '8 contingent upon their own behavior, and is defined in terms ofexpectancyvalues. Expectancyvaluesreferto the probabilitytlnt aparticuhrreinforcement willoccurasaresultofaspecificbehavior,orone’sabilityto anticipatehissuccess. Inother words,wantingtoachieveagoal'snotenough;onemustalsoconsidertheexterrtto whichan 34 individual can expect to succeed (Rotter, 1966). Specifically, locus of control refers to where individuals find the decision making factors that influence life (Swick & Graves, 1986). The concept involves an individual’s view of who or what is in charge of events in their life and ranges fi'om internal to external (Rotter, 1966). Internal locus of control is the belief that an individual can determine his/her own fate within limits. External locus of control is the belief that he/she is controlled by forces outside him/herself (Lefcom't, 1976). Locus of control orientations are developed and reshaped throughout the life span based upon an individual’s life experiences. The concept is not a fixed trait, but rather one thatsis continually shaped and influenced as a product of the individual within the context of his or her own environment (Rotter, 1966). As a result of these lifelong experiences, a fi'amework for decision making is configured, centered within the construct of locus ofcontrol (Swick & Graves, 1986). While most people lean toward one end of the continuum or the other, locus of control is often situation specific and people tend to function in a rmnner that reflects both internal and external beliefs (Smith, 1985). Factors Related to Locus of Control Orientation Socioeconomic status Inadditionto the lifeexperiencestlmt develop and refine locusofcontrol orientation, therearennnyoflrervafiableswlfichlnvebeenlhrkedtothisconsfiuctmthe literature. Researchers suggest that locus of control orientation is a function of socioeconomic status. Individuals fi'om lower socioeconomic status groups tend to score as extemals, while 35 individuab fi'om middle and upper socioeconomic groups generally score as internals. For example, mastudyexammnrgflremhfionshipbetweenbcusofmnmlscoresamlschool achievement among black andwhite students, Shaw& Uhl(l971) found that lower class bhcksmflmeslndsigmficmfilyhiglmemamlworesflmuppernuddkchssbbcksam whites. It ins been theorized that this association between lower socioeconomic status and externallocusofcontrolorientationisaresultofarealisticresponsetotheexternalcontrol factors of poverty (Phares, 1976). That is, the lower socioeconomic groups tend to perceive themselves to be more externally controlled than more advantaged socioeconomic groups. ThispremiseissupportedbyAdleretal. (1994)who suggestthatpersonalcontrolisa‘higher ordervariable”whichindividuals fiemhighersocioeconomic groupsdevelopasaresultoftheir having rmre frequent and/or more significant opportunities to influence those events which aflbcttheirlives. Moreover, the work ofGore and Rotter(l971) and Gurinand Epps (1975) suggeststlntmiddleincomeAfiicanAmericanstendto expressrnoreconfidence intheirabilityto influencesocialenvironrnent, whichrmyresult indevelopingcopingskillsto externalstressorsandleadtoamoreintermlorientation. EM! Resemehaslnvealsosuggefledflmflremisamhfiomlfipbetweenraceandetmncuy andlocusofcontrolorientation. IngeneraLwhitestendtobermreintemalintheir orientations than blacks and other minorities (Lefcomt, 1976; Luster, 1987). Numerous strdhslnveexamhndbcmofmnfiolorflafiomofAfiicanAnericmsandwhnes Consistently, Afiican Americans have yielded more external reports (Dyal, 1984). In addition, 36 several meta-analyses have been conducted regarding race and locus of control orientation. ResearchersagreethatthelikelihoodthatblacksintheUnited Statesarermreextemally oriented than whites in the United States is high. While socioeconomic status is probably confoundedwithraceinsome instances, stmliesthathavecontrolled for SES supporttlre hypothesis that this is generally not the case (Lefcourt, 1982). However, numerous investigators have found internal orientations within minority groups. For example, Palisi (1988) examined the locus of control orientations of 42 non- working, low income black and Hispanic mothers with children in Head Start. Although their meanscoresweremore externalthanmeanscores for white low income rnotherswithchildren in Head Start, mean scores for both minority groups were internally oriented. Furthermore, within group research of 177 Afiican American suburban women found evidence ofintemality, andsuggeststhatinternalcontrolisachieved followinganacceptanceofblackidentityand successfirltransitiomthroughthe racial identityprocesswhichallowsthemto experience control over their lives (Martin & Hall, 1992). These findings support Carter’s (1991) premise that a limitation of the research regarding minorities and locus of control, especially among AfiicanAmericans, isanassmnptionthatallmembersofaparticulargroup interpretsituations andlifeeventssimilarlyandfailto studywithingroupdifl’erences. Omparticularcriticismofthe locus ofcontrolresearchregarding minoritygroupsis the ambiguity in distinguishing between an individual’s external locus ofcontrol orientation and hisperoeptionofanenvirorunent irrwhichhemay have arelative lack ofcontrol. Intact, Gurin & Epps (1975) postulate that an external orientation rmy be adaptive for minorities as it helps them fiom intermlizing the effects ofracisrn. More specifically, a healthy external locus of 37 wnnoloriemmbnrecognizesflntalflmughmnflividualmyhaveuiedherbesttoattainher goahfleremesocietalresfiahusflmfluemenherabflnytownuollerenvuomemm ultimately achieve. When factors beyond her control prevent her from achieving her goals, an extamloriafiafionvfiflaflowhertomtacceptmsponsrbflityforpersomliaihue. Ontheother hand, an individual with an unhealthy internal orientation may feel that she should be able to overeomeallibrcesofracismandmaybeplacinganmrtairburdenuponherselfifhergoalsare notachieved. Inflesesituatiomthebhrmishnermfizedbecauseflemdividualbelievesflre faultiswithinhermotsociety. mgr IbeisdsoambsMrfiialbodyofresemehwhichsuggestsMeflmfityisafinnfion ofgender.Theresearchonwomen, especiallythosewhoareAfiicanAmerican,hasofien lbcusedontherole ofpowerarxlcontrolintheirlives(Pirxlerhughee, 1982, 1989; Solomon, 1976; Weick & Vandiver, 1982). Women tend to score in the external direction while men generally score more intermily (DeBrabander & Boone, 1990; Dyal, 1984; & Mwamwenda, 1995). In an examination of the relationships among locus of control, gender, and academic achievermnt, Kanoyetal.(1990) fomrdthatwomenwhoreportedinternalorientatiomwere abhmwcembhrmformsuwessfiflmadenncemerhmeshumtmkecrednforacadennc success. Ontheotherhandnrentypicallyreporttaking creditfortheiracademicsmcess (Meeser, 1972). DeBrabanderandBoone (1990) hypothesizethat women’s tendency toward extermlityrmyberelatedtotheideatlntfemalesareinfluencedbytheirperceptionsofwhatis 38 sociallyacceptable. Intheirresearch, fennleanswers onassessmentstendedto reflectthe societalpereepfionflntwonnnarennmdependMuponextermlfactorsflnnmen This hypothesisseemstobeconsistentwith theworktinthasbeendoneintheareaoffemale development. For example, Gilligan (1995) suggests tint in general, women are not encomaged to articulate their need to achieve power, although this is particularly true for women within lower socioeconomic groups. In addition, the female personality is much more likely to define itselfin terns of a comrection to others than the nnle persornlity (Carter, 1991; Chodorow, 1988). The value tint women place upon interpersonal connectedness and caretaldng beinviors nny nnke it difficult for them to perceive themselves as individuals outside of their relatiornhips with others. Therefore, the critical transition in the development of adult women is tint they begin to acknowledge themselves as deserving of the corniderations they grant others as oppowd to defining “goodness” in terms of selfsacrifice. The capacityforwomento nnkethistransitionennnates fiomanunderstanding ofthe destruction tint continuous self sacrifice breeds within thermelves (Gilligan, 1977). In contrast, other investigators exploring the relationship between gender and locus of control orientation inve not found significant differences. For example, Lee & Dengerink (1992), in a replication ofa 1972 study examining locus ofcontrol in relation to gender and rntiornlity, foruxlmsigrnficamdifi‘eremesbetweenwhitennlesandfennlesmtheUnned States. The authors suggest tint these results nnybe a reflection ofsocietal changes irrthe United States inthe last decadewhich may lnvedirninished some ofthe moreuaditiornl socialimtion processes, and eliminated gender differences in the locus of control construct. Fumermom,mastudyofnnddknmmeAfiicanAnnficannnlesandfermksbetweenthe 39 ages of 23 and 45, there were no significant differences in locus of control orientations, and both genders reported a general tendency toward interrnlity (Cain, 1994). According to the author, this overall propensity to irrterrnl orientations may be attributed to the successfirlness of this sampleand aclearreflection oftheir sense ofpersonal powerand responsibility. In conclusion, investigators would agree that there are nnny variables tint affect an individual’s locus of control orientation. The disagreement revolves around which of those variablesseermtobethemost influential Ingeneral, theiocusofcontrolconstructis influencedbytherealworld. Groupswho possessrnininnlsocialpowerinterrnsofclass, race, or gender, tend to score higher in the exterml control direction (Lefcourt, 1976; Solonnn et al., 1971). Conversely, those individuals who have more opportunities to achieve positive outcomes,whetheritbearesuitofgroupmembershiporsociaiposition,aremore likelytobe internally oriented (Lefcourt, 1976). Ultimately, one must consider tint there are multiple ecological factors that influence individual development. Locus of control orientations among andudtifingrormswfllvmywmrdhrgtoflnrmgemdonmesofhflividualpasoml experiences within the context of one’s environment (Carter, 1991). Significance of Internal Orientation man Intennllocusofcontrolhasbeenrelatedto avarietyofpositive behaviorsandpersornl clnracteristics. For example, Lefcourt (1982) suggests tint individuals who are interrnlly oriewedmeabkmseekomandufifizemfonmtionmomeflecfiwlythannflividmbwhome externally oriented, even if they rrny have access to the same inforrrntion. In addition, Phares 40 et al., (1968) found tint those who are internally oriented exhibit a greater willingness to addreesapotentialproblemassuggestedbyoonstmctive feedbacktinnthosewithrnore exterml orientations. Moreover, interrnl locus of control has been associated with better mental health. Specifically, investigators inve found evidence of negative associations with depression (Haworth, et al., 1997; Landau, 1995) and positive associations with life satisfaction (Kopp & Ruzicka, 1993; Landau, 1995). Internality ins also been associated withthe initiation and nnintenance of behavior change in areas such as weight loss, smoking cessation, athletic performance, and medical and substance abuse treatment programs (Lefcomt, 1982). Additionally, numerous studies have also found a positive relationship between locus of control and academic achievement (Lefcourt, 1991). In their review, Stipek & Weisz (1981) suggest tint students with internal orientatiorn receive higher grades and higher scores on achievement teststhanotherstudents. Inaddition, otherdimensionsofinterrnlcontroltintlnvebeen associated with productive living include a progressive ideology toward life, an optimistic view towardthe firture, effective problernsolving skills, apositive outlook, andaperception of difficult situations as creative challenges (Schaefer, 1983). EM Conversely, external locus of control has been linked to depression (Benassi et al., 1988), psychopathology (Lefcomt, 1976), poor selfconcept (Goodman et al., 1994), and low academic achievement (Lee & Dengerink, 1992). However, while exterrnlity has been associatedwithadverseclnracteristics, some investigators hypothesize tint exterrnl locusof controlappearstobeaprotective factorwhichallowsanindividualtorefiainfi'omfeelingsof 41 faihrre, remorse, or blame after experiencing a negative outcome (Rotter, 1966; Smith, 1985). The idea has been controversial in the literature, however, since many others inve linked exterrnl locus of control orientations with feelings of helplessness, hopelessness, and self deprecation (Goodrrnn et al., 1994). Rehtionship Between Locus of Control and Other Related Constructs Self esteem Ahhoughbcusofmnfiollnsbeenassocntedwnhseveralpersomfitywnmuctsmfln literature, two tirat have received considerable attentionare selfesteemand selfeflicacy. To betterunderstandthe locusofcontrolconstruct, itisnecessaryto describetheseother corntructsaswell. Selfesteemhasbeendefinedastheassessrrentindividualsrrnkeand nnintain about thennelves. The degree of self esteem describes the extent to which a person perceives hirnselfto be capable, successful, and worthy (Goodmanet al.,1994). Researchers suggest tint locus of control is one component of selfesteem (Wood, Hillrrnn & Sawilowsky, 1996). To finflnrmmpoflflrispreflsewarbusmsemehmsmifizingavmietyofmeasnesinve consistently found significant yet srrnll relationships between locus of control orientation and selfesteem (Dyal, 1984). Those who report higher interrni orientation scores also report higher self esteem, and reports of external locus of control orientation inve been linked to low self esteem (Eager, Howerton & Cobbs, 1994). In a qualitative study of black, urban mothers with Mndienpped sons, locus ofcontrol was positively associated with selfesteem and feelings ofpersonal empowerment (Morris, 1992). Moreover, Goodnnn et al. (1994) report tint women with the lowest selfesteem and the most extennl locus of control orientations inve the 42 worst view of themselves compared to other women in the sample. Selfe Locusof controllnsalsobeeneloselytiedtotheconstruct of selfeflicacy. Self eflieacyisanirxiividual’sperceptionofhowwellheorsheexpectstocopeinaparticular situation (Bandura, 1977). The original theoretical definition of locus of control as defined by Roflasuggestsflntmumnbelnvbrisdefinedmtenmofermectmcyvahnswhichisfln probabifityflntaparticularrehrforcemerflwifloccmasaresuhofaspecificbehavior. Perceptions of selfeflieacyinfluencehow much effort peoplewillexpendandhow longthey willpersistinadversity orwhenconfrontedwithchallengesmarxlma, 1982). Perceptions of selfeflicacyareinfluencedtiuoughrelatiomhipswithothers (Troutrrnn&Cutrorn, 1986). Theideatlnttiresetwoconstructsarerelatedisclearuponcarefirlreviewoftheirconceptual definitions. InfieLthetwoareoftenusedintheliteratmeintercinngeably. Aswithits counterpartlocusofcontrol, selfefiicacylnsalsobeenpositivelyassocntedwithavarietyof belnviors. Shifts in Locus of Control Orientation Becauseofthesubstantialbodyofevidencewhich suggeststhatinternallocusof control orientation is related to a variety of positive behaviors, and the notion tint individual orbrtafiommennlbabkthouglthhehfespmtmsemehemeindsigmficmtmtaefim determining if locus of control orientation can shift toward interrnlity (Nowicki & Duke, 1983). A number of investigators utilizing a variety of interventions lnve been successful in 43 documenting cinnges towards interrnlity with adults (Braton, 1981; Newsome & Foxworth, 1980; Roueche & Mink, 1976; Tait, 1976). For exanrple, in a study of thirty black mothers, a systenntized counseling program over the course of six weeks was effective in facilitating a significant change to a more interrnl orientation (Andrews & Gregoire, 1982) Locus of Control and Parenting Thereisconsiderableevidenceintheliteratmetint suggeststhereisarelationship between locus of control orientations and parenting. A theoretical fi'amework tint has received substant'niattentioninthe field ofparentingand supportsthisprernise isJay Belsky’smodelof the determirnnts of parenting (1984). This model origirnted in the literature of child uniteamnranggestSMpmemalfimiomisdetemfimdbyflueedormhnz characteristics of the child, contextual sources of support and stress, and the psychological mm of the parent. Psychological mm are the interaction of one’s developmental historyandpersornlitycharacteristics. One suchpersonalityvariable wirichinsbeenidentified as a critical component to parental functioning is locus of control (Stevens, 1988; Swick & Graves, 1986). Researchins linkedparerrtal locusofcontrolto avarietyofoutcomesinrelationto parenting. For example, in a study examining social support, locus ofcontrol, and parenting behaviors, Stevens (1988) postulates that an inmortant predictor of parenting skill for black and white adult mothers was locus of control. Specifically, interrnl scores for African American rmthersonanabbrev'nted version ofthe ANSIE wereassocntedwiththemother’sabilityto provide a stimulating environment for children as evidenced by the HOME measure. Researchers inve also found a conelation between parental locus of control and perceptions of children’s behavior problems (Harris & Natinn, 1973). That '5, pments who had external locus of control scores believed tint their children’s belnvior problems were a result of external influences, while parents who related their children’s problem to parental behavior had significantly lower exterrnl scores. The authors conclude those parents who are more likely to believetintlifeandlifeeventsaredeterminedbyfate,arealsonnre likelyto believethatchild remmgandpmentmgwnsequemesarebawduponiateorcinmeandactaccordmgly. Moreover, some suggesttintthoseparerrtswho feelpowerlessnessintheirownlivesallow flnsefeelingstofiarrscerrdirrtotireirroleasparerrts($wick&Graves,1986)andchildrenwho fivemmdexperbmeflresepowerlessmvuommmmendtonndelmesarmmuihnesmflnn own attitudes and behaviors (Kempe & Kernpe, 1978). Acwrdmgmflnfiteratmemmentsinvebeenidarfifiedasflnbasicmlenndelforthe development of locus of control orientation (Swick, 1984). External child orientations have bemassochtedwhheflamlernlofiafiafiommflmowcfivecmhrwmgsMedBmfing, 1982). The authors contend tint this association makes sense as protective mothers allow their childrento havelessopportunityto experience powerandcontrol inlife. Onthecontrary, nnthersofinternalchildrenscoredsignificantlymore interrnllythanmothersofexterrnl children and there were positive significant correlations between children’s and mother’s 'nrterrnlity/exterrnlity scores (Chandler, Wolf; Cook & Dugovics, 1980). The investigators abodocunnmdflmmtermlchildrenhadpammswhoenmbyedamhofinfivenmhodsof discipline, were accepting, nomestrictive and rewarding of independence, and used positive rather than negative verbalimtion techniques. 45 Parenting W The literature has been clear in documenting parenting behaviors that lead to healthy outcomes for children. Evidence strongly supports that stimulating, organized and safe home environments, supportive social networks, age appropriate developmental expectations, and positive discipline strategies are among the significant predictors of competent parenting. For example, researchers have found evidence which supports that what parents know about child development is positively related to parenting skill (Cook, 1991; Stevens 1984). Parental knowledge of child development has been found to be positively associated with mother’s responsivity (Stevens, 1984), and positive parent child interactions (Chandler, Wolf, Cook, & Dugovics, 1980). Fulton et al., (1991) found that increases in teenage mothers’ knowledge of infant development were positively correlated to increases in knowledge regarding appropriate parent-child interaction. In addition, parents’ knowledge of child development has been negatively associated with punitive discipline (Johnson, 1993). Findings suggest that parents who have abused or neglected their children are less knowledgeable about child development (Twentyman & Plotkin, 1982) and have poorer problem solving skills (Azar, Robinson, Hekimain & Twentyman, 1984). In other words, abusive and /or neglectful parents were found not only to have expectations for the child’s behavior which were inappropriate for the child’s ability but also less able to select adaptive strategies in challenging situations. Furthermore, parents who were not considered to be abusive but did admit to a punitive child rearing philosophy were also less knowledgeable of child development, had less social support, 46 and were more depressed than non-punitive parents (Reis, Orrne, Barbera-Stein & Herz, 1990). Relationship between conmetent parenting and child outcomes An abundance of empirical evidence suggests that children who are parented by psychologically healthy adults, who provide safe, stimulating environments, and interact in responsive, nurturing ways experience a host of positive outcomes. For decades, investigators have been documenting significant correlations between parenting behaviors and child outcomes. For example, in their review of the literature, Maccoby & Martin (1983) concluded that warm, involved, nurturing parents had children who were competent, responsible, independent, confident, achievement oriented, and able to control aggression. Furthermore, parental sensitivity to task, level of aggravation towards child, and extent to which independence is encouraged have been linked to secure attachments and children’s problem solving skills (Easterbrooks & Goldberg, 1984). Also, for over 30 years, researchers have linked developmental outcomes of children, especially in the areas of cognitive and social emotional growth, to the quality of the home environment (Gottfried, 1984). Evidence supports that characteristics such as the provision of appropriate play nnterials, the adequacy of available space, the availability of learning nnterials, and number of books are positively associated with cognitive measures such as achievement test scores and IQ scores (Bradley & Caldwell, 1976; Wachs, 1992). Clearly, the literature demonstrates that warm, responsive, sensitive parents who are in tune with their children’s capabilities and developmental tasks promote emotional security, 47 behavioral independence, social competence and intellectual development in their children (Belsky, Lerner & Spanier, 1984). Ethnicgy' and menting A risk associated with the identification of competent parenting behaviors is that often the standards tint are accepted and observed among majority families are assumed to bethesameasthose for minority families, aswell. Whentheserrnjoritystandardsare not adopted by the minority group, differences are often described in terms of deficits (Kelly, Power & Wirnbush, 1992). This idea of a deficit focused frame of reference is particularly true for Afiican American families whose parenting styles have been characterized as harsh, rigid, and strict (Alvy, 1981). Cultural and contextual perspectives are now being examined, however, to better understand Afiican American parenting practices that have been characterized as more authoritarian and parent-focused than those used by middle class whites. For example, Afiican American mothers have been described as failing to consider child needs and having high expectations for obedience toward parents and parental authority (Baumrind, 1972). However, more recent research suggests that contextual factors influence parenting style and has begtm to attribute this difference in parenting styles to the community context (Baumrind, 1991). For instance, Afiican Americans in the United States are more likely to be living in poor and dangerous communities than whites. Some suggest tint authoritarian parenting practices are a protective function for youth who are more likely to be living in poor and dangerous cormnunities (Baumrind, 1991). Brody et al., (1998) have defined a phenomena 48 referred to as “No Nonsense Parenting” among Afiican American parents. This parenting style is characterized by high levels of parental control, including physical punishment, and high levels of affectionate behaviors. It is based upon the work of those who have investigated black families and the idea that parents try to raise their children in order for them to fimction effectively in their environments. Researchers suggest this method is an adaptive parenting approach for parents living in dangerous cormnunities, and is intended to communicate a vigilant concern for the child’s welfare where disobedience could have sober consequences. These stringent parenting practices are believed to promote children’s self regulatory competencies and protect them from danger by discouraging disobedience and antisocial activities (Kelly, Power & Wirnbush, 1992). While this strict parenting style has been correlated with negative outcomes for non-minority youth, it appears as if authoritarian parenting is less deleterious for minority children (Baldwin, Baldwin & Cole, 1990), suggesting that minority youth living in a dangerous community may