QUALITY IN FAMILY CHILD CARE: THE VOICE OF THE FAMILY CHILD CARE PROVIDER 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. _____________________________________ Amy No?l Newell Certificate of Approval: ______________________ ______________________ Jennifer L. Kerpelman Ellen Abell, Chair Professor Associate Professor Human Development and Human Development and Family Studies Family Studies ______________________ ______________________ Jacquelyn Mize George T. Flowers Professor Dean Human Development and Graduate School Family Studies QUALITY IN FAMILY CHILD CARE: THE VOICE OF THE FAMILY CHILD CARE PROVIDER Amy No?l Newell 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 May 9, 2009 iii QUALITY IN FAMILY CHILD CARE: THE VOICE OF THE FAMILY CHILD CARE PROVIDER Amy No?l Newell Permission is granted to Auburn University to make copies of this thesis at its discretion, upon request of individuals or institutions at their expense. The author reserves all publication rights. _____________________________ Signature of Author _____________________________ Date of Graduation iv THESIS ABSTRACT QUALITY IN FAMILY CHILD CARE: THE VOICE OF THE FAMILY CHILD CARE PROVIDER Amy No?l Newell Master of Science, May 9, 2009 (B.A., Johns Hopkins University, 2004) 125 Typed Pages Directed by Ellen Abell The primary purposes of this study are to explore the provider perspective on quality in family child care (FCC) and patterns of relationship between the provider perspective and the research-based perspective on quality in FCC. Sixteen FCC providers from the Family Child Care Partnerships mentor-driven, quality improvement program participated in the study. Each provider completed a telephone interview during which time she described quality in her own words, prioritized components of quality derived from the research literature, and completed a self-report measure of quality. Qualitative analyses as outlined by Strauss and Corbin (1990) were used to identify themes and patterns of relationship between the provider and professional perspectives on quality. Most providers described quality using both professional and non-professional terminology. Overall, there was evidence of consistency across study measures, findings that suggest a positive relationship between the provider perspective and observed quality v (as measured by the Family Child Care Environment Rating Scale [FCCERS]). The relationship between self- and observer-reported quality was particularly noteworthy when comparing providers who had been identified as improving on the FCCERS with those who were not. Implications are discussed for research, practice, theory, and policy. vi ACKNOWLEDGEMENTS I would like to thank Dr. Ellen Abell for her generous spirit in sharing encouragement, support, wisdom, guidance, and patience throughout this process, and for enriching my experience as a graduate student beyond what I could have imagined. I have learned so much from my time working with you and appreciate the opportunities you have made available to me. I will carry these things with me on the rest of this journey. And to the other members on the committee, Dr. Jennifer Kerpelman and Dr. Jacquelyn Mize, thank you for your feedback and guidance. I would also like to thank my friends and family for their love, encouragement, prayers, support, phone calls, and patience throughout my life, and to the friends I have met in Auburn who helped make this place home. You are each so dear to me. And a special thanks to my dad and aunt Bunny for graciously and thoroughly proofreading this document. vii Style manual used: American Psychological Association Publication Manual, 5th Edition Computer software used: Microsoft Word viii TABLE OF CONTENTS LIST OF TABLES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . x LIST OF FIGURES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xi I. INTRODUCTION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 II. LITERATURE REVIEW . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11 Research-Based Definitions of Quality in Family Child Care . . . . . . . . . . . . . . . 11 Provider Definitions of Quality in Child Care . . . . . . . . . . . . . . . . . . . . . . . . . . . 16 The Family Child Care Partnerships Program . . . . . . . . . . . . . . . . . . . . . . . . . . . 22 Summary and Aims of the Current Study . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25 III. METHOD . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28 Participants . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28 Procedure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29 Measures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31 Data Analysis Plan . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39 IV. RESULTS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 42 Demographics and Child Care Quality . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 42 Caregiving Priorities . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 47 Provider Descriptions of Quality . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 49 V. DISCUSSION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .66 The Provider Perspective on Quality . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 67 Implications for Future Research and Theory . . . . . . . . . . . . . . . . . . . . . . . . . . . 75 Implications Practice and the FCCP Program . . . . . . . . . . . . . . . . . . . . . . . . . . . 77 Implications for Policy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 79 Limitations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 80 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .81 REFERENCES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 83 ix APPENDICES Appendix A ? FCCERS Items by Subscale . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 99 Appendix B ? Berkeley-Yale Telephone Interview . . . . . . . . . . . . . . . . . . . . . . . . . . . . 101 Appendix C ? Transcript of Open-Ended Responses . . . . . . . . . . . . . . . . . . . . . . . . . . . 106 Appendix D ? Open-Ended Response Coding Key . . . . . . . . . . . . . . . . . . . . . . . . . . . . 111 x LIST OF TABLES Table 1: Table of Measures of Quality in Child Care . . . . . . . . . . . . . . . . . . . . . . . . 88 Table 2: Subscales on the FCCERS and other Measures of Quality . . . . . . . . . . . . . 89 Table 3: Subscales on the FCCERS and Quality in the Scientific Literature . . . . . . 90 Table 4: Demographic Characteristics by Group, T-test and Chi-Square Analysis . . 91 Table 5: Mean FCCERS Total and Subscale Scores by Group . . . . . . . . . . . . . . . . . 92 Table 6: Individual Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 93 Table 7: Most Common Key Words . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 95 Table 8: Theme Groups and Other Study Measures . . . . . . . . . . . . . . . . . . . . . . . . . 96 xi LIST OF FIGURES Figure 1: Word Cloud of Open-Ended Responses (All Words) . . . . . . . . . . . . . . . .97 Figure 2: Word Cloud of Open-Ended Responses (Words Said at least Twice) . . . 98 1 I. INTRODUCTION Approximately 60% of children in the United States under the age of five receive some form of nonparental care (Johnson, 2005). For many families, child care is necessary to allow both parents to participate in the workforce or to allow parents to fulfill welfare requirements. There are four types of child care recognized in the United States: center-based care; family child care; family, friend, and neighbor care; and at- home care. These types of child care are categorized according to regulatory status, provider-child relationship, and/or setting. Center-based child care (CCC) offers regulated, non-relative care that is run out of a building created and designed for professional use. Family child care (FCC) is also regulated, but is based out of the provider?s own home. Family, friend, and neighbor care (FFN) is also based out of the provider?s own home but is unregulated. Finally, at-home care (AHC) is unregulated and the child is cared for in his or her own home. FCC is commonly chosen by families as an early child care and development setting. Approximately 25% of children under the age of five attend FCC at some point, spending an average of 31 hours per week there (Johnson, 2005). Parents may choose FCC because it typically costs less, allows for more schedule flexibility, and provides a more home-like atmosphere than center-based care (Bromer & Henly, 2004; Cryer, 2003; Kontos, Howes, Shin, & Galinsky, 1997). 2 There are several features that distinguish FCC from other types of child care. For example, unlike center-based programs, the FCC provider runs a small business directly out of her own home, typically with little help or supervisory support. This makes her the primary, and often only, person responsible for developing the program and caring for the children. She has the additional responsibility of carrying out activities related to running her own business. Being in a home rather than a professional building distinguishes FCC from center-based care. While center-based buildings are specifically made for child care programs, providers? homes are designed to be living spaces. Many parents who choose to have their children in FCC specifically prefer the home-like setting of FCC programs (Bromer & Henly, 2004). FCC also typically includes a wider range of ages in the same group than center-based care. Unlike FFN or at-home care, FCC is regulated in most states, meaning that there are minimum standards for the quality, safety, and provider characteristics of FCC programs. These standards vary by state, but may include nutritional standards for meals, quality of the child care environment, quantity of developmentally appropriate materials, and the providers? level of education or training. These distinguishing characteristics of FCC can pose unique challenges for providers with respect to providing high quality care. Because the provider is operating out of her own home, she must balance the space used for child care and that used for her personal life. The low-cost and flexible hours that draw parents to FCC programs mean longer hours and lower pay for FCC providers than for those who work at child care centers (Morrissey, 2007). The wide age range of children served makes designing and implementing a developmentally appropriate program more complicated for the family 3 child care provider, because it requires a breadth of child development knowledge more extensive than that held by the average child care center teacher. Research suggests that these challenges to providing high quality care in the FCC setting can be addressed through regular training on a variety of topics and professional development activities (Arnett, 1989; Fukkink & Lont, 2007; Kontos, Howes, & Galinsky, 1996; Norris, 2001; Pence & Goelman, 1991; Weaver, 2002). Indeed, FCC providers themselves have reported a desire for more training opportunities (Buell, Pfister, & Gamel-McCormick, 2002; Clarke-Stewart et al., 2002; Gable & Halliburton, 2003; Rusby, 2002). Furthermore, training may act as a moderator of stress, reducing turnover among FCC providers, which is seen as beneficial to the children in their care (Deery-Schmitt & Todd, 1995). These findings have motivated many states to require a minimum number of training hours per year in order to continue operating as a licensed FCC home. Defining Quality in Family Child Care The definition of higher quality in child care is usually based on both child outcomes and literature from the fields of child development and parenting. Researchers report that most FCC programs in the United States provide mediocre or low quality care, with only about 8-10% of programs considered to be of good or high quality (Austin, Lindauer, & Rodriguez, 1997; Kontos et al., 1997). The assessment of quality from a research-based perspective is made possible by generally agreed upon definitions and measures. Professionals and researchers typically refer to three types of quality: structural, process, and global (Cryer, 2003). 