AN EXPLORATION OF PARTICIPANT-LEVEL FACTORS ASSOCIATED WITH THE SUCCESS OF THE BEGIN EDUCATION EARLY PROGRAM Except where reference is made to the work of others, the work described in this thesis is my own or was done in collaboration with my advisory committee. This thesis does not include proprietary or classified information. ____________________________________ Jennifer Ann Wells Certificate of Approval: _______________________________ ______________________________ Carol B. Centrallo Ellen Abell, Chair Associate Professor Associate Professor Consumer Affairs Human Development and Family Studies _______________________________ ______________________________ Jennifer Kerpelman Stephen L. McFarland Associate Professor Dean Human Development and Graduate School Family Studies AN EXPLORATION OF PARTICIPANT-LEVEL FACTORS ASSOCIATED WITH THE SUCCESS OF THE BEGIN EDUCATION EARLY PROGRAM Jennifer Ann Wells Permission is granted to Auburn University to make copies of this thesis at its discretion, upon request or individuals or institutions at their expense. The author reserves all publication rights. _______________________________ Signature of Author _______________________________ Date of Graduation iii AN EXPLORATION OF PARTICIPANT-LEVEL FACTORS ASSOCIATED WITH THE SUCCESS OF THE BEGIN EDUCATION EARLY PROGRAM Jennifer Ann Wells A Thesis Submitted to the Graduate Faculty of Auburn University in Partial Fulfillment of the Requirements for the Degree of Master of Science Auburn, Alabama August 8, 2005 iv VITA Jennifer Ann Wells, daughter of Herman Joseph Wells and Edna Carolyn Lewis Wells, was born February 20, 1974, in Pascagoula, Mississippi. She enrolled in the Psychology Program at Tuskegee University in August 1994. In 1997, while still pursuing her bachelor?s degree she joined the staff of East Alabama Mental Health in Opelika as a paraprofessional. Her primary responsibility was to work with mentally challenged youth and adults by teaching them basic living skills and coping techniques. She graduated from Tuskegee University, in July 1998 with a Bachelor of Science degree in Psychology. In January 2001, she joined the Tuskegee University Cooperative Extension Program as a County Agent for Macon and Montgomery Counties. Her responsibilities in this capacity were to design and implement family and individual well- being programs for Alabama residents. In April of 2005, she joined the Alabama Cooperative Extension System as a Regional Extension Agent for Macon, Montgomery, Lee, Barbour, Bullock, Russell, and Elmore County. Her specialty area remains Family and Child Development. v THESIS ABSTRACT AN EXPLORATION OF PARTICIPANT-LEVEL FACTORS ASSOCIATED WITH THE SUCCESS OF THE BEGIN EDUCATION EARLY PROGRAM Jennifer Ann Wells M.S., August 8, 2005 (B.S., Tuskegee University, 1998) 85 Typed Pages Directed by Ellen Abell This study explored the participant-based factors identified in Cummings? (1999) organizational framework of factors associated with the successful implementation of family-based, educational programming, and examined the relationship between these factors and the differential effectiveness of six independent implementations of the Beginning Education Early (BEE) program. The BEE program is a 10 week parenting education and early childhood education program that is conducted through home visitation and targets children between the ages 2-5. The six West Alabama Counties where the BEE program is implemented are characterized by high unemployment, illiteracy, and poverty. Face-to-face interviews were conducted with 27 mothers or primary caregivers randomly selected from the families who had participated in the program in each county. The transcribed interviews were qualitatively analyzed to vi identify themes that had been noted in prior research, as well as any themes not previously noted in the literature, and to determine whether participant-level factors are useful for understanding program effectiveness. Results indicate that some of the participant?level factors identified in the literature are present in the BEE program data. Attitudes toward program topics, reasons for participation, and perceptions of program staff were the most identifiable themes about which participants talked. Further examination of the data to determine how these factors may be related to program effectiveness across counties, however, found few discernible patterns among the participant-related factors that might predict differences in program effectiveness. Demographic differences was one exception: Counties in which younger, unemployed mothers who had at least one other adult living with them in the household appeared to be more effective in their implementation than other counties. The implications of these findings for future practice and research with hard-to- reach audiences served by program like the BEE program are discussed. Additional research is necessary that would analyze the factors associated with program effectiveness by examining their interaction across the various levels of the model. Patterns of interactions among factors at different levels of the model should be more visible and predictive of strong and weak program implementations than patterns found at any one level. vii ACKNOWLEDGEMENTS The author would like to thank Dr. Ellen Abell for her invaluable help in organizing this thesis and agreeing to chair the thesis committee, and sticking with me through it all. Thanks are also due to Dr. Carol Centrallo and Dr. Jennifer Kerpelman for their support in working on this thesis and agreeing to be members of the thesis committee. Special thanks go to Dr. Nii O. Tackie for his valuable comments during the organization of the study. Particular appreciation is extended to my parents, Herman and Edna Wells for molding me into the person that I have become and always being my source of security and stability. Gratitude is extended to my sisters, Kimberly, Heather, and Ashley for supporting me in the critical periods of my life. Heart felt appreciation goes to Asia, Gabrielle, and Nexton Marshall, III for providing me motivation and giving me the extra push I needed to complete this study. Finally, I extend my appreciation to God, for without whom none of this would be possible, and all others who contributed to the success of this study in one way or another. viii Style manual or journal used Style Manual for American Psychological Association. 19xx. Nth Edition. City: Publisher Computer software used Microsoft Word and NUD*IST ix TABLE OF CONTENTS LIST OF TABLES?????????????????????????.??xi LIST OF FIGURES??????????????????????????xii CHAPTER: I. INTRODUCTION??????????????...??????1 II. LITERATURE REVIEW????????????..?????10 Individual and Social Characteristics Social Support Factors Program Specific Characteristics III. METHODS???..???????????????..??25 Sample Data Collection Data Analysis IV. RESULTS????????????.?????..???.36 Individual and Social Characteristics V. DISCUSSION??????????.??????..??.?47 Participant-Related Factors and the Organizational Framework Participant-Related Factors and their Relation to BEE Program Effectiveness Implications for Practice and Research Limitations and Recommendations for Future Research REFERENCES???????????????????????????.... 59 APPENDICES??????????????????????????..?.66 Appendix A. Information Sheet Appendix B Interview Schedule Appendix C. BEE Adult Subject Data Sheet x LIST OF TABLES Tables 1. County Demographics by Larger Sample and Smaller Sample ??.???.27 2. Reasons for Participation, Attitudes Toward Parent and Child Topics, and Perceptions of Program Staff by County?????????38 xi LIST OF FIGURES Figure 1. Organizational Framework of Factors Affecting Family-based Program Implementation????????????????????.2 2. Coding Scheme for Two-Dimensional Variables???????????34 1 CHAPTER I INTRODUCTION As the problems faced by families, youth, and children persist and increase, the need for effective family support programming continues to grow. There are many programs that are developed to help families deal with several issues in family life. Parenting education, child abuse prevention, and early child education programs are some of the more common family support programs. Head Start, Even Start, and Women, Infants and Children (WIC) are examples of successful early intervention programs that are funded at the federal level. As issues around family life become increasingly critical, successful policies and programs are essential. However, successful programs are not easily developed. As a result, family scientists and other family life practitioners have tried to delineate those aspects of programs that may have contributed to a program?s success or failure. Through extensive program evaluations, a number factors have been identified that can enhance or obstruct a program?s success. For example, Cummings (1999) summarizes the factors associated with successful program development by incorporating them into a schematic model (see Figure 1). This model consists of five levels of factors that affect program implementation: program design, program participants, program staff, organizational climate, and community. The model is represented using concentric circles, similar to an ecological approach, in order to indicate that the various levels in the Figure 1. Organizational Framework of Factors Affecting Family Based Programs 2 model interact, affecting and being affected by other levels in the model. Ideally, a full understanding of program effectiveness would require that the model be examined in its entirety. The first level of the model affecting program implementation is the program design level. Research suggests that programs that are grounded in theory and research and follow an ecological contextual framework usually result in successful outcomes. That is, these programs see the child as developing through the family and see the family developing through the community and larger society (Bogenschneider, 1996; Dumka, 1995; Hughes, 1994; Schorr, 1988). Moreover, Schorr (1988) through her evaluations of community-based family support programs found that comprehensive programs are more successful than those that offer fewer services. Wasik (1993) notes that family support programs can be a source of stress for staff members. This is especially true for programs that circumvent traditional meeting places and times such as home-visiting programs. Although non-traditional community-based programs, such as those programs that meet in non-traditional places (e.g., homes, churches, unsafe areas, and parks) and at non-traditional times (e.g., weekends, after school, early morning or late evening) are able to reach a more diverse audience, staff turn over can be an obstruction to program success if the program design has no way to counteract or address these concerns. The issue of program participants, the second level in the framework, is discussed in detail later in the chapter. 3 Another level of the model of factors affecting program success is the level of program staff. Schorr (1988) contends that program staff should be committed to the program and its goals. They should be flexible and willing to go beyond traditional, professional bureaucracies and red tape in order to meet the needs of the participants. Likewise, she maintains that staff members should be trustworthy and have respect for, interest in, and concern for participants in a relationship built on equality. Similarly, Wasik (1993) reports through her evaluation of home visit programs that staff members should be culturally sensitive and have respect for the values and beliefs of people from various backgrounds. It is also important that the staff members respond to participants with sensitivity and appropriateness. Furthermore, Abell et al. (1999) find that it is essential for program staff to have the ability to convey information in a non-threatening manner. The professional background, educational experience, skill level and training of staff members also play a vital role in program success (Pecora, et al. 1995, Schorr, 1988; Wasik, 1993;). Still another level of the model deals with the organizational climate of the program. Zaltman (1973) found that the level of openness and trust among staff is a contributing factor in program success. Also attributed to program success are the staff?s ability to handle and resolve conflict and the ability to take risks when needed. Steckler et al. (1992) argue that employees? motivation and satisfaction with their jobs, supervisors, and their role in decision making is critical in program implementation and success at this level. The final level of the model deals with issues at the community level. Thompson and Kinne (1990) suggest that the local values, norms, and behavior 4 patterns in the community are key factors relating to implementation. The needs and constraints in the community should also be identified and met (Lefebvre, 1990; Usher, 1995). Wandersman et al. (1996) posit that racial and ethnic diversity within the community, past historical dynamics, concerns of the local government, local and national economic trends, and the political and economic consequences are all equally important when considering program success at the community level. The level of factors on which this study will focus is the program participant level. From the literature, the factors of the program participant level can be summarized in three groups: social and individual characteristics, social support, and program specific characteristics. When addressing the social and individual characteristics of participants, Hughes (1994) finds that programs that match their participants? ages, gender, race or ethnicity, and social class with their program?s design may have a better chance of reaching the intended audience. Likewise, skills, knowledge, and intentions to comply with norms also have been identified as factors that may affect whether or not participants are willing and able to change undesirable behaviors, which in turn will affect program success (McLeroy et al., 1988). In addition, Abell et al. (1999) suggest that successful implementation also may be affected by the participants? reasons for participating as well as his or her willingness to talk about sensitive issues. These program developers also identified participants? attitudes about specific program goals and related topics as issues playing a role in the programs? ability to reach its objectives. 