benefit fiom stricter parenting. An explarntion may be tint since the cultural and community context is influential in determining norms of parenting practices, children may perceive “No Nonsense Parenting” to be evidence of parental involvement and concern, reducing the likelihood that the parent is considered to be excessively harsh or punitive (Lamborn et al., 1996). Parent Education The significant predictors of competent parenting include the ability to provide stimulating, organized and safe home environments, establish supportive social networks, 49 develop age appropriate developmental expectations, and engage in positive parenting behaviors. However, researchers have substantiated that parents’ ability to care for and mature their children may be jeopardized for a variety of reasons. Living in a stressfirl or chaotic environment, having limited support networks, and lacking knowledge of normal child development may limit parents’ capacity to foster their child’s optimal development (Seitz, Rosenbaum & Apfel 1985). Thus, targeted outcomes of parent education programs often include the development of the qualities that promote competent parenting and/or the amelioration of those factors that jeopardize families. Fmthermore, the important connection between competent parenting and child development has long been supported by the literature (Maccoby & Martin, 1983). Evidence suggests that child development is multiply determined, dependent upon farme and community influences, and any intervention tint focused solely on the child is likely to have limited success (Belsky, 1984). Parent education programs have the potential to directly and indirectly influence the child by creating positive parent, family and contextual changes. These changes within the family system and community context can lead to significant and sustainable impacts for children which are more likely to continue long afier the intervention has subsided. (Black, 1994). Admittedly, the noble goal of helping all parents to create change through enhancing personal development and/or strengthening parent-child interaction and parenting skills can not always be achieved by a parent education program However, there is a great deal of evidence which exists suggesting that parent education programs are effective in creating a host of positive changes in parents and in the children of parents 50 who have participated (Black et al., 1994; Brems, Baldwin & Baxter, 1993; Powell, 1983). Changes in children as a result of their parents’ participation in parenting education that have been documented include improvements in cognitive ability (Gray & Ruttle, 1980; Madden, O’Hara & Levenstein, 1984; Slaughter, 1983), social skills (Mischley, Stacy, Mischley & Dush, 1985; Seitz, Rosenbaum & Apfel, 1985), achievement (Phamrenstiel & Seltzer, 1989), more positive affect when interacting with the parent (Giblin et al., 1984), and child resiliency (Wyrnan et al., 1999). In general, parents who participate in parent education report higher levels of control in their lives and their children’s lives as a result of such programs (Brown & Swick, 1979; Gordon, 1977; Levenstein, 1977; Watson, Brown & Swick, 1983). Outcomes for parents involved in parenting education include positive changes in parental attitudes (Anchor & Thomason, 1977; Dembo et al., 1985: Taylor and Beauchanrp, 1988; Telleen, Herzog & Kilbane, 1989), enhanced knowledge of child development (Field, 1981; Fulton, Murphy & Anderson, 1991; Roosa, 1984; Stevens, 1988; Taylor and Beauchamp, 1988), and the development of positive child rearing skills (Brems, Baldwin & Baxter, 1993; Fox, Fox, & Anderson, 1991; Mischley, Stacy, Mischley & Dush, 1985). For example, when compared to a waiting list control group, an experimental group of parents improved parenting perceptions and discipline practices following a 4 week, 10 hour parent education program. Changes in discipline practices were sustained through a 6 week follow up assessment (Gross, Fogg & Tucker, 1995). Additiornlly, Head Start mothers were observed in their homes to nnke significantly fewer critical remarks and cormnands, to use less harsh discipline, and to be more positive and competent in 51 parenting practices when compared to a Head Start control group not receiving the program. The intervention was an 8 week parent education program designed to teach effective parenting skills to families with young children (Stratton, 1998). Home Visitation Parent Education While parent education is delivered througha variety ofstrategies, an increasingly popularwaytoreachfamilieswithyoungchildrenisthroughhomevisitationservices,or programstintgointoconnnmritiestoreachfamilieswhereflreylive. Someresearchers arggefltlnthresponeflmlnnnvnnmbnmogrannandearlymmnsWhigh levelsofsuccessassociatedwithsuchprograms(01ds&Kitzrrnn, 1993). Thisresearch renfiedmanmflrmofremmeesaflocmedtohonnservicennavemiomandflnusarflsof programs acrossthe United States (Gomby, Cuhoss& Behrnnn, 1999). Ingeneral, caution mistbeeneMedwhenamlyzmgflnefiecfiveressofmnnvisimfionshnehdemrflresa deliverystrategyrathertinnaprogram. Withinthisstrategythereisagreatdealofdisparity amngimiividualprogrannandtheirprinnryactivities. Clearly,thesecontextualdiflerences lead to a host ofdifi‘erent outcomes, variable success rates and mixed findings (Gomby, Cuhoss &Be1nnnn,l999). However,whiletheresultslnvebeenmixed,thereisevidencetlnt suggeststhathonnvisitationpmgramscanbeanefi'ecfivewaytoreachfannheswithyomrg children(Carrpbell, 1994;Oldset a1, 1986; Hardy& Street, 1989). For example, Oldsetal., (1986; 1999) demonstrated positive results of honre visitation using outcomes rented to child nnlfieafinerminchrdingchfldabuseandinjmymrxlnnternalhfecomse. Usingprofees'ornl nmsesashonefisfiommemwasmSWAdecreasemsubstmuntedabuseaMmgbamsesm 52 thefirsttwoyearsoflifeinthetreatmentgroupwhencompmedto acontrolgroup sample. Moreover, fifteen years later, results indicated the home visited women had fewer subsequent pregnancies, fewer months on welfare, fewer arrests and convictions, and less alcohol and drug problems. Fmflrerrmrefireprogrmnlndthegreatestbernfittothennthersnnstatrisk. Hardy and Street (1989) concluded tint there was a significant reduction in children’s hosphalimfiomaslnrpmdmfionmwspeaedchfldabusemdmgbcuandamdmfionmthe substantiatedreportsofchildabuseandneglect irrahomevisitedgroupcomparedtoacontrol groupwhodidnotreceivetheirrtervention. Otherinvestigatorsexaminingahomevisitation program for families with newborns at risk for poor outcomes found tint mothers in the experimental group reported munerous positive outcomes including improved parenting efficacy, decreased parenting stress, and increased use of non-punitive discipline techniques whencomparedto acontrolgroup notreceiving services(Dugganet. al.,1999). Firnlly, after 18 months of bi-weekly home visits to an extremely high risk population of single, low income, AfiimnAnnrhanwonnnwhhmbstameahrseMstorhsflnexpahnanalgmupwasnnm likely to provide responsive and stimulating home environments than a randomized comparison group (Black, 1994). While each of the above interventions had different goals and program specific activities to address them, all of the authors suggest that the key to helping vulnerable families through home visitation interventions is in the supportive relationship between the home visitor and the program participant (Black et al., 1994; Duggan et aL, 1999; & Olds et aL, 1999). It is through the connection between the staff member and the family tint change is expected to occur. Although there seems to be agreement on the potential that 53 home visitors have in influencing a parent’s feelings of self and his or her behaviors, there is a great deal of controversy regarding whether or not home visitors should be professionals or paraprofessionals. Thus far, evaluations of both models have yielded incomistent results. However, there does appear to be consensus that regardless of experience and background, home visitors must be well trained and supervised to effectively serve families (Gomby, Cuhoss, & Behrman, 1999). Ingeneralresearchersagreeflntflnremeoflnrmrponamdennntsofsuccessfifl home visitation programs, as well. For example, differences in life cireunntances, emotiorni status and individual needs ofpotential clients require tint prograrrn are willing to vary content and duration (Barnard et aL, 1987). Parents should inve choices so that the program is individualized to their own needs as much as possrble. In addition, while the program’s curricuhmrisanirrportantelement, hornevisitorsnnyneedto setitasideto dealwitha family’s more critical concern or unanticipated situation. The ability of staff to respond timely andsemifiwlymfanflyisnesiswidelywcemedasomofflnsnmgthsofmnnfishafion (Gomby et al., 1999) and should be incorporated into the program’s design. Another element of successfirl home visitation prograrm is related to program dosage. Dosage refers to both the intensity and duration of the intervention. Models vary widely in theiprwficesmndwnnstrfliessuggefisflflflmsewhorwefiemmvisfisreapmmbemfim (Black et al., 1994; Olds et al., 1999; Wagner & Clayton, 1999). While a definite number of visitsoriengthofanintervention'snot known, some have speculatedtintaminirmunof4 visits, or three to six months of participation, is needed before change can occur. In addition, it iSposeibIetlntprogramswithlowintensitylevels, suchasrrronthiyvisitsnnynotprovidean 54 appropriate level of vrsrtmg to expect changes within the target population (Gomby, Culross, & Behrman,]999). A firnl recommendation fiom those analyzing research on several national home vkimtbnmogrammdekisflntmbeefieefive,mogrannmusthnegrmesocnlsupmnwim infornntion. Ingeneral, therelationsiripbetweenthe home visitorand theparent iscritical to thecourseofchange forfamilies,andtheefl‘ectivenessofhomevisitationprograms. Yet, if familiesareto achievesuccessirrcinngingattitudes, sldflsandbehaviorsflresupportmustbe mkMedeacmrkuhmMmovidesmfbnmfionpmfimladvbeorothaassistanw (Gomby, Culross & Beimrnn, 1999). Summary of the Literature This comprehensive review of the literature serves multiple purposes. First it intends to provide a thorough definition ofthe locus ofcontrol construct, and a comparison between locus of control and other related concepts. Secondly, it discusses the various factors which develop and refine individual locus of control orientations. This review also descnhes the significance of interrnl orientations in relation to hunnn behavior and personality cinracteristics. Inaddition, findingsfiomsnrdiesarepresentedwhichdennnstratethelinkage between interrnl locus of control and effective, responsive parenting. Finally, the literature base onparentingisexplored, aswellasparenteducationandhomevisitationservicedelivery modelsto betterunderstand their abilityto influenceand shapetireattitudesand behaviors of rnotherswithyotmgchildren. 55 The overall goal, however, is to provide a rationale for this research examining the locus of control construct in relation to parenting belnviors. Studies are presented which support the notion tint internal locus of control is related to positive behaviors, tint parental banofmmolhnpmmthepmenfingproceesandchiflrenarflflntnnavenfiominvebeen successful in influencing locus of control orientation. Furthermore evidence documents tint parenteducationprograrrnareeffective increatingpositive cinngesinfamilies, andhome visitationcanbeaneffective deliverymodel Therefore, empiricalevidenceinsbeenpresented whichsupportsthe studyoflocusofcontrolwithinthecontext ofparentingarxlparent ' education. Program Information Thissectionpresentsinfornntion abouttheBuilding Strong Families home visitation, parerfieducation program. Building Strong Families isdesigned for limited resource and/or limitedliteracyparemswithchildrennewbomtiu'oughtiueeyearsofage. Oneoftheguiding primiplesofthisstudyistodeterminetheirmactthatBSFlnsuponthelivesofthosemothers whoparticipateintheprogram. Build_rr_rg° Strong, Families cmriculum The Building Strong Families curriculum consists of multi-cultural, cartoon style flipcharts and short videotapes. The flipcharts and videos present scenarios that parents often encounter with their young children and prompt discussion of behavioral choices tint parents can make. Supplemental learning activities are used in conjunction with the 56 flipcharts. These activities are experiential in nature, and support key concepts found in the curriculum. In addition, concept sheets, or lesson summaries, are left in the home with the participants to reinforce the visit in the absence of the parenting instructor. The curriculum was designed specifically to meet the needs of a limited resource and/or limited literacy audience. The materials nnke no assumptions about parents’ support systems, extended family networks, spousal support, or ability to read. The curriculum has four units tint work together to present concepts that will help the parent, as well as the child, develop. The units are: How Kids Develop, which explores nornnl stages of child development; Helping Kids Behave, which presents positive discipline alterrntives tint are consistent with the stages of development; Playing to Learn, which focuses on the irrrportanee of positive parent-child interactions; and Smart Living, a process focused, personal development unit for the parent which addresses individual strengths and goal setting. Theprogramisastrengthsbasedprograrninthatitisdesignedtoreinforcethose things tint parents are already doing well, and introduce new infornntion and ideas in areas tint present an opportunity for growth. Parents are encouraged to determine their own starting point in the curriculum by identifying their personal parenting concerns early in the program. In addition, the BSF curriculum was designed to be fluid and flexible, and participants and instructors are encouraged to move between and among units as needed. 57 Buil ° Stron Families deliv The BSF program is delivered through trained paraprofessionals who are recruited from the communities in which they live and work, and have similar backgrounds and life experiences to the target population. The program is delivered primarily in the participant’s home, and occasionally in snnll group settings. The parenting instructors present the BSF curriculum and serve as a resource for additional parenting infornntion assistance, as well as a resource for referrals to other family support programs in their community. Typically the sessions are one to one and a half-hours in length and the intervention is short term, lasting between 8 and 12 weeks. Eligibility criteria for emoliment in the program is mininnl. The program is available to any caregiver, nnle or fennle, who has at least one child three years of age or younger and lives in a county tint offers the BSF program. The recommended income guideline for participation is 185% of the poverty level, which is consistent with other state eligibility assistance programs. Participation is vohrntary, and parents are at h'berty to choose whether or not they will enroll and/or graduate from the program. While many participants are referred as a result ofa court rrnndated ruling, BSF is not a regulated program, and participants are flee to choose. However approximately 90% of the families referred enroll, and of those 90% complete the program. In addition to mandated sources, other referral sources include juvenile court, WIC clinics, health maintenance organizations, Head Start, homeless shelters, emergency room teams, community mental health agencies, jails, enrployability programs, hot lines, word of mouth and self. Clearly, 58 the level of need and the intensity of services required for families referred through this assortment of comnnmity partners is extremely diverse. Along with participation in BSF, many clients also participate in substance abuse, mental health, infant mental health, and other therapeutic treatment as well as needs for perrrnnent and/or safe housing, enrployment, education, literacy, and/or other community services. Program goals, objectives, and activities The BSF program intends to indirectly impact the lives of children as a result of directly influencing parenting attitudes and behaviors. The program’s mission statement is “to provide caregivers with the knowledge and skills necessary to help children reach their potential”. Below are the program’s goals, objectives and activities. Goal 1: The program will enable parents to create positive and safe environments for children ages 0—3 years. Objective: Parents will irrrprove the physical enviromnent of their hone over the course of tie program Activity: Teach Childproofing and safety concepts that are presented in the How Kids Develop and Helping Kids Behave fiipcharts, videotapes, concept sheets and supplemental activities. Objective: Appropriate parent-child interaction will improve as a consequence of the program Activity: Teach positive parent-cth interaction concepts which include skills to 59 enhance play, touching, verbal and nonverbal communication, positive discipline, and care taking covered in the How Kids Develop, Helping Kids Behave and Playing to Learn flipcharts, videotapes, concept sheets and supplemental activities. Goal 2: The program will facilitate the personal development and self care of the parent. Objective: Parents’ social support network will improve as a result of the program Activity: Refer parents to appropriate community agencies; teach parents to identify and access infornnl and forrnai support systems through supplemental activities included in the curriculum. Objective: Parents will experience changes in self-efficacy or locus of control as a result of the program. Activity: Teach participants the importance of planning, goal setting, resiliency, and identifying personal strengths as highlighted in Smart Living. Objective: Parental feelings about selfwill change as a consequence of the P708131"— Activity: Teach participants to apply planning and goal-setting concepts taught in Snnrt Living to their own lives. Goal 3: The program will enable parents to respond to children in ways that are appropriate to the developmental stage of the child. Objective: Parental expectations of children will change as a result of the program. Activity: Teach participants developmental milestones for children newborn 60 through three years using the How Kids Develop flipchart, video, and concept sheets. Objective: Parents will improve their ability to respond to children in ways that are appropriate to the age/stage of the child as a result of the program Activity: Help participants apply concepts regarding child development, positive discipline, and parent-child interaction that are covered in How Kids Develop, Helping Kids Behave, and Playing to Learn in their interactions with their children. Progzarn values Investigators analyzing research of several home visitation program models implemented rntionally inve deducted that to be effective, programs must integrate social support with information. The relationship between the hone visitor and the parent is critical, yet it is important that the program content, or cmriculum, is delivered as intended as well (Gomby, Culross, Behrnnn, 1999). The Building Strong Families program and stafi’ value both the content of the materials and the process of program delivery, and strive to integrate these elements throughout the intervention. These core values were identified early in the program’s development and served as a benchrrnrk for program planning, implementation, and evaluation decisions. In general, the literature suggests that parental sensitivity can be enhanced by reducing stress, improving social support, increasing parents’ knowledge about child development, enhancing mothers’ self esteem and selfeflicacy, and providing practical assistance (Cmic et al., 1983; Culp et al.,l998), and serves as a guiding principle for the program. 61 In regard to content, the curriculum was based upon empirical evidence from the parent-child interaction, family relations, child development and abuse and neglect literature that supports healthy parenting principles and positive child outcomes. For example, parents’ knowledge of child development has been found to be positively associated with mother’s responsivity (Stevens, 1984), and positive parent child interactions (Chamberlin, Szurnowski & Zastowny, 1979) and negatively associated with punitive discipline (Johnson,1993). In addition, warm, nurturing parents with clear expectations and reasonable control had children who were competent, responsible, independent, confident, and able to control aggression (Baumrind, 1991; Maccoby & Martin, 1983). In light of these findings, information on appropriate expectations regarding the nornnl course of child development, positive discipline strategies, and positive parent-child interaction were included as key elements in the BSF curricula. The process of the BSF program is valued as an integral component of the program as well. Snnrt Living, a process-oriented unit intended to facilitate participants’ self care and personal development, was created to assist in building the paraprofessional- client relationship and guiding clients through a personal growth and development sequence. In addition, a thorough review of the adult learning, extension education, home visitation and mentoring literature was conducted to determine the most effective delivery strategy. Studies suggested that models integrating support and education were an effective design, and early results of home visitation evaluation indicated it is an effective way to reach families with young children (Olds et al., 1986). Also, while the literature yielded inconsistent results regarding the effectiveness of a paraprofessional versus a 62 professional model, several researchers have found an indigenous peer instructor to be effective working with similar audiences (Bradley & Martin, 1994). These findings, in conjunction with a 25-year history of home visitation programming within the organization, were the foundation for developing a paraprofessional, hone visitation model for BSF. Bufltlg Strong Families proggam evaluation A pilot evaluation of the BSF program was completed in October 1995. An independent evaluator provided leadership to the evaluation team that consisted of direct service and administrative staff involved with the program in Wayne, Gingham, and Jackson counties and the state level staff of MSU Extension. The pilot study focused on three major areas: changes in parents’ perceptions of their interactions with their children; parents’ satisfaction with and perceptions of the program; and instructors’ assessment of the impact of the program on the farniiy. Data to assess changes in parents’ perceptions of the parent-child interaction was collected at the beginning and the end of the program Parent satisfaction and instructor assessment questionnaires were administered at the completion of the program Pre and post assessnents on parents’ perceptions of parenting behaviors were analyzed and significant results were found. Specifically, after participating in the program, parents were more likely to: talk and listen to their child; make up games and play with their child; encourage their child to do things with his/her innds; encourage their child to play with other children; encourage their child to do things on his/her own; and let their child make choices; help their child be safe and secure; 63 encourage their child to play with other children; praise their child; discipline without spanking; give their child time to cairn down timing tantrums; and set appropriate limits for their children. In addition, parents reported being satisfied with the program and believed it to be helpfirl. Instructor assessments supported evidence that the program was effective for participants as well. CHAPTER III RESEARCH METHODOLOGY This section provides infornntion on the research design used for this study, including the sampling scheme, the instrumentation, data collection and data analysis. Research Design This study on the influences of a home visitation parent education program on locus of control orientation and parenting behaviors of limited resource women is the quasi-experimental design, Pretest-Posttest Non-equivalent Comparison Group. Figure 6 illustrates the study’s research design. Figure 6: Research Design Quasi-Experimental Design GROUP PRETEST INTERVENTION POSTTEST Experimental X X X Comparison X X Non—equivalent comparison group No random assignment 65 A true control group was not feasible, as local BSF adnrinistration did not feel comfortable creating groups for experimental and control conditions through random assignment from the pool of program referrals. To eliminate other explanations for any changes in the experimental group, however, a group as similar as possible on key variables was needed to serve as a comparison. A