4 Structural quality includes the basic framework within which the program operates. This type of quality is comprised of features such as group size, provider-child ratio, and the education and experience of child care providers. These features of child care are easy to measure and typically are included in the standards for regulation. In FCC, better child outcomes and higher quality care are associated with licensure and/or regulation of programs (Kontos et al., 1996). Process quality refers to the aspects of the child care program that the children experience. These aspects include provider-child and child-child interactions, materials, and activities. These features can be more difficult to measure than structural features. Typically, independent observers assess process quality by closely observing the child care program for several hours. Observations can be scored based on qualitative, open- ended accounts or using a scale or checklist. Programs rated high on process quality measures rate higher on positive interaction scales and lower on negative interaction scales. They are also characterized by a variety of developmentally appropriate materials and activities. Global quality combines features of both structural and process quality. An interview or checklist may be used to assess structural measures, while more in-depth observation and interviews are used to assess the process quality features, issues related to parent-provider relations, and aspects of the structure of the child care environment. It follows that programs scoring higher on structural quality and process quality items are considered higher quality from a global perspective. The most commonly used measures to assess global quality in FCC are the Family Day Care Rating Scale (FDCRS; Harms & Clifford, 1989) and its recent revision, the Family Child Care Environment Rating Scale 5 (FCCERS; Harms, Cryer, & Clifford, 2007). The scale is completed by a trained, independent observer over the course of approximately three hours. Programs are evaluated on items such as furnishings, materials, and interactions. Scores on several subscales are averaged to create an overall ?global? assessment of the program?s quality. The definitions and measures discussed above are useful for researchers, practitioners, policymakers, and others. It is important, however, to recognize that they represent only one perspective on quality in FCC, the research-based perspective. Family child care providers also have their own working definitions of quality in child care, and it follows that these definitions may influence their priorities and personal standards of caregiving. Some researchers have attempted to collect and describe provider perspectives. For example, in a qualitative study of Vermont FCC providers, the women were more likely to align their role with that of a mother than a professional (Nelson, 1990). They emphasized their relationships and interactions with children, as well as other behaviors, in ways that were more consistent with mothering, and tended to reject a ?professional? definition of their roles. In addition to using their own definitions of quality child care to guide their caregiving practices, parents? definitions of quality in child care may also influence child care providers? caregiving behaviors. Ultimately, parents are the providers? clients and if the parents are not satisfied, the provider will not have children to care for. In a study across several different stakeholders in child care, researchers found that parents and child care providers were often in agreement on general priorities for quality in child care (Harrist, Thompson, & Norris, 2007). Whereas agency staff and policymakers placed an emphasis on training, education, and other staff characteristics, parents and caregivers 6 emphasized the needs and safety of children, as well as the importance of an acceptable work/life balance. Others have identified priorities for parents as an environment that is safe, sanitary, and supportive (Rosenthal, 1991). The unique characteristics of FCC may best suit the needs of those families that choose to use FCC for the care of their children (Bromer & Henly, 2004). For example, parents report that they consider FCC to be more relational than center-based care (Gable & Halliburton, 2003), and many FCC providers also identify the relational aspects of their role as important (Gable & Halliburton; Nelson, 1990; Taylor, Dunster, & Pollard, 1999). Attitude and behavior theory (Bentler & Speckart, 1981) emphasizes the connection between attitudes and behaviors, and how understanding one can give insight into the other. Therefore, understanding providers? definitions of quality child care is important because those definitions should be related to their behaviors. Only a few studies have examined this assumption. The Vermont-based FCC providers interviewed in Nelson?s study (1990) reported that their behaviors were consistent with how they defined their role as a FCC provider. Rosenthal (1991) also connected the child-caregiver interactions of Israeli FCC providers with the providers? beliefs about their level of influence. Finally, in a study of Israeli FCC providers, Isralowitz and Saad (1992) found that providers reported more engagement in behaviors they believed to be more important to child care. In addition to its connection to caregiving practices in the FCC setting, an understanding of the provider?s definition of quality may also be important to understanding provider behavior related to participation in training and professional development opportunities. Although research has shown that training is related to higher 7 quality in FCC (Fukkink & Lont, 2007; Kontos et al., 1996; Norris, 2001; Pence & Goelman, 1991; Weaver, 2002), providers (especially in FCC) often do not take advantage of training opportunities. One of the barriers to seeking training opportunities most often cited by child care providers is the perceived lack of relevant and/or high quality training opportunities (Gable & Halliburton, 2003; Rusby, 2002; Taylor et al., 1999). It is logical to expect that a provider?s definition of quality child care may be related to her training and learning priorities. Other obstacles to participating in professional development opportunities, especially for rural family child care providers, include the lack of availability and accessibility of training opportunities, the cost (both financial and time) required for attendance, inconvenient scheduling, lack of transportation, and the need to travel long distances to reach training centers (Bailey & Osborne, 1994; Gable & Halliburton, 2003; Rusby, 2002; Taylor et al., 1999). In addition, FCC providers have unique needs related to the features of FCC that distinguish it from other forms of child care. Topics such as business practices and parent-provider interactions are not often addressed in general child care training programs (Taylor et al.). In the year 2000, the Family Child Care Partnerships program (FCCP) began operating in Alabama specifically to address the availability, accessibility, quality, and relevance of training opportunities for Alabama?s family child care provider workforce. Funded by the Alabama Department of Human Resources, FCCP currently offers over 200 providers weekly or biweekly in-home visits by mentors trained to tailor their services to providers? specific needs. Program participation is voluntary and free of charge. 8 The primary purpose of the FCCP program is to increase quality caregiving practices and professionalism among family child care providers. With social cognitive theory as a guide (Bandura, 1986), mentors use a combination of direct instruction, modeling, and coaching strategies to help family child care providers develop competencies based on national accreditation standards. As a result of the services provided by FCCP, 53 providers have achieved these standards and been accredited by the National Association of Family Child Care (NAFCC). During FCCP?s first seven years of operation, mentors were trained to use a standardized assessment of quality designed for the family child care setting. FCCP used the FDCRS to assess a provider?s initial global quality, strengths and weaknesses; to develop goals with the provider based on this assessment; and to measure progress toward meeting those goals. Even for those providers not achieving national accreditation, analyses of mentors? quarterly assessments of providers? global quality over the duration of their program participation have indicated that a majority of providers significantly improved their child care practices (Abell, Miller, Keiley, and Ma, under review). These gains are notable; however, a continuing concern among FCCP staff is that there is still a sizable minority of providers who neither substantially improve their caregiving quality nor sustain improvements made (E. Abell, personal communication, January 2008). The purpose of the proposed research is to consider possible reasons for the disconnect between FCCP training efforts and provider quality improvement. This will be done by examining the relationships among providers? definitions and self-assessments of quality caregiving practices and mentors? assessments using a standardized measure of 9 global quality. Questions to be asked include the following: How do providers describe and prioritize quality practices? How do providers? self-reported definitions of and priorities for quality caregiving practices relate to mentors? assessments of their quality? What is the relationship between providers? self-assessments of quality and mentor- reported quality ratings of their practices? Can providers be grouped meaningfully based on patterns of relationships across these measures? This study is unique in that it will use both self-reported and observer-reported assessments of quality to better understand the relationship between priorities in child care and caregiving behaviors.. The results of this study are expected to contribute both to the practitioner and to research literatures on family child care. First, the results from this study may inform programmatic aspects of the FCCP?s quality enhancement program. Decisions about FCCP mentoring content are based on quality practice standards as outlined by the FCCERS and assessed by FCCP mentors. Understanding if and how providers? perspectives on quality may influence their receptivity to FCCP mentoring content could result in the development of alternate or revised training protocols. In addition, exploring the possibility of using provider self-assessments as an additional measure of quality could be useful in future program evaluation efforts. Second, many researchers who have contributed to the FCC literature specifically call for more research that evaluates and examines the links between training efforts, such as FCCP, and changes in provider practices. Determining how provider beliefs about quality may explain variations in the efficacy of training designed to enhance quality caregiving practices would inform future research in this area. Some questions researchers might examine include: What is common among the definitions of providers 10 who offer high quality care? Are there providers who offer high quality care without those components as a part of their definitions? Do provider definitions change over time, and under what training conditions? Are changes in the definition of quality related to changes in quality of care? 11 II. LITERATURE REVIEW This section reviews the literature related to defining and assessing quality in family child care. This review includes a discussion of both the research-based, scientific perspective and provider perspectives on quality in child care. There is also a description of the Family Child Care Partnerships program, a program that uses one-on-one mentoring to improve quality among licensed family child care providers in Alabama. Finally, this section will describe the aims of the current study. Research-Based Definitions of Quality in Family Child Care Definitions of quality in child care have been developed using research-based findings on child outcomes, literature from the fields of child development and