5 The next group of participant level factors is the participants? social support network and the effect those networks have on the program. McLeroy et al. (1988) draw a link between the existence of social support networks and the network members? acceptance of program goals as well as participation in the program. Social support networks are necessary but not sufficient factors for program success. If the participant?s network does not agree with and help foster the program?s goals then, receptivity of those goals by the participant may be challenged. These researchers argue that support networks are key because they provide, ?emotional support, information, access to new social contacts and social roles, and tangible assistance in fulfilling social and personal obligations? (McLeroy, et al., 1988, p. 351). As a result, people?s individual decisions, attitudes, and behaviors can be heavily influenced by their social networks. Therefore, it is critical to program implementation that social support networks receive some attention. The final sub-category of participant level of the model includes program specific characteristics. The family support literature points out that it is important that programs seek out, meet, and adapt to the needs of the participants? whom the program is intended to serve. The needs and constraints of participants, such as the lack of transportation, geographical isolation, scheduling conflicts or timing, and child care may present some potential barriers to participation, which may in turn affect program implementation. Powell (1993) maintains that a program?s capabilities and willingness to address and adapt to such needs influences that program?s possible success. In addition, Abell et al. (1999) suggest that other key issues at this level are 6 getting the participants to accept the program?s goals and getting the participants to trust the staff members. These family scientists contend that a program?s goals are more readily attainable when the participants agree with the program?s goals and trust that the program?s staff genuinely care about them and their families. A particular program for which the participant level factors identified by Cummings (1999) can be evaluated is the Begin Education Early (BEE) program. The BEE program is a 10-week, parenting and early childhood education program that serves families with children ages 2-5. The goals of the BEE program are to increase parenting skills and knowledge with regard to healthy child development, the use of positive guidance strategies, and the provision of appropriate support for children?s intellectual, social, and emotional needs (Abell et al., 1999). The BEE program has been active in six counties in Alabama (Choctaw, Marengo, Perry, Sumter, Wilcox, and Tuscaloosa) where the larger than average proportion of the residents are disadvantaged by low-income status, low educational attainment, and/or lack of regular employment. Strategies designed to attract these hard-to-reach families included a number of different methods. In some counties BEE staff and County Extension Agents drove down county roads and went door to door. In other counties more traditional methods were used such as flyers being placed in grocery stores and announcements placed in Head Start newsletters and church bulletins. Local organizations such as County Department of Human Resources were sometimes used as a referral source. 7 Because targeted program participants are geographically isolated and lack transportation, the BEE program was designed to be delivered through home-visits. A van renovated to be a classroom on wheels travels to the participants? homes once a week. Inside the van is an early childhood classroom, complete with a table and chairs as well as toys and materials needed for developmentally appropriate children?s activities. Two paraprofessional educators are trained to work one-on-one with the parent and child, respectively. The child educator carries out semi-structured lesson plans with the preschooler while the parent educator discusses parenting topics with the parent. The educators are trained to model positive child guidance behavior during the sessions and instruct parents on providing developmentally appropriate children?s activities that foster school readiness. Educator-participant pairs work simultaneously in the van doing their respective lessons. The parent educator discusses parenting topics, using the ? Principles of Parenting? curriculum, and school readiness issues with the parent. The educator also addresses concerns that come up during the sessions and ties those concerns to program goals. Typically, concerns most often mentioned by participants are various child development issues, discipline, and stress. Approximately ten minutes before the end of the hour-long visit, the parent, child, and BEE program educator unite for a ?show and tell.? This final session gives the parent the opportunity to model, practice, and review the principles brought forth during the lessons. The ?classroom on wheels? serves several purposes: it fills the need participants have for transportation; it alleviates any distress participants may have 8 hosting sessions in their homes; it facilitates participant retention by maintaining the parents? interest in the program through the creative opportunities it offers their children. Finally, the van allows educators to structure the educational setting and to have more control over distractions such as phones, televisions, visitors, etc. (Abell et al., 1999). Several studies examining the impact of the BEE program have suggested that participating parents have experienced short-term changes in their knowledge and attitudes about parenting and child development. For example, parents? reports of their use of positive guidance strategies show a significant increase from pre-test to post-test (Bartoszuk & Abell, 1999). Also, significant increases were seen in parents? reports of behaviors associated with school readiness, such as engaging in arts and crafts activities 9 with children, reading to them, and teaching them nursery rhymes (Duffie, 2000). While these are encouraging from an outcome evaluation perspective, they do not address individual differences in the success of BEE program implementations at the individual county level. Questions about the successful replication of the BEE program outside the pilot county arose early on as a result of examining process evaluation data. For example, lower than expected levels of parental satisfaction, an increase in families who did not finish the program, and decreases in the percentage of parents who reported changes in their parenting behaviors were found in one of the first counties attempting program replication (Abell, 1998). The purpose of this study is to explore the participant-based factors identified in Cummings? (1999) organizational framework and to evaluate the relationship between these factors and the effectiveness of the BEE program as represented by separate effectiveness rankings of the six counties which implemented it. Qualitative responses of parents interviewed about their experiences after having participated in the BEE program are analyzed to determine whether participant-level factors are relevant or useful for understanding program effectiveness. 10 CHAPTER II LITERATURE REVIEW A fundamental component of program success is a fit between the program design and the characteristics and context of the program participants. Knowing the audience for which a program is intended will decrease the likelihood of problems arising with implementation, while increasing the probability that the program will be delivered to the appropriate participants (Bogenschneider, 1996; Dumka, 1995; Hughes, 1994; Spoth & Redmond, 1996). Cummings (1999) identifies three sets of participant-level factors in her organizational framework of factors affecting family- based program implementation: (1) individual and social characteristics, (2) social support, (3) program specific characteristics. This section will review the literature related to the factors outlined in this level of the framework, discussing the relationship between these factors and program success. Individual and Social Characteristics 11 Hughes (1994) develops a framework for development of family life education programs in which he maintains that certain information should be gathered about the target audience. A program being designed with children in mind, for example, should consider the age of children, developmental level, gender, social class, family type, and race or ethnicity. Likewise, programs aimed at adults should be prepared to discuss differences in life stage, family stage, family type, age, gender, ethnicity and social class. 12 In general, all programs should consider the information on age, gender, race or ethnicity, social class, and family type. Participants? Age and Developmental Level Programs should consider the age of the target audience to avoid delivering content that is too difficult for the target audience to understand (Bogenschneider, 1996; Hughes & Ong, 1995). Another reason why it is important to consider the target audience?s age is because it will help alleviate a developmental mismatch from occurring. A developmental mismatch occurs when the program content is not congruent with the development period in which the target audience is experiencing (Hruska, 1998). For example, a parenting education curriculum designed for adults is not likely to be effective when administered to adolescent parents. Often times when an adolescent becomes a parent he or she is expected to think, behave, and respond like an adult parent would. A program for adolescent parents must consider the developmental tasks of adolescence as well as the developmental tasks of parenthood. Programs that consider the participant?s age and developmental period may be more effective programs than programs that do not. As a result, Hruska (1998) conducted a training session for professionals working with pregnant or parenting adolescents. It was hypothesized that the training would increase the parent educators? basic knowledge of adolescent development (with specialized knowledge regarding developmental mismatch) provide knowledge of teaching strategies and class content to address the developmental mis-match, and increase effective interaction between the educators and the adolescents. After conducting the one-day, 6 hour training, the study showed that 82% agreed that the 13 training would make a difference in the lives of the teens that they worked with. Likewise the results showed that 67% of the participants agreed that they had learned something new about adolescent development, in particular the developmental mismatch. Finally, the study indicated that 96% of the participants agreed or strongly agreed that the workshop would positively influence their teaching effectiveness. Gender Gender is another factor that can affect a program?s success. Although not intended, the majority of people participating in family support types of programs are female (Kltizner et al., 1990; Lengua et al., 1992; Spoth, 1996). Many of the programs today are ?parent-child? programs and in an overwhelming number of cases the parent tends to be female, especially in low-income or teen parent families (Lengua et al., 1992). It is uncertain whether or not programs purposely seek out women or if women are more likely to participate than men. Lengua et al. (1992) and Klitzner et al. (1990) studies looking at gender differences in program participation found that mothers were more likely to participate than fathers and that mothers were more active than fathers. The study did uncover that fathers? and mothers? participation varied in kind. Mothers were more likely to participate directly in carrying out group activities, whereas fathers were more active in community relations and providing linkages and networks with businesses and civic organizations. Likewise in a study of adolescent drug abusers and their families in treatment, Szapocznik et al. (1988) reported that a dramatic 57.7% of fathers did not show up for the initial intake interview. The study also reported that among those family members 14 who resisted treatment 16% were fathers. Furthermore, in a study of barriers to program participation, Spoth and Redmond (1996) also reported that mothers were more likely than fathers to be involved in program participation. In addition, mothers were less likely to present spousal refusal to participate as a reason for themselves not participating in a program. Fathers, however, reported that the primary reason that they did not participate in programs was the fact that their spouses did not want to participate. These researchers suggest that recruitment strategies should maximize the mothers? role in the participation decision. Race and Ethnicity Another factor influencing program implementation is cultural sensitivity. An ethnic group?s experiences are affected by their varying socio-historical and cultural backgrounds, which in turn shape their opinions, values, attitudes, and beliefs about parenting, education, family, religion, as well as other ethnic groups. Therefore, it is important for programs to be sensitive to the participant?s racial and ethnic heritage (Bogenschneider, 1996; Hildreth and Sugawara, 1993; Hughes, 1994; Hughes and Ong, 1995; Jenkins, 1987; Pecora et al., 1995). If the program content, materials used, or information shared brings about negative images of a particular ethnic group, then individuals with those characteristics may be less likely to participate especially if the information being shared is viewed as offensive or damaging. Similarly, families may decline participation if they feel that the some components of the program are not the norm for their particular ethnic group. For example, as Hughes (1994) argues, many family life programs (parent education) are 15 intended for only white middle class families. As with many things, it is assumed that what is true for the majority has to be true for the minority. Family life issues are presented in blanket form. Families with an ethnicity other that white and a social class other than middle class may find themselves not being able to relate to the program content. For instance, because African Americans and Hispanics comprise a large portion of the U.S.?s disadvantaged persons (Robinson, 2000), family support programs for limited