comparison group would help to isolate not only the effects of the experiment but also the effects of events that occur outside of the experiment (Babbie, 1992). Therefore, it was considered appropriate in this situation to design a study which utilized a non-equivalent conrparison group as opposed to a true control group with random assignment. Research Sample The unit of analysis for this study is limited resource mothers. The sample size is 50 mothers in the experimental group and 50 in the comparison group. This figure was based on a power analysis by Pecora et al., (1995), which recommended a sample size of 50 per group for an alpha level of .05 and a nroderate effect level of .80. The sampling scheme for the experimental group is all of the anthers 18 years, who ins at least one child three or younger or is pregnant, participating in the Building Strong Families (BSF) program in a large, urban, Midwest city. Building Strong Families is one ofthe community bawd educational progranrs offered in the city through the land grant C00perative Extension System The Building Strong Families program is a home visitation, parent education intervention for families with children newborn through three years of age. The program is delivered by trained paraprofessional staff indigenous to the community on a one-to- 66 one basis in the client’s home. Approximately 150 mothers in this city participate in the program each year. The referral sources vary, but most participants are referred by health care providers, substance abuse treatment centers, the Family Independence Agency, schools, head start, and child protection teams. Referrals are based on a range of circumstances and a continuum of participant needs. For example, some families self enroll with an interest in learning nrore about parenting; some are referred following a visit to the local Emergency Room after a child’s injury; some are referred by doctors who inve concerns about parents’ knowledge or skills; and some are referred as a result of a court order for substantiated child abuse or neglect. However, Building Strong Families is a voluntary program That is, each participant has the choice of enrolling in and completing the program, regardless of the situation that elicited the referral, or the referral source. Beginning in April 1999 through August 1999, each mother 18 years and older, who had a child three years of age or younger or was pregnant, tint enrolled in the BSF program was asked by her parenting instructor to participate in the study. Women under the age of 18 were excluded fiom participation. This decision to exclude mothers under the age of 18 was based upon the literature which suggested that locus of control orientation is influenced by age. Teens have been associated with more external orientations when compared to adults, which is probably a result of the high levels of parental, educational and societal controls imposed upon them (Morganti et al., 1988). Because of the propensity for teens to score in the external direction, they were excluded fiom the sample to avoid skewing the data toward externality. All fifty five of the women approached agreed to participate in the study. Each wonnn was given a brief 67 description of the study, asked to sign a consent form (Appendix A) and complete enrollment paperwork (Appendix B). After the enrollment procedures were finished, the parenting instructor asked each participant to complete two pretests for this study as part of the initial home visit. Each participant was assigned a family number by her instructor that was recorded on all of her assessments. No other identifying information appeared on any of tie forms which were sent to the investigator. The parenting instructors maintained a separate list in their files, which linked each family number with the participant’s name and address. The non-equivalent comparison group for this study was identified from clients of the Women, Infants, and Children (WIC) program, a federally funded supplemental food and nutrition education program for limited resource wonen and their children, administered locally by the Urban league. WIC’s target population is pregnant wonen, and women with young children. An examination of WIC records revealed that their caseload matched the BSF participants on the key demographic variables such as ages of children, ethnicity, income, family composition, and educational level. In addition, the Urban League’s WIC program provided services within the same geographic area that was served by BSF. Since the target population and service areas were similar, the WIC Urban League program was selected as the most appropriate recruitment site for the comparison group. To recruit volunteers, this investigator sat in the WIC waiting room on 6 separate dates between March 1999 and August 1999. As the women signed in for their quarterly WIC appointment, the investigator asked if they would be interested in participating in a research project. Only those mothers who were not participating in the 68 Building Strong Families program and 18 years of age or older who were pregnant or had at least one child three or younger were eligible to volunteer. Mothers were offered a $10 gift certificate to a local grocer in exchange for participation. Fifty nine mothers were asked to participate in the study. Fifty eight agreed to enroll. Volunteers were given a brief description of the study and asked to sign a consent form (Appendix C). Following the completion of their consent and enrollment forms (Appendix D), the mothers were assigned a number by the investigator for identification purposes. The numbers were in chronological order and the corresponding number was placed on all forms that were completed by participants. Those who volunteered to participate completed their pre assessments during their WIC appointment on the day of recruitment. According to WIC guidelines, participants must return to the WIC clinic approximately 12 weeks later for their next food coupon pick up. At tint appointment, vohmteers completed the post assessment and received their stipends. Fifty wonen were to be included in the comparison group, but 58 were invited to participate in the study to compensate for those who did not follow up with their WIC appointments. Fifty of the 58 women enrolled completed both the pre and post testing and were included in the comparison group sample. Research Instruments Based upon this research's purpose and hypotheses, the outcome measures or dependent variables for this study are locus of control orientation and parenting behaviors. The following instruments are used in this research: (1) The Adult Nowicki- 69 SH (Pl li—TJ Va it St Strickland Internal-Exterrnl Control Scale (AN SIE), (2) Parenting Behavior Assessment (PBA), and (3) Family Record Form. The AdultNowicki-Stricliand Internal-Enema] Control Scale (AN SIE) The Adult Nowicki-Strickland Intemal-External Control Scale (AN SIE) is a self- adnrinistered, self report, pencil and paper questionnaire (Nowicki & Duke, 1974). There are 40 items requiring yes or no answers. Unlike other existing locus of control instrunrents, the AN SIE is specifically designed to assess locus of control orientation in terms of internality versus extemality among noncollege adults (Nowicki & Strickland, 1983). The reading level is less dennnding and the true-false format is more easily understood than for other instruments (Lefcourt, 1991). The instrument was developed through an adaptation of the Nowicki-Strickland Intemal-External Control Scale for Children (CNSIE) and was selected due to its suitability for noncollege adults, construct validity, and known reliability. The questions were developed to measure one’s perceptions of the connection between his/her behavior and its consequences. Examples of questions include “Do you believe that most problems will solve themselves if you don’t fool with them?”, “Are some people just born lucky?”, and “Most of the time, do you feel that you can cinnge wint might happen tomorrow by what you do today?”. A copy of the AN SIE is included as Appendix E. Available information on the validity of the AN SIE states that the construct validity is based on information fiom more than 400 studies that have used the Nowicki- Strickland tests. A number of factor analyses have been computed for various 70 populations. These reports suggest that some factor structure overlap exists among the three nnjor scales for nreasuring locus of control orientation in adults: Rotter (1966), James (1973), and the AN SIE, and presents evidence that the instrument is a parallel scale constructed for different populations (Nowicki & Duke, 1983). Available information suggests that the AN SIE has an acceptable reliability level as well. Reliability for internal consistency was measured with split half reliability indexes, mostly between .74 and .86 (Nowicki & Duke, 1983). Test-retest reliability figures have varied from .65 with a 7 week interval to .83 with a 6 week interval and .56 with a one year interval (Lefcourt, 1991 ). The AN SIE was normed on American Caucasian college adults, although subsequent scales have been adapted for and tested on other populations as well. Various investigators' results suggest tint similar findings are reported when subjects are drawn fiom various cultures and comparable socioeconomic levels although variations in mean scores across groups have been reported (Nowicki & Duke, 1983). In addition, ANSIE scores have been found to be relatively free of social desirability bias and unrelated to intelligence scores or gender. The instrument is scored by assigning one point for each external response tint is selected. Exterrnl responses are indicated on an answer key, which accompanies the instrurrent. A score is assessed by adding together the total number of external responses. Therefore, the higher the number, the more external the orientation. Possible scores range fiom 0-40. 71 brill lmem effecli helm; 3Will it lUK review reflect- llltludt popula CollSltll Parenting Behgvior Assessment (PILA) The Parenting Behavior Assessment was developed as an adaptation of the Q—Sort Inventory of Parenting Behaviors. The PBA was created specifically to evaluate the effectiveness of the BSF program It measures parental perceptions of parenting behaviors. It is designed to assess parent-child interaction, as well as the consistency and appropriateness of parent behavior in relation to child development principles. There are no known reliability or validity measures on the PBA. However, an expert panel of reviewers verified the content validity of the measure to ensure tint items in the measure reflected positive parenting practices. Additiornl reviewers verified that all of the items included on the assessment were taught in the BSF curriculum. To be consistent with the principles of the program and the needs of the target population, the learning styles of limited resource/limited literacy audiences are considered in the adnrinistration of the instrunrent. There are thirty-two cards, each one representing a parenting behavior addressing the physical, intellectual, social, and emotional development of the child. Each of the behaviors identified on the cards can be directly linked to infornntion presented in the BSF curriculum. Examples of statements are: “I provide things for my child to play with”, “I help my child feel safe and secure”, “I let my child make choices”, and “I discipline my child without spanking”. A copy of the instrument is included as Appendix F. An innovative Likert Scale is created to assist the participant in making her perception of each of the particular behaviors. Five envelopes are labeled “Like me—all of the time”, “Like me—most of the time”, “Like me—some of the time”, “Like are—hardly ever”, “Like me—never”. After reading a card tint identifies a behavior, the mother selects the envelope which best reflects her l’-Y'1 behavior and places the card inside the envelope. This process is continued until all 32 behaviors are assessed and placed in one of the 5 envelopes. The cards rennin in the envelopes until the parenting instructor is able to code and enter the scores. A numeric value represents each of the envelopes, with five being the most fiequent behaviors and one being the least. The corresponding values are entered upon the score sheets and the figures are added to produce an overall score for all 32 behaviors. Composite scores range from 32, the lowest possible score, to 160, the highest. In addition, each of the 32 behaviors can be assessed individually as well. The PBA has been used to evaluate the effectiveness of the Building Strong Families program in relation to parental perceptions of changes in parenting behaviors. Program participants complete the instrument on their first visit prior to any educational intervention, and then again on the last. Each of the individual items is analyzed for movement between pre and post assessment. Reports are generated on an annual basis to describe program outcorres for funders and other program stakeholders. Currently, there are over 1,600 families who are in the PBA database managed at the university. FLmily Record Form The Family Record Form collects demographic infornntion (Appendix H). Parenting instructors interview mothers to conrplete the instrument. Information includes, gender, age, education level, incone level, family conrposition, race, household members and ages, residency, and relationships with community organizations. This infornntion is used to describe the demographic distribution of those who participate in 73 the program at the state and county level. The participant profiles are included in reports to funders and other program stakeholders. Data Collection Procedures The investigator met with the four Building Strong Families program staff assigned to the target community in November 1998 to request their assistance in data collection. All four parenting instructors agreed to cooperate. Each of the parenting instructors had been trained to administer the PBA and Family Record Form and was collecting this data as part of their current prograrmning responsibilities. The staff participated in additional training to administer the AN SIE and received written instructions to support their efforts during the data collection procedure. To ensure tint the instrument was sensitive to the gender, ethnicity, socioeconomic level, and literacy needs of the population, the ANSIE was pilot tested with BSF program participants, prior to the beginning of the study. Each instructor was asked to administer the instrument to her clients to learn if there were any confusing or offensive sentences or words on the instrument or in the directions. No concerns were identified. Beginning in February 1999 and lasting until September 1999, each mother 18 years of age and older who enrolled in the Building Strong Families program was given a brief description of the study and asked to participate. All of the mothers approached agreed to participate. Following a verbal commitment to participate, each participant was asked to sign a consent form, and complete the Family Record Fornr. Next they were asked to conrplete the AN SIE. Instructions for completion were written at the top of the questionnaire. Parenting instructors were permitted to assist with contextual clarification, 74 but were not permitted to engage in any dialogue or discussion which could influence a client’s response while she was completing the AN SIE. In addition, staff were instructed to strongly encourage mothers to select one response for each of the forty items. It took approximately 10 minutes to complete the AN SIE. Next, participants completed the Parenting Behavior Assessment. Each client was instructed to think of only one of their children who were three years of age or less, and record that child’s age at the top of the page. To complete the instrument, mothers reviewed each of the 32 cards that identified a particular parenting behavior and selected the response envelope that best described the frequency of their behavior. Each of the response envelopes represented a number along a Likert scale. At the conclusion of the assessment process, the envelopes were clowd. The parenting instructor coded and entered the responses on a data sheet upon returning to the office. This procedure took approximately 20 minutes. The entire pre-testing process lasted 40 minutes. Staff reviewed all forms after completion and submitted to the local program coordirntor who verified the records were complete and accurate. Data was nniled to the investigator on a monthly basis. Participants in the experimental group received the BSF program following the initial interview. A parenting instructor visited the client in her home weekly and provided support and education regarding nornnl child development, positive discipline techniques, appropriate parent-child interaction, and parental goal setting. The intervention ranged in length from 7 to 15 weeks, with the mean being 8 weeks. At the conchrsion of the program, the mothers repeated the Parenting Behavior Assessment and the Adult Nowicki-Strickland Internal-External Scale. In all, 55 mothers agreed to enroll 75 in the study and 50 conrpieted. In general, attrition can be attributed to participant relocation, staff turnovers, and changes in participant’s life situation. The non-equivalent control group for this study was recruited from clients of the Wonen, Infants, and Children (WIC) program, administered by the Urban League. The oflice was located less than two miles from the Extension office and provides services to wonen living throughout the same city. On 6 separate occasions, beginning in March 1999 through August 1999, the investigator sat in the WIC waiting room and solicited vohrnteers fi'om the women as they signed in for their quarterly WIC appointment. A sign was posted behind the investigator announcing the opportunity to earn $10, which elicited interest among the women. As the women registered for their WIC appointment, they were given a brief description of the study, and all women were asked personally to participate. If interested in participating, they were asked to sign a consent form Following the completion of a consent form, each mother was assigned a number in chronological order for identification purposes. The volunteers were given a clipboard that contained photocopies of the Family Record Form, the PBA, and the AN SIE. They received verbal instructions and completed their forms while they waited for their WIC appointnent. In addition, written instructions appeared at the top of the PBA and the ANSIE for reference. Due to logistical and staffing limitations at the WIC clinic, the mothers completed the PBA as a pencil and paper assessment, using a numerical Likert scale as opposed to an experiential activity like the experimental group. Although the process varied slightly, participants were still required to rate their perceptions of their parenting behaviors in relation to a specific child. It took approximately 20 minutes to administer the instruments. Before leaving the clinic, the volunteers were assigned a date 76 for their next appointment, which was approximately 12 weeks later. This date was shared with the investigator and recorded upon the Family Record Form for appropriate follow up. One week prior to her next appointnent, each mother was mailed a letter fiom the WIC staff reminding her of her scheduled visit. The investigator went to the WIC clinic at the participants’ scheduled appointment times and distributed the post assessments to the volunteers as they arrived. Post testing began in June of 1999 and lasted until October 1999. Again, each mother received a clipboard with the PBA and the AN SIE attached and received verbal instructions for completion. The participant completed her instruments in the WIC waiting room and returned them to the investigator when finished. At this time, the investigator verified that all information was complete. Tire volunteer was thanked and received her $10 gift certificate. In the event tint the investigator could not be present for a follow up appointnent, a BSF parenting instructor went to the WIC clinic for the appointment instead. In spite of the fact that clients receive their appointment dates well in advance and tint they are reminded of the appointment one week prior, 28 of the 58 women who agreed to participate in the comparison group failed to nnke their WIC appointment. Immediately upon missing their scheduled appointments, the investigator sent them letters in the mail, along with copies of the post assessments and self addressed stanrped envelopes. A copy of the letter is included as Appendix G. Participants were asked to complete and return their questionnaires immediately to guarantee their receipt of their gift certificates. Twenty eight of the 58 mothers missed their scheduled appointnent and were contacted through nnil to complete their posttests. Of those who received their 77 instruments in the mail, twenty rettu'ned their posttests to the investigator. Eight women dropped out of the comparison group without completing the post assessment. In sum, 55 wonen emolled in the study and completed pretests in the experimental group. Fifty of those women completed posttests, as well. Fifty eight wonen enrolled in the comparison group and completed pretests, Fifty of them conmleted their posttests. Therefore, the total sample size for data analysis is 50 in the experimental group and 50 in the conrparison group. Data Analyses Data was scored, entered, and analyzed using the Statistical Package for the Social Sciences (SPSS) Version 9.0 by the investigator. Statistical analysis for this research consisted of both descriptive and inferential statistics. Descriptive statistics were used to describe population parameters in terms of central tendency and variability. Means and standard deviations were reported for interval level variables, and cell counts and percentages for ordirnl and nominal variables. Inferential statistics were arnlyzed using T-tests and Pearson product moment correlation coefficients. To test tint the population mean of locus of control and parenting behavior scores were the same for both groups, the independent t-test was used. The independent t-test is used to test the equality of two neans for interval and ratio level variables. This procedure tested if the experimental and comparison groups were similar in their AN SIE and PBA scores prior to the educational intervention. It was also used to test if there were differences between groups on key demographic variables. A second analysis used in this study was the paired sample t-test. This procedtue tests to determine 78 if the mean difference on matched populations is 0. It was used to test the mean difference for each subject on AN SIE and PBA scores in both the experimental and comparison groups on a before and after basis. This analysis tests the null hypothesis in terms of differences in scores as a result of treatment status, or receiving an educational intervention. Pearson product moment correlation was used to examine the relationship between AN SIE scores and PBA scores. Pearson product moment correlation is the statistical analysis appropriate for testing the linear relationship between interval level variables. Coefficients range fi'om -—1 to +1 to describe either a positive or negative association. This analysis tested the hypothesis which examined the strength of the relationship between locus of control and parenting behaviors. Ethical Considerations An application was submitted to the University Committee on Research Involving Hurrnn Subjects (UCRIHS), the Institutional Review Board of Michigan State University, to conduct this research. Approval to conduct the study was granted by UCRIHS prior to the beginning of data collection. To ensure that the rights and welfare of the participants are protected throughout this study, several ethical considerations are taken into account. First of all, participants in the Building Strong Families experimental group are assured that they will receive the Building Strong Families program, regardless of their participation in the research study. Furthermore, they could elect to discontinue the study at any time for any reason, and still receive the BSF program. A consent form is distributed to each participant, to ensure that 79 she completely understood her rights regarding participation in the BSF program and the intent of the research. Appendix A is a copy of the consent form used with the experimental group. Voirmteers of the comparison group fiom the WIC program were also provided a consent form which explained the purpose of the research and their rights. They were assured that they would receive their WIC benefits even if they chose not to participate in the research and they could discontinue with the study at any time. In addition, the consent form describes their compensation for participation. A copy of the WIC consent form is attached as Appendix C. Privacy of the subjects is protected through confidentiality. Each participant in the experimental group is assigned an identification number by the BSF instructor at enrollment. Prior to submitting completed instruments to the researcher, all identifying information such as name and address is removed. Clients are identified solely by family number. Each BSF instructor maintains a master list linking participant names and case numbers in the event that it is necessary to correct irnccurate or missing information. The comparison group is assigned numbers as well, in chronological order upon recruitment. Like the experimental group, all identifying infornntion is removed from their paperwork, and participants were referred to only by number. A master list that recorded each participant’s name with her assigned number was kept in a locked file and referred to only for follow up on missing information. Privacy was further protected in tint results were only reported for the aggregate, with no reference to individual subjects or responses. 