parenting, professional input on best practices, and practical input from child care providers themselves. These definitions have been used to develop standardized measures to assess quality in child care. While there is some variation in definitions of quality within the child care literature, researchers typically recognize three types of quality: structural, process, and global. Cryer (2003, p. 37) provides a definition for structural quality based on a review of the literature. Structural quality is defined as the ?regulable? features of child care. Originally, this definition included group size, provider-child ratios, provider education, and experience. The definition has been expanded to include staff wages, turnover, and parent fees. In other words, structural quality is the framework within which child care 12 programs operate. Structural quality is usually measured by self-report questionnaires or survey data obtained from agencies. Researchers have found that, on average, higher quality child care centers are those with smaller group sizes, lower provider-child ratios, and better educated, more experienced providers (National Institute of Child Health and Human Development Early Child Care Research Network [NICHD ECCRN], 1996; Tout, Zaslow, & Berry, 2006). In FCC, researchers have found high quality to be associated with being licensed or regulated (Kontos et al., 1996) and higher levels of provider education (Tout et al.). Process quality refers to the features of child care that children actually experience (Cryer, 2003). These features include caregiver-child and child-child interactions, space and furnishings, activities, materials, and personal care routines (such as meals and toileting). Process quality is typically assessed using standardized observer reports. The most commonly used instruments for measuring process quality are the Caregiver Interaction Scale (CIS; Arnett, 1989) and the Observational Record of the Caregiving Environment (ORCE; NICHD ECCRN, 1996). The CIS was developed based on findings from parenting socialization practices and their effects on child development (Arnett, 1989; Maccoby & Martin, 1983). It is a 26-item scale assessing four types of provider-child interactions: positive, permissive, punitive, and detached. An independent observer indicates how often each interaction occurs between the caregiver and children during the observation (not at all, somewhat, quite a bit, or most of the time). Higher quality programs are those characterized by a higher scores on the positive interactions subscale and lower scores on the Punitive and Detached subscales. 13 The ORCE is another process quality measure used to describe what the child experiences in child care. The measure includes items related to caregiver responsiveness, affective states, and the interactions experienced by one of the children in her care. It uses a coded observational procedure completed by an independent observer to record both the frequency of behaviors and the quality of interactions during discrete time periods. Higher quality programs are those with higher frequency and quality scores for caregiver sensitivity, responsiveness, and fostering of exploration, and lower scores for detachment, flat affect, and intrusiveness. When researchers assess global quality in child care, they combine both structural and process quality features. Global quality measures typically are comprised of an interview with the caregiver to obtain information related to structural quality as well as an independent observer report. The observer report usually includes process quality items in addition to items related to provider-parent relations, professional development, and the child care environment. Scores calculated based on these measures represent the overall (or ?global?) quality of the child care program. Child care programs that score high on global measures of quality are safe environments that promote optimal physical, cognitive, social, and psychological development for children in their care. The scales most often used to assess the global quality of family child care programs are the Child Care Home Observation for Measurement of the Environment (CC-HOME; Caldwell & Bradley, 1984) and the Family Day Care Rating Scale (FDCRS; Harms & Clifford, 1989). The CC-HOME was adapted for use in child care homes from the Home Observation for Measurement of the Environment Inventory (HOME; Caldwell & 14 Bradley, 1984), which was originally developed to measure the quality of care provided by parents in the home. The CC-HOME inventory has six subscales, taken from the scientific literature on parenting, child development, and child outcomes: Emotional and Verbal Responsiveness, Acceptance of the Child, Organization of the Environment, Provision of Appropriate Play Materials, Involvement with the Child, and Variety in Daily Stimulation. The measure is a checklist in which the observer marks each item as either present or absent, with higher scores indicating better quality care. Higher scores on the HOME measure (especially Provision of Appropriate Play Materials) between the ages of 2.5 and four years have been found to be related to intelligence and achievement scores in first grade (Bradley & Caldwell, 1984). Another widely used measure of global quality in family child care is the Family Day Care Rating Scale (FDCRS; Harms & Clifford, 1989). The FDCRS was developed based on child care and child development literatures, as well as input from practitioners and child care providers. It is specifically designed to assess quality in family child care settings serving infants through school-aged children. The measure consists of six subscales: Space and Furnishings for Care and Learning, Basic Care, Language and Reasoning, Learning Activities, Social Development, and Adult Needs. A trained professional completes an in-depth observation during a visit to the child care site over the course of several hours. Higher scores on the subscales indicate higher quality. A global score is calculated by averaging the scores across all items. Recently, the FDCRS was revised based on information from relevant research in the field of child development, the content of similar measures, questionnaire feedback from FDCRS users, and the developers? experiences using the FDCRS. The revision was 15 renamed the Family Child Care Environment Rating Scale (FCCERS; Harms et al., 2007) to reflect the current terminology in the literature. The scoring of the FCCERS is virtually the same as the FDCRS, and the content and format are very similar. The most significant revisions to create the FCCERS were to add a Listening and Talking subscale, revise the criteria for some items based on growing knowledge, and to make scoring decisions easier for the rater. The Listening and Talking subscale includes items related to books, helping children to understand language, and helping children to use language. To aid the observer in accurately scoring each item, the authors expanded the clarification sections to include more notes, examples, and questions. Negative wording was eliminated to avoid confusion. While there have been no published studies comparing scores on the FDCRS and FCCERS, a personal communication from one of the authors indicated that scores on the FCCERS are generally lower than those on the FDCRS, reflecting the more stringent standards of the FCCERS (R. Clifford, personal communication, August 2007). Holloway and colleagues (Holloway, Kagan, Fuller, Tsou, & Carroll, 2001) developed the Berkeley-Yale Telephone Interview (BYTI), a self-report measure of quality based on the professional standard of the FDCRS. The BYTI was created to provide a measure of global quality that could be administered quickly and over the phone. The BYTI consists of 25 multiple-choice items that are related to, although not directly taken from, items on the FDCRS scale. The items on the BYTI were intentionally worded so that answers reflecting lower quality are not easily recognized, thus decreasing responses based on social desirability. Overall, the authors concluded that the BYTI was a valid and reliable measure of quality. 16 Summary High quality child care from a research-based perspective is defined as an environment and caregiving practices that support optimal child development. Standards of quality draw from the child development and parenting literatures. Measures of quality depend in large part on ratings by trained observers. Providers? professional development activities are often included in measures of quality, as they have been found to relate to the quality of the child care program. High quality programs are defined as those in which children are involved in a variety of engaging activities, have access to a wide variety of developmentally appropriate materials, and enjoy generally positive interactions with their caregivers and other children. While there are many ways to measure quality in FCC, the FCCERS is the most widely used measure. It is especially valuable to research because it has been recently updated to include recent advances in knowledge of how different aspects of care affect children. Provider Definitions of Quality in Child Care While the research-based perspective on quality in child care is important, it is not the only perspective. Child care providers also have their own working definitions of quality that may include child development in addition to other concerns. Attitude and behavior theory says that attitudes directly affect behavior (Bentler & Speckart, 1981). Applying this theory to family child care, one would expect providers? own attitudes toward quality to directly affect their behaviors. While no research studies published to date have directly examined this link, there are several studies that offer insight into the relationship between provider attitudes and their behaviors. 17 In a sample of 48 family child care providers in Israel, Isralowitz and Saad (1992) examined the relationship between the reported priorities of FCC providers and the attention they gave to those same tasks. Providers were given a list of 13 job-related tasks (for example, Education Enrichment Planning, Parent Relations) and asked to report how important they were on a scale of 1-5, with lower scores indicating higher priority. They were also asked to rate their level of behavior associated with each priority on a scale of 1-5, with lower scores indicating more behavior. By ranking tasks in order of the average provider ratings, highest priorities were usually those tasks that were also associated with the highest levels of behavioral involvement. This supports predictions based on attitude and behavior theory, showing that providers? priorities (?attitudes?) were related to their behaviors. This study also revealed that, among this sample of Israeli FCC providers, education enrichment planning and parent relations were top priorities, while relations with other professionals and budget management were ranked the lowest. A major limitation of this study was the lack of an objective measure of provider behaviors. Sole reliance on self-report can be problematic. Rosenthal (1991) examined the relationship between caregiver-child interactions and providers? beliefs about their level of influence. The sample included 41 sponsored day care homes in Israel. A time-sampling technique was used to record interactions between the child and caregiver. Researchers also conducted structured interviews with caregivers to determine how much influence over the children?s development the providers believed they had. It was found that those caregivers who believed they had a lot of influence over children?s social development interacted with children in a more 18 positive way and spent less time in group interaction than those providers who believed they had less influence over the children?s social development. A study by Nelson (1990) gives insight into how FCC providers construct their roles as caregivers. The author completed extensive interviews with 70 FCC providers in Vermont. These providers were generally representative of Vermont FCC providers