resource families may have a higher participation from Black and Hispanic families (depending on the region) than not. It is important to program implementation that racial and ethnic differences be approached with respect and cultural sensitivity. When participating in a program that prohibits spanking, African American parents may feel overwhelmed by the difference in what their culture identifies as acceptable and what the program identifies as acceptable. Parents may exhibit the culturally acceptable parenting style in the presence of those for whom the behavior is accepted and refrain from exhibiting those behaviors in the presence of program staff, which leads to inconsistent parenting styles. Dumka and Roosa (1995) further agree that African Americans as well as Hispanics are more likely to participate in programs where cultural differences are valued and variations disciplining styles are not diminished. Social Class As stated in previously many of the programs that are designed for parents and children are based on research and information gathered from middle class families (Dumka, 1996; Hughes, 1994; Hughes & Ong, 1995; Spoth, 1996; Schorr, 1988). This presents problems when applied to low-income families. It is important 16 that families are able to relate to the content and materials. Written activities, handouts, visual aids, or anecdotes used to reinforce topics and program goals during sessions should be consistent with the life experiences of the participants (Hughes, 1994). If a parent exhibits concerns with structuring free time for her children while she works, it would be inconsiderate and inappropriate to mention possibilities where class and income differences are differentiated. For example, it would be off target to suggest activities where money is an issue such as ballet, dance, ice skating, piano lessons, karate, or the like as a vehicle for structuring children?s after school, summer or weekend activities. These activities are often categorized as middle class luxury and require money that many low-income families do not have to spare. As a result, suggesting such activities as possibilities for extra curricular programs for low-income parents may prove to be futile and leave the participants frustrated and somewhat depressed about their economic dispositions. Extensiveness is another programmatic concern with interventions for families of varying social classes. Family scientists and practitioners contend that the frequency, intensity, and length may vary greatly among programs intended for participants from an array of social classes (Dumka and Roosa, 1995; Price et al., 1989; Schorr, 1988). For instance, interventions for middle class families may only need to meet on a monthly basis. However, programs that are initiated for low- income families with multiple risk factors may need much more frequent contacts and for longer period of time. Due to the multi-faceted nature of their concerns, high-risk families often require one on one individualized attention. The frequency and 17 intensity of the intervention are also subject to change over the course of participation because low-income families may be prone to cycle through more crises, whereas middle-class families may experience more stability. Participants? Family Types Hughes (1994) argues that the majority of family life programs are designed to address the needs of middle class European Americans, but, if programs are to be delivered to a diverse audience, particular attention needs to be paid to the ways in which different family types are portrayed. He gives the example of African American families frequently being used as examples of single parenting or low- income families. He states that programs need to be sensitive to various family types. Moreover, family type should be considered because the family life cycle does not only exist in traditional form. Not many families follow the trajectory set in the family life cycle (Carter and McGoldrick, 1999). Changes in the family cycle, such as adolescent pregnancy, have lead to the increase visibility of varying family structures. There has been an increase in multi-generational families where grandmother, mother, and child share one roof. As a result, the care-giving role may not always be filled by the biological parents (Abell, et al., 1999). Blacks and Hispanics, as noted in the literature have strong ties to extended family members, some of whom reside in the same household, and rely on familial ties for caregiving (Crawley, 1996; Dumka, 1996; Hines, 1999; Hunter, 1997;). Additionally, more and more are being reared by grandparents, uncles, aunts, siblings, and fictive kin. As Dumka (1996) and Crawley (1996) both suggest, African American and Hispanics are more likely to be attracted to programs where extended family ties are valued and 18 seen as a strength or support system. Also, the traditional family life cycle can be interrupted by life changes and choices beyond teenage pregnancy such as divorce, death, remarriage, formal and informal adoption (Carter and McGoldrick, 1999). Knowledge of the program participant?s family types will help decrease information being delivered to an audience that it is not meant for. Attitudes about Program Goals and Related Topics In addition to the preceding individual attributes, socially constructed characteristics also affect program success such characteristics include the participant?s attitudes about program goals and related topics. In a program developed by Dumka and Roosa (1995), Parent?s Road to Successful Children, parents in the focus group reported that they were more interested in topics addressing their immediate concerns such as gangs, safe neighborhoods, drugs, or reduction of school problems. Although these topics were not a part of the core content, the program allowed flexibility between the core content and the requested content. The program developers decided to present the Parent?s Road to Successful Children?s program as an asset in equipping parents with the skills and knowledge needed to help support academic achievement and guard against the influence drugs and gangs. Likewise, responses to a mail survey examining the reasons parents declined participation in a family based intervention targeting risky adolescent behaviors revealed that parental attitudes and the program topics and goals were inconsistent 19 (Spoth and Redmond, 1996). The parents reported that the intervention would not benefit their families because they did not think their adolescents were at risk for the behaviors targeted by the program. The authors support the idea that interventions are more readily accepted when the program goals and related topics are of value and interest to the participants. Program participants must see a link between the intervention, the topics, and their interests before they become committed to a program. Bogenschneider et al. (1998) found that parents usually deny that issues affecting teens at the national level affected their teens locally. For example, data collected from families in rural, suburban, and urban communities of one Wisconsin county showed that 30% of the 8 th to 12 th grade children sampled admitted to drinking alcohol at least once a month where as only 13% of mothers and 12% of fathers reported that it was likely that their child had engaged in drinking behaviors. Once the program developers presented the information to the parents, and community, participation was mobilized. Willingness to Talk about Sensitive Issues Abell et al. (1999) argue that the extent to which a participant exhibits the willingness to discuss sensitive issues may present some challenges with successful implementation. In a program on child rearing, the authors report parents? initial unwillingness to elaborate on their child rearing experiences and practices. Indeed, many parents feel uncomfortable disclosing personal and sensitive information for fear that it will be used in an incriminating manner against them. For example, if physical punishment is used as the disciplinary method, then the parent may be unwilling to openly talk about it because of fear that the discussion will somehow 20 lead to allegations of child abuse. In addition, participants may not be willing to discuss other sensitive issues such as substance abuse, domestic violence, self-esteem, or sexual behavior. A therapist may see no progress in clients who refuse to discuss the traumatic experience of being raped or sexually molested and in couples who refuse to discuss the issue of physical abuse. This can be especially problematic when sensitive issues are not the presenting problems. Reasons for Participating in Program According to Abell et al. (1999) participants? reasons for participating in a program will have some effect on program success. BEE program developers made it easier for parents to participate by adhering to the responses they received from parents, prior to implementing BEE, regarding reasons to participate in the program. Many parents stated that they would participate in a program that provided opportunities for their children. The parents may have chosen not to participate had the program developers not provided some activities for their children during the program. If parents do not see a need or reason to participate they will not. For example, in an evaluation of parent-led adolescent drug prevention programs Klitzner et al. (1990) reported that the most common reason cited for program participation was fear that their teens were using or at risk of using drugs. The parents that were highly involved may have received more from the program than those who were less involved and participated for other reasons. 21 Moreover, people have various motivating factors or reasons for participating in a program and the reasons may or may not be consistent with what the program goals are. Two examples in this regard are: (1) many states are now mandating divorcing parents to attend parenting education classes before finalizing divorce, and (2) in some areas the courts are mandating people who are drug users and/or drug abusers to complete drug rehabilitation programs. The participant may be resistant to any type of change because the participant is only participating because it is required; in programs such as these the participant may not be open to ideas expressed in the sessions. Also, when people participate in programs for reasons other than facilitation of program or individual goals some problems with success can occur. Recognizing the need for comprehensive services, several family support programs provide participants with a wide range of services (Schorr, 1988). In programs such as these, it can be quite difficult to identify whether a person is participating because they want to benefit from the program goals or because have underlying motives. For instance, some parenting education programs provide families with referrals to community resources, transportation to appointments, donations of food, clothes, toys pampers, formula and the like. In such programs it may be difficult to identify whether or not the participant is in the program because they want to know more about parenting education or because they do not have transportation. The success of the program is compromised when participants take advantage of one service and neglect participation in another. Skills, Knowledge and Intentions to Comply with Norms 22 McLeroy et al. (1988) stress other important factors influencing program implementation are the participant?s skills, knowledge, and intentions to comply with norms. They report that drug use prevention programs aimed at preventing drug use among teens focused on increasing adolescent?s skills and knowledge which would in turn decrease the likelihood of giving in to negative peer pressure. Programs such as these also tend to focus on changing the perception adolescents have about drug use. They emphasize that increasing knowledge is a basic element of all prevention programs. Programs that seek to increase the knowledge of the program participants will be more effective if there is an awareness of the participant?s knowledge base. In drug education programs, participants are informed of the harmful side effects of drugs; the damages caused to others; and the risks associated with drug use. It is assumed that an increase in knowledge will result in a change in behaviors. However, information does not guarantee an increase in skills. Furthermore, they contend that because of their developmental periods and cognitive abilities many teens are not able to problem solve or decision make. Therefore, drug education, sex education or parenting education programs will be more effective if they increase the problem solving skills and decision-making skills of the program participants. Knowledge is useless unless the participants have to skills to apply it. When teens find themselves in situations where they are pressured to have sex or do drugs, having knowledge about drugs or sex may not be as important as having the skills needed to get out of the situation. 