80 Additionally, both the experimental and comparison groups were advised of the intent of this research. Firnlly, the participants, the BSF staff and the WIC staff were advised that results of the study would be available to them upon request. 81 CHAPTER IV RESULTS This chapter contains the results of the data analysis. The infornntion is organized into three sections. The first section utilizes descriptive statistics to report aggregate demographic characteristics of both the experimental and control group samples. The second section focuses on the use of inferential statistics to draw inferences about the population based upon probability and statistical significance. The final section presents a brief conclusion of the results. Descriptive Statistics Demoggphic data A total of 100 wonen completed the study by providing data at pre and post assessment. Mothers were recruited for the experimental group through an urban county extension service tint provides a home visitation, parent education program for families with young children. Comparison group mothers were recruited from WIC clinics servicing the same geographic area and target population. Demographic data was obtained using the Family Record Fornr, which collected information about mother’s age, family composition, education, ethnicity, monthly income, number and ages of children, and household members. A summary of the demographic characteristics is presented in Tables 1 and 2. Table 1 presents cell counts and percentages of the nominal level variables. Table 2 presents means and standard deviations for the ordirnl and interval variables. 82 TABLE 1 Cell Counts and Percentages for Experimental and Comparison Group Demographics EXPERIMENTAL COMPARISON GROUP GROUP n=50 n=50 n '/o n % Ethnicity Afiican American 40 80 49 98 Caucasian 5 10 1 2 Hispanic 1 2 0 0 Other 1 2 0 0 Education High School Graduate 23 46 24 48 Non High School Grad 24 48 26 52 Family Structure Single Parent 28 56 35 70 Two Parent 15 30 10 20 Extended Family 6 12 5 10 Foster Fanrily 0 0 0 0 Other 0 0 0 0 Monthly Income $800 or less 26 52 24 48 3800-81000 12 24 17 34 31000-31200 3 6 3 6 $1200 or more 2 4 2 4 Age 18—23 19 38 17 34 24-29 15 30 17 34 30-35 9 18 7 14 36+ 4 8 7 14 Number of Children 0 2 4 1 2 l 25 50 13 26 2 12 24 12 24 3 8 16 11 22 4 2 4 10 20 5 0 0 2 4 6+ 0 0 1 2 First Time Parent Yes 27 54 14 28 No 22 44 36 72 83 TABLE 2 Means and Standard Deviations for Experimental and Comparison Group Demographics VARIABLE MEAN RANGE STD. DEV. Exp. Corn. Exp. Com. Exp. Corn. Education 11.26 1 1.40 8-14 8-14 1.34 1.35 Age 26.02 27.09 18-47 18-51 6.58 7.40 Number of Children 1.65 2.50 0—4 0-7 .95 1.38 Exp. = Experimental Group Com. = Comparison Group 84 The total number of mothers who completed the study in the experimental group was 50. The age of the anthers ranged from 18-47 years, with a mean of 26 years. Thirty eight percent (19) were between the ages of 18 and 23, 30% (15) were 24-29, 18% (9) were 30-35, and 8% (4) were 36 years of age or older. Or, one third of tie women were under 23 and two thirds were under the age of 30. Eighty percent of the women were Afiican American (40), 10 % (5) Caucasian, 2 % (1) Hispanic and 2 % ( 1) Multi-cultural. Data revealed that 56 % (28) of the sample were single mothers, 30 % (15) were fiom two parent families, and 12 % (6) lived in extended family arrangements. Over half, or 52 %, (26) were not high school graduates and 12% (6) had participated in educational programs beyond their high school diplomas. Seventy six percent (38), or three quarters, of the women earned $1000 or less per nnnth. Within the current sample, 66 % (33) were also emolled in Women Infants and Children (WIC), 54 % (27) received financial assistance and 58% (29) received food stamps. The mean number of children per family was 1.6 (SD = .95), with a range of no children (pregnant) to 4 children. Over lnlf (27) of the sample was first time parents. Five women in the experimental group dropped out of the Building Strong Families program prior to completing the study. Their information is not included in the summary of denngraphic characteristics. Fifty women completed the study in the comparison group. The age of the mothers ranged fiom 18-51 years, with a mean of 27 years. Thirty four percent (17) were between the ages of 18 and 23, 34% (17) were 24—29, 14% (7) were 30-35, and 14% (7) were 36 years of age or older. Again, approximately one third of the mothers were under 23 and two thirds were under 30. Ninety eight percent (49) of the women were Afiican American and 2 % (1) Multi-cultural. Data revealed that 70 % (35) of the sample were 85 single anthers, 20% (10) were from two parent families, and 10 % (5) lived in extended family arrangements. Nearly half, or 48 %, (24) were not high school graduates and 12% (6) had participated in educational programs beyond their high school diplomas. Eighty two percent of the women earned $1000 or less per nnnth. All of the mothers were emolled in Women Infants and Children (WIC), 34 % (17) received financial assistance, 34% (17) received food stamps and none reported participating in a parent education program The mean number of children per family was 2.5 (SD = 1.38), with a range of no children (pregrnnt) to 7 children. Approximately one quarter (14) of the sample was first time parents. Eight women in the comparison group declined to continue in the study following the initial assessment. A series of independent t-tests revealed that when equal variances are assumed, there were no significant differences between those who completed the study and tinse who did not. Table 1 shows that the distribution on the variables education, monthly income, and maternal age were nearly identical for the experimental and comparison groups. For both groups, approximately half were high school graduates, two thirds were under the age of 30, and three quarters earned $1000 per month or less. There were some observed differences between groups, however. Just over half of the experimental group were single parents, compared to over two thirds of the comparison group. Additionally, the entire comparison sample identified themselves as a member of a minority group, corrrpared to only 80% of the experimental group. Also, half of the experimental group had only one child, or were first time parents, conrpared with one quarter of the comparison group. In addition, 20% (10) of the comparison group had 4 or more 86 ch cle cor (Si 4,7 gm exp C0ll children in contrast to only 2 mothers, or 4%, with 4 or more children in the experimental group. Table 2 firrther describes the contrasts and similarities between groups on demographic variables. Both groups reported a mean educational level of just above eleventh grade (M = 11.26, SD = 1.34 experimental group; M = 11.40, SD = 1.35 conrparison group). The mean age of the experimental group was 26.02 years (SD = 6.58), compared to 27.09 (SD = 7.40) for the comparison group, with range of 18- 47 and 18-51 respectively. In general, the distributions show that the experimental group and comparison group were similar on most of the demographic background characteristics. However, nnthers in the comparison group were nnre likely to represent a minority ethnic group and inve more children than mothers in the experimental group. Also, nnthers in the experimental group were more likely to be first tinre parents than mothers in the comparison group. Pre-intervention analyses A series of independent t-tests were conducted on each variable to determine if there were any statistically significant differences between nnthers in the comparison group and mothers in the experimental group on any of the demographic characteristics. Table 3 presents the results, confirming that there were significant differences in the denngraphic data between the experimental and control groups in number of children (t = 3.679, p< .000), first time parent status (t = -2.736, p< .000) and ethnicity (t = 2.419, p< .018). Because of the small cell counts in some of the ethnic categories, the variable 87 TABLE 3 Testing Differences Between Group Means at Pretest on Key Denngraphic Variables Using Independent T—tests VARIABLE N T-TEST P VALUE Family Composition 99 -1.090 .279 Ethnicity 97 2.419 .018“ Monthly Income 89 .420 .676 Education 97 .458 .648 Enrolled in TAN F 99 -.897 .372 Maternal Age 95 .803 .424 Number of Children 99 3.679 .000” First Time Parent 98 -2.736 .000“ * statistically significant at the .05 level " statistically significant at the .000 level 88 was recoded into two categories, minority and non-minority. There were no significant differences in the other key demographic variables (family composition, t = -l.090, p< .279; monthly income, t = .420, p< .676; education, t = 458; p< .648; enrolled in TANF, t = -.897; p< .372; nnterrnl age, t = .803, p< .424). Tire pre-intervention analysis of denngraphic variables suggests that the experimental and control group are reasonably equivalent in terms of identifying characteristics, prior to the administration of treatment. Inferential Statistics Pre-intervention analysis of group means Independent t-tests were run to determine if the study’s experimental and comparison groups were similar in terms of the locus of control and parenting behaviors variables prior to treatment. Tables 4 and 5 present the results of the independent t-test arnlysis. Table 4 shows that the nean of the experimental group on the locus of control (ANSIE) scores was 15.04 (SD = 5.77), and the mean of the comparison group was 13.64 (SD = 5.60). The experimental group scored higher, but it was not statistically significant (t = -1.23, p< .221). These slight differences in group mean scores reflected that the experimental group scored nnre towards externality. Because the differences were not statistically significant, this finding indicates that the experimental group was similar to the comparison group in terms of locus of control orientation prior to receiving the Building Strong Families intervention. Table 5 shows tint the mean of the parenting behavior score (PBA) for the experimental group was 125.86 (SD = 21.46). The mean score for the comparison group was 132.50 (SD = 24.70). These scores also reflected slight differences in group means, 89 TABLE 4 Testing Differences Between Group Means of AN SIE scores at Pretest for Experimental and Comparison Groups Using Independent T-tests Pretest t-test for differences between group means Group N Mean SD Experimental Group 50 15.04 5.77 t = -1.23 Comparison p = .221 Group 50 13.64 5.60 TABLE 5 Testing Differences Between Group Means of PBA scores at Pretest for Experimental and Comparison Groups Using Independent T-tests Pretest t test for differences between group means Group N Mean SD Experimental Group 50 125.86 21.46 t = 1.44 Comparison p = .155 Group 50 132.50 24.70 r statistically significant at the .05 level ** statistically significant at the .000 level with the comparison group more likely to identify self with positive parenting behaviors, yet the differences were not statistically significant (t = 1.44, p< .155). Therefore, prior to treatment, the experimental and comparison groups were also similar in terms of the parenting behavior variable. Data regarding pretest scores on the locus of control and parenting behavior variables from Tables 4 and 5, in combination with Tables 1 and 2, indicate that the experimental and comparison groups were equivalent prior to the administration of treatment conditions. H1: Post-intervention differences in locus of mml ori_e_ngtion between gmup neans The primary purpose of this study is to determine the extent to which a home visitation, parent education program increases internal locus of control orientation annng nnthers. The level of measurement of the locus of control construct is interval and the aim of the hypothesis is to analyze the differences between two group neans. Therefore, the t-test is the statistical analysis that is selected for this study. It was hypothesized that antlers who complete the Building Strong Families pment education program will show an increase in internal locus of control orientation wiren corrrpared to the comparison group. This hypothesis is based on the premise tint the BSF materials include a process oriented, personal development component for mothers which should influence the degree to which they perceive reinforcement as contingent upon their own behavior and enhance their sense of personal control. A core element of the program is to help nnthers identify personal strengths, and set and achieve small, realistic goals. Once attained, the successful experience should shape their perceptions regarding the ability to influence their own lives. Moreover, the hypothesis 91 TABLE 6 Pre- and Posttest Results for Locus of Control Scores for Experimental and Comparison Groups Using Paired Sample T-tests t-value Pretest Posttest Difference differences between Group N Mean SD Mean SD Mean SD group means Experimental Group 50 15.04 5.77 12.60 4.99 2.44 3.03 t=5.69" Comparison Group 50 13.64 5.60 13.82 5.31 -.18 2.35 =-.54l TABLE 7 Pre- and Posttest Results for Parenting Behavior Scores for Experimental and Comparison Groups Using Paired Sample T-tests t-value Pretest Posttest Difi'erences differences Group between N Mean SD Mean SD Mean SD group means Experimental Group 50 125.86 21.46 139.92 15.15 -14.06 15.61 t=-6.37“ Comparison Group 50 132.5 24.71 133.22 20.58 -.72 24.87 t=-.21 * statistically significant at the .05 level ** statistically significant at the .000 level 92 also draws upon the studies that link a home visitation, parent education nndel of service delivery to increased self efficacy in program recipients (Black, 1994; Mitchel & Donnelly, 1993). Also, research has confirmed that parent education in and of itself can lead to participants experiencing changes in perceptions of personal power and control (First & Way, 1995). Furthermore, Swick (1986) postulates that meaningfirl relationships within the family and the larger community context can facilitate the development of a productive parental control orientation and BSF aims to establish those meaningful community relationships by utilizing the bone visitation delivery model. Therefore, evidence exists which suggests that feelings of power and control can be shaped through interactions with others. One would expect tint mothers participating in a hone visitation parent education program that emphasizes the client-staff relationship, and includes a personal development component, should experience shifts in their locus of control orientation toward internality. Paired t-tests were used to compare the experimental sample’s pre- and posttest scores on the locus of control scales. Table 6 presents the results of the analysis. As expected, the AN SIE scores decreased for the experimental group fiom pre to post assessment, reflecting a change toward more interrnl locus of control orientation by the endofthe program. The meanpretest scorewas 15.04 (SD = 5.77) andthe mean posttest score was 12.60 (SD = 4.99 ). There were no changes in the ANSIE scores for the comparison group, with a nean of 13.64 (SD = 5.60 ) onthe pretest and 13.82 (SD = 5.31) on the posttest. Table 6 shows that paired differences in group means for the experimental group were statistically significant (M = 2.44, t = 5.69, p< .000). The conrparison group showed no significant differences in group mean scores at post 93 assessment (M = -.18, t = - .541, p< .591). As stated previously, there were no statistically significant differences between the experimental and comparison group scores on the AN SIE at the pretest. Because the groups were similar at the initial assessment in terms of AN SIE scores and denngraphic characteristics, one may attribute differences between group scores at the posttest to the intervention with a reasonable degree of confidence. H2: Post-intervention differences in parenting behaviors between group means It was hypothesized tint mothers who complete the Building Strong Families home visitation parent education program will be more likely to identify self with positive parenting behaviors at the end of the program when compared to a comparison group. Tint is, parental perceptions of the consistency and frequency of positive parenting behaviors will increase for those who receive BSF, but not for those in the comparison group. This hypothesis is based upon the literature, which has found parent education interventions to be effective in changing the attitudes and behaviors of parents who participate (Brems, Baldwin & Baxter, 1993; Gross, Gogg & Tucker, 1995; Stevens, 1988). For example, outcomes include, eninnced knowledge of child development (Stevens, 1988), improved discipline practices (Gross, Fogg & Tucker, 1995), and positive child rearing skills (Brems, Baldwin, & Baxter, 1993). Parenting belnvior scores were tested for group difi‘erences using paired sample t- tests. The results of the analysis are presented in Table 7. The PBA scores increased for the experimental group fiom pre to post assessment, reflecting a change toward more positive parenting behaviors by the end of the program. Mean scores at pretest were 94 125.86 (SD = 21.46) and 139.92 (SD = 15.15) at posttest. There was an extremely small cinnge in the rrean PBA scores for the comparison group, from 132.12 (SD =24.82) at pretest to 132.83 (SD = 20.67) at posttest. Table 7 illustrates the differences in group means between the experimental (M = - 14.06, t = - 6.37, p< .000) and comparison (M = - .72 mean; t = - .21; p< .839 ) group following the treatment conditions at post assessment. There were no statistically significant differences between the experimental and comparison group scores on the PBA at the pretest. This evidence suggests tint the Building Strong Families program is effective in pronnting nnther’s perceptions of positive parenting behaviors in limited resource populations and further supports previous BSF evaluations which document statistically significant changes in perceptions of parenting behavior scores as a result of the program R§l_ations_hip between locus of con_trol and pa_r§nting behaviors Pearson product moment correlations were computed for the locus of control and parenting behavior variables to determine the strength of the linear association between the two. It was hypothesized that there was a negative correlation between locus of control scores and parenting behavior scores. That is, it was expected that as locus of control scores decreased, reflecting a more internal orientation, parenting scores would increase, reflecting nnre positive, responsive parenting. This hypothesis was grounded in the scientific literature which suggests there is a significant relationship between a parent’s belief tint she can influence her own life and responsive, competent, and nurturing parenting styles (Belsky, 1984; Lefcourt, 1986; Stevens 1984). 95 TABLE 8 Correlation Between Locus of Control and Parenting Behaviors at Pretest PrePBA PreANSIE Correlation PrePBA Coefficient 1 .0 .03 8 N 100 100 Correlation .038 1 .0 PreANSIE Coefficient N 100 100 TABLE 9 Correlation Between Locus of Control and Parenting Belnviors at Posttest PostPBA PostANSIE Correlation PostPBA Coefficient 1.0 -.168* N 100 100 Correlation -.168* 1.0 PostANSIE Coefficient N 100 100 * statistically significant at the .05 level Since the hypothesis predicted a negative correlation between the two variables, it was possible to utilize a one-tailed correlation arnlysis. Contrary to expectations, the data yielded inconsistent results. No significant correlations were found between PreANSIE scores and PrePBA scores (r = .04). However, a significant negative correlation was found between PostANSIE scores and PostPBA scores (r = -.168). This figure is a negative value correlation and suggests an inverse relationship, or that as AN SIE scores decrease, PBA scores increase. In other words, nnthers with internal locus of control orientations are more likely to identify self with positive parenting behaviors. Tables 8 and 9 provide a sumnnry of the Pearson product moment correlation coefficients for the variables locus of control and parenting behaviors. Infigroup analysis of locu_s of control The literature reports that there are nnny variables which influence the development of locus of control orientation. Examples of variables that have been shown to be associated with locus of control in the literature include socioeconomic status, ethnicity, and educational level. In general, the construct is influenced by multiple factors and those individuals who have nnre power to achieve positive outconres, whether it be a result of group membership or social position, are nnre likely to be internally motivated (Lefcourt, 1976). For these reasons, it was expected that there would be sone intra- group differences in relation to the locus of control construct on key demographic variables. The variables tint were of particular interest in this study were monthly income, education level, number of children and first time parent status. 97 Independent T-tests for Intra-group Differences among TABLE 10 Locus of Control Scores and Demographic Variables Variable Pretest Posttest Mean SD t p Mean SD t J) Family Comp. Single Parent 15.27 5.63 1.473 .144 14.00 4.97 1.448 .151 Two Parent 13.24 6.30 12.24 5.55 Ethnicity Minority 13.97 5.48 2.376 .019“ 20.00 6.63 1.625 .108 Non-minority 12.97 5.09 16.80 5.72 Monthly Income $1001 and over 10.50 5.04 -2.077 .041* 9.40 4.50 -2.397 .019“ $1000 and less 14.54 5.88 13.47 5.12 Education H.S. Grad 12.90 5.13 -2.550 .012‘ 11.82 4.94 -2.773 .007" Non H.S. Grad 15.80 6.09 14.66 5.14 Enrolled AFDC Yes 14.75 5.82 -.639 .524 13.48 4.67 -.495 .622 No 14.00 5.67 13.76 5.44 Age <= 23 years 12.58 4.42 2.182 .032“ 12.00 4.35 1.649 .103 >=24 years 15.14 6.10 13.76 5.44 Numb. Children 3 or more 13.30 5.85 -1.225 .223 13.06 5.48 -.142 .887 2 or less 14.72 5.54 13.22 5.04 1" Time Parent Yes 14.95 5.81 -.819 .415 13.27 4.83 -.071 .944 No 14.00 5.58 13.19 5.41 Age of Child <18 months 14.28 5.67 -.088 .930 13.02 4.72 .329 .743 >= 19 months 14.18 5.79 13.37 5.80 * statistically significant at the .05 level ** statistically significant at the .01 level *" statistically significant at the .001 level 98 Independent t-tests for intra-group differences demonstrated some significant differences, yet fewer than expected. Table 10 shows the results of the analysis for locus of control scores. Within the entire sample, there were significant differences between those who received $1000 a nnnth, and tinse who reported less than tint in terms of locus of control orientation. In general, more money was associated with internality. The mean ANSIE scores for those earning more than $1000 per month was 10.5 (SD = 5.04) at pretest and 9.4 (SD = 4.50) at posttest. For those earning less than $1000 per month, the mean ANSIE scores was 14.54 (SD = 5.88) and the posttest was 13.47 (SD = 5.12). The differences between income groups were statistically significant at the p< .05 level at pretest (t = - 2.077, p = .041) and posttest (t = - 2.397, p = .019). Significant difi‘erences were also detected within the education level vau'iable in relation to locus of control. Those mothers who graduated fiorn high school were nnre internal in their locus of control orientations than those who did not. The mean AN SIE pretest score for high school graduates was 12.9 (SD = 5.13) compared to 15.8 (SD = 6.09) for non-high school graduates. This difference was statistically significant at the p< .05 level (t = - 2.550, p = .012). At posttest, the scores were 11.82 (SD = 4.94) for high school graduates and 14.66 (SD = 5.14) for non graduates. These differences were statistically significant at the p< .01 level (t = - 2.773, p = .007). Together these findings suggest tint those who graduate from high school are more internally oriented than those who do not. Also, differences between groups were detected within the financial assistance variable, but only for those enrolled in the experimental group. The mean score at pretest for those who reported receiving financial assistance was 16.9 compared to 13.4 of those mothers who reported not receiving assistance. These differences were reflected at the posttest, as well, with the mean scores being 14.6 and 10.8 for those who receive assistance and those who do not. The results were statistically significant at the p< .05 level. These findings suggest that within the experimental group mothers who adnrit to receiving financial assistance such as ADC or TANF are nnre externally oriented than those who do not. Finally, there were intra-group differences on the age variable, although the results were surprising. Contrary to expectations, mothers older than 24 years scored more towards the external end than their counterparts in the group of mothers 23 or younger. Mothers 23 and younger had mean ANSIE scores of 12.58 (SD= 4.42) at pretest and 12.0 (SD = 4.35) at posttest. Mothers 24 and over, however, scored 15.14 (SD = 6.10) at pretest and 13.76 (SD = 5.44) at posttest. These findings suggest that mothers under the age of 23 were nnre internally oriented than mothers over the age of 24. There were no significant differences within groups for the renninder of the variables at pretest (family composition, t = 1.473, p = .144; enrolled AFDC, t = - .639, p = .524; number of children, t = - 1.225, p = .223; first time parent status, t = - .819, p = .415; and age of child, t = - .819, p = .930) and at posttest ( family composition, t = 1.448, p = .151; ethnicity, = 1.625, p = .108; enrolled AFDC, t = - .495, p = .622; age, t = 1.649, p = .103; number ofchildren, t = - .142, p = .887; first time parent status, t = - .819, p = .415; and age of child, t = .329, p = .743). By posttest, most of the intra-group differences had disappeared, leaving statistically significant differences for only two variables, monthly income and education level. 