in that most were married women with an average age of 34.5 years, most had children of their own, and about half cared for their own children in their day care programs. The interviews consisted primarily of open-ended questions and follow up questions were tailored to each participant. For example, one question was ?What do you want to be offering to the children in your care?? (p. 589). Providers? perspectives about a wide range of topics related to FCC were sought in these interviews. Several common themes arose related to providers? constructions of their roles as family child care providers. Most providers aligned their roles with mothering as opposed to professional practice. They stressed the importance that the children in their care should feel ?at home? in the provider?s home. Providers also emphasized their relationships and interactions with children, as well as other behaviors associated with being a mother, such as completing housework. Providers said that over time, they had developed strong feelings toward the children in their care, comparable to feelings toward their own children. They also reported that their caregiving behaviors were consistent with how they defined their roles as child care providers. Although they tended to ignore the professional side of their roles in response to the initial questions, upon further questioning and prompts, many providers did recognize that limited responsibility for the children in their care, limited authority, and the 19 financial realities of their role were several aspects that distinguished child care from mothering. In order to effectively deal with those aspects of caregiving that differed from mothering, providers reported practicing ?detached attachment? with the children in their care. That is, they had to control the level of emotional involvement they engaged in with the children in their care. Approximately one decade later, Gable and Hansen (2001) conducted a study that assessed providers? opinions about the content and type of training and education necessary for providing quality child care. Twenty-five center directors, 19 center providers, and 26 family child care providers from suburban and urban areas participated in focus groups. Individuals were asked to identify the three to five types of training or education they considered the most important with respect to providing quality child care. Participants were also asked what level of training and education they believed child care workers needed. Responses were coded based on 15 a priori defined categories. These categories spanned a wide range of topics, including administration, developmentally appropriate practices, personal attributes (such as patience), and professionalism. There was evidence of convergence across groups as well as role-specific priorities. For example, all participants rated training or education in child development as one of their ?Top 3-5? priorities. However, while both directors and FCC providers identified training in administration as important, no center providers included this topic. On the other hand, whereas both center and FCC providers identified health, safety, and nutrition as a priority, directors did not select this topic. The top priority for center providers was level of education, while only three directors and three FCC providers selected this topic as one of their priorities. 20 One very recent study sought to understand what different stakeholders? opinions were of the components of quality in child care (Harrist et al., 2007). Randomly selected child care center directors and owners (n = 19), parents (n = 27), child caregivers (n = 30), policymakers (n = 16), and social service providers (n = 8) participated in 11 focus groups discussing that which they believed to be important for quality in child care. Six components of quality were consistently identified across groups: 1. Communication and rapport: positive working relationships and emotional climate. 2. Caregiver practices: behavior, work habits, and attitudes of caregivers. 3. Staff characteristics: such as training, education, child-caregiver ratios, and turnover. 4. Finances and resources: funds and costs for equipment and supplies. 5. Visibility and involvement: salience of program in the community and of stakeholders in the program. 6. Professionalism: provider and public perceptions of child care as a valued profession. These findings are consistent with those of Gable and Hansen (2001), who found a significant amount of agreement across directors and providers, and that those aspects valued by directors and providers were generally consistent with a professional definition of quality. The study also found some important differences across groups. Caregiver definitions of quality were very similar to parents? definitions and tended to focus on interactions and developmentally appropriate activities. In contrast, policymakers and 21 social service professionals focused on staff characteristics (e.g., training, turnover, and ratios) and visibility and involvement (e.g., parent involvement). Although the literature related to child care givers? perspectives on quality is very limited, taken together, these studies do give insights that may help us better understand the provider perspective on quality. We know that child care providers (especially FCC providers) tend to align their roles with parenting, and that the quality definitions offered by parents and providers are very similar (Harrist et al., 2007; Nelson, 1990). This is consistent with the nature of their work and the fact that, ultimately, parents are the providers? clients. As business owners, it is important for FCC providers to provide programs based on the needs and expectations of their clients (Bromer & Henly, 2004; Gable & Halliburton, 2003; Harrist et al., 2007; Shlay, Tran, Weinraub, & Harmon, 2005). Child care providers serve not only parents, however, and (as with the research- based perspective) take into consideration the experiences of the children in their care. Gable and Hansen (2001) showed that the educational priorities of child care providers are consistent with important components of a professional definition of quality. In addition, the top priorities identified by Israeli FCC providers (Isralowitz & Saad, 1992) are similar to those identified by caregivers, policymakers, and social agency professionals in the United States (Harrist et al., 2007). Specifically, these priorities include an emphasis on educational activities as well as relations with parents. Although these studies do not directly ask providers how they define quality child care, they do help us to understand the perspective caregivers have on quality and their priorities in caregiving. 22 Summary Research has shown that structural features of quality (e.g., licensure) and process features of quality (e.g., provider-child interactions) are related to global quality. To date, however, there are no known published studies that have examined how providers? own definitions relate to their global quality as reported by an observer using a standardized measure. This neglect is problematic ?because providers are the link between children?s experiences, their beliefs are a critical part of designing effective educational initiatives and policies? (Gable and Hansen, 2001, p. 40). A provider?s perspective on quality may give unique insight into why she does what she does (Isralowitz & Saad, 1992; Nelson, 1990; Rosenthal, 1991), and may have implications for training and quality improvement. A better understanding of the provider perspective and how it relates to the research- based perspective on quality is important and has implications for training, practice, policymaking, and interventions. The Family Child Care Partnerships Program In the year 2000, the State of Alabama raised its licensing standards to require a minimum of 20 training hours per year for FCC providers. With a significant portion of the FCC workforce located in rural areas of Alabama, it was anticipated that many of these providers would have difficulty complying with the increased minimum of 20 training hours per year. FCC providers in rural areas, in particular, face significant barriers to training including inconvenient scheduling, lack of transportation, cost for attendance, and the long distances required to reach training centers (Bailey & Osborne, 1995; Gable & Halliburton, 2003; Rusby, 2002; Taylor et al., 1999). The Family Child Care Partnerships program (FCCP) was funded by the Alabama Department of Human 23 Resources to increase the availability and quality of training for FCC providers in order to help them meet the new requirements. The FCCP program was designed to provide quality training and, in addition, had as its goal to advance the quality of family child care practices to national accreditation level standards, as outlined by National Association for Family Child Care (NAFCC). The primary method used by FCCP to increase quality and professionalism among FCC providers is the delivery of one-on-one training from mentors during weekly or bi-weekly in-home visits that occur during the providers? normal operating hours. Mentors use modeling and coaching strategies to help family child care providers develop competencies based on national accreditation standards. Mentors use the FCCERS (Harms et al., 2007) to assess initial quality, strengths, and weaknesses; to develop goals with the provider based on this assessment; and to measure progress toward meeting those goals. It is expected that the process of professional socialization will help FCC providers internalize professional standards of quality. As providers align their standards of caregiving with those identified by researchers and professionals in the field of early childhood as those most beneficial for children?s development, their behavior may change accordingly, thereby increasing the quality of their child care practices. The mentoring relationship is also expected to increase provider receptivity to training. Providers have identified a lack of respect from other professionals as one barrier to their participation in training (Taylor et al., 1999); in the FCCP program, mentors are considered ?quality partners? rather than experts. In addition to quality improvement through the mentoring relationship, the FCCP program offers support for additional professional development activities. FCCP mentors 24 help providers connect with other child care professionals and opportunities for professional development. For example, mentors may help FCC providers start provider organizations in their communities. Mentors also make providers aware of upcoming professional development opportunities. The FCCP program also addresses financial barriers related to improving quality and obtaining accreditation. Participating providers are given a stipend to use toward materials for their child care programs and FCCP covers the cost of the process of accreditation through NAFCC. Through mentoring relationships, additional professional development opportunities, and financial assistance, the FCCP program seeks to improve quality, increase professionalism, and encourage the pursuit of national accreditation for family child care providers in Alabama. In fact, many child care providers participating in FCCP do improve the quality of their child care. For example, upon FCCP program entry, 8% of providers scored at the highest level of overall quality (i.e., a total FDCRS score between 6 and 7). This is in contrast to the 21% who were performing at the highest level upon their departure from the program (Abell et al., under review). Upon FCCP program entry, 40% of providers were rated as engaging in minimal or inadequate quality practices. At program departure, those rated as engaging in less than adequate quality practices dropped to 26% (Abell, et al.). While this drop is desirable, still fully a quarter of FCCP providers failed to adjust their caregiving practices to what is professionally considered adequate quality. 