23 The authors maintain that programs may be more effective when they increase the participants? skills and knowledge while simultaneously increasing the positive perceptions of norms that teens have about drugs and sex. Program success may depend on perceptions of norms and whether or not the participant has the intentions to comply with those norms. Using the above example, teens may be provided with the knowledge and skills needed for resistance, but if the participants perceive that ?everybody? is ?doing it,? then they may be more likely to engage in negative behaviors. The perceptions that individuals, especially teens, have about norms may be relative to their peer groups or social networks. Social Support Factors As observed in the previous section, the literature suggests that there are also social factors at the program participant level that influence a program?s effectiveness in reaching its goals. The make up of the participants? social network and the extent to which the members network encourage or discourage the facilitation of the programs goals have an effect on program outcome (Spoth and Redmond, 1996). As stated previously, African American families are more likely to live in extended families. The attitudes and behaviors of social support network may not always be in the best interest of those family members participating in family-based programs. For instance, it is common for parents to rely on parenting advice from family members, especially if they are members of the same household (Taylor et al., 1993; Burton, 1990). The parenting advice provided to participants concerning child rearing may be inconsistent with the program goals; as a result, program success may be compromised. Consider the example of an adolescent parent who lives with her 24 mother and grandmother and participates in a teen parenting program. The young mother learns varying ways of managing her child?s behavior that do not include physical punishment. At first chance the adolescent mother begins to discuss with her mother and grandmother disciplining alternatives to physical punishment such as time out or redirecting. The mother and grandmother firmly tell the adolescent that spanking is the only form of acceptable punishment for their household. The adolescent?s continued participation in the parenting program may be jeopardized as well as the goal of increasing positive discipline. Program Specific Characteristics Finally, the literature on program participants suggests that a third category may affect a program?s success. The specific characteristics of a program such as the design or delivery method may have an influence on whether or not the program is effective in obtaining the goals set forth. Participants? perceptions of staff concern for them also play a role in program success. When participants feel that staff members genuinely care about them, they may be more receptive to information that is being conveyed during sessions. Likewise, programs that seek, identify and then adapt to the needs of the target audience will be more successful than those that do not (Bogenschneider, 1996; Dumka and Roosa, 1995; Hughes and Ong, 1995). Often times family support programs target disadvantaged populations where needs such as transportation exists. Programs that adapt and accommodate to the need for transportation will be more successful in reaching its goals. Many family support interventions such as welfare to work programs arrange for participants to be picked up and brought to the center 25 where the programs are hosted. For example, the Alabama Satellite Child Care Project (ASCCP) is a federally funded welfare-to-work program that allows TANF recipients the opportunity to become entrepreneurs (Findlay and Johnson, 2000). Unlike many traditional welfare-to-work programs, ASCCP includes an entrepreneurship component. Families are trained to become licensed child-care providers. The program goes beyond traditional programming by not only providing transportation to and from the training sites but assuming all the costs associated with transforming one room in the participant?s home to a family daycare. Similarly, many parent education or early intervention programs are delivered through home visitation. Programs that seek to reach the needs of participants by nontraditional means like home visitation or by scheduling visits after hours or during the weekend may be more likely to produce desired outcomes than programs that do not. The factors discussed in this section will be examined for their relevance to an understanding of how the BEE program is effective in the six counties in which it has been implemented. The next chapter will deal with such an examination. 26 CHAPTER III METHODS Sample The subjects of this study were selected from families who had participated in the Beginning Education Early (BEE) program in six west Alabama counties. Of the six counties where the BEE program has been activated, four are known as Black Belt counties (Choctaw, Marengo, Sumter, and Wilcox). Currently, Black Belt counties are characterized by a relatively high concentration of blacks, high unemployment, high illiteracy rates, and high poverty rates (Diabate et al., 1995). Counties were given pseudonyms to preserve the anonymity in future discussions of program effectives of the participants, staff members, supervisors, and community leaders interviewed for the study. Counties are subsequently referred to as Blue, Gray, Green, Orange, Red, and Violet. The BEE program is a 10-week parenting education program that focuses on parent behaviors that promote school readiness. The program targets limited-resource, rural families whose primary caregiver reports low educational attainment and whose children are between the ages of 2-5. At the time of data collection a total of 331 families had enrolled in the BEE program over the course of 3 years. Of the 331 families reporting gender, race, education, employment, and marital status, 96% were female, 27 70% had completed high school, and 29% had some education beyond high school. Likewise, 67% reported that they were unemployed and 88% reported being low-income. Eighty percent were African American, 12% were White, and 3% were Hispanic or multiracial. Thirty-seven percent of BEE parents reported being single or never married; however, 70% of the participants identified one or more adults living in the household with them. The six counties began their respective implementations of the BEE program at different times, as the grant funds necessary to operate them became available. The data analyzed in the present study were collected at a time when the BEE program had been in operation in these six counties for periods ranging from 8 to 36 months. The modal length of program operation was 12 months. Twenty-seven BEE program adults participated in the study. Of those reporting demographic information, 100% were female, the average age of the group was 28, 88% were African-American, 12% were White, 78% were unemployed, 19% had education beyond high school, and 54 % had at least one other adult living with them. 28 Table 1. Demographics by Lager Sample and Smaller Sample N=331 N= 27 Race 88% African American 9% White 3% Hispanic/Other 88% African American 12% White Gender 96% Female 100% Female Education 70% Up to High School 19% Up to High School Employment 67% Unemployed 78% Unemployed Marital Status 37% Never Married DNA* Income Level 88% Low Income DNA* Family Type 30% Single Adult Household 60% Single Adult Household * Data Not Available Data Collection Data collection was completed as part of a larger evaluation of the implementation and sustainability of the BEE program. Face-to-face interviews were conducted with the BEE program adults, that is, those adults who had previously participated in or graduated from the BEE program. All 27 interviews were conducted by the same interviewer. To recruit participants, the BEE program educators and county agents compiled a list of previous participating or graduating families in the program. From this list, ten adults from each county were randomly selected. In addition, a list of alternates was also randomly selected in case individuals on the initial list were unable to be located or unwilling to participate. 29 Twenty-seven BEE program adults agreed to participate in the study. The educators provided BEE program adults with packets containing consent forms and information sheets (see Appendix A), and explained carefully and in detail all aspects of the study to make sure the BEE program adults had full comprehension. After obtaining the adult?s agreement to participate, the BEE program educator scheduled an appointment for that adult to meet with the interviewer. Semi-structured interviews were conducted with each participating adult. The interviewer was given a one-page survey of demographic questions to go over with the BEE program adult, as well as an interview schedule which the interviewer used as a guide for a tape-recorded interview session (see Appendix B). The interviewer was encouraged to follow up and probe statements that arose beyond those appearing on the interview schedule. Interviews lasted approximately 45 to 90 minutes. Among other things, adults were asked about their attitudes toward education and school readiness, their social networks, and the school systems in their counties. They were also asked to describe their experiences in the BEE program, including their relationship with program staff. They were asked what lessons they learned as well as what lessons their children learned. Other questions touched on community involvement and community support. County Effectiveness Ranking Prior to the interviews, a ranking of the effectiveness of each county?s BEE program was obtained from the BEE program developer. This ranking was based on the relative effectiveness of each program at meeting program objectives, as indicated 30 by data collected as of June 1998, on short-term outcomes and user satisfaction (Bartoszuk and Abell, 1999), and on the number of clients served and retained (Abell, 1998). The ranking is used in the present study as the basis for exploring how participant-level factors may be influencing the effectiveness of the BEE Program. Data Analysis Data were analyzed in three stages. First, the 27 interview transcripts were analyzed as a group to identify or delineate themes common across all counties. Second, these themes were examined with regard to the factors related to program effectiveness outlined in the literature. Those themes that have been noted in past research, as well as those not previously noted, were identified and discussed. In order to make cross-county comparisons between participant-level factors possible, qualitative material related to specific themes was transformed into numerical codes. Interrater reliability was established, and data were examined for indications that particular variables or combinations of variables might distinguish lower functioning BEE programs from higher functioning programs. Identification of Themes with the Interview Data The interview transcripts were analyzed using the software package, NUD*IST. As stated earlier, NUD*IST is a tool for indexing, searching, and theorizing about non-numerical, unstructured data. Interviews and surveys were prepared for import in NUD*IST by separating the data into topic areas based on questions asked on the interview guide. A subset of seven interviews were initially read and data were put into one of three major groups of participant factors as outlined by the organizational framework: social and individual characteristics, social 31 support, and program specific characteristics. Reports were generated for each node. As the interviews were thoroughly examined each of the three broad nodes were refined to smaller subcategories. Comparing Themes across Counties In order to compare how these themes were expressed across counties differing in the effectiveness of the BEE programs they were implementing, numerical scores were assigned to represent variations in how these themes manifested in each county. Two raters (the author and program developer) worked together to define the coding definitions for each variable. Then they individually coded the data and an interrater reliability percentage was computed. For codes on which raters disagreed, raters discussed their reasoning until both agreed on the appropriate code. Coding scheme for Attitudes Toward Parent Topics Parent changes were defined as a change in attitude, or an increase in skills, knowledge, or behaviors such as learning to care for self, understand, guide, nurture, motivate and/or advocate as a result of program content or program goals. These six categories of parenting behaviors are the foundation of the National Extension Parenting Education Model, or NEPEM (Smith, Cudaback, Goddard, & Myers-Wall, 1994). The number of parent changes per family were counted, individually and summed for a total frequency per county. The total frequency per county was then divided by the number of possible families to get the mean number per county. The interrater reliability for this code was 86%. Examples of this code are statements like, 32 ?She really taught me how to make sure each child have the attention they really need,? and I learned to cope. There is a basic thing to how to cope with your children, what to expect, how to deal with not yelling at them all the time and that is very difficult not to yell at your child, it is. And uh, some days when I have bad days, when I just have a bad day and I seem to be yelling at them, I go back through all my papers, I read them, it makes me happier, it makes me calm down, it makes me realize that children are doing what is normal. Coding Scheme for Attitudes toward Child Topics Nonacademic child changes were defined as an increase in social skills such as learning to share, having a better relationship with parent or siblings, learning not to be afraid, having better eye contact, learning to speak well, having increased concentration skills, increased motor skills, or showing an increased interest in books as a result of program content or program goals. The total number of nonacademic child changes per family were counted individually and then summed for a total frequency per county. The total frequency was then divided by the number of possible families to get the mean number of non academic child changes per county. The interrater reliability for this code was 80%. Examples of this code are statements like, ?Oh yeah, he get along with kids a lot better because normally he wouldn?t play with nobody, and stuff like that, but since he was in the program he got better,? and 33 Um, some colors, uh the rest of her alphabet, she finished up learning those. How to make different creative things, like sometimes she?ll just put together, like she was shown how to make a butterfly then she will try to go back and copy how to do this and uh for the reading, I?ll see her now, she?ll go and pick up her book and read it like they showed her, or try to go back and point out things they have shown her in her books and things like that. Coding Scheme for Participant?s Perception of Program Staff?s Concern for Them The program staff variable was defined as the quantity and quality of responses that indicated the extent to which the program participants perceived the program staff as having a genuine concern for them and/or their children. A numerical score from 1 to 6 was assigned to counties, reflecting a combination of the frequency and quality of statements parents in each county made about BEE program staff. By looking at the distribution of statements across the whole sample and then across counties, frequency was considered ?high? if three or more positive statements were made or ?low? if positive statements