100 Intrigroup analysis parenting behaviors Empirical evidence suggests that individuals parent differently. Many factors have been linked to these differences and Belsky’s (1984) model on the determinants of parenting presents the domains which influence parental firnctioning and behaviors. He posits tint parenting is multiply determined from forces within the parent, forces within the child, and forces within the social network in which the parent-child relationship is embedded. It was expected that as a result of those multiple influences, there would be significant differences within the groups on key demographic variables in terms of perceptions of parenting behaviors. Independent t-tests were run to determine if there were within group differences among the sample on the parenting behavior variable. Contrary to expectations, results indicated that there were very few differences. Table 11 shows the findings. At pretest for both groups, older women reported more positive parenting behaviors. The mean PBA score at pretest was 133.80 (SD = 15.21) for wonen 24 and older, and 122.69 (SD = 30.75) for those under the age of 23. These findings were statistically significant at the p< .05 level (t = 2.347, p = .021). At posttest, however, there were no significant differences between age of participant and parenting behaviors. Also, first time mothers were less likely to identify themselves with positive parenting than women with nnre than one child. PBA scores were a mean of 119.39 (SD = 29.84) at pretest for first time nnthers and 135.31 (SD = 15.16) for tinse who were not. These differences were statistically significant at the p< .001 level (t = 3.548, p = .001). The differences were not significant at posttest, however, with mean scores of 134.56 (SD = 15.97) for first time parents and 137.56 (SD = 19.94) for others (t = .797, p = .427). F irnlly, consistent 101 TABLE 11 Independent T-tests for Intra-group Differences among Parenting Behavior Scores and Demographic Variables Variable Pretest Posttest Mean SD t p Mean SD t p Family Comp. Single Parent 129.46 23.13 .400 .690 134.87 20.95 -.870 .387 Two Parent 127.24 24.32 127.24 24.33 Ethnicity Minority 130.42 23.07 -1.367 .175 136.54 18.57 -.656 .514 Non-minority 1 16.00 21.07 131.00 14.00 Monthly Income $1001 and over 134.40 18.90 -2.077 .041" 136.70 13.69 .172 .864 $1000 and less 127.50 24.58 135.61 19.41 Education H.S. Grad 125.40 28.36 -1.537 .128 136.44 14.84 .026 .979 Non H.S. Grad 132.66 16.10 136.34 21.80 Enrolled AFDC Yes 129.40 20.48 -.183 .855 137.38 15.19 -.452 .652 No 128.53 25.02 135.68 20.19 Age 23 and younger 122.69 30.75 2.347 .021" 136.66 13.95 -.019 .985 24 and older 133.80 15.21 136.60 20.74 Numb. Children 3 or more 137.11 15.18 2.528 .013“ 139.47 10.66 1.227 .223 2 or less 125.54 25.42 134.74 21.08 1" Time Parent Yes 119.39 29.84 3.548 .001“ 134.56 15.97 .797 .427 No 135.31 15.16 137.56 19.94 Age of Child <18 months 123.53 26.47 3.513 .001” 135.47 15.03 .329 .743 >= 19 nnnths 139.03 9.56 137.63 22.19 * statistically significant at the .05 level ** statistically significant at the .01 level *** statistically significant at the .001 level 102 with the previous findings, women with more children reported more positive parenting behaviors when compared to women with fewer children. Specifically, women with 3 or more children recorded mean scores of 137.11 (SD = 15. 18) at pretest compared to 125.54 (SD = 25.42) for women with 2 or less children. These findings were statistically significant at the p< .05 level (t = 2.528, p = .013). While slight differences appeared in scores at posttest, the mean score for the experimental group was 139.47 (SD = 10.66) and 134.7 (SD = 21.08) for the comparison group. Mothers with more children were nnre likely to identify self with positive parenting practices, although the differences were not statistically significant (t = 1.227, p = .223). There were no significant differences for the renninder of the variables at pretest (family composition, t = .400, p = .690; ethnicity, t = -1.367, p = .175; education, t = -1.537, p = .128; enrolled in AFDC, t = -.l83, p = .855). At posttest, there were no significant differences within any of the variables regarding parenting behavior scores. (family composition, t = -.870, p = .387; ethnicity, t = -.656, p = .514; nnnthly income, t = .172, p = .864; education, t = .026; p = .979; enrolled in AFDC, t = -.452, p = .652; age, t = -.019, p = .985; number of children, t = 1.227, p = .223; first time parent, t = .797, p = .427; age of child, t = .329, p = .743.) Conclusion Overall, the findings from this research support the existing body of knowledge that describes an association between internal locus of control orientations and positive parenting behaviors. In general, mothers in this study reported listening to their children, rennining calm dming temper tantrums, hugging and kissing their children daily, playing 103 with their children and other parenting behaviors which imply responsive, sensitive, and nurturing parenting. As the literature suggests and this research reinforces, these warm, accepting and guiding parenting characteristics are consistently linked with internal orientations (Belsky, 1984; Mondell & Tylers 1981). In other words, women with internal orientations are nnre likely to interact with their children in developmentally appropriate, empathic, and nurturing ways. Results also indicate that mothers in the study are more likely to identify self with positive parenting beinviors at the end of the program when compared to a comparison group. This measure of self reported perceptions is intended to reflect the frequency and consistency of developmentally appropriate and responsive parenting behaviors. Mothers who completed the BSF program were likely to experience changes in their perceptions of their own parenting behaviors to reflect more positive parenting practices. Mothers in the comparison group experienced no changes. This evidence suggests that BSF is an effective way to promote competent parenting among limited resource nnthers and confirms previous research that has documented the success of the BSF intervention in influencing parenting behaviors. In addition, there was some evidence to support that both locus of control and parenting behaviors are influenced by forces from within the parent, within the child and contextual factors. Locus of control orientation appeared to vary within the groups according to educational attainment, monthly income, and age of mother. Moreover, parenting behaviors were found to be affected by mothers’ age, first time parent status and number of children. Any statistically significant differences detected in parenting behaviors at pretest, however, were no longer present at posttest, suggesting the BSF 104 intervention may buffer factors which present as a potential risk to positive parenting practices. Finally, the other significant finding of this study is, when compared to a comparison group receiving no parenting education treatment, the experimental group experienced an increase in their internality from the beginning to the end of the BSF program In general, those women who were enrolled in and completed the BSF program were likely to experience shifts in their locus of control orientations in the internal direction. Women in the comparison group, however, experienced no changes in orientation. The literature has consistently documented that the correlation between internality and competent parenting is consequential. As Belsky (1984) hypothesizes, while parenting is multiply determined by characteristics of the child, contextual sources of stress and support, and parental psychological resources, the nnst important element of the system is parental psychological resources. The likelihood that parents are capable of providing Optimal care to children is the least when parental psychological resources is the weakest system While the significance of internal locus of control orientations has long been established in the theoretical and empirical literature, this study provides evidence to support that individual orientations are malleable within the context of a parent education program Given what is known about the importance of internality in relation to parenting, this is a significant finding. Furthermore, if limited resource mothers are learning new ways to think about themselves and their personal sense of power and control as this study suggests, then the potential for these effects to spread beyond pmenting and throughout multiple areas of maternal life is tremendous. 105 CHAPTER V DISCUSSION, SUMMARY, CONCLUSIONS, IMPLICATIONS AND SUGGESTIONS FOR FUTURE RESEARCH Discussion of Results The results of the study will be summarized according to the three research hypotheses and statistical analyses. Hypgthesis 1: Mothers who complete the bone visitation, parent education program will show an increase in internal locus of control orientation when compared to the comparison group. The results of the data analysis support this hypothesis. Paired t-tests reveal that mothers who participate in the Building Strong Families home visitation, parent education program experience statistically significant increases in their locus of control orientations fiom the beginning to the end of the program Mothers fiom the comparison group receiving no parenting education treatment, however, experience no changes in locus of control orientation fi'om pre to post assessment. In other words, nnthers are more likely to believe that they can determine their own fate within limits after completing the program, when compared to mothers not participating in the program It is important to note that there were no significant differences between groups in locus of control scores at the pretest. Therefore, it is possible to attribute differences in group mean scores at posttest to the Building Strong Families program with a reasonable degree of certainty. 106 There are several possible explanations to account for this increase in internality. First of all, the Building Strong Families program includes the “Snnrt Living” tmit which was designed to facilitate the self care and personal development of the mother. Its purpose is to assist nnthers in identifying their own personal strengths, recognizing sources of fi'ustration, and setting and achieving realistic goals. While the unit’s concepts are introduced early in the program, they are a continuous theme throughout. Parenting instructors are trained to support mothers to nnke life changes and actualize goals they have set for themselves during the course of the program Through this process, program participants may realize they inve the ability to shape and alter their life course, within reasonable limits. This relationship between internality and feelings of personal empowerment has been substantiated in the literature (Morris, 1992). In general, individuals who have more Opportunities to achieve positive outcomes tend to be more internally oriented. Additionally, experts agree that locus of control orientations develop and reshape throughout life based upon individual experiences (Lefcourt, 1976). Thus, it is plausible to conclude tint the Smart Living unit of the BSF curriculum has an impact on feelings of personal power. Another potential explanation for the increase in internal locus of control orientations in nnthers who participate in the program is the process of the intervention itself. Investigators document tint effective firmily support programs integrate elements of information and support (Gomby, Culross & Behrman, 1999) and that home visitation programs provide a valuable resource in enhancing social support networks (Baker et al., 1999). Research on home visitation programs suggests that the key to helping families is in the supportive relationship between the home visitor and the program participant 107 (Black et al., 1994; Duggan et al., 1999; Olds et al., 1999). Moreover, research around the concept of self efficacy, a construct often used interchangeably with locus of control in the literature, has demonstrated strong linkages to the idea of perceptions of self being influenced by others. Bandura (1982) identified four sources of beliefs regarding self efficacy. Two of them, vicarious learning and verbal persuasion, originate in interactions with others and infer that others clearly influence one’s perceptions of self efficacy. It makes sense that because the BSF staff are working one-on-one with mothers in a mentoring role, supportive and trusting relationships develop. Through these relationships, parenting instructors encourage nnthers to identify their strengths and make changes in their lives, ultimately influencing the mothers’ perceptions of themselves to realize they have power within limits to influence and regulate their own lives. To firrther support this notion, researchers suggest tint it is necessary to have useful relationships within the larger community context in order to develop a healthy locus of control orientation (Swick, 1984). The BSF program provides the opportunity for mothers to establish that positive relationship within the community, which nny be helpfirl in encouraging mothers to realize their potential to influence their own lives. Finally, previous research has documented that parenting education in general has elicited changes in women’s life views regarding empowerment and individual sense of control (First & Way, 1995). The authors speculate that the parent education process nntivated women to question their basic values and beliefs regarding childrearing. This process appeared to stimulate participants to think critically about themselves and their life situations. As a result, the women evidenced a major shift in their orientations from reactive to proactive, unempowered to empowered. This reformation in thinking and 108 behavior is labeled transformative learning (Apps, 1991). Transformative learning implies tint as a result of the parenting program, wonen are learning new ways to be responsive, loving, and nurturing parents in addition to new ideas regarding their personal power and ability to control their own lives. (First & Way, 1995). The critical thinking and reflection skills associated with transformative learning are embedded throughout the BSF curriculum and program process, and serve as a possible explanation for the changes program mothers experience in locus of control orientation. What this discussion illustrates is that numerous questions remain about the critical elements tint predict the effectiveness of the BSF program, in spite of the fact that significant results were found. As previously stated, there are many plausible explanations for the program’s success. Because this study only sought to determine if mothers experienced changes toward internality following the program, much still needs to be learned about why and how significant effects were achieved. Namely, it is still not known if the effectiveness is attributed to the curriculum of the BSF program that includes the parental self care component, to the home visitation model that enhances social support networks and encourages personal growth, to parenting education in general, which has been known to transform the way participants’ view themselves and their lives, or to the interaction annng all of these issues. Future research with alternative designs would be most helpful in beginning to learn the information necessary to formulate answers to questions such as these. Most likely, each of the components contributes to achieving significant findings. However, after observing the interaction between mothers enrolled in the program and their parenting instructor, this investigator speculates tint the relationship between the 109 participant and staff which is cultivated in the home visitation model, is at the core of the findings. Admittedly, the curriculum provides a necessary tool that permits staff to guide mothers through the personal development process. Yet the warmth, encouragement, support and assistance that staff provide to their clients on a consistent basis, is a testimony to the self worth of many of the mothers participating in the program who may not have received that kind of support or attention in their lives fi'om any other source. This investigator postulates tint evidence to support this connection might be reflected in the fact that strong effects are seen in the experimental group following a relatively short intervention. Researchers have suggested that home visitation programs should expect a minimum of 4 to 6 months of visits, prior to program participants experiencing any degree of change (Gomby, Culross & Behrman, 1999). The reality is that the BSF intervention is on average 2 months, a much shorter duration than those recommended. Perhaps these strong effects on locus of control orientation are a consequence of the influence tint the paraprofessional relationship has upon the nnthers, and the staff’s ability to affect participants’ perceptions of selfand personal power. Regardless of the reason that facilitates change, what is clear is tint nnthers who participate in the Building Strong Families program are experiencing vital changes within themsleves and their attitudes about personal power after completing the program Statistic analysis confirmed at the p< .000 level that mothers who complete the BSF program have a shift in their locus of control orientation toward internality after at the end of the program when compared to a comparison group not receiving treatment. These findings suggest that a short-term intervention can be successful in influencing perceptions of selfand personal control in limited resource mothers. 110 Hypgthesis 2: Mothers who complete a home visitation, parent education program will show an increase in perceptions of positive parenting behaviors when compared to the comparison group. The results of the data analysis support this hypothesis. Paired t-tests revealed that mothers who participated in the Building Strong Families parent education program experienced statistically significant increases in their perceptions of positive parenting behaviors fiom the beginning to the end of the program Mothers from the comparison group receiving no parenting education treatment, however, experienced no changes fi'om pre to post assessment. In other words, women were more likely to identify themselves with positive parenting behaviors after completing the program than those receiving no treatment. The results were statistically significant at the p< .000 level. Furthermore, there were no statistically significant differences between the experimental and comparison groups at the pretest in terms of parenting behavior scores. Therefore, it is possible to attribute differences in group mean scores at posttest to the Building Strong Families program with a reasonable degree of certainty. These results are consistent with the literature which has demonstrated that parent education interventions are effective in creating a host of positive changes in parents who have participated (Brems, Baldwin & Baxter, 1993; Powell, 1983). Specific outcomes for parents include positive changes in parental attitudes (Telleen, Herzog & Kilbane, 1989), enhanced knowledge of child development (Fulton, Murphy & Anderson, 1991; Stevens, 1988), and the development of positive parenting skills (Fox, Fox & Anderson, 1991) and are consistent with the concepts presented in the BSF curriculum Moreover, 111 investigators have documented changes in parental discipline practices and improved parenting perceptions as a result of short term, parent education interventions (Gross, F ogg & Tucker, 1995; Stratton, 1998). In addition, the findings fiom this study are consistent with findings over the last four years which have demonstrated that Building Strong Families is effective in influencing parental perceptions of parenting behaviors. Previous analysis of results on the PBA has confirmed that changes are statistically significant on 30 of the 32 items (n=1600). Further, client exit evaluations and instructor assessments confirm that changes are occurring within the mother and family as indicated by the changes in PBA scores. Overall, the findings from this research support that, in general, nnthers in this study reported listening to their children, rennining cairn during temper tantrums, hugging and kissing their children daily, playing with their children and other parenting behaviors which imply responsive, sensitive, and nurturing parenting. Interestingly, however, despite the high prevalence of positive perceptions regarding parenting behaviors in both groups, the nnjority of women admitted to using spanking as a discipline technique at least sone of the time. While this information seemed to be in conflict with the positive reports of parenting behaviors, it could in fact be reflective of culturally competent parenting. Some suggest that authoritarian parenting practices are a protective function for youth who are more likely to be living in poor and dangerous communities (Baumrind, 1991) which is characteristic of the communities in which this sample resides. Brody et al., (1998) have defined a phenomena referred to as “No Nonsense Parenting”, a common parenting style among Afiican American parents. No Nonsense Parenting is characterized by high levels of parental control, including physical 112 punishment, and high levels of affectionate behaviors. It is based upon the premise that parents try to raise their children in a manner that prepares them to function effectively in their environments. Researchers suggest this method is an adaptive parenting approach for parents living in dangerous communities, that is intended to communicate a vigilant concern for the child’s welfare where disobedience could have sober consequences. These stringent parenting practices are believed to pronnte children’s self regulatory competencies and protect them fi'om danger by discouraging disobedience and antisocial activities (Kelly, Power & Wirnbush, 1992). Since the sample was predominantly Afiican American women living in poor, dangerous, urban communities, it is possible that this concept of “No Nonsense Parenting” needs to be explored to better understand and educate mothers on effective discipline practices for young children. Hyp_othesis 3: As locus of control scores decrease, parenting behavior scores increase. Internal locus of control has been associated with a variety of positive behaviors including responsive, stimulating parenting. Specifically, internal locus of control has been linked to parent child interaction, the reduced incidence of child abuse, and more stimulating home environments to nanre a few (Chandler, Wolf, Cook & Dugovics, 1980; Ellis & Milner, 1981; Stevens, 1988). In addition, internal children have been associated with internal parents, and internal parents have been associated with authoritative parenting styles, acceptance, positive verbalization, and nomestrictive independence (Chandler et al., 1980). Researchers inve also found a correlation between parent orientations and perceptions of children’s behavior problems (Harris & Nathan, 1973). 