25 Summary and Aims of the Current Study In summary, much research has been done to determine the components of high quality child care. There are several standardized, research-based measures available that assess the degree to which child care practices meet quality standards. These measures tend to focus on the specific behaviors of child care providers, how the child care environment is set up, and the materials and experiences children have in the child care program. It is also known that training is an important part of professional development and can help providers increase the quality of the care they offer. Many of the providers in the FCCP program have improved their quality, although others have shown little or no improvement. Few studies have been done that give insight into providers? perspectives on quality. Generally speaking, a provider?s understanding of quality in child care reflects influences from personal experience, parents, and research-based standards. It has also been shown that those behaviors providers believe to be important are related to their self-reported level of involvement in those behaviors (Isralowitz & Saad, 1992). Rosenthal (1991) found that providers? beliefs about their influence over children were related to caregiver-child interactions. None of these studies, however, examined the relationship between providers? definitions of quality and observed quality as assessed by standardized, research-based measures. The current study addresses four main research questions. The first question is ?What is the provider perspective on quality in FCC?? Providers were asked an open- ended question about quality in FCC, allowing them to define it in their own words. Qualitative methods were used to identify themes and analyze the responses. To give 26 further insight into providers? priorities and to bridge the gap between provider- and research-based definitions of child care quality, providers also prioritized research-based standards of quality care. These two indicators comprised the ?provider perspective.? The second question this study addresses is ?What patterns of relationship exist between the provider perspective and observed quality?? Observed quality (as measured by the provider?s mentor using the FCCERS) was compared to the provider perspective to identify patterns of relationship. Essentially, this comparison looked for patterns that suggested how the provider perspective was reflected (or not) in the quality behaviors providers engaged in, as observed by their mentors. The third question this study addresses is ?What is the relationship between the provider?s self-report of child care quality, as measured by the BYTI (Holloway et al., 2001) and a mentor?s observations of her quality, as measured by the FCCERS (Harms et al., 2007)?? Programmatically, understanding this relationship is important in order to determine if the BYTI still functions as a useful, additional measure of quality when compared to the FCCERS, a similar but more stringent standard than the FDCRS (on which the BYTI is based). It is also possible that there are reasons other than measurement issues to consider when examining the strength of the relationship between the BYTI and the FCCERS with a sample of FCCP providers. Because the BYTI is a reflection of the provider?s subjective experience of the quality she offers, other factors affecting that subjective experience (for example, the provider?s own definition of quality and whether or not she believes she measures up to that standard) may be influencing the relationship. 27 Finally, this study addresses the question ?Are any patterns of relationship addressed by this study different for providers who are improving versus those who are not improving on quality (as measured by the FCCERS)?? Distinguishing improving and not improving providers is especially relevant for the FCCP program and other quality improvement programs. Identifying different patterns of relationship for those providers who improve and those who do not may inform practitioners and give insight into how programs may be tailored to be more effective with providers who do not improve over time. The answer to this question may also help to improve understanding of the patterns identified by the first three questions. For example, a provider?s own definition of quality and whether or not she believes she measures up to that standard may be influencing the relationships between other study measures. The results of this study are then discussed in terms of how they fit into the current body of literature on quality in FCC and future directions for research. Implications for theory, practice, and policy are also discussed. 28 III. METHOD This section describes the participants in the study, the process used to complete the study with each participant, the measures used, and the data analytic techniques that were employed. Participants Participants were a subset of the 208 family child care (FCC) providers who were currently participating in the Family Child Care Partnerships (FCCP) program at the time of selection. Providers enroll in FCCP voluntarily to receive regular in-home mentoring and training, and they live throughout the state of Alabama. It is assumed that a key reason that providers participate in the program is to improve the quality of the care they offer and to receive guidance and support in the pursuit of national accreditation through National Association for Family Child Care (NAFCC). In addition, participation in the FCCP program is one way to fulfill the state licensing requirement of completing 20 training hours per year. One of the aims of the current study was first to identify providers whose quality was improving (as measured by the Family Child Care Environment Rating Scale [FCCERS]) and those who were not improving. Out of all 208 providers, 136 were identified as potential participants in the current study because they had been in the FCCP program for at least one year and had at least two complete FDCRS and/or FCCERS evaluations from their FCCP mentor over the past year. Of those providers, 101 were 29 considered improving because their most current FCCERS scores showed improvement of at least 0.5 point over past FCCERS total scores. Attempts were made to reach all of these providers by phone, and 36 were successfully contacted. Of those 36 providers, 13 agreed to be in the study and eight of them completed the interview and returned the consent form. Thirty-five of the 136 potential participants identified were not improving on quality, that is, their most recent total FCCERS scores were within 0.5 point of past scores or they showed evidence of decline over time. Several attempts were made to reach all of these providers by phone. Seventeen were successfully contacted, and 12 of them agreed to participate in the study. Nine completed the interview, but one failed to return her consent form after repeated opportunities and reminders to do so. This process yielded a total of 16 useable interviews, with half of them from FCCP providers who were improving on quality, and half from FCCP providers who were not improving. Procedure After identifying a subset of potential participants from all FCCP providers, providers were contacted over the phone by the principal investigator. If it was a convenient time to talk, the investigator described the study and provided an opportunity for the provider to ask questions. If the provider was unable to talk, a better time was agreed upon and she was called back at that time. Providers were told that if they participated, they would be entered into a drawing to win one of three $50 gift certificates to Lakeshore Learning, a company that sells developmentally appropriate materials for children. If a provider indicated she was interested in participating in the study or would 30 like to think about it further, she was mailed the IRB-approved consent form and the caregiving priorities form (see Measures for a description). An appointment was made for approximately one week later for the investigator to call the provider and complete the interview. Interviews were conducted by calling the provider at the appointment time. If it was convenient to talk, the investigator proceeded with the interview, if not, a new appointment time was agreed upon. During the first part of the interview, the investigator reviewed the consent form with the provider and gave her the opportunity to ask questions. After giving consent to participate in the study, the provider was asked to sign and date the consent form and set it aside. Providers were then asked to answer the open- ended question about quality in family child care. Responses were recorded verbatim at the time of the interview by the principal investigator using a word processing program. Following the open-ended question, providers were instructed to open the caregiving priorities card. The investigator then instructed the provider on how to fill out the card, creating an ordered list of caregiving priorities. The provider was then asked to place both the signed consent form and the caregiving priorities card in the return envelope and mail it to the principal investigator. Then the BYTI was completed, with the investigator recording the provider?s responses directly onto the form. Providers were then asked if they had any questions, reminded to return the study materials by mail, and told that they would be entered in the drawing when the study materials were received. All interviews took approximately half an hour. Eight of the 25 providers who said they were interested in participating in the study completed the interview at the time scheduled during the initial contact. The other 31 providers were more difficult to reach, and up to five attempts were made at different times of the day and on different days of the week to contact these providers. Eight providers completed the interviews following additional contact. Nine of the providers still had not completed the study after five additional attempts to contact them following the initial phone contact. If the study materials were not received within one week of the interview for those who completed the study, providers were contacted and reminded to mail them back. If the provider needed a new copy, it was sent to her in the mail along with a return envelope, instructions, and contact information should she have any questions. There were a total of 16 providers who completed the interview and returned all study materials, including the consent form. Measures Demographic Data Demographic data for the study sample were already on file. The demographic variables of interest were age, years of experience in family child care, time in the FCCP program, ethnicity, level of education, annual household income, and annual child care income. Age, years of experience in FCC, and time in the FCCP program were continuous variables. Ethnicity, level of education, annual household income, and annual child care income were categorical variables. Ethnicity was collapsed into two categories: Caucasian and minority. Level of education had three possible options: high school or GED, some college but no degree, and post-secondary degree. There were four options for level of household income: less than $20,000; $20,001-$40,000; $40,001-$60,000; and greater than $60,000. Annual income from FCC alone had three categories: less than $20,000; $20,001-$40,000; and greater than $40,000. 