numbered less than three. With regard to quality, counties were considered to be high, medium, or low as reflected by their perception of how well the staff person in their respective counties interacted with them during the course of the program. If respondents perceived the staff person to show extra care and concern for their families and to do things beyond what was necessary then the family was considered ?high,? for example, 34 Well, I loved the people, Ms. C and uh, we just loved the uh, the coordinator, she was like a teacher and I loved Ms. B. and they just seemed to uh, have patience and they listened to me, even with problems that were not even in the BEE program, I?d say, I?m having problems with potty training, you know, she?s like the next time I come in I will have the information on potty training. I mean they helped me, they almost became like family, just not a teacher like program, they become family. A ?medium? rating was given if the respondent perceived the staff person?s demeanor as more than general niceness and could elaborate on how those things arose during the sessions. An example of a medium quality statement is They were very nice. When uh, Ms. P first started coming here my son wouldn?t even talk or nothing but the more she started teaching him and reading books to him he got to know her and better and he started acting like he wanted to know something, it was very nice, I liked her very much. A ?low? rating was given if the respondent perceived the staff person?s demeanor as general niceness. An example of this code is, ?I enjoyed the way she taught my daughter.? Counties were assigned a ranking of 1 to 6 based on the combination of these two dimensions (see Figure 2), with more weight given to the quality of responses over the frequency of positive responses. For example, parent statements reflecting 35 both high quality and high frequency of positive perceptions about program staff received a score of 1, while counties in which parent statements that were low in frequency, although high in quality received a 2, and low-frequency, low quality received a 6. Interrater reliability for this code was 81%. Figure 2. Coding scheme for two-dimensional variables. Frequency Quality Code Assigned High High 1 Low High 2 High Medium 3 Low Medium 4 High Low 5 Low Low 6 Coding Scheme for Reasons for Participation Reasons for participation was defined as the extent to which participants viewed participating in the BEE program as investing in their children?s futures. A numerical score from 1 to 6 was assigned to counties, reflecting a combination of the 36 frequency and quality of statements parents in each county made about their reasons for participating in the BEE program. With regards to frequency, counties were considered ?high? if three or more families responded and ?low? if one or two families responded. With regards to quality, counties were considered to be high, medium, or low as reflected by their reason for participating in the BEE program. ?High? quality responses included reasons such as wanting their children to do well in life and seeing the value in starting education early. For example, ?Education is important, very important. Because the way things are changing now?you need your education to be on top of things? if you don?t then you?ll be left behind or low in society.? Responses were coded as ?medium? if the individual believed that having their children participate in a program regardless of the program?s goals as something positive. For example, ?Anything that benefits me and my family, you know, and we can learn from that, you know, is good.? A score of ?low? was assigned to respondents who participated because they saw the program as something to do. An example of a low quality statement is ?Well she wanted to go over there with her friends so I just gone sign her up.? Counties were assigned a ranking of 1 to 6 based on the combination of these two dimensions (see Figure 2), with more weight given to the quality of responses over the frequency of positive responses. For example, counties that reflected both high frequency and high quality reasons for participation received a score of 1, while counties that low frequency but high in quality received a 2, and low-frequency, low quality received a 6. Interrater reliability of this code was 84%. 37 38 CHAPTER IV RESULTS Factors identified in the organizational model are presented in three subcategories: individual and social characteristics, program specific characteristics, and social support. Program participants were cooperative in the interview sessions, however they did not often elaborate on their answers. The types of issues they discussed include: intent to comply with norms, reasons for participation, attitudes toward parent topics and attitudes toward child topics. With regards to the social support factor, few comments were made except for statements indicating that counties lack community support. In some cases the BEE program was viewed as the only existing support available to them. For example, one Gray County participant said, ?I think they need to keep it [BEE program] going because it?s the only thing we?ve got in our area.? Like social support, there were several factors identified at the participant level of the model that participants did not discuss in detail; consequently, cross county comparisons could not be conducted. As a result, the following information contains only those factors where cross county conclusions can be made. Individual and Social Characteristics Demographic Characteristics 39 The literature suggests that the demographic variables of age, gender, race, class, etc. affect program effectiveness through congruence of these factors with program design. In these data, however, parent responses did not yield information that explicitly addressed how they may have experienced the BEE program as related to any of these characteristics. Although participants did not explicitly address these characteristics in their interviews, demographic information available from their surveys is presented in Table 2. In the table, the number appearing in the age row represents the average age of parents in each county. Younger parents were interviewed in Blue and Gray, while participants in the less effective counties were, on average, older. Race is represented by the percentage of African Americans to total number of respondents in each county. Race showed very little variance with regard to program effectiveness. Family type is represented by the percentage of multiple-caregiver households to the total number (multiple is defined as having at least two adult caregivers). Of the 13 families representing the relatively more effective counties-Blue, Gray, and Green, nine have at least two adult caregivers while only 5 of the 14 families representing the less effective counties (Orange, Red, and Violet) have multiple caregivers. Social class or socio-economic status (SES) is represented by the percentage of unemployed respondents to the total number of respondents and the percentage of respondents with education beyond high school to the total number. There were more unemployed parents in the more effective counties than the less effective ones. Gender does not appear in the table because all subjects interviewed were female. 40 Table 2: Demographics by County BLUE N=5 GRAY N=4 GREEN N=5 ORANGE N=5 RED N=4 VIOLET N=5 Age[median] 29 27 31.5 36 36 37 Race 80% AA 100% AA 100% AA 100% AA 50% AA 100% AA Family Type 40% Multi 100% Multi 75% Multi 40% Multi 50% Multi 20% Multi Employment 100% UE 75% UE 100% UE 80% UE 50% UE 60% UE Education (beyond high school) 20% 25% 25% 20% 25% 0% Individual Concerns Individual concerns discussed in the literature include the participants? skills, knowledge, and intent to comply with norms; reasons for participating; willingness to talk about sensitive issues, and attitudes about BEE topics. Of these four factors parents did not mention their willingness to talk about sensitive issues nor did they offer any information that would allow for any conclusions on program effectiveness. The literature suggested that participants? skills, knowledge, and intent to comply with norms are related to program effectiveness because they are connected with the participants? ability and desire to change. Skills, knowledge, and intent to comply with norms. Although the participants did not speak explicitly about their level of parenting skills and knowledge, they did talk about their intent to comply with norms. Across all counties, parents expressed 41 personal expectations for their children?s successful progression through school. They also spoke explicitly about how they intended to help their children reach expected goals through parental involvement. Comments found in the data regarding this factor revealed that parents? ideas and perceptions of success ranged from matriculating into college to staying away from drugs to being independent. For example, a Green County mother argued: I expect them to go to the limit and for me, the limit for my children are to graduate from high school, to graduate from college in a good field and accomplish something. I?ll help them as much as I can but I want them to learn it on their own and all I am hoping I can do is be a good role model to them. Other parents expressed hopes that their children avoid negative outcomes: Oh yes, oh yes, I would like for em to get a high school diploma and get them a nice job and you know, do what?s right. At least please don?t get drugs, I don?t go for that, I don?t go for even smoking cause I don?t smoke myself. I really like for em, that?s why I really enjoyed the program because it always try to help children, you know grow up and be more?A lot of things you can do to uplift kids, like, if you try to help them to grow up, that is you have to teach them, you don?t need to be hollering at them, you can just take time out and you know, 42 be with em and let em understand things?as long as you help em along they will grow. 43 There were also parents who measured success against their own personal accomplishments. Open and honest statements like ?I don?t want him to be like me.? or ?I want my baby to do better than I did? epitomize parental norms and values reflecting the notion of wanting children to have a better life. A Green County parent candidly states: I hope he?ll do good, very well cause I want all three of my kids to finish, even though I didn?t. I want them to be, you know have a good job, don?t be like me, sitting over there at home waiting for a husband to take care of them, I am serious, you know. I want them to finish, do the best they can to finish. Reasons for Participation. Participants? reasons for participating in a program are deemed important to program effectiveness because participating in programs for reasons other than goal facilitation may compromise program success. Respondents discussed their reasons for participating in the interview data in varying degrees of enthusiasm and commitment. There were parents who enrolled themselves and their children because they identified with the benefits of early childhood education programs such as the BEE program. These parents believed that by participating in the BEE program they were giving their children a head start in life. Statements in the interview data, such as ?it is important to give children a good start when they are little so they will do good later on?, show that some parents perceived that early childhood education initiatives have a 44 positive impact on future education. For example, one Gray County parent stated ?Oh, I love it because like I say it?s a gift to start children at an early age. This had never been before, you know, that I can remember and to me it?s a good start in life.? Other parents participated in the program because they perceived participation in any child centered program as something positive for their children. ?I am for anything that my family can learn from,? said one Blue County parent. Likewise, one Violet County mother said she enrolled her daughter and herself in the program simply because ?it would teach her a lot of things.? There were also some parents who participated to satisfy a need. They viewed the BEE program as a good alternative to Head Start since some children ?didn?t get a chance to go.? The data further revealed that some parents participated as an alternative to doing nothing. For example, a Orange County mother confessed Well, [parent educator and child educator] came to my house one day and they? asked me if I would like to participate in it and I told em yes it would be fine with me. I wasn?t doin nothing at the moment so I decided to participate with em?. Table 2 shows the differences among counties in the quality of the reasons parents gave. Blue, Gray, and Red County had a score of medium, while Green, Orange, and Violet were rated as low. Orange and Gray counties have the highest number of parent and child behavior changes representing attitudes toward parent topics and attitudes toward child 45 topics. As result of averaging and then combining the quality and frequency of two- dimensional codes the highest rating given to a county was medium with regards to reasons for participation. It is important to note that most counties had at least one high quality reason for participation. Table 3. Reasons for participation, Attitudes Toward Program Topics, and Perceptions of Program Staff by County BLUE GRAY GREEN ORANGE RED VIOLET Reasons for Participation ( 1= highest/ 6= lowest) 4 4 6 5 4 5 Attitudes Toward Parent Topics (avg. # of changes/county) 1.75 2.75 2.25 2.8 2.25 .2 Attitudes Toward Child Topics (avg. # of changes/county) 1.25 1.75 1 1.8 1.25 .4 Perceptions of Program Staff ( 1= highest/ 6= lowest) 4 3 6 1 3 No Data Attitudes about Program Topics. The literature suggests that the participants? attitudes about program topics are important because they determine whether or not they are likely to maintain their attendance in the program and achieve program goals. Participants are more likely to participate in, enjoy and learn from programs lessons if the topics are issues that relate to them. Families across all counties had positive attitudes about the topics discussed during BEE sessions. While parents didn?t explicitly address 46 their own attitudes, one of the themes drawn from the data dealt with the changes parents described for themselves and for their children due