113 Investigators suggest that parents who had external locus of control scores believed that their children’s behavior problems were a result of external influences, while parents who related their children’s problems to parental behavior had significantly lower extenral scores. The authors conclude that those parents who are more likely to believe that life and life events are determined by fate, are also more likely to believe that child rearing and parenting consequences are based upon fate or chance and act accordingly. Therefore, it was hypothesized that internal locus of control scores would be associated with positive parenting behavior scores. Data analysis for this hypothesis yielded inconsistent results. Using pretest scores from the AN SIE and PBA, the correlation coefficient was quite small (r = .04), indicating that a linear relationship between the two variables did not exist. These findings appear to contradict findings fiom previous studies presented earlier which describe a correlation between interrnl locus of control and positive parenting behaviors (Chandler et al., 1980; Stevens, 1988). A plausible explanation for the insignificant correlation coefiicient at pretest may be a limitation of the parenting measure used in this study. Data was collected from both the experimental and comparison groups using a self report measure. While the investigator has no basis for judgement regarding the experimental group as others collected the data, it appeared as if the comparison group often over-evaluated themselves in terms of their parenting behaviors. Because there was the opportunity to observe the mothers in the clinic with their children for a one to two hour period of time as they waited for their WIC appointments, inconsistencies between their self ratings and actual behaviors were clear. Previous research has described common sources that pose a 114 potential threat to the validity of results and suggests that one of them is the process of testing itself. Often, the administration of a test can either sensitize subjects to a measure or a concept and influence responses, or subjects provide answers tint describe behaviors which present the most positive images (Campbell & Stanley, 1963). Perhaps the notion of testing effects might be contributing to what appears to be the over-evaluation of parenting behaviors and contaminating the ability of the statistical analysis to detect any correlation between the locus of control and parenting behavior variables. On the contrary, however, results fiom the data analysis on the locus of control and parenting behaviors variables at posttest dennnstrate an inverse relationship between locus of control orientations and positive parenting behaviors as expected. In general, decreased locus of control scores, reflecting a shift toward internality, were associated with increased parenting behavior assessment scores, suggesting increases in positive parenting behaviors. Based upon the extensive literature base that consistently docunents this association, it was not surprising to discover a statistically significant correlation. The surprising finding is that these results were not consistently documented at both pretest and posttest. While an explarntion for the inconsistent results was offered, the correlation between parenting behaviors and locus of control orientation is a relationship that should be firrther explored in firture research with additional or more sensitive measures. However, there is an alternative explanation for the significant correlation between parenting behaviors and locus of control after the intervention. That is, perhaps the BSF intervention is successfirl in influencing both variables. As a result of their experience in the program, mothers become more internally oriented and improve their 115 parenting behaviors. In other words, the pre and post measures upon which the correlation coefl'rcient is based nny be better evaluating the effectiveness of the program in relation to two distinct measures as opposed to describing a relationship between two variables. The fact that a correlation between locus of control and parenting behaviors in general could not be documented in the earlier analysis may further support this contention. Although evidence was presented to document a significant correlation between locus of control and parenting behaviors, a cautionary note must be extended. A significant correlation does not suggest that changes in one variable are causing changes in the other variable. Many investigators have found internal orientations and parental psychological resources to be integral components of healthy parenting (Blesky, 1984; Ellis & Milner, 1981; Stevens, 1988). However, while there is a significant association between locus of control and parenting, there is no empirical evidence in this study, and very little in the literature, to document the direction of the relationship. For example, a correlation could be explained by the presumption that as a mother’s sense of internality increased, she would be more highly nntivated and better able to influence her children in a positive manner. On the other hand, another explanation for a correlation might be that as an individual becomes more effective, she feels as if she has more control. Within the context of parenting, this idea might be that as a mother gains more experience applying positive parenting practices and appreciating the consequences of those behaviors, she may begin to feel a greater sense of control in her life in general Thus, these results pose the question: Is it that mothers who practice positive parenting behaviors feel more personal power as a result of their interactions and/or skills, or that 116 mothers who feel more personal power are more motivated or skilled to parent in ways that are considered to be empowering or positive? Clearly, more work is needed to investigate these questions further and obtain data that is better able to address causality between locus of control and parenting behaviors, instead of just correlation. Intra-gmup analysis The literature suggests that both parenting and locus of control are influenced by numerous factors and been shown to be shaped and molded over time (Belsky, 1984; Lefcourt, 1976). In addition to the major research hypotheses, this research also examines both the locus of control and parenting behavior constructs in relation to several key denngraphic variables to determine if there are differences in the dependent variables within groups. In general, those who have more power to achieve positive outcomes, whether it be a result of group membership or social position, are more likely to be internally nntivated (Lefcourt, 1976). Because there is a substantial body of literature tint describes in detail those factors which influence and affect orientation, it was expected tint there would be significant differences within groups in terms of this study’s key variables. While there were some intra-group differences, there were fewer than expected. At pretest, approximately half of the variables demonstrated significant differences within groups, all at the p< .05 level. Those variables were ethnicity, monthly income, education, and age. Except for age, the findings were predictable. Tint is, more external orientations were associated with the group considered to have less power. Minorities were more external than non-minorities, those earning less than $1000 per month were more external than those earning more, and non-high school graduates were 117 more external than high school graduates. For age, mothers 24 years and older were found to be more exterrnl than those 23 and younger, which was in the opposite direction as expected due to evidence in the literature (Morganti et al., 1988). By posttest, only 2 variables, monthly income and education, found statistically significant differences. Intra-group analysis was also conducted for parenting behavior scores and denngraphic variables. Because of the evidence that suggests that parenting is multiply determined (Belsky, 1984), it was expected that there would be intra-group variation within the variables regarding parenting. Contrary to expectations, however, only 5 of the variables at pretest were found to be statistically significant, and four of those are suspected to be closely related. The variables were monthly income, mother’s age, number of children, first time parent status, and age of child. The four similar variables are mothers age, number of children, first time parent status and age of child. The reason these variables are tlnught to be related is that, most likely, younger mothers have fewer children so far, are nnre likely to be first time parents, and have younger children. The results of the analysis indicate that women earning less than $1000 per month, younger mothers, those whose children are 18 nnnths and younger, those with 2 or less children, and first time parents are less likely to report positive parenting behaviors than their counterparts. By posttest, however, there were no significant differences within any of the variables to suggest any intra-group differences. In conclusion, intra-group analysis suggests that there is some variation in terms of locus of control and parenting behaviors among demographic variables. However, there were fewer differences than anticipated. Additionally, most ofthese differences dissipate by the end of the program In general, there is very little movement on either 118 locus of control and parenting behavior scores within the comparison sample. Because these significant differences all but disappear by posttest, and very little change occurs within the comparison group mean scores, one may speculate that the majority of movement is occuring within the experimental population. This idea presents further evidence to support the effectiveness of the BSF intervention, and suggests that the program acts as a bufl'er to ameliorate those conditions which may pose a risk to an internal locus of control orientation and/or perceptions of parenting behaviors. Summary Researchers agree that parents’ personal and psychological clmracteristics are of great significance to the parenting process and developmental outcomes of children (Belsky, 1984; Swick, 1984). In his theoretical model describing the determinants of parenting, Belsky posits that parenting is multiply determined by characteristics of the child, contextual sources of support and stress, and the psychological resources of the parent. He hypothesized that to optimize parental functioning, all three domains should be operating in the supportive mode. However, he argues that the most critical component to parent-child interaction is parents’ psychological resources, and speculates that parents who are most capable of responding to children in sensitive, nurturing and empathic ways are mature, psychologically healthy adults. While several personality dimensions encompass the traits known collectively as psychological resources, one construct which is critical to consider in relation to parenting is locus of control, or one’s beliefs about his or her ability to influence the outcomes of life. Locus of control orientations range fiom internal, the belief that one can determine his own fate within 119 limits, to external, the belief that he is controlled by forces outside himself (Lefcourt, 1976). The purpose of this study is to examine the locus of control construct within the context of parent education and parenting behaviors. Specifically, it is designed to determine if mothers’ locus of control orientations shifi toward internality and perceptions of parenting behaviors improve as a result of a home visitation, parent education program. In addition, the correlation between locus of control and parenting behaviors is explored. One hundred mothers with children three and younger, living in a large, Midwest, urban city participated in the study in 1999. F ifty of the mothers were enrolled in the experimental group through the Building Strong Families program and fifty volunteered to participate in the non-equivalent comparison group after being recruited through their participation in the Women, Infants and Children (WIC) supplemental food and nutrition education program. Data was collected on a pretest-posttest basis using the following research instruments: The Adult Nowicki-Strickland Internal-External Control Scale (AN SIE), The Parenting Behavior Assessment (PBA) and the Family Record Form (FRF). Group differences were tested using t-tests and correlations between key variables were tested through the Pearson product moment correlation. The findings around the study’s hypotheses are summarized below. It was hypothesized that mothers who complete the Building Strong Families Program, a home visitation, parent education program for families with children three years of age and younger, would show an increase in interrnl locus of control orientation when compared to the comparison group. Generally, findings indicated that, in fact, 120 participation in BSF was related to significant increases in internal locus of control orientation. Those mothers who were enrolled in and completed the BSF program were likely to experience shifts in their locus of control orientations in the internal direction following the program while women in the comparison group were not. These findings were statistically significant at the p< .000 level. Secondly, it was hypothesized that mothers who conrplete the program would be more likely to identify selfwith positive parenting behaviors when compared to the comparison group. As expected, results indicated that participation of BSF was related to significant increases in maternal perceptions of positive parenting behaviors. Mothers who completed the program were more likely to identify themselves as engaging in more frequent and consistent positive parenting practices following the program while women in the comparison group were not. These findings were statistically significant at the p< .000 level, as well. Third, it was also hypothesized that there was an inverse correlation between locus of control orientation and parenting behaviors. As locus of control scores decreased, it was expected that parenting behavior scores would increase. A decrease in locus of control scores reflected a more internal orientation and an increase in parenting behavior scores described more positive parenting behaviors. The results, however, were inconsistent in supporting this hypothesis. Pearson product correlations between locus of control and parenting behaviors at pretest revealed no significant correlation between the variables. The same statistical analysis performed using posttest data however, suggested that there was a statistically significant correlation. One reason for this inconsistent finding may be that participants overestimate their parenting behaviors at pretest and 121 scores are not an accurate representation of their actions. Over inflated scores would then pose a challenge to determine actual correlations and effects. Conclusion Generally, the findings fiom this research support the existing body of knowledge which describes an association between internal locus of control orientations and positive parenting behaviors. Mothers in this study reported positive parenting practices that describe responsive, sensitive, and nurturing parenting styles. As the literature suggests and this research reinforces, these warm, accepting and guiding parenting characteristics are consistently linked with internal orientations (Belsky, 1984; Mondell & Tylers 1981). That is, internal locus of control orientations are associated with developmentally appropriate, empathic, and nurturing parent-child interactions. Moreover, the results of the study suggest that locus of control orientations can be shaped through a home visitation, parent education program Admittedly, the home visitation literature has been inconsistent in demonstrating positive effects (Black et al., 1994). However, investigators have documented that home visitation programs provide a valuable resource in enhancing social support networks and are a promising strategy to promote healthy parenting (Baker et al., 1999; Black et al., 1994). Research around the concept of self eflicacy, a construct often used interclmngeably with locus of control in the literature, has demonstrated strong linkages to the idea of perceptions of self being influenced by others. Bandura (1982) identified four sources of beliefs regarding self efficacy. Two of them, vicarious learning and verbal persuasion, originate in interactions with others and infer that others clearly influence one’s perceptions of self efficacy. 122 Finally, studies have suggested that it is necessary to have useful relationships within the larger community context in order to develop a healthy locus of control orientation (Swick, 1984). At the core of the Building Strong Families program is the expectation that staff establish strong, trusting relationships with program participants, and interact as mentors. The staff are trained to encourage and support women through a personal journey as they identify strengths, set and attain achievable goals, and strengthen parenting skills and family relationships. Based upon empirical evidence which suggests ways in which perceptions of self are shaped, Building Strong Families staff affect the beliefs mothers have about their ability to influence and regulate their own lives. Therefore, the findings of this research demonstrate with careful optimism that a home visitation, parent education program can influence limited resource, Afiican American mothers living in an urban community towards more internal locus of control orientations and greater feelings of personal power, and perceptions of positive parenting behaviors. The findings of this study also provide support to the operational map presented earlier in this study (pg. 29), integrating Belsky’s model of The Determinants of Parenting and Rotter’s Social Learning Theory, with key concepts of this research. Belsky’s model describes the ways in which parenting is influenced as a result of forces within the parent, forces within the child, and forces within the social context in which the parent-child relationship is embedded. Social learning, on the other hand, describes personality and behavior, recognizing that both occur within the context of the environment and experience. Together, these fiameworks provide a description regarding the influences on personality and behavior, and ways in which this is applied within the context of parenting. When the key variables of this research were integrated 123 into the operational map, a visual representation of previously documented and anticipated findings was created. The map implies that developmental history has been shown to influence personality, personality has been shown to influence social network, and the child, the social network, and personality have been associated with parenting. In addition, it was expected that results of this study will demonstrate influence in terms of personality, or locus of control. Figure 7 revisits the operational map and documents relationships that have been supported through this study. While not an explicit hypothesis of this research, evidence was found that validated the conceptual model which posits that aspects of a mother’s developmental history can influence personality, or locus of control orientation. Furthermore, child characteristics defined as child’s age were found to be associated with parenting behaviors, although the relationship was inconsistent. Confirming findings fi'om previous studies, personality, or locus of control, was correlated with parenting behavior, and social networks, or participation in a home visitation parent education program, was associated with changes in parenting behaviors. Finally, results demonstrated that social network via a parent education program can also influence personality, or locus of control orientation. This information provides further support for integrated conceptual model and reinforces the need for additional work to establish linkages among the key elements for program and policy support. A limitation of this report, however, is its inability to address the long term effectiveness of the BSF intervention. Many reports indicate short term clmnges in parenting knowledge and/or behaviors although there appears to be little evidence of long term changes. Glanville and Tiller (1991) suggest that until long term effects can be 124 Al A Baosafi 3320352 308600803 $800093 beam 320m .025 efiaeom .II; D ”HOMO!— ” fl _ , ease: 3&5 ewe £20 . Same—505200 ~23th azmhzmmeQ £0323 9:823 ¢O~><=Hm $35.6; 3.580 02230 .«0 038:2 - “ mesa. . «:23 08: “mam .8056 $2530 9.6502 own 9.5502 mmm @6502 >¢Ohm=m _m iszdzmggwd 030..an EugeneQ 00330300 6.5000 m0 300— 98502 .. Prat—«V7593: 038.6; 303:00335 game:— 028260 328 003383 080m EOEHZ 300m All-Ir .3398 038080800 05 beam £5 m0 920 3:03.230 K 9:3:— 125 demonstrated, multiple follow ups may be necessary to promote changes in attitudes and behaviors. Future studies should incorporate a time series design that would follow participants after the intervention to determine if changes in locus of control orientation are maintained over time. Another limitation of the findings is that while it is reasonable to attribute changes in locus of control orientations to the BSF program, it is not possible to deduct if the changes are a result of the program process, program content, or the interaction between the two. Future research might compare women enrolled in Building Strong Families with other county extension programs that utilize a paraprofessional, home visitor approach but do not use the personal development materials. Findings fi'om this type of research could be helpful in determining if it is the process or the content that is most effective in influencing locus of control towards internally. Finally, the results are not generalizable to all parents enrolled in a home visitation, parent education program nor all families receiving the BSF program. This study evaluated the effectiveness of one particular intervention, within a specific community. Therefore results are generalizable only to limited resource, African American women living in this particular area. Future research examining BSF’s ability to influence locus of control orientation across program populations would be helpful, as well as the results of impacts on locus of control fiom parent education in general. Future Research Since the evaluation of the Building Strong Families is in its infancy, several recommendations can be made for future studies. For example, based on literature that 126 exists regarding the locus of control construct, further research is needed regarding the rehtionship between predictors of locus of control orientation and changes in the outcome measure as a result of the program. Furthermore, research exploring the relationship between perceptions of parenting behaviors and locus of control orientation would also be beneficial. Since a significant correlation between the two variables was established, further research that explores causality would be meaningful. Another implication for firture research is a longitudinal study to address the long term effectiveness of the BSF program in terms of locus of control orientation. Future studies should incorporate a time series design which would follow participants after the intervention to determine if changes in locus of control orientation are maintained over time. It would also be meaningful to follow participants over time to see if; in fact, women were able to translate more internal locus of control orientations into changes in maternal life course. Olds et al., (1999) in a 15 year follow up to a home visitation program using nurses to visit families fi'om pregnancy through the child’s second birthday, found tint the experimental group experienced several long term benefits when compared to the comparison group. Those visited by nurses had fewer subsequent births, postponed subsequent children longer, fewer months on welfare and receiving food stamps, fewer arrests, and less substance abuse. While the BSF intervention is much less intense and of a shorter duration than the program implemented by Olds and colleagues, it would be interesting to learn if the women who completed the BSF program were more likely to return to school, return to the workforce, leave welfare, be reunited with their children, and other attainments relative to maternal life course, when compared with a similar group not receiving the program. 