32 Assessments of Caregiving Quality Self reported quality. The Berkeley-Yale Telephone Interview (BYTI; Holloway et al., 2001) is a self-report measure of quality in child care. This measure was included for two purposes. The primary reason was to evaluate whether the BYTI would be a useful addition to the measures the FCCP program currently collects from participating providers. The second reason was to examine the relationship between the BYTI and the FCCERS. The BYTI was developed by Holloway and colleagues (2001) to address the need for a measure of quality that could be administered over the phone, quickly, and with little training. The measure is composed of 25 multiple-choice items that are related to, but not directly taken from, the FDCRS. The items cover space and furnishings, basic care, language-reasoning experiences, learning activities, social interactions, and parents and staff. The items on the BYTI were carefully worded so that answers reflecting lower quality were not easily recognized, thus decreasing responses based on social desirability. The creators administered the scale to a sample of 89 family child care providers who also had FDCRS scores. Chronbach?s alpha was reported as .78 for the original sample, indicating adequate reliability. The creators regressed the FDCRS total score on provider background characteristics (group size, education, experience, training, and membership in a professional organization) and the BYTI items. Adding the BYTI items significantly improved the model. The BYTI explained 49% of the variance in FDCRS scores. The authors also completed a discriminant function analysis using the BYTI to predict categorical quality based on the FDCRS (poor, adequate, or good). This revealed that the BYTI could categorize providers based on their FDCRS score with 92% 33 accuracy. Overall, the authors concluded that the BYTI was a reliable and valid measure of quality in family child care. In the current sample of 16 FCC providers, Chronbach?s alpha for the BYTI was low, reaching a maximum of .54 after excluding two items (minutes spent reading to the children and satisfaction with space). According to the reliability analyses, the reliability would not be increased by the inclusion or exclusion of any other items. Observer reported quality. The Family Child Care Environment Rating Scale (FCCERS; Harms et al., 2007) is used to measure the quality of different aspects of caregiving. It is the newest version of the FDCRS, the most widely used measure of FCC quality in research. The FCCERS was included in this study in order to have an observer rating of quality in addition to the self reported rating (BYTI). A secondary purpose for inclusion was to examine the relationship between the FCCERS and the BYTI. The FCCERS is completed in the provider?s home by a trained observer over the course of three to four hours. The measure has 38 items that are divided into seven subscales (Space and Furnishings, Personal Care Routines, Listening and Talking, Activities, Interaction, Program Structure, and Parents and Provider). Each of the 38 items on the FCCERS is scored on a scale of one to seven, with higher scores indicating higher quality, and lower scores indicating lower quality. In order to score each item, there are several indicators which are marked as either met or unmet. The indicators correspond to four scores: one, three, five, and seven. In order for an item to receive a certain score, all the indicators of one score, in addition to those of any lower score(s), must be met. For example, to receive a score of five on any item, all of the indicators for one, three, and five must be met. If all of the indicators for the lower score(s) are met and 34 at least half (but not all) of the indicators for the next higher score are met, then the even number between the lower and higher numbers is given as the score. The average of the items in each subscale yields a subscale score. The average across all items yields the FCCERS total score. In order to score high on the Space and Furnishings subscale (?Space/Furnishings?), providers must have a wide variety of child-appropriate furniture in good condition, and ?keep the children in mind? when organizing the space (e.g. colorful pictures on the walls are kept at children?s eye level). Programs scoring high on the Personal Care Routines subscale (?Personal Care?) follow stringent standards of hygiene and safety, and routine care (e.g. diapering and meal times) is pleasant and appropriate for the children in the care environment. A high score on the Listening and Talking subscale (?Listening/Talking?) indicates that children have many opportunities throughout the day to hear and use language with others and in different contexts. This subscale also covers the accessibility children have to books and literacy experiences (e.g., being read to by an adult). In order to score high on the ?Activities? subscale, providers must have a wide variety of materials that are accessible to children throughout the day. Children must have a significant amount of choice and independence in choosing what to play with, when, and how long to enjoy it. The provider should be involved in, but not controlling of, play. A high score on the ?Interaction? subscale indicates that the provider designs the program so that children have many opportunities to interact with the provider and the other children, and that most of these interactions are pleasant. This subscale also covers tone of voice and how conflicts are handled by the provider. 35 A high score on the ?Program Structure? subscale indicates that the children are allowed significant independence in choosing activities. Children should not be required to participate in group activities, and alternative materials/activities are available for children who choose not to participate. A high score on the Parents and Provider subscale (?Parents/Provider?) indicates that the provider has made herself and relevant information (in the form of articles, fact sheets, etc.) available to parents. She is also involved in professional development activities beyond those currently required in Alabama?s minimum standards for licensure. In the FCCP program, mentors were trained by the Managing Director to 85% interrater reliability on the FCCERS, in accordance with the guidelines described by the authors of the measure. After reviewing how to use the measure and score each item, several practice items were completed. Mentors then scheduled and completed as many in-home FCCERS observations as necessary to achieve 85% reliability with the Managing Director. Once reliability is achieved, the mentors complete FCCERS observations of new providers within one month of their enrollment in the program and twice per year thereafter. In the current study, the FCCERS total and subscale scores that were used for analysis were from the most recent mentor-completed FCCERS available for all of the providers in the study (completed approximately six months prior to data collection for the current study). For the purposes of this study, mentor-reported assessments of quality are reported and discussed in terms of categories of quality, rather than simply in terms of raw scores. Quality categories (e.g., Austin et al., 1997; Campbell & Milbourne, 2005; Holloway et al., 2001; Kontos, Howes, & Galinsky, 1996) are defined as follows, 36 Inadequate: average total FCCERS and average subscale scores of less than 3; Minimal: 3 to 3.9; Adequate: 4 to 4.9; Good: 5 to 5.9; and Excellent: 6-7. Assessments of Caregiving Priorities The caregiving priorities measure consisted of two parts: (1) the provider?s response to an open-ended question, and (2) the provider?s ranking of components of quality in terms of their importance. Open ended question. The purpose of the open-ended caregiving question was to collect the providers? descriptions of quality in their own words. The question and prompts were derived in consultation with the Executive Director of the Family Child Care Partnerships program (E. Abell, personal communication, August 2008). The providers were given the following request: ?In a few sentences, please describe to me what quality child care means to you.? Additional prompts were used if the interviewer determined that the provider?s first response was too vague for the purposes the study. These prompts were ?How do you make sure?? or ?What do you mean by?? For example, if a provider answered that qualit y child care meant that children are safe and loved, the prompts were: ?How do you make sure children are safe?? or ?What do you mean by ?the children are loved??? Responses were recorded verbatim using a word processing program. Ranking priorities. Following the open-ended portion of the interview, providers were asked to rank a list of statements about quality child care in terms of their relative importance. The inclusion of the ranking task served two purposes. First, it ensured that the caregiving priorities measure was not simply a test of how well the provider could articulate her beliefs about quality in child care. It is possible that for providers who have 37 not had much experience expressing their views, the open-ended task would be difficult. The ranking task gave the providers an opportunity to elaborate on the priorities they may or may not have communicated in the open-ended question in words more similar to those expressed in the professional literature. Second, this measure acted as a bridge between provider- and research-based definitions of quality. It used concepts from a research-based perspective, but allowed the provider to rank them in a way that was meaningful to her. The list of components of quality was constructed for the purposes of the current study from a review of standardized measures of quality, state regulatory standards, and the scientific literature on quality in family child care. Specifically, the resources included the minimum standards for licensing by Alabama?s Department of Human Resources, the most widely-used standardized measures of quality in family child care (the FCCERS, the ORCE, and the CC-HOME), and components identified by four peer- reviewed studies that are relevant to quality in FCC (Gable & Hansen, 2001; Harrist et al., 2007; Isralowitz & Saad, 1992; Nelson, 1990). Table 2 lists the components of quality identified by measures of quality compared to the FCCERS subscales. Table 3 lists the components of quality identified by the peer-reviewed studies compared to the FCCERS subscales. Examining the components identified by each resource yielded seven general components of quality in family child care. Six mentors in the Family Child Care Partnerships program provided feedback about the relevance of the components as well as the wording, and changes were made based on this input. Thus, a review of the literature by the author of this study, in consultation with an expert in the field of family child care (E. Abell, personal 38 communication, July 2008), and feedback from mentors with direct experience providing technical assistance to family child care providers yielded a list of seven components of quality in family child care. The final version of the list of components of quality was: 1. Children are listened to and talked with often. (?Listening/Talking?) 2. Children?s learning is stimulated in planned activities. (?Activities?) 3. Provider is responsive and discipline is sensitive to each child?s needs. (?Responsive/Discipline?) 4. Providers and children engage in healthful behaviors and practices. (?Healthful Behaviors?) 5. Providers seek out professional development and networking. (?Professional Development?) 6. Providers adjust their care practices to parents? needs. (?Parents? Needs?) 