to the program. The changes are understood as indicators of having a positive, receptive attitude toward program topics. Some topics were mentioned more than others. The majority of respondents noted their enjoyment of topics on positive discipline, stress, and child development. Participants also discussed how they learned from these topics. For example, one Blue County parent said: ?I enjoyed reading about the spoiled children, so I know now I have spoiled children? Some of their discipline ideas were pretty good.? Similarly, a Red County participant expressed what she learned from the stress topic: Yes, I learned to cope. There is a basic thing to how to cope with your children, what to expect, how to deal with not yelling at them all the time and this is very difficult not to yell at your child, it is. And uh, some days when I have had bad days, when I have a bad day and I seem to be yelling at them, I go back through all my papers, I read them, it makes me happier, it makes me calm down, it makes me realize that my children are doing what is normal and you don?t , you just don?t understand how it helps. On the other hand, some parents discussed in more detail topics that their children learned from such as an Orange County parent who said her child learned, 47 um some colors, uh the rest of her alphabet, she finished up learning those. How to make different creative things, like sometimes she?ll just put together, like she was shown how to make this butterfly then she will try to go back and copy how to do this and uh for the reading, I?ll see her now, she?ll go and pick up her book and read it like they showed her, or try to go back and point out things they have shown her in her books, and things like that. As seen in Table 2, Gray and Orange counties described the most changes as indicated by the attitudes toward parent topics and child topics. Given Blue?s position as the most effective program, it is interesting to note that parent attitudes about parent and child topics were not as high as some of the less effective programs. Program Specific Characteristics Variables Program Staff. Participants? comments do not show concern about program features such as program implementation, design, and delivery as noted in the literature. Participants did not talk about aspects of the program that were not tangible or outside their immediate attention. The program staff was the only factor mentioned under this category. Across all counties participants viewed the program staff as ?the program.? The participants? experience with the program is informally connected to the program staff. Participants talked about the program staff in terms of how well the staff persons from their counties behaved toward them, their children, or their families. Some 48 participants mentioned outstanding behaviors modeled by staff, spoke of how well their children responded to the staff, and described how the staff person referred them to other social service agencies for resources. For example, a Red County participant stated: Well,?we just loved the, uh, the coordinator she was like a teacher and okay, I loved [the parent educator] and they just seemed to, uh, to have patience and they listened to me, even with problems that were not even in the BEE program, I?d say, I?m having problems potty training, you know, she?s like, next time I come in I will have the information on potty training. I mean they helped me, they almost became my family, just not a teacher, they become family. Likewise one Orange County participant said, ? ? I am a single parent and I needed some tips on childcare. She gave me a number to help me with that and a person who I can talk to, to help me with financial things?.? Other participants elaborated on general behaviors expected of family support workers. Comments such as these were given a rating of medium. For example a Blue County resident had this to say about the program staff, ?The most thing I enjoyed how they worked with my child and sat down with her and made her, you know, helped her learn and everything like that,? and one Gray County participant offered this, ?Well she 49 explained everything, you know, she had a positive attitude about everything, she wasn?t negative about nothing.? However some participants? perceptions of program staff were not as elaborative. Their comments reflected positive but basic behavior exhibited by the staff which received a rating of low. For example, one Red County participant response was, ? Oh, she did pretty good, you know? She act pretty good,? and another Green County participant said, ?She was nice.? Table 3 shows that Orange County, a less effective county ranked highest on perceptions of program staff. Orange County, ranked fourth in overall effectiveness, had the most parent and child changes and had very high perceptions of BEE staff. 50 CHAPTER V DISCUSSION The purpose of this study was to explore the participant-based factors identified in Cummings? (1999) organizational framework and to evaluate the relationship between these factors and success of the BEE program. The qualitative data provided through interviews with parents who had participated in the BEE program offered a valuable perspective on families? experiences while enrolled in the program. Parents involved in the BEE program valued norms and ideas relative to the success of their children such as making good grades and staying out of trouble. Most of the families viewed the BEE program as beneficial, had positive relationships with staff, and did not want the BEE program to end. However, there were no clear patterns found among the counties to support the idea that program effectiveness could be predicted at the participant level of the organizational model. This discussion will begin by reviewing what the data suggests about the factors proposed in the organizational framework and how those factors appeared to relate to the relative effectiveness of the six counties. Specific findings will be discussed and limitations of the research will be noted. The study will conclude by addressing the implications of the findings for the model and the BEE program and suggestions for future research. 51 Participant-Related Factors and the Organizational Framework An integration of the literature about participant-related factors suggested that a number of individual factors--such as race, age, gender socioeconomic status, and family type?are relevant to understanding program effectiveness. From a program design standpoint, knowing and understanding the target audience has always been considered a critical part of developing effective programs (Hughes, 1994). It is necessary to have a programmatic match between the participants and the design to avoid barriers to participation and to enhance the transfer of information and the acquisition of skills, knowledge, and behaviors. The BEE program was designed to counter many of the barriers to participation that low-income, rural parents of young children often have with commonly delivered, classroom-based parenting classes, for example, through its innovative van-based approach and inclusion of the child in the educational setting. BEE participants were not expected to elaborate on these individual factors, but their demographic data suggest that parents in the more effective counties were relatively younger, more likely to be unemployed, and more likely to be living with at least one other adult in the household. More effective counties appear to have attracted or recruited younger parents with time at home and another adult to support them in their tasks. These characteristics may be associated with being more open to the information and educational goals of the programs. With the exception of program staff, most of the factors identified under the program specific category of the model (delivery, implementation, adaptation to meet the needs and barriers of the intended audience) are not content about which parents 52 spoke in the interviews. Program participants did not speak about valuing the benefits of the ?classroom on wheels? or the home visits. Because many participants perceived the BEE program as being the only program available to them and their families, they had little apparent basis for comparisons that might have led them to do so. If the BEE program is reported as being the only support available in many of the counties, then participants are drawing from only one experience--the BEE program. The fact that these families had limited experience with family support programs did not prevent them from analyzing BEE staff. Although the participants may have limited experience in programs like BEE, they do have experience in human interaction. Because staff is tangible, immediate, and visible on a regular basis, this factor is more concrete than the others, and participants can discuss their perceptions about staff more readily. Experiences dealing with people in general allow them the ability to express, in varying degrees, their perceptions of staff behavior. Relatively all counties had positive things to say about how the staff interacted with them and their children. To say the least, the BEE educators ?were? the BEE program. Descriptions of social support, the last of the three categories of participant- related factors, also did not appear in the words of the parents interviewed. The research literature maintains that the people closest to the participants have some influence on their attitudes about program participation, their likelihood to remain in a program, and their likelihood to adopt new behaviors as a result of the program. The parents interviewed did not speak about whether or not their friends and family agreed with them participating in the BEE program. When asked questions 53 concerning social support many individuals didn?t mention family members, even though 70% of participants reported having at least one other adult caregiver living with them in the household. This could be due to the cultural norm that families not seek advice or help from people outside the family, or it could be explained by an implicit expectation that other members of their family ?are suppose to help because that?s what families do.? Therefore it could be possible that the perception of having no support is maintained by the fact that there is not outside support, though not necessarily lack of familial support. Perhaps a lack of more specific questions related to social support as considered in the literature had something to do with the responses received. More direct questions such as ? Is there anyone in your household who helps you with your role as a parent?? or ?Who assists you with the day to day activities associated with child rearing?? and ?What do these people think or say about the BEE program?? may elicit responses that reflect social support factors found in the literature. Social support as considered in the model is key to program success; nevertheless, these data do not capture it. Social support in these data comes out in community support and faith in God, rather than kinship or friendship. Participants stated that God is who they turn to when times are difficult, and their faith in Him carries them through hardships and isolation. The participants? perceptions of social support were reported as low or nonexistent. This could be because families who are geographically isolated, low income, and disenfranchised often times need a lot more than the organizations in their areas provide. Therefore, it may be possible that because the needs of the 54 families and the resources of the community aren?t equal, the families begin to feel even more isolated and neglected. In fact, many families expressed their appreciation for the BEE program because they recognized it as being ?the only thing that?s around for families.? It may be worth noting that the BEE families? perception of the BEE program as being the only program available may be the result of the hands-on recruitment strategies of the BEE staff (e.g., driving the van up to the house and asking if the family would like to enroll in the program). Participants acknowledge that if there are other programs going on in the community they don?t know about them because as one parent states ?the only way you know what?s goin? on is to be at the schools.? The majority of BEE parents have children ages 2-5 and haven?t attended any formal educational institution like Head Start. Participant-related Factors and their Relation to BEE Program Effectiveness The model suggests that there is a relationship between program staff and participants, and successful programming. Generally, when there is a positive relationship between the staff and the participants, the participant will be more receptive to the goals of the program. However, Orange County was ranked as one of the lower counties in the overall program effectiveness ranking, yet was ranked highest in terms of parents? attitudes toward parent topics, attitudes about child topics, and perceptions of program staff. At the same time, Blue County, ranked as the most effective county in overall program effectiveness, ranked relatively low in the identical categories. Orange County participants frequently had high praise for their 55 staff, and this may have directly affected their attitudes toward parent topics and child topics. This unexpected finding may have something to do with the nature of the overall effectiveness ranking--obtained from a combination of short term outcomes, user satisfaction, and number of clients served and retained- rather than ranking in this study. Short-term outcomes were assessed pre- and post-program and consisted of closed questions regarding changes in parents? attitudes about discipline, the emotional support children need, and school readying behaviors. Responses could be ?agree,? ?disagree,? and ?unsure.? In the face-to-face interviews, questions were open-ended, allowing for parents to provide a more full account of their activities and lessons. Perhaps the close-ended questions did not capture other benefits of the program that parents were able to recount weeks or months after finishing the program. Secondly, Orange County was one of the younger programs and in its first year had problems recruiting (Ellen Abell, personal conversation) bringing down their overall effectiveness ranking. The most effective counties, such as Blue and Gray, had served and retained more families than Orange. Violet County ranks