127 Although BSF is currently limited to short term intervention, a body of literature is emerging that suggests that family support programs need to increase their intensity or duration to be more effective (Howing et al., 1989; Upshur, 1988). While some researchers have documented sustained behavior changes in parents who participate in a short term intervention (Goss, F ogg & Tucker 1995; Stratton, 1998), short term programs have yielded inconsistent results. Future research which examined program duration and intensity in terms of outcomes for parents and children could greatly benefit this promising program This study was limited to mothers since the literature has been clear in documenting gender differences regarding locus of control. In general, women tend to score in the external direction while men score more internally (DeBrabander & Boone, 1990; Dyal & Mwamwenda, 1995). Because of the small sample of men currently enrolled in this community and the gender differences documented in the literature, men were omitted fiom this study. However, further work is needed to determine if changes in locus of control orientation occur for men as well. Also, firture research with this program should incorporate a qualitative research component. Because the AN SIE and PBA are self report questionnaires, measures which include multiple perspectives and qualitative data would add depth, detail and meaning to the quantitative results. Recommended additions include an exit interview for both the program participant and instructor with open ended questions addressing locus of control and parenting, and any changes as a result of the program. It also may be helpful for staff to incorporate an observational assessment such as the HOME to support or refirte PBA scores. While qualitative analysis is not intended to test casual interpretation, data based 128 speculation can allow researchers to make conjectures about which things appear to lead to other things, what produces what effects, and how processes lead to certain outcomes (Patton, 1987). This richness and detail will complement the inferential statistics and provide a more conclusive interpretation of the impact that the BSF program furs on locus of control and parenting. Finally, future studies should incorporate a more rigorous scientific design. This study utilized a quasi-experimental design due to existing program and staff resources. Although the groups were similar at pretest based upon statistical analysis, and differences between groups was found at posttest, there are some potential threats to the validity of the findings inherent in quasi experimental designs. A randomized control group design, however, would be much more rigorous and any differences between the experimental group and control group at posttest could be attributed to the Building Strong Families intervention with greater certainty. The previous suggestions for future research focused specifically on the Building Strong Families program. However, there are also several recommendations for further studies that could contribute to the overall research base in the area of home visitation and family support programs. The recommendations can be categorized in the three discrete groupings which address either program, staff or participant issues. Results fi'om this sort of study pose significant potential for contributions to the empirical literature base. An overview of recommendations within each of the categories is highlighted below. Program information that would provide meaningful contributions to the literature includes research that studies if a particular formula for program intensity predicts the 129 level of success among families. Information that addresses not only the optimal length of the intervention but also the most appropriate schedule for home visits might be useful in allocating appropriate staffmg resources in the program planning and implementation process. Clearly, it is crucial to have the right balance when devising a program model. If there is a point in time in which home visitors have saturated their potential to facilitate change, or, on the other hand, if changes in families do not occur until a specific threshold has been achieved, then that information would be extremely advantageous to programs interested in utilizing their resources most effectively. Moreover, home visitation programs often serve multiple purposes ranging from parent education, to community referrals, to health care services. Information that evaluated the various components of home visitation programs to learn if there are specific dimensions that lead to successful outcomes for families more often than others would be helpful. The second category of research on home visitation and family support programs that is critical to study focuses on staffing issues. While there is currently a great deal of controversy arguing if paraprofessional or professional delivery models are the most effective for home visitation programs, there is consensus that regardless of credentials or previous experience, extensive training and supervision are vital to any program’s success (Gumby et al., 1999). Research that examines the elements of successful training and supervision programs would have significant implications for those wanting to prepare and support staff working within the field. Moreover, it would be interesting to learn if there are certain personality characteristics that are predictive of “good” staff, or if the more critical predictor is a “goodness of fit” between program participants and staff. Information which more clearly describes if there are particular types of people 130 who are more effective with particular clients, or if there are individual attributes and qualities within staff that account for the majority of success has implications for programming in terms of staff recruitment and training to client intake and staff assignments. Evidence linking stafling issues to measures of success and outcomes for families has great potential for improving program infrastructure to more skillfully support the participants programs intend to serve. Finally, there is great opportunity to learn more about those who participate in home visitation programs. It would be worthwhile to know if there is a specific client profile that best describes those who are most likely to benefit fiom home visitation. Research in this area would offer insight regarding any relationships which exist between program effectiveness and variables such as child’s age, parent skill level, maternal age, education level, and others. If information suggests that efl‘ectiveness is related to certain parent or child characteristics, then that knowledge would be helpful not only in developing and adapting programs, but also in targeting appropriate referrals. Additionally, studies that explore if there is a window of opportunity, or a teachable moment, when participants are more likely to enroll or accept new information and/or ideas, would allow program staff to time the interventions more effectively. Lastly, a significant source of data is often lost to programs, in those families who choose to discontinue services. In general, very little is known about this subgroup of the target population. Knowledge that describes if there is a tendency in which either the most skilled participants or the most needy drop out of programs would make a significant contribution to this area of work. 131 Clearly, there are numerous research opportunities yet to be investigated within the field of home visitation. Additional findings would provide the necessary data and details describing the effective elements and the predictors of successful home visitation programs. Future research focusing on any one of these issues presents significant opportunities for families, staff and stakeholders. This information would be meaningful not only to the Building Strong Families program but also others involved in the development, implementation, and evaluation of home visitation and family support programs. Implications Parenting is affected by several domains of parental functioning such as knowledge, skills, attitudes and capacity, each of which is affected by child characteristics and contextual support (Belsky, 1984; Pecora, Fraser, Nelson, McCroskey & Meezan, 1995). One of the nrajor domains that appears to get less evaluation attention than the others is the parents’ capacity to provide adequate care. The importance of incorporating this component in program planning and evaluation activities is directly associated with Belsky’s model (1984) which states that personal characteristics of the parent impact parental functioning. While investigators claim that improving parental capacity is an important objective, one must be cautious in concluding that improved competency will directly result in reducing child maltreatment (Fink & McCloskey, 1990). However, in future studies and programs, all determinants of parenting, in addition to child characteristics and contextual support, may be the direct or indirect 132 focus of an intervention and must be evaluated in family based programs (Percora, Fraser, Nelson, McCroskey, & Meezan, 1995). Throughout this study, other investigators and their scholarly works have provided the foundation for why locus of control is an important personality construct. Further evidence was presented which documented the role of locus of control and its influence on parenting. This research demonstrated not only that there is a significant correlation between parenting behaviors and locus of control but also that locus of control can be shifted toward internality as a result of an educational intervention. These findings present a substantial impact for the field of parent education and program planning. As Belsky (1984) hypothesized, while parenting is multiply determined, the most critical element in determining whether or not parents are capable of providing optimal care to their children is psychological resources. The likelihood that parents can respond and interact with their children in responsive and nurturing ways is the least when the parent is not psychologically healthy. While some investigators may disagree with the importance Belsky places upon parental capacity, his theory has been widely embraced by the scholarly community. Therefore, if, in fact, parental psychological resources are the most critical element of the parenting process, then it nukes sense that this concept of increasing or supporting parental capacity becomes an indirect or direct focus of family support and parent education programs. If it is true that the most important predictor of competent parenting is attributed to the personal development of the parent, it seems as if the parent, rather than the child, should be a focus of parenting education programs. Programs need to begin an intervention with parental resources in mind, finding ways to support parents’ growth and development into mature, psychologically healthy adults and 133 then move to parenting skills, child. development principles, and so on, the potential to have long term impact may increase. This change in program development would represent a significant paradigm shift. Yet, the work of First & Way, (1995), further supports the idea that focusing parent education on teaching parents new ways of thinking and viewing the world as opposed to teaching only parenting skills, creates the opportunity for meaningful transformation in multiple areas of the parents’ life. This suggests that perhaps parent education is rrrost effective when thought of in the context of adult education. In other words, the focus of the intervention should be to facilitate change within the adult, in general, rather than concentrating upon parenting or some other narrowly defined topic. It is time that those involved with program development and implementation accept that any intervention directed exclusively toward the child will probably be short lived (Black et al., 1994). Only when there is an intentional shift toward developing interventions that focus upon strengthening parents, fimilies, and communities will we build long term capacity and promote long term change. Another significant implication of this research is the impact its findings may inve upon the growing interest in and understanding of early brain development in children. Neurologists and other experts are advocating that the early care and nurturing children receive has significant impacts on their brain development. Since parents are typically the ones who are the primary caregivers of young children, parents have great opportunities to influence their children’s brain development and intellectual capacity. Studies on the quath of the caregiving environment indicate that stimulating, responsive, nurturing care contributes to optimal brain development in young children. Interestingly, these same descriptives are those associated with parents with internal locus of control 134 orientations. If parent education programs can be redesigned to incorporate opportunities to build parental capacity and develop parental resources, including shifts toward internal locus of control orientations, the impact on parenting behaviors, and ultimately the brain development of young children could be great. Clearly, this research and others with similar findings have significant findings and serious implications for families and children. Results must be taken thoughtfirlly and translated into program and/or policy recommendations. Another implication of this study is, based upon limitations that surfaced through the course of this study, a significant contribution might be to revise the Parenting Behavior Assessment, (PBA), which is currently used as the evaluation instrument in the Building Strong Families program. The instrument was developed specifically to evaluate the effectiveness of the BSF program by measuring changes in parental perceptions of parenting behaviors. While the instrument has been useful in documenting program effectiveness, concerns regarding the reliability of the instrument to measure change have arisen. The concern is that while the instrument includes 32 items which are reflective of responsive, nurturing parenting, the behaviors are those more common among parents of toddlers and preschoolers than parents of infants. For example, items include “I encourage my child to use her hands”, “I take my child outside to play”, “I encourage creative play”, “I encourage my child to play with fi'iends his own age”, to name a few. If a participant Ins an infant, she may not be engaging in those behaviors because it may seem inappropriate for the age of her child. However, 3 to 4 months later when she completes her posttest, her child is at a different developmental stage and she has started to report engaging in these types of parenting behaviors more frequently. 135 While the mother in this example would experience significant increases in her parenting behaviors from the beginning to the end of the program, it is not clear if these changes are a result of her learning that these are important behaviors, or that her child is older and it makes sense to do these kinds of activities at the child’s current developmental stage. Therefore, the program is taking credit for increases in parenting behaviors which might be better explained by maturation. An additiomrl measure, based upon the indicators of responsive, nurturing parenting behaviors of parents of children less than 12 - 15 months of age is needed to more accurately report program effects. Another major implication of this research is the meaning it ins for the Building Strong Families program and the university land grant system that supports it. For 9 years, program and administrative stafi' have been hearing anecdotal reports about and observing changes in those who have participated in the program. Staff has speculated that there was something significant about the BSF program which facilitated change in participant’s attitudes about self and life, but never had the empirical evidence to substantiate this. The findings from this study present great validity to the work that the staff are doing and the much needed feedback to those who are involved with the program at many levels. Moreover, within the university system in general, these findings pose questions about the successful elements of educational programming and suggest that perhaps BSF can serve as the model throughout the organization for effective planning, management, implementation and evaluation of paraprofessional, home visitation programs. 136 APPENDICES 137 APPENDIX A Michigan State University Extension Children, Youth and Family Programs Building Strong Families Parent Consent Form Name of parent: Family number: We are conducting an evaluation to better serve parents of young children throughout Michigan. Your participation in this project is voluntary and services are sill available should you choose not to participate. Information collected will be kept confidential and your name will not be used in any way when reporting the results of this project. You have the right to drop out of the project at any time. By participating you could help other parents like yourself. Signature of parent: Date: 138 APPENDIX B Building Strong Families Participant Record Form Instructor's Name: County: Date of Enrollment: Participant Number: Participant'sName: Type of Instruction: Address: 0 Group Cllndividual 0 Both TW' ( ) D Other Sex of caregiver who participates Family composition: Ethnicity: in the program: 0 Single parent 0 White 0 Male 0 Two parent C1 African-American 0 Female 0 Extended family 0 Hispanic C1 Pregnant 0 Foster Parent C1 American Indian C1 Breastfeeding C] Other: 0 Asran C] Multi-cultural Residence: Education: 0 Other (SPOCW) C1 Towns under 10,000 & rural Last grade completed: non-farm Family participation at entry: - . D FIP C1 WIC C1 Towns & Cltles 10,000-50,000 Total Monthly . . C1 Suburbs of Cities over 50,000 Income: 5 D Commodities 0 0th“ C1 Central Cities over 50,000 0 Food Saunas C1 Head Start Family /Homehok1 Members Date of A Relationship to Sex . . . Bith ge Participant (Inclrde the Partcrpant on Lne #1) M F 1- Self 7. 8. Form R2/6-99 139 APPENDIX C Michigan State University Extension Children, Youth and Family Programs WIC Parent Consent Form Name of parent: Family number: We are conducting an evaluation to better understand and serve parents of young children in Wayne County, Michigan. Your participation in this project is voluntary and WIC services are sill available should you choose not to participate. lnforrnation collected will be kept confidential and your name will not be used in any way when reporting the results of this project. You will complete 3 surveys today and 2 surveys 12 weeks from now. In total, it should take you less than one hour of your time. If you complete both surveys, you will receive a $10 gift certificate to your local grocer. However, you have the right to drop out of the project at any time. By participating you could help other parents like yourself. If you have any questions regarding the study or your participation, please call Dawn Koger at (313)833-3414. Signature of parent: Date: 140 APPENDIX D instructor's Name: County: Date of Enrollment: ParticipantNumber: Participant'sNamc: Type of instruction: Address: 0 Group Olndividual 0 Both 0 Other Telephone: ( ) Sex of caregiver who participates Family composition: Ethnicity: in the program: 0 Single parent 0 White 0 Male Cl Two parent 0 African-American _ 0 Female 0 Extended family 0 Hispanic 0 Pregnant Cl Foster Parent 0 American indian C] Breastfeeding C1 Other: 0 Asian 0 Multi-cultural Residence: Education: 0 Other (SNOW) Cl Towns under 10,000 & rural Last grade completed: Homfam Family participation at entry: 0 Towns & Cities 10.000-50.000 To“. Mommy 0 F1? _ . 0 WIC C1 Suburbs of Cities over 50,000 Income: 3 0 Commodities 0 0th“ 0 Central Cities over 50,000 0 Food Stamps C1 Head Start Family/Household Members Date of A Relationship to Sex . . . Birth 8" Participant (Inchde tire Partmpant on Late # 1) . M |- Self 2. 3. 4. 5. 6. 7. 8. Form R2/6-99 141 APPENDIX E Adult Nowicki-Strickland Internal—External Control Scale Directions: Please circle the answer which best describes how you feel for each question. Yes No 1. Do you believe that most problems will solve themselves if you just don’t fool with them? Yes No 2. Do you believe that you can stop yourself from catching a cold? Yes No 3. Are some people just born lucky? Yes No 4. Most of the time, do you feel that getting good grades meant a great deal to you? Yes No 5. Are you often blamed for things that just aren’t your fault? Yes No 6. Do you believe that if somebody studies hard enough, he or she can pass any subject? Yes No 7. Do you feel that most of the time it doesn’t pay to try hard because things never turn out right anyway? Yes No 8. Do you feel that if things start out well in the morning, it’s going to be a good day for you no matter what you do? Yes No 9. Do you feel that most of the time parents listen to what their children have to say? Yes No 10. Do you believe that wishing cam make good things happen? Yes No 11. When you get punished, does it usually seem it’s for no good reason at all? Yes No 12. Most of the time, do you find it hard to change a friend’s opinion (mind)? Yes No 13. Do you think that cheering more than luck helps a team to win? Yes No 14. Did you feel that it was nearly impossible to change your parents’ minds about anything? Yes No 15. Do you believe that parents should allow children to make most of their own decisions? Yes No 16. Do you feel that when you do something wrong, there is very little you 142 Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes No No No No No No can do to make it right? 17. Do you believe that most people are just born good at sports? 18. Are most of the other people your age stronger than you are? 19. Do you feel that one of the best ways to handle most problems is just not to think about them? 20. Do you feel that you have a lot of a choice in deciding who your friends are? 21. If you find a four leaf clover, do you believe that it might bring you good luck? 22. Did you often feel that whether or not you did your homework had much to do with what kind of grades you got? 23. Do you feel that when a person your age is angry at you, there’s little you can do to stop him or her? 24. Have you ever had a good-luck charm? 25. Do you believe that whether or nOt people like you depends on how you act? 26. Did your parents help you if you asked them to? 27. Have you felt that when people were angry with you it was usually for no reason at all? 28. Most of the time, do you feel that you can change what might happen tomorrow by what you do today? 29. Do you believe that when bad things are going to happen, they just are going to happen no matter what you do to stop them? 30. Do you think that people can get their own way if they just keep trying? 31. Most of the time do you find it useless to try to get your own way at home? 32. Do you feel that when good things happen they happen because of hard work? 143 Yes Yes Yes Yes Yes Yes Yes Yes No No No No No 33. Do you feel that when somebody your age wants to be your enemy there’s little you can do to change matters? 34. Do you feel that it’s easy to get fiiends to do what you want them to do? 35. Do you usually feel that you have little to say about what you get to eat at home? 36. Do you feel that when someone doesn’t like you there’s little you can do about it? 37. Did you usually feel that it was almost useless to try in school because most other children were just plain smarter than you? 38'. Are you the kind of person who believes that planning ahead makes things turn out better? ' 39. Most of the time, do you feel that you have little to say about what your family decides to do? 40. Do you think it’s better to be smart than lucky? 144 APPENDIX F .902 up 0.0200 0 00E..00.0w ”m 0000 ”v m>0>>.< “m “0.00m .>00>.0>0 0.20 >0. 0.00 .0000 0:0 .00... . No .>.0.00 0.0.0.0 000 0000.0 050. o. 0.00 >0. 00050000 . or .8258 505.; 2.8 >0. 02.0.8.0 _ .0 .800 c... 80; 2.6 >0. 0258 _ 9 .8208 9.00. 2.6 >0. .2 . on .2955 .08.... 2.6 >0. 602 9 >0 . 3 0:00 >0. 003 000.00 >0.0 _ mm .00. 0. 2.0. .0 000000 00.00. 0. 2.00 >0. 00050000 . 9 86002800022358.2323. 00 .apaeceeeafiaufiaeeceaaeoeegepfieecfi. 0. 03000020. 00 0003 2.00 >E 0.0000 . mm 0.00 >0. 00.0.0 . S 0:00 >0. 00..» 0E. 00.00000 >0_00 . cm .030 .00 00 000.0. 00 0. 0:00 >0. 00050000 . or 2.00 >0. 2 020.. . mm >00 0... 5208.5 2.00 >0. 9 2.0. _ 0 8.0.98 .00 o. 8:000 0 2.00 >0. 2.0 . «0 >00 0.500 c. 2.6 >0. 5.; 0029,... .00 . 0 05.000. .0062 0 000 2.00 >0. 0003 0..00 U0 . aw 0020.0 >00 o. 2.00 >0. 