7. Space and furnishings are appropriate for children. (?Space/Furnishings?) To administer this measure, providers were given the following directions: Please open your Caregiving Priorities card. There you see a list of seven statements that most people agree are all an important part of quality child care. Let me read them aloud to you as you read them on the card. Of those statements, which do you believe is the most important for quality in child care? Please place a one next to that statement. Now, there are six statements left that do not have a number next to them. Which do you believe is the most important of those statements? Please place a two next to that statement. There are now five statements left. Of those five, which do you believe is the least important? Please place a seven next to that statement. Of the remaining four statements, which do 39 you believe is the least important? Please place a six next to that item. There are now three statements remaining. Please place a three next to the one that is the most important, a five next to the least important of those three statements, and a four next to the remaining statement. This process yielded a list of quality statements that were ranked in order of relative importance, from one to seven. In order to determine the relative priorities of groups of providers, an average priority ranking was calculated. This was done by first assigning a value to each priority equivalent to the rank given by the provider (for example, if the provider ranked Listening/Talking as her top priority, it was assigned a value of one; if she ranked Space/Furnishings as least important, it was assigned a value of seven). For each priority, these values were aggregated across providers and an average rank was computed. This yielded a list of priority means that could be ordered numerically from most important to least important (the most important priority had the lowest mean, and so on). Data Analysis Plan Quantitative Data Analysis Quantitative data analysis methods were used to analyze the demographics and FCCERS scores of the study sample and compare them to all FCCP providers as a whole. For continuous variables (i.e., age, experience, and FCCERS scores), t-tests were run to compare means between all current FCCP providers and the study sample. For categorical variables (i.e., ethnicity, educational attainment, household income, and FCC income), chi-square tests were run to determine if the groups differed. 40 Quantitative data analysis methods were also used to calculate the study sample?s priorities as a group, and to compare the FCCERS and the BYTI. Because the sample was small, a simple correlation was calculated to determine the relationship between these two measures of quality. Qualitative Data Analysis Because the sample was small, a qualitative method was the primary approach to address the research questions of the study. Qualitative data were analyzed in consultation with the Executive Director of the Family Child Care Partnerships program. In order to understand how providers in the study described quality in FCC, Strauss and Corbin?s (1990) guide to qualitative research methods was employed. Open coding was used to identify all key words, defined as any ?instance of phenomena? that could be labeled (Strauss & Corbin, 1990, p. 61). The percent of responses that included each key word was recorded. The key words were used to create a ?word cloud? as a visual representation of their frequency. In a word cloud, the font size changes relative to the frequency with which each word is used (i.e., more frequently used words appear larger and vice versa). Software to generate a word cloud was found on the internet, www.wordle.net. Each instance of the key words were copied and pasted onto the website, and then a word cloud was generated from that information using a random format. The tools allowed one to change the orientation, font, and color of the words in the word cloud. The word cloud could then be printed directly from the website. During the next phase of analysis, axial coding and constant comparison strategies were used to identify themes based on the key words. The goal was to identify themes based on related key words that could be used to describe and compare the providers? 41 responses to each other, without losing too much of the richness and complexity of the individual responses. The strategy of constantly comparing the new data to that which were already analyzed meant that the definitions of the themes evolved throughout the process. The final version of the coding key was established based on the data from all responses, and all responses were re-evaluated based on this final version. Theme groups were then examined in light of other study variables. To establish interrater reliability, the Executive Director of the FCCP program coded responses for themes using the final version of the coding key. Interrater reliability between the Executive Director and the author of this study was 98.36%. The one discrepancy was discussed and resolved. 42 IV. RESULTS The following chapter describes the results of the study. The first section characterizes the study sample based on their demographics, performance on the observer-rated Family Child Care Environment Rating Scale (FCCERS), and self- reported child care quality on the Berkeley-Yale Telephone Interview (BYTI). A closer look is taken at those providers identified as improving and those identified as not improving on the FCCERS. The second section describes the results that are related to the research question ?How did the family child care (FCC) providers in this study prioritize different components of quality?? This section also includes a closer examination of providers whose priorities differed from the sample as a whole. The final section addresses the primary goal of this study: to describe how these FCC providers defined quality in their own words. The key words used by the study sample in their open-ended responses are described. Themes are defined and theme groups are described in terms of the other study measures (demographics, FCCERS, BYTI, and priorities). Demographics and Child Care Quality Demographics Table 4 shows the demographic information for all currently enrolled FCCP providers and the study sample. The providers in the study were in large part very similar 43 to all current FCCP providers on demographic measures. Study participants were an average of 50 years old, with a range of 35-76 years. On average, they had approximately 12 years of experience as FCC providers, with a range of 2-20 years. T-tests comparing these averages to all FCCP providers were not significant, indicating that the study sample did not differ statistically from all FCCP providers on age or experience. The educational attainment of the study sample showed that one-fourth had only a high school diploma or GED, approximately half completed some college coursework but had not received a degree, and about one-fourth had a post-secondary degree. Approximately one-fourth of the study sample had a household annual income of $40,000 or less, while half earned between $40,001 and $60,000, and 21% earned more than $60,000 annually. Reported annual income from FCC was less than $40,000 for most (93%) of those reporting, with about 43% earning less than $20,000 annually. Six of the providers reported that their FCC income accounted for most or all of their household income. With respect to ethnicity, 75% of the sample were Caucasian and 25% were African American. In contrast, approximately half of all FCCP providers are Caucasian and half identify themselves as belonging to a minority group. The chi-square analysis was significant for ethnicity (p < .05), indicating that the study sample was less diverse than FCCP providers as a whole. Chi-square analyses for educational attainment, household income, and FCC income were not significant, indicating that the study sample did not differ statistically from all FCCP providers on those measures. 44 Observer-Reported Quality The FCCERS was included in this study to obtain a measure of observer-rated child care quality for the providers in the interviewed sample. Scores on this measure were used to address two objectives of the current study: (1) to determine the relationship between observer-reported quality (as measured by the FCCERS) and self-reported quality (as measured by the BYTI), and (2) to examine the relationship between observed child care quality and providers? self-reported caregiving priorities. Table 5 shows the average FCCERS total and subscale scores for all currently enrolled FCCP providers (n = 197) and the study sample. The mean total FCCERS score for all current FCCP providers as well as those in the interviewed subsample was in the adequate range (4.42 and 4.75, respectively). On average, the study sample provided adequate care on most subscales: Space/Furnishings, Personal Care, Activities, and Program Structure. They provided a good level of quality care in the areas of Listening/Speaking and Parents/Provider. Differences between all FCCP providers and the interviewed subsample were found on the Activities and Listening/Speaking subscales. On the Activities scale, the average level of quality for all FCCP providers was minimal, whereas the study sample offered care in the adequate range. On the Listening/Speaking subscale, FCCP providers as a whole offered an adequate level of care, whereas the study sample average was in the good range. Because these differences represented a categorical difference between the groups, they are considered ?observable differences? (Campbell & Milbourne, 2005; Kontos et al., 1996). This level of difference indicates that an independent observer would be able to see the difference between the higher and lower quality care. The 45 difference between groups on Listening/Speaking trended toward statistical significance (p = .07). Perhaps those providers who scored higher on the listening and speaking subscale were also more apt to participate in a research study that required a half hour phone interview. One of the original objectives of the study was to identify providers who were improving or not improving on the FCCERS in order to compare them on study measures. While half of the study sample was identified as improving and the other half was not improving, it was later determined that the not improving group actually consisted of two types of providers: those who had high scores on the FCCERS (in the good or excellent range) and those who had lower FCCERS scores. A ceiling effect may be influencing the high scorers because their initial quality scores started high, leaving littler room for improvement, whereas the low scorers have room for improvement but have not been improving. There are four providers in the not improving, high scoring group (NI-H) and four in the not improving, low scoring group (NI-L). While these numbers are too small to make meaningful quantitative comparisons between groups, whenever meaningful patterns do seem to emerge, they are discussed qualitatively. Provider-Reported Quality The BYTI was included in this study to obtain a measure of self-reported child care quality. This measure was used to address three objectives of the current study: (1) to examine the relationship between observer-reported quality (as measured by the FCCERS) and self-reported quality (as measured by the BYTI), (2) to examine the relationship between providers? self-reported child care quality and caregiving priorities, 46 and (3) to provide evidence that may be used to evaluate the usefulness of including the BYTI as an additional measure of child care quality in the FCCP program. Scores on the BYTI ranged from 44 to 67, with higher scores indicating higher self-reported quality. BYTI scores are located in the final column of Table 6 which lists the results for individual providers on this and other key study variables. The average BYTI score was 55.25, the range was 44 to 67, and the standard deviation was 5.30. Although the BYTI had low reliability in this sample (Cronbach?s alpha = .54), it was moderately correlated with the FCCERS (r = .63, p < .01). This indicates a moderate overlap between observed quality ratings and self-reported ratings. For example, as seen in Table 6, Provider A had the highest score on the FCCERS and also the highest score on the BYTI; Provider G had the lowest score on the FCCERS and the lowest score on the BYTI. There were, however, a few exceptions. Providers C, D, and I ranked in the lower half of observed quality scores, but were in the upper half of self-reported quality scores. In fact, the observed quality score for Provider D ranked her 15th out 16 on the FCCERS, yet her score on the BYTI was the third highest in the sample. Conversely, Providers E and F ranked in the higher half of the observed quality scores, but their BYTI scores were in the lower quality half of the sample. For example, Provider E?s observer quality score (FCCERS) ranked her third in this sample, but her self-reported quality was ranked at 10th overall. A fairly consistent pattern is seen in the relationship between observed quality rankings) and self-reported quality rankings for the Improving and Not Improving groups. Seven of the eight providers whose FCCERS scores were improving had BYTI 47 scores that were ranked in the top half relative to the other providers, with five of these providers ranking themselves higher relative to the sample than their observer-reported quality scores. This was true for improving providers regardless of how high their observer-rated quality scores were. Conversely, seven of the eight Not Improving providers ranked themselves in the lower half relative to the other providers. Only one of the providers who was not improving ranked herself higher on the BYTI than the FCCERS, and six of the eight providers who were not improving (regardless of low or high scores on the FCCERS) ranked themselves lower relative to the sample than their observer-reported quality scores. In other words, all but one of the Improving providers put themselves in the top half relative to the other providers in the study, whereas the scores of all but one of the Not Improving providers put them in the lower half relative to the other providers in this study. This seems to suggest a positive relationship between the providers? subjective experience of the quality she offers and her progress as evaluated by her mentor using the FCCERS. This association is noteworthy in light of the fact that these providers have been participating in a quality improvement program and receive regular feedback about their progress. Caregiving Priorities Another primary aim of this study was to understand how FCC providers prioritized different components of quality in FCC. Each provider ranked a list of seven caregiving priorities in terms of her belief in which were the most important for quality. Most providers commented during the phone interview that this task was very difficult because they thought all of the components were so important for quality. 