as the least effective county over all and on attitudes toward parent topics and attitudes toward child topics, and there was no data at all on perceptions of program staff. (In fact, this county ranked the lowest across all categories with the exception of reasons for participations where they ranked as the second lowest.) Most telling, no one in Violet County responded to questions regarding program staff. The silence of parents in Violet County about BEE educators 56 may be an indicator of negative perceptions of staff. According to the model, negative staff perceptions can lead to negative attitudes about parent topics and child topics which then lead to little behavior change for the parents or the children. It may be that parents did not favor the staff in Violet County and chose to avoid the question rather than say anything negative. Lowry?s (2002) analysis of the staff level of the model confirms that there were problems in Violet County between the parent educator and the child educator in being able to work together. These problems may have filtered down into the educators? interactions with the families. Another interesting finding was that the more effective counties had more participants who were unemployed, younger, and who potentially received help from other adults in the household than in the less effective counties. Societal norms justify economically stable mothers delaying or avoiding the work force so that they are able to spend quality time bonding with their children. This might suggest that parents who are younger, who are unemployed (by choice or by force) or are able to be at home (perhaps with help from others) may have more time, energy and attention to devote to the positive development of their children. Implications for Practice and Research It is a commonly assumed that individuals who are low income, isolated, unemployed and who have little or no education don?t value the same norms as those who aren?t low income, connected, employed and educated. However, these data did show that the BEE participants valued education. They valued the same norms as others in society, such as graduating from high school, going on to college, getting a good job, and staying out of trouble and away from drugs. Participants wanted their 57 children and other children in their families to be more successful than they had been. Statements like ?I want them to do what I didn?t do? are found throughout the interviews. Participants had a clear vision of which direction they wanted their children to go, even if they did not necessarily have the knowledge and skills that would lead to these positive outcomes. Even though families initially enrolled their children in the BEE program because of its perceived concentration on early childhood education and not parenting education, participants? comments depict positive attitudes about the parenting topics. The BEE parent educator used a variety of topics which were derived from the categories of the National Extension Parenting Education Model or NEPEM: Care for Self, Understand, Guide, Nurture, Motivate, and Advocate (Smith, et al, 1994). Comments from the BEE interviews showed that participants found some topics more useful than others. Because the individuals participating in the study were more concerned with day-to-day living and managing immediate needs, they paid more attention to and learned more from lessons on stress, discipline, patience, child development, age appropriate behavior, and parental involvement. Similarly, the selection and training of BEE educators and program staff was another positive program design element mentioned by participants. To tackle the difficulties in presenting sensitive information to hard-to-reach individuals, the BEE program developers had to be selective in the hiring process, making sure to hire para-professionals who exhibited the basic behaviors important in any family service worker: compassion, sensitivity and a passion for helping others (Abell, Mize, & Shields, 1999). As trained home visitors, BEE educators realized the importance of 58 understanding the culture, dynamics, and mentality of isolated and disenfranchised individuals and communities. In the eyes of the participants, the BEE staff represented what the BEE program was all about. The BEE educators ?were? the BEE program and made optimistic impressions on the lives of the BEE participants by building trusting relationships, respecting the point of view of the caregiver, and expressing a genuine concern for the families. These data show the importance of staff in program effectiveness. Program administrators should be selective in their choices for staff, employing only those individuals who exemplify the qualifications and behaviors necessary for building trusting and empathetic relationships with program participants. Effective interpersonal skills are key in shaping positive perceptions of program staff. It is recommended that the BEE programs, and programs such as BEE, incorporate a ?customer service? or interpersonal skills component into their existing trainings for staff. In reality programs are businesses that provide a service to customers/ participants. Just as the quality of customer service that staff provide in commercial business helps determine user satisfaction, the quality of customer service program staff provide also helps determines user satisfaction. Program staff should be trained in providing excellent customer service to program participants to help ensure a return investment of positive perceptions of program staff, program goals being met, and overall program satisfaction. Limitations and Recommendations for Future Research 59 Despite participants? descriptions of the positive impact of the BEE program, firm conclusions on the program?s effectiveness as it relates to the model cannot be drawn. A number of limitations must be considered. First, in their descriptions of their experiences with the BEE program, information appeared to be limited by the lack of contact participants had with other programs. As a result of what was perceived as insufficient and non-existent family-based community programs, the participants had nothing else to compare their experience in the BEE program to. Secondly, many of the BEE participants were unable to articulate their experiences. They were concrete in their responses and did not tend to elaborate on the open-ended questions asked in the interviews. In addition, the small sample size, with less than 10% of the total population of BEE families who had completed the program interviewed, adds to a sense that additional information would have provided a fuller picture of the context needed to understand the participant-related factors affecting program implementation. In the future, it is recommended that interview procedures for audiences such as BEE participants be less formal and evaluation instruments more concrete, and include a mixture of open-ended and closed-ended interview questions. When participants lack the knowledge base, experience or skills to articulate and elaborate on program specifics, it is key that effective tools be designed to give the best reflection of the participants? experiences. Future research will be necessary to examine the best evaluation techniques and instruments for gathering information from individuals with similar characteristics. 60 Furthermore, it must be remembered that this study only looks at one level of the model. In order to examine program effectiveness as a whole, the factors at the participant level must seen in the context of all levels of the model. For example, Violet County wasn?t only deficient at the program participant level of the model. In studies by Cumming (1999) and Lowry (2002) looking at the community level and staff level, respectively, significant problems such as racial division, a lack of community support, and problems between the two BEE educators were noted. The findings in this study suggest that the relatively less effective counties have older parents than the more effective counties. It is recommended that this age group be targeted with a different approach. Because of the incongruent life cycle stages of senior adults rearing young children, there are certain challenges that may interfere with program effectiveness. Issues such as health conditions, living on a fixed income, lack of energy, feelings of resentment, and stress may present barriers to achieving program goals. The BEE program should incorporate curricula such as those used in family support programs targeting grandparents as parents to address issues around managing the tasks associated with being a senior adult while at the same time managing the tasks associated with being the parent of young child. The results of this study show the importance of program developers understanding the target audience. Program designers should not only be able to identify characteristics and demographics of the individuals they are planning for, but also what those characteristics and demographics mean for programming and evaluation. Program evaluators need to comprehend how the characteristics of their audiences translate into 61 program design. There may be demographic characteristics that are particular to, or associated with, successful programs. If this is the case then these demographics should be targeted. Research and analysis is essential from the beginning of the planning phase and should be ongoing throughout the evaluation phase. Because BEE program developers analyzed how the circumstances of low-income, under-educated, unemployed, disenfranchised adults might create barriers to success and incorporated features into program design to address these barriers, they were successful in reducing the prospective threats to effectiveness. The knowledge derived from this study should be extended by incorporating it into an analysis of program effectiveness that examines all levels of the model at once. Patterns of interactions among factors at different levels of the model may be more visible and predictive of strong and weak program implementations. In addition, research would be useful that examined variations in evaluation techniques and instruments used with individuals who have limited experience in assessing key dimensions of the family-based programs they are enrolled in or who are less likely to analyze or to comment on the more abstract aspects of their experiences in a family based program. Finding ways to tap the experiences of program participants as a means of evaluating program effectiveness in hard-to-reach populations, would be one more way the BEE program could show its ability to ?meet people where they are.? 62 REFERENCES Abell, E. (1998, November). Factors related to the replication of a parenting and early childhood development education program for rural families. Presented at the annual conference of the National Council on Family Relations, Milwaukee, WI. Abell, E., Mize, J. and Shields, E. (1999). Challenges to providing parenting education in rural, low-resource communities: Lessons from the Begin Education Early program. Family science review, 12 (2), 87-100. Barry, C. A. (1998). Choosing qualitative data analysis software: Atlas/ti and Nudist compared. Sociological research online, 3 (3), 1-22. Bartoszuk, K. and Abell, E. (1999, February). Providing parental education in rural, low- resource communities: Results from the Begin Education Early (BEE) program. Presented at the annual conference of the Southeastern Council on Family Relations, Atlanta, GA. Bogenschneider, K. (1996). An ecological risk/protective theory for building prevention programs, policies, and community capacity to support youth. Family relations, 45 (2), 127-138. 63 Bogenschneider, K., Ming-Yeh, W., Raffaelli, M., Tsay, J. (1998). Other kids drink, but not my kid: Does parental awareness of adolescents alcohol use protect adolescents from risky consequences. Journal of marriage and the family, 60 (2) 356- 373. Burton, L. M. (1990). Teenage child bearing as an alternative life-course in multigeneration black families. Human nature, 1 (2) 123- 143. Buston, K. (1997). NUD*IST in action: its use and its usefulness in a study of chronic illness in young people. Sociological research online, 2 (3), 1-18. Carter, B. and McGoldrick, M. (1999). Overview: The expanded family life cycle: Individual, family and social perspectives. In B. Carter and M. McGoldrick (Eds.), The expanded family life cycle: Individual, family, and social perspectives. (pp. 1-24). Boston:Allyan and Bacon. Crawley, B. H. (1996). Effective programs and services for African American families and children: An African ?centered perspective. In S. L. Logan (Eds.), The black family (pp. 112-130). Cummings, R. (1999). An organizational framework of factors affecting family based program implementation: Exploration of community level factors associated with the success of the Begin Education Early program. Unpublished master?s thesis, Auburn University, Alabama. Denzin, N. K. and Lincoln, Y.S. (2000). The Discipline and Practice of Qualitative Research. In N. K. Denzin and Y.S. Lincoln (Eds.), Handbook of Qualitative Research, 2 nd ed. (pp. 1-28). Thousand Oaks, CA: Sage Publications. Diabate, Y. (1995). The impact of human resource factors on economic growth in 64 Alabama black belt counties. In R. Zabawa, N. Baharanyi, and W. Hill (Eds.), local communities and sustainable development. (pp. 45-56). Tuskegee: Tuskegee University. Duffie, S. J. (2000). Parental perceived enactment of school readiness behaviors among families enrolled in the begin education early (BEE) program. Unpublished master?s thesis, Auburn University, Alabama. Dumka, 1996 missing. Dumka, L. E. and Roosa, M. W. (1995). Using research and theory to develop prevention programs for high risk families. Family relations, 44, 78-86. Findlay, J. and Johnson*** Missing**** Hildreth, G. J. and Sugawara, A. L. (1993). Ethnicity and diversity in family life education. In M. E., Arcus, et al. (Eds.), Handbook of family life education, vol. (pp. 162-188). Hines, P. M. (1999). The family life cycle of African American families living in poverty. In B., Carter and M., McGoldrick (Eds.), The expanded family life cycle:Individual, family and social perspectives 3 rd ed. (pp.327-345) Boston: Allyan- Bacon. Hughes, R., Jr. (Jan. 1994). A framework for developing family life education programs. Family relations, 74-80. Hughes, R., Jr. and Ong, B. (1995). Effective prevention programs for children. Human development and family life bulletin, 1(2). Columbus, OH: The Ohio State University. Hunter, A. G. (1997). Living arrangements of African American adults: Variations by 65 age, gender, and family status. In R .J. Taylor, J. S. Jackson, and L. M. Chatters (Eds.), Family life in black america. (pp. 262-276). Hruska, J. L. (1998). Working with pregnant and parenting adolescents : A developmental approach to parenting education. Unpublished master?s thesis, Auburn University, Alabama. Jenkins, S. (1987). Ethnicity and family support. In S. L. Kagan, et al. (Eds.), America?s family support programs (pp. 282-294). Klitzner, M., Bamberger, E., Gruenewals, P. J. (1990). The assessment of parent-led prevention programs: A national descriptive study. J. Drug Education, 20 (2) pp 111-125. Lefebvre, R. C. (1990). Strategies to maintain and institutionalize successful programs: A marketing framework. In N. Bracht (Ed.), Health promotion at the community level (pp. 209-228). Newbury Park, CA: Sage. Lengua, L. J., Roosa, M. W., Schupak-Neuberg, E., Micheals, M. L., Berg, C. N., and Weschler, L. F. (1992). Using focus groups to guide the development of a parenting program for difficult- to- reach, high -risk families. Family relations, 41, 163-168. Lindop, E. and Cannon, S. (2001). Evaluating the self-assessed support needs of women with breast cancer. Journal of Advanced Nursing, 34 (6) 760-771. 