00050000 _ n .2300 000 £00 .00. 0..00 >E 0.00 . Nu 0.0000 .0000 0000.0 030090.00 00 0. 0.00 >0. 0.00 . 0 2.00 >0. 5.; 9.03 .0 v.8. . 0... >20 9 2.5 >0. .0. $0.00. a: 9.0... . m 000 n. 9. 80.... 0......0 .o .280 00. 2 2.00 >0. 9.0. _ on 0.80.... 08. o. 2.00 >0. 000588 _ e .0300 6.00 0. 00.0 2.00 >0. 020 . 0.. 00.200 .0000 0...; >0.0 0. 0:00 >0. 00050000 . m .2...0 >0. .2 2...... .8 _ 2 5.; >0... 2 2.6 >0. .2 was... 8.5... _ m 0800.. 80.5 .80 .o 0.8. 000.5 so: 580 00.0 >0. 2 2.0. . t .8000 .9. 5.; 00.0. on o. 2.00 >0. 000585 . F 5000080 .02 .80 or. 8000080 6: ”0.00 «00.5000 “0.00 «00.0.0.0 02.80.. 2.00 .o 00< 000802 0000.00.00 :00“. 00000.0, .0_>000m 00000.00 145 APPENDIX G YOU CAN GET SRO—FREE! Dear My name is Dawn Koger and I am the person who talked to you at the Gateway WIC clinic a couple of months ago about earning a $10 gift certificate to Farmer Jack’s. I wasn’t able to see you at your last WIC appointment, but I don’t want you to miss out on your $10. Since you have already done more than half of the survey a few weeks ago, I only need you to fill out 2 more forms. It should take about 10 minutes. There are two things to remember when you are filling out the forms. On the sheet that talks about how you act with your children, pick the number at the bottom the page that is most like you for each one and write it in the box. On the sheet that you circle yes or no, remember there is a front and a backside. Make sure you do both! I have sent you a stamped envelope addressed to me so that you can easily return the forms when you are finished. Thanks again for your help. As soon as I get your forms back and make sure everything is complete, I will send you your $10 gift certificate. If you take 10 minutes, fill out the surveys and mail them today, you should have your money in a week! Please page me at (248)966-6716 if you have any questions. Sincerely, Dawn Koger 146 BIBLIOGRAPHY 147 BIBLIOGRAPHY Adler, N.E., Boyce, T., Chesney, M.A., Cohen, S., Folkman, S., Kahn, KL. and Syme, S.L. (1994). Socioeconomic status and health. The challenge of the gradient. American Psychologist. 49(1), 15-24. Alvy, KT. (1981). Effective parenting. Studio City, CA: Center for the Improvement of Child Care. Anchor, K. & Thomason, TC. (1977). A comparison of two parent training models with educated parents. Journal of Community Psychology, 5, 134-141. Andrews, J.E. & Gregoire, E. (1982). Locu_s of control between black parents and their children: Applied educationg research and evaluation. A paper presented to Nova University in partial fulfillment of the requirement of the Ed.D. degree. Apps, J. (1989). Foundations for effective teaching. In E.R. Hayes (Ed.), Effective Etching Styles. New Directions for Continuing Education, No. 43. San Francisco: Jossey-Bass. Azar, S.T., Robinson, D.R., Hekimain, E., & Twentyman, CT. (1984). Unrealistic expectations and problem solving ability in maltreating and comparison mothers. Journal of Consulting and Clinical Pachology, 52, 687-691. Babbie, E. (1992). The Practice of Social Research. Belmont, CA.: Wadsworth Publishing Company. Baker, A.J.L., Piotrkowski, C.S., & Brooks-Gum, J. (1999). Home visiting: Recent program evaluations. The Future of Children. 9(1), 134-151. Baldwin, C., Baldwin, A., & Cole, R. (1990). Stress-resistant families and stress- resistant children. In J. Rolf, A Masten, D.Ciccheti, K. Neuchterlin, & S.Weintraub (Eds), Risk and protectiveiactors in the development of fiychopathology (pp. 257-280), New York: Cambridge University Press. Barnard, K.E., Hammond, M.A., Sumner, G., Kang, R., Johnson-Crowley, N., Snyder, C., Spietz, A., Blackburn, 8., Brandt, P., & Magyary, D. (1987). Helping parents with preterm infants: Field test of protocol EarlLChild Development and Care 2_7,, 255-290. Bandura, A. (1977). Self efficacy: Toward a unifying theory of behavioral change. Pachological Review. 84. 191-215. 148 Bandura, A. (1982). Self-efficacy in human agency. America Psychologist, 37, 122- 147. Barling, J. (1982). Maternal antecedents of children’s multidimensional locus of control beliefs. Journal of Genetic Psychology, 140, 155-156. Baumrind, D. (1972). An exploratory study of socialization effects on black children: Some black-white comparisons. Child Development, 42, 261-267. Baumrind, D. (1991). The influence of parenting style on adolescent competence and substance abuse. Family Relations, 11, 56-95. Belsky, J. (1984). The determinants of parenting: A process model. Child Development, 55, 83-96. Belsky, J., Lerner, RM., & Spanier, GB. (1984). The child in the family. New York, NY: Random House Benassi, V.A., Sweeney, P.D. & Dufour, CL. (1988). Is there a normal relation between locus of control orientation and depression? Journal of Abnormal Psychology, 22, 357-367. Black, M.M., Nair, P., Kight, C., Wachtel, R., Roby, P., & Schuler, M. (1994). Parenting and early development among children of drug abusing women: Effects of Home Intervention. Pediatrics, 94(4), 440-448. Bradley, R.H. & Caldwell, BM. (1976). Early home environment and changes in mental test performance from 6 to 36 months. Developmental Psychology, 12, 93-97. Bradley, P.J., & Martin, J. (1994). The impact of home visits on enrollment patterns in pregnancy-related services among low-income women. Public Health Nursing, um), 392-398. Braton, MB. (1981). The effects of assertiveness training with married subjects on passiveness, assertiveness, agggssiveness and marital satisfaction. Unpublished doctoral dissertation, university of Missouri, Kansas City. Brems, C., Baldwin, M., & Baxter, S. (1993). Empirical evaluation of a self psychologically oriented parent education program. Family Relations, 42, 26-30. Brody, G.H., & Flor, D.L. (1998). Maternal resources, parenting practices and child competence in rural, single African American families. Child Development, 62(3), 803-816. 149 Bronfenbrenner, U. (1986). Ecology of the family as a context for human development: Research perspectives. Developmental Psychology, 22(6), 723-742. Brown, M., & Swick, K. (1979). The neighborhood as a support system for children and families. The Clearinghouse, 54, 350-354. Cain, RA. (1994). Perception of power/control among African Americans: A developmental approach. The Western Journg of Black Studies 18, 164-174. Campbell, D., & Stanley, J. (1963). Experimental and mfii-Expefimentfiesaigns for Research Chicago, IL: Rand McNally. Campbell,T.L. (1994). Impact on prenatal/early infancy home visitation on family health. Family System Medicine. 12. 81-85. Carter, RT. (1991). Black racial identity and psychosocial competence: A preliminary study. Journal of Black Psychology, 23, 58-73. Chandler, TA, Wolf, F.M., Cook, B., & Dugovics, DA. (1980). Parental correlates of locus of control in fifth graders: An attempt at experimentation in the home. Merrill-Palmer Quarterly- 26. 183- 195. Chodorow, N. (1988). “Family Structure and Feminine Personality.” In M. Rosaldo and L. Lamphere (Eds). Women. Culture. and Society. Stanford, CA: Stanford University Press. Benasich, A.A., Brooks-Gunn, J ., & Clewell, BC. (1989). How do mothers benefit from early intervention programs? Journal of Applied Developmental Psychology, 13, 3 1 1-3 62. Cook, B. (1991). Thinking and knowledge in underlying expertise in parenting: Comparisons between expert and novice mothers. Family Reiations, 40, 3-13. Cmic, Gsham, Ragozin, N.M., Robinson, N.M., & Basham, RB. (1983). Effects of social support on mothers of premature and full-term infants. Child Development. 54, 209-217. Culp, A.M., Culp, R.E., Blankemeyer, M., & Passmark, L. (1998). Parent education home visitation program: Adolescent and nonadolescent mother comparison after six months of intervention. Infant Mental Health Journal. 192(2), 111-123. Cutrona, C.E., & Troutman, BR. (1986). Social support, infant temperament, and parenting self efficacy: A meditational model of post partum depression. Child Development, 57, 1507-1518. 150 Daro, D.A. & Harding, KA. (1999). Home visiting: Recent program evaluations. T_he Future of Children. 9(1), 152-171. DeBrabander, B. & Boone, C. (1990). Sex differences in perceived locus of control. Jomnal of Social Psychology, 90, 271-272. Dembo, M., Sweitzer, M. & Lautizen, P. (1985). An evaluation of group parent education: Behavioral, PET, and Adlerian programs. Review of Educational Research. 55. 155-200. Duggan, A.K, McFarlane, E.C., Windham, A.M., Rohde, C.A., Salkever, D.S., Fuddy, L., Rosenberg, L.A., Buchbinder, S.B., Sia, C.C.J. (1999). Home visiting: Recent program evaluations. m Future of Children. 9(1), 66-90. Dyal, J .A. (1984). Cross Cultural Research with the Locus of Control Construct. In H.M. Lefcourt (Ed.). Reseaach with the Locu_saf Control CoastructLVolume 3: Extensions and Limitations. Orlando, FL: Academic Press. Easterbrooks, M.A. & Goldberg, W.A (1984). Toddler development in the family: Impact of father involvement and parenting characteristics. Child Development, a; 740-752. Ellis, RH., & Milner, IS. (1981). Child abuse and locus of control. Psychological Rep_orts, 48, 507-510. Enger, J .M., Howerton, D.L., & Cobbs, CR. (1994). Internal/external locus of control, self esteem, and parental verbal interaction of at-risk black male adolescents. Journal of Social Psychology, 134, 269-279. Field, T.M., Widmayer, S.M., Stronger, S., & Ignatoff. (1981). Teenager, lower class, black mothers, and their preterm infants: An intervention and developmental follow up. Child Development, 51, 426-436. First, J.A. & Way, W.L. (1995). Parent education outcomes: Insights into tranformative learning. Family Relatiops, 44, 104-109. Fox, RA., Fox, T.A., & Anderson, S.L. (1991). Measuring the effectiveness of the STAR Parenting Program with parents of young children. Psychological Ramrts, 6_8_, 35-40. Fulton, A.M., Murphy, KR, & Anderson, S.L. (1991). Increasing mothers’ knowledge of child development: An intervention program. Adolescence 26(101), 73-81. Gatz, M., Grease, E., Tyler, F.B., & Moran, J. (1982). Psychosocial competence characteristics of Black and White women: The constraining effects of triple jeopardy. The Black Scholar, 13. 5-12. 151 Giblin, P.T., Starr, RH., & Agronow, SJ. (1984). Affective behavior of abused and control children: comparisons of parent-chfld interactions and the influence of home environment variables. Jomf Genetic Psychology, 144, 69-82. Gilligan, C. (1977). In C. Gilligan, J.V. Ward, J.M.Taylor with B. Bardige (Eds), Mapping the morgdomain: A contribution of women’s thinking to psychological meory anal education. Cambridge, MA: Harvard University Press. Gilligan, C. (1995). In C. Gilligan, J.M. Taylor, & A.M. Sullivan (Eds), Between voice m silence: Women and girls, race and relationship. Cambridge, MA: Harvard University Press. Glanville, C.L., & Tiller, CM. (1991). Implementing and evaluating a parent education program for minority mothers. Evaluation and Progpam Planning, 14, 241-245. Gomby, D.S., Culross, P.L., Behrman, RE. (1999). Home visiting: Recent program evaluations. The Future of Children. 9(1), 4-26. Goodman, S.H., Cooley, E.L., Sewell, D.R., & Leavill, N. (1994). Locus of control and self esteem in depressed, low income, African American women. Community Mental Health Journal. 30, 259-269. Gordon. (1977). Parent Education: A Plethora of Approaches. Gainsville, FL: Institute for Human Development, University of Florida. Gore, P. & Rotter, J. (1971). “A Personality Correlate of Social Action”. In R. Wilcox (Ed.), The Psychological Comuences of Being a Black American. New York: John Wiley. Gottfried, AW. (1984). Home Environment and Early Cogmt' ive Development: Longitudinal Research. New York: Academic Press. Graves, S. (1986). The Relationship of Parental Locu_s of ContaoL Intemrsonal Supmrt, and the Young Child’s Level of Developmental Functioning, Unpublished doctoral dissertation, University of South Carolina, Columbia. Gray, S.W. &Ruttle, K. (1980). The Family-Oriented Home Visiting Program: A longitudinal study. Genetic Psychology Monoggphs, 102, 299-316. Gross, D., Fogg, L., & Tucker, S. (1995). The efficacy of parent training for promoting positive parent-toddler relationships. Research in Nursing and Health. 18(6), 489- 499. Gurin, A., & Epps, E. (1975). Black consciousness, identrgz' and achievement. A study of students in historically black colleges. New York: John Wiley & Sons. 152 Hardy, J.B. & Street, R. (1989). Family support and parenting education in the home: An effective extension of clinic-based preventive health care services for poor children. Journal of Pediatrics 115, 927-931. Harris, S.L. & Nathan, RE. (1973). Parents’ locus of control and perceptions of cause of children’s problems. Journal of Clinical Psychology, 29, 18-184. Haworth, J.T., Jarman, M., & Lee, S. (1997). Positive psychological states in the daily life of a sample of working women. Journal of Applied Psychology, 27, 345-370. Johnson, D.L. (1993). leac ' low-income m_others to teach_er their children. Paper presented at the Biennial Meeting of the Society for Research in Child Development (70*, New Orleans, LA, March 25-28, 1993). Kanoy, KW., Wester, J., & Lata, M. (1990). Understanding differences between high and low achieving women: Implications for effective placement and teaching. Journal of College Student Development, 31, 133-140. Kelly, M.L., Power, T.G., & Wimbush, DD. (1992). Determinants of disciplinary practices in low income Black mothers. Child Development, 63, 573-582. Kempe, R, & Kempe, C. (1978). Child abuse. Cambridge, MA: Harvard University Press. Kleemeier, C.A.P. (1977). The relationship between observed mother-child interactions and attending misbehaviors in children (Doctoral dissertation, Emory University, 1976). Dissertation Abstracts Internatiofl, 37, 36788. Koeske, G.F., & Koeske, RD. (1990). The buffering effect of social support on parental support. American Journal of Orthopsychiany, 60, 440-451. K0pp, R.G. & Ruzicka, M.F. (1993). Women’s multiple roles and psychological well being. Psychological Reports. 72. 1351-1354. Lamborn, S.D., Dornbusch, S.M., Steinberg, L. (1996). Ethnicity and community context as moderators of the relations between family decision making and adolescent adjustment. Child Development, 67, 283-307. Landau, R. (1995). Locus of control and socioeconomic status: Does internal locus of control reflect real resources and opportunities or personal coping abilities. Social Science and Medicine, 41, 1499-1505. Lee, V.K & Dengerink, HA. (1992). Locus of control in relation to sex and nationality: A cross-cultural study. Journflf Cross Cultural Pachology, 23, 488-498. 153 Lefcourt, HM. (1976). Locus of Conarol: Current Tram in Theory and Research. Hillsdale: Lawrence Erlbaum Associates, Inc. Lefcourt, HM. (1982). Locus of Control: Current Trends in Theory anflesearch, Volume 2. Hillsdale: Lawrence Erlbaum Associates, Inc. Lefcourt, HM. (1991). Locus of Control In J.P. Robinson, P.R Shaver, & L.S. Wrightrnan (Eds), Maasures of Personality and Social Psychological Attitudes. San Diego, CA: Academic Press. Lerner, RM. & Lerner, J .V. (1985). Temperament-intelligence reciprocities in early childhood: A contextual model. In M. Lewis (Ed.), Origins of intelligence, (2nd edition). New York: Plenum. Levenstein, P. (1977). The mother-child program. In M.Day and R Parker (Eds). Ih_e Presclpol in Acti_on: Exploring_Early Childhood Programs QM edition). Boston: Allyn and Bacon. Luster, T. (1987). The Antecedents and Correlates of Parental Perceptions of Efficacy. Paper presented at the Biennial International Conference on Infant Mental Health (2"d , Chicago, IL September 20, 1986). Maccoby, E.E. & Martin, J.A. (1983). Socialization in the context of the Emily: Parent- child interaction. In E.M. Hetherington (Ed.), Handbook of Child Psychology: Volume 4. Socialization parsonality, and social development (pp. 1-101). New York: Wiley. Madden, J., O’Hara, J. & Levenstein, P. (1984). Home again: Effects of the Mother- Child Home Program on mother and child. Child Development, 55, 636-647. Martin, J.K & Nagayama-Hall, G.C. (1992). Thinking black, thinking internal, thinking feminist. Journal of Counseling Psychology, 39, 509-514. McCroskey, J., & Meezan, W. (1997). Family preservation and family functioning. Washington DC: Child Welfare League of America. Melson, G.F., Ladd, G.W., & Hsu, HC. (1993). Maternal support networks, maternal cognitions, and young children’s social and cognitive development. Child Development. 64, 1401-1417. Messer, SB. (1972). The relation of the internal/external locus of control to academic performance. Child Developme_nt. 43(4). 1450-1462. Mischley, M., Stacy, E.W., Mischeley, L. & Dush, D. (1985). A parent education project of low income families. Prevention in Human Services. 3. 45-57. 154 Mitchel, L., & Donnelly, A.C. (1993). Healthy Families America: Building a national system. The APSCA Advisor. 6(4), 9-11. Mondell , S. & Tyler, F. (1981). Parental competence and styles of problem solving/play behavior with children. Developme_ntal Psychology, 17, 73-78. Morganti, J.B., Nehrke, M.F., Hulicka, I.M. & Ctaldo, J.F. (1988). Life-span differences in life satisfaction, self-concept, and locus of control. International Journal of Aging and Hungn Development¢26(l), 45-56. Morris, J.K. (1992). Personal power in black mothers of handicapped sons. Affilia Journal of Women and Soc'1alWont. 7L72-92. Mwamwenda, TS. (1995). South Afi'ican graduate students locus of control, gender differences, and academic performance. Paychological Reports. 77(2), 629-631. Newsome, R. & Foxworth, L. (1980). A study of the impact of specific teaching strategies on locus of control of adult students. Adult/Continuing Education (College of Education, North Texas State University), 31-35. Nowicki, S. Jr., & Duke, MP. (1974). A locus of control scale for college as well as noncollege adults. Journal of Personality Assessment, 38, 136-137. Nowicki, S. Jr., & Duke, MP. (1983). The Nowicki-Strickland Life-Span Locus of Control Scales: Construct Validation. In H.M. Lefcourt (Ed.), Research with the Log of Control Construct. Volume 2. New York: Academic Press. Olds, D., Henderson, C.R., Chamberlain, R., & Tatelbaum, R. (1986). Preventing child abuse and neglect: A randomized trial of nurse home visitation. Pediatrics. 78. 65-78. Olds, D. & Kitzman, H. (1993). Review of research on home visiting for pregnant women and parents of young children. me Future of Children. 3. 53-92. Olds, D., Eckenrode, J., Henderson, C.R., Kitzman, H., Powers, J., Cole, R., Sidora, K., Moris, P., Petitt, L., & Luckey, D. (1997). Long term effects of home visitation on maternal life course and child abuse and neglect. Journal of the America_n Medical Association. 278. 637-643. Olds, D., Henderson, C.R., Kitzman, H.J., Eckenrode, J.J., Cole, R.E., & Tatlebaum, RC. (1999). Home visiting: Recent program evaluations. fle Future of Children 2(1), 44-65. Ollendick, DO. (1979). Parental locus of control and the assessment of children’s personality characteristics. Journal of Personality Assessment. 43. 401-405. Palisi, B.J. & Canning, C. (1988). The perceived control of well-educated women. m Socirl Science Journal. 25. 337-351. 155 Patton, M.Q. (1987). How to Use Qualitative Meth_ods in Evaluation. Newbury Park, CA: Sage Publications. Pecora, P.J., Fraser, M.W., Nelson, K.E., McCroskey, J. & Meezan, W. (1995). Evaluatmg' FamilnLBased Services. New York: Aldine de Gruyter. Phannenstiel, J.C. & Seltzer, D.A. (1989). New parents and teachers: Evaluation of an early parent education program. Early Childhood Research Quarterly. 4. 1-18. Phares, E.J. (1976). Locu_spfccmtrol in personalfl' . Morristown, NJ: General Learning Fairway Press. Pinderhughes, E. (1982). Afro-American families and the victim system. In M. McGoldrick, J. Pearce & J. Giordano (Eds), Ethnicity and Family Therapy. New York: Guilford. Pinderhughes, E. (1989). Understanding race, ethnicity, and mwer. New York: Free Press. Powell, DR. (1980). Persoml social networks as a focus for primary prevention of child maltreatment. Infant Men_t_al Hea_lth Journal, 1, 232-239. Reis, J .S., Orme, Barbera-Stein, L., & Herz, E.J. (1986). A baseline evaluation of family support programs. Journal of Community Health, 11, 122-136. Roosa, M.W. (1984). Short term effects of teenage parenting programs upon the knowledge and attitudes. Adolescence. 75. 659-666. Rotter, J. (1954). Social learning and clinical pgchology. New York, NY: Prentice Hall. Rotter, J. (1966). Generalized expectancies for internal versus external control of reinforcement. P§ychological Monoggphs, 80. Roueche, J.E. & Mink, O.G. (1976). Locus of control and success expectancy (a self study unit). In J .E. Roueche & O Mink (Eds), anroving StudenL Motivation. Manchaca, TX: Sterling Swift. Schaefer, E. (1983). Parent- rofessional interaction: Researc aren rofessional and pplicy finspgctives. Unpublished manuscript, University of North Carolina, Chapel Hill. Schaefer, E., Edgerton, M. & Hunter, M (1983). Childrearing and child development correlates of maternal locus of control. Paper presented at the meeting of the American Psychological Association, New Orleans, LA. Seitz, V., Rosenbaum, L.K. & Apfel, NH. (1985). Effects of family support intervention: A ten year follow up. Child Development. 56. 376-391. 156 Shaw, R.L. & Uhl, NP. (1971). Control of reinforcement and academic achievement. Journal of Educational Resfearch. 64. 226-228. Slaughter, D.T. (1983). Early intervention and its effect on maternal and child development. Monogmphs of the Society for Research in Child Development, 48(4, Ser. No. 202). Smith, EM. (1985). Ethnic minorities: Life stress, social support, and mental health issues. LhaCounaeling Psychologist, 13, 537-579. Soloman, D., Houlihan, R.D., Busse, T.V. & Parelius, R.J. (1971). Parent behavior and child’s academic achievement, achievement striving, and related personality characteristics. Gene_ta: Psychology Monognlphs, 83, 173-273. Soloman, B. (1976). Black empowermenLSoc'jral work in oppressed communities. New York: Columbia University Press. Stevens, J .H. (1984). Child development knowledge and parenting skill. Family Relations 33, 237-244. Stevens, J.H. (1988). Social support, locus of control, and parenting in three low income groups of mothers: Black teenagers, black adults, and white adults. Child Developm_e_nt. 59. 635-642. Stipek, D.J. & Weisz, J.R. (1981). Perceived personal control and academic achievement. Review of Educational Research. 51. 101-137. Stratton, CW. (1998). Preventing conduct problems in Head Start children: Strengthening parenting competencies. Journal of Consulting and Clinical Psychology, 66(5), 715-730. Swick, KL (1984). Inviting parents into the youngchild’a world_. Champaign, IL: Stipes. Swick, K.J., & Graves, SB. (1986). Locus of control and interpersonal support as it related to parenting. Childhood Edgation. 9. 41-50. Tait, NH. (1976). The effect of assertion training in selected personality dimensions of women. Unpublished doctoral dissertation, University of Missouri, Kansas City. Taylor, D.K. & Beauchamp, C. (1988). Hospital based primary prevention strategy in child abuse: A multi level needs assessment. Pediatrics 12, 343-354. Taylor, J.M., Gilligan, C. & Sullivan, A.M. (1995). Between voice an! silence: Women and girls, race and relationship. Cambridge MA: Harvard University Press. 157 Telleen, S., Herzog, A. & Kilbane, TL (1989). Impact of a family support program on mothers’ social support and parenting stress. American Journal of Orthopsychiatry. 59. 410-420. Twentyman, C.T. & Plotkin, RC. (1982). Unrealistic expectations of parents who maltreat their children: An educational deficit that pertains to child development. Journal of Clinical Psychology, 38, 497-503. Upshur, CC. (1988). Measuring parent outcomes in family program evaluation. In Weiss, H.B., and Jacobs, F .H. (Eds), Evaluating Family Prom, (pp. 131- 152). Hathorn, NY: Aldine de Gruyter. Wachs, T.D. (1979). Proximal experience and early cognitive development: The physical environment. Merrill-Palrner Quarterly. 17. 283-317. Wagner, M.M. & Clayton, S.L. (1999). Home visiting: Recent program evaluations. fie Future of Children. 9(1). 91-115. Watson, T., Brown, M. & Swick, K. (1983). The relationahip between home supp_ort and neighborhood supp_ort and the achievement of children enteringthe first ggagle. Unpublished doctoral dissertation, University of South Carolina, Columbia. Weick, A. & Vandiver, S. (Eds) (1982). Women. power. and change. Washington DC: National Association of Social Workers. Wood, P.C., Hillman, S.B. & Sawilowsky, SS. (1996). Locus of control, self concept, and self esteem among at risk African American adolescents. Adolpacence. 31. 597-623. Wyman, P.A., Cowen, E.L., Work, W.C., Holt-Meyers, L., Magnus, K.B., & Fagen, DB. (1999). Caregiving and developmental factors differentiating young, at-risk urban children showing resilient versus stress affected outcomes: A replication and extension. Child Development. 70(3), 645-659. 158 "lllllllllllllllllf