66 Lowery, E. (2000). Thesis: Program Staff Factors Impacting the Effectiveness of the Begin Education Early Program. Unpublished master?s thesis, Auburn University, Alabama. McLeroy, K., Bibeau, D., Steckler, A., and Glanz, K. (1988). An ecological perspective on health promotion programs. Health education quarterly, 5, 315-330. Meadows, L. M. and Dodendorf, D.M. (1999). Data management and interpretation: using computers to assist. In W. L. Miller and B.F. Crabtree (Eds.), Doing qualitative research (pp. 195-218). Thousand Oaks, CA: Sage Publications. Mechanic, D. and Meyer, S. (2000). Concepts of trust among patients with serious illness. Social science and medicine, 51 (5) 657-668. Miller, W. L. and Crabtree, B.F. (1999). Clinical research: a multimethod typology and qualitative roadmap. In W. L. Miller and B.F. Crabtree (Eds.), Doing qualitative research (pp. 3-30). Thousand Oaks, CA: Sage Publications. Pecora, P., Fraser, M. W., Nelson, K. E., McCroskey, J. and Meezan, W. (1995). Evaluating: family- basedservices. New York: Aldine DeGruyter. Powell, D. R. (1993). Inside home visiting programs. The Future of children 3(3), 23-38. Price, R. H., Cowen, E. L., Lorion, R. P., Ramos-McKay, J. (1989). The search for effective prevention programs: What we learned along the way. American Journal of OrthopsXchiatrics, 49-58. 67 Qualitative Solutions and Research. Ltd. (1997). QSR NUD*IST. Thousand Oaks,CA: Sage Publications Software. Robinson, R. (2000). The debt: What America owes to blacks. New York: Dutton. Ryan, G. W. and Bernard, H. R. (2000). Data management and analysis methods. In N. K. Denzin and Y.S. Lincoln (Eds.), Handbook of Qualitative Research, 2 nd ed. (pp. 769-802). Thousand Oaks, CA: Sage Publications. Schorr, L. (1988). The lessons of successful programs. In ??Within our reach (pp. 256-283). New York: Doubleday. Smith, C.A., Cudaback, D., Goddard, H. W., & Myers-Wall, J. (1994). National extension parenting education model. Manhattan, Kansas: Kansas Cooperative Extension Service. Spoth, R., and Redmond, C. (1996). Barriers to participation in family skills preventative interventions and their evaluations. Family relations, 45, 247- 254. Steckler, A., Goodman, R. M., McLeroy, K., Davis, S., & Koch, G. (1992 Measuring the diffusion of innovative health promotion programs. Americanjournal of health promotion, 6(3), 215-255. Szapocznik, J., Perez-Vidal, A., Brickman, A. L., Foote, F. H., Santisteban, D., Hervis, O., and Kurtines, W. M. (1988). Engaging adolescent drug abusers and the families in treatment: A strategic structural systems approach. Journal of consulting and clinical psychology, 56, 552-557. Taylor, R. J., Chatters, L. M., and Jackson, J. S. (1993). A family profile of familial 68 relations among three generation black families. Family relations, 42, 332- 341. Thompson, B. & Kinne, S. (1990). Social change theory: Applications to community health. In N. Bracht (Ed.), Health promotion at the community level (pp. 4565). Newbury Park, CA: Sage. Usher, C. (1995). Evaluating changes in systems. In Pecora, P., Fraser, M. W., Nelson, K. E., McCroskey, J., and Meezan, W. (Eds), Evaluating family- based services (pp. 17-21). New York: Aldine DeGruyter. Waldrop, D. P. and Weber, J. A. (2001). From grandparent to caregiver: the stress and satisfaction of raising grandchildren. Families in society: the journal of contemporary human services, 82 (5), 461-472. Wandersman, A., Valois, R., Ochs, L., de la Cruz, D.S., Adkins, E., Goodman, R. M. (1996). Toward a social ecology of community coalitions. American journal of health promotion, 10 (4), 299-307. Wasik, B. (1993). Staffing issues for home visiting programs. The future of children 3 (3), 144-157. Weitzman, E.A. (2000). Software and qualitative research. In N. K. Denzinand Y.S. Lincoln (Eds.), Handbook of Qualitative Research, 2 nd ed. (pp. 803-820). Thousand Oaks, CA: Sage Publications. Zaltman, G., Duncan, R., and Holbeck, J. (1973). Innovations and organizations. New York: John Wiley. 69 APPENDICES 70 APPENDIX A INFORMATION SHEET 71 INFORMATION SHEET FOR Beginning Education Early Replication Study You are invited to take part in a study of how well the Beginning Education Early (BEE) program works in the State of Alabama. This study will be conducted by Dr. Ellen Abell, Extension Specialist with the Alabama Cooperative Extension, Auburn University. This study has three goals. First, we hope to learn more about how to actually put the BEE program and other family programs in place. Second, we hope to identify factors that affect how the program runs. Third, we want to suggest a model of the factors related to how a program runs. This model will help the BEE program run more smoothly. It may also help other programs that serve families. From this study, we hope to learn how to make the program work well in your county and in other counties in the state. You were selected as a possible participant because of your participation in the BEE program. Your name has been selected from a list of families in 6 counties who have gone through the BEE program. Your name was randomly selected along with nine others from your county to share your experiences with the BEE program. If you decide to take part in this study, Jermaine Duffie, a graduate student in the Department of Human Development and Family Studies at Auburn University, will be arranging a time and place to talk with you about your thoughts on your child, school-readiness, and the BEE program. This conversation will last for about one and one-half hours and will be audio-taped. Whether you decide to take part in this study or not, the type and amount of services you receive from your county Extension office will not change. We will keep the information you share with us a secret. BEE educators and county extension agents will not hear or read what you have told Jermaine Duffie. Your decision whether or not to participate will not affect your future relations with Auburn University or the Alabama Cooperative Extension System. Information that you share will help us make the BEE program better. Your thoughts and experiences may help the BEE program be more successful in working with families in the future. What you share may be important in helping future BEE families get more out of the program. The results of this research will be published in scholarly journals. No names or identifying information will be included in these publications. As appreciation for your participation, you will receive a book for your child. 72 73 If you have any questions we invite you to ask the researchers Ellen Abell at (334) 844- 4480 or Jermaine Duffie at (334) 844-3229. We will be happy to answer all your questions. For more information regarding your rights as a subject you may contact the Office of Research Programs, Ms. Jeanna Sasser at (334) 844-5966 or Dr. Leanne Lamke at (334) 844-3231. HAVING READ THE INFORMATION PROVIDED, YOU MUST DECIDE WHETHER TO PARTICIPATE IN THIS RESEARCH PROJECT. IF YOU DECIDE TO PARTICIPATE, THE DATA YOU PROVIDE WILL SERVE AS YOUR AGREEMENT TO DO SO. THIS LETTER IS YOURS TO KEEP. 74 APPENDIX B Interview Schedule 75 BEE Program Adult Interview Guide TURN TAPE ON and read through Informed Consent as adapted for audio-taping. Before I ask you your opinions and observations, let me ask you first for some basic information about you and your family. This information, along with your opinions and observations, will of course be kept private. For this part of the interview I will turn off the tape and write down your answers. TURN TAPE OFF (Complete Bee Program Adult Subject Data Sheet.) TURN TAPE ON First, I=m going to ask you about the BEE program. We want to understand how the BEE program works for families. To do this, we not only need your opinions about the positive things that may have happened, but also we need to know about the less positive things. So please don=t hesitate to tell us these things. Your voice is important to us. ? Tell me about your family=s experiences with the BEE program. Prompts: How did you hear about it? What made you decide to enroll? What were the sessions like? Tell me about what you did. Tell me about what your child did. ? What memories do you have of the BEE program? Prompts: What about it did you enjoy? What about it did you not enjoy as much? ? Tell me about some of the lessons in the BEE program. Which ones do you remember most? Prompts: Do you remember any of the lessons the child had with the educator? Can you tell me some about what your child learned? Do you remember any of the lessons you had with the parent educator? ? Tell me about how the child educator behaved with your child. What did s/he do? How would you describe her manner? ? Tell me about how the parent educator behaved with you. What did s/he do? How would you describe her manner? ? What things did you learn while participating in the BEE program? 76 ? How do you think what you learned from the program will be good for your family? Prompts: Can you think of any ways it may have helped you? Can you think of any ways it may have helped your child? Can you think of any ways it may have helped your relationship with your child/ren? ? Can you think of any ways in which the BEE program has made things harder? The next questions I=m going to ask are about your opinions about education in this county and in your family. ? How would you describe the school system in your county? Prompts: What do you think it does well? What doesn=t it do as well? ? What was your school like when you were a child? Prompts: What did you enjoy about it? What didn=t you enjoy so much? ? In your opinion, what do people in your community think about education in general? What do you think about it? Follow-up: What do people you know think about school-readiness and education for young children? Follow-up: How do people feel about child care or Head Start or other early childhood programs? ? What do people in your neighborhood or family think about these things? ? What are things you think people need to do with young children to prepare them for school? Prompts: What do you believe you need to teach your child so that s/he goes to school ready and able to learn? What does a child who is ready for school look like? What does s/he know? What can s/he do? ? What expectations do you have for your child when s/he gets to school? 77 Prompts: How well do you expect your child to do in school? How far do you expect your child to go in school? ? What role do you expect to play in helping your child succeed in school? Prompts: Do you expect to be able to help your child with schoolwork? How else might you support your child when s/he is in school? These final questions are about the kind of support available for families in your community . ? What is there for families with young children to do in your area? Prompt: Are there special places or activities? ? Where do you turn when things aren=t going so well or you need someone to talk to? Prompts: Outside the family, where do you go to? ? Where do families in this area ago when things aren=t going so well? Prompts: Who can get things done around here? Who do people listen to? ? What are some common problems families in this area face? ? What kinds of support or services are available in this area for dealing with these problems? ? Is there anything else you would like to say about the BEE program, education, or the area you live in? 78 Appendix C Bee Adult Subject Data Sheet BEE Adult Subject Data Sheet Subject ID# ______________________ Date of first BEE lesson: ________________ Date of last BEE lesson: ________________ Number of lessons: ____________________ Graduated: _____ yes _____ no County Code: ____________________ Interviewer Initials: _______________ Date of Interview: ________________ Time of Interview: ________________ 1. How long have you lived in this area? ___ 1-5 yrs ___ 6-10 ___ 11-20 ___ over 20 (Interviewer check appropriate category, unless ethnicity is unclear, then ask.) Ethnicity: ____ african-american ___ asian american ___caucasian ___ hispanic ___other 2. Work Status: ____ unemployed, not seeking employment (inc. homemaker, student) ____ unemployed seeking employment ____ employed less than 25 hours per week ____ employed more than 25 hours 3. Counting yourself, how many people live in your household? ________ 4. (Ask the adult the following information in the table below about each of the members of her household, including herself. Write down the answers. Be sure the number of people in the table is equal to the number of people in question #3.) Relationship to BEE adult Age Gender Grade, if in school (or years completed, if not) Job description, if employed Self 4. Which child was enrolled in the BEE program? (Put a star next to the line in the table above that contains this child=s information.) 5. How many meetings did you have with the BEE educators? _____ 6. Where did your lessons take place? ___ on the BEE van ___ in my home ___ both ___ in another place: (please describe) ________________________________________ 79