48 As a group, the study sample ranked listening to and talking with children as the most important of the seven components of quality. All 16 providers listed that component as one of the top three most important. The other top priorities were Healthful Behaviors (top three for 10 [62.5%] of the providers), Responsiveness/Discipline (top three for nine [56.25%] of the providers), and Activities (top three for eight [50%] of the providers). Conversely, adjusting to parents? needs ranked the lowest, with all 16 providers placing it in the bottom three relative to the other components of quality. The other lower priorities were Professional Development (bottom three for 12 [75%] of the providers) and Space and Furnishings (bottom three for nine [62.50%] of the providers).For the sample as a whole, the average relative rankings of priorities were (high to low): Listening/Speaking, Healthful Behaviors, Responsiveness/Discipline, Activities, Space/Furnishings, Professional, and Parents. Not all providers, however, followed this general pattern. These rankings show some consistency and some inconsistency between provider-reported priorities and observer-rated quality. For example, the Listening/Speaking and Interaction FCCERS subscale means were in the good range for the sample as a whole. These subscales corresponded to the Listening/Speaking and Responsiveness/Discipline priorities, both of which were ranked in the top three priorities. The Parents/Provider subscale mean, however was also in the good range, but Parents and Professional Development was ranked by providers as the least important component of quality. 49 Provider Descriptions of Quality Key Words Following Strauss and Corbin?s (1990) guidelines for qualitative data analysis, open-ended responses were first analyzed by identifying key words used by providers to describe what quality in child care means to them. Table 7 lists the percent of responses that included each of the most common key words and phrases. Figure 1 is a word cloud in which the frequency of key words in the provider responses is represented by the size of the font (words mentioned more often are larger and vice versa). Figure 2 is another word cloud that includes only those key words that were mentioned more than once. All of the providers used some form of the word ?child? in their open ended responses. Different forms of the words love, safe, and teach were each mentioned by seven different providers (43.75% of the sample). Six different providers (37.50%) listed specific basic needs (such as food or diapering), and different forms of the words care, talk, and parent were also mentioned by six different providers. Meeting needs, caring for others? children like they were their own, and teaching morals were each mentioned by five providers (31.25%). School, a homelike environment, activities, and standards or professionalism were each mentioned by four providers (25.00%). Most of the responses included more than one key word. The open-ended response section was the first quality-related task of the interview. In light of this fact, it is important to highlight that many of the key words the providers produced on their own, without being prompted or primed with professional verbiage, were closely related to the FCCERS subscales and the priorities created for this study that were derived from research-based sources. For example, key words related to 50 the environment could be found on the Space/Furnishings subscale of the FCCERS, and key words related to basic needs could be found on the Personal Care subscales. These concepts were also related to the priorities of Space/Furnishings and Healthful Behaviors. These and other relationships between FCCERS subscales, priorities, and relevant key words are shown in Table 8. While many of the key words could be found in the FCCERS and priorities, not all of the key words mentioned by the providers were related to the research-based definition of quality. For example, many of the concepts providers mentioned (e.g. love, warmth) would be difficult to operationalize and are not used in the FCCERS. There was also a noticeable absence of a few important concepts from the FCCERS. Aside from soothing physical touch, none of the providers mentioned any other key words that would correspond to the behaviors on the Interaction subscale (e.g. children?s interactions with each other, tone of voice). All but one provider failed to mention anything related to encouraging child-directed exploration of the environment or the concept of accessibility of materials. These concepts are emphasized on half of the FCCERS items (Space/Furnishings, Activities, and Program Structure subscales). Only one provider mentioned any key words related to the Program Structure subscale. Although she did not go into detail about her concept of the ideal program structure, her score on this subscale (7.00) indicated that her idea of a structured environment was consistent with the FCCERS standard. Overall, the provider perspective offered by the providers reflected a combination of many concepts similar to the research-based perspective as well as elements that were more specific to FCC. Figures 1 and 2 show that the safety and needs of the children are 51 an important part of quality, in addition to teaching and the learning environment. These concepts are all consistent with the research-based definition of quality in child care. The FCC provider perspective also included concepts that are not typically emphasized by the research-based perspective, but that were particular to FCC. For example, many providers expressed desire for the child care environment to be ?like a home? and serving parents as a central concern. This combination of professional elements and concepts specific to FCC was what one might expect from providers participating in a quality enhancement professional development program designed especially for FCC providers. Themes The next step in analyzing the open-ended responses was to group the key words into themes. By grouping related key words, six general themes emerged. Dividing providers based on the themes mentioned in their responses generated six overlapping groups (any provider who mentioned more than one theme was present in more than one group). All but one of the providers mentioned more than one theme. On average, a provider?s response included the mention of about three to four different themes (M = 3.44). Table 8 characterizes each theme group in terms of years experience, years in the FCCP program, FCCERS total score, the score of the FCCERS subscale that corresponds to the theme group, the priorities that correspond to the theme group, and the priority rankings. Custodial care needs. Custodial Care Needs (?Custodial?) was a common theme. Found in 13 (81.25%) of the responses, this theme included any form of the general term ?basic needs? as well as referring to specific needs (e.g. toileting, nutrition), any form of the word safety, using the word ?environment,? and any reference to the organization of the 52 child care environment. For example, the first sentence of one of the responses addressed the idea of meeting the children?s basic needs, ?Making sure that they are comfortable, fed, given proper nutrition, and any other needs are being met.? The following excerpt was from another provider who focused more on the environment, ?Someone who can provide a safe, loving, structured environment for the children.? The importance of Custodial Care for these providers was reflected in their priority rankings. The priorities related to theme (Healthful Behaviors and Space/Furnishings) were both ranked higher compared to the sample as a whole, while Activities (not related to the theme) was ranked lower compared to the sample. Custodial Care was not reflected in the FCCERS scores of the women who mentioned it. The providers, on average, offered an adequate level of quality on the FCCERS subscales related to the theme (Space/Furnishings and Personal Care), whereas quality was in the good range on unrelated subscales. Emotional Climate. Another common theme was Emotional Climate. This theme was mentioned in 10 (62.50%) of the responses and included the concepts of love, warmth, affection, nurturing, and soothing physical touch. The word ?care? was also included in this theme if it was used in a nurturing sense, as opposed to basic care needs. The following is an excerpt from the response of one of the providers in this group, ?? sing songs about how they are loved, activities that emphasize love and kindness.? The importance of the Emotional Climate was reflected in these providers? caregiving priorities, which emphasized the ?caring? aspects of quality over the teaching aspects. These providers ranked Responsiveness/Discipline as the second most important component of quality (higher than the sample and all theme groups with the exception of 53 Professionalism). The importance of the Emotional Climate was also reflected in the FCCERS scores of these providers. The mean level of quality offered on the Interaction subscale of the FCCERS was good, one level of quality higher than the total FCCERS (adequate). Engagement. Engagement was mentioned in nine (56.25%) of the responses and included words such as talking to, communicating with, being involved with, quality time, and playing with the children. One provider put it this way, ?Playing with them, involved, not just putting them in a room and letting them play all day.? The importance of Engagement was generally reflected in providers? priority rankings. Listening/Talking was ranked as the most important component of quality and it was the most closely related to the theme of engagement. On the other hand, among these providers, Engagement was not reflected in the corresponding FCCERS subscale. The mean level of quality offered by these providers on the Listening/Speaking subscale was adequate, one level of quality lower than the sample and all other theme groups. Parents and Families. The theme of Parents and Families (?Parents/Families?) was also mentioned by nine providers (56.25%). Responses in this theme mentioned parents or families either implicitly or explicitly. Many of the providers said it was important to them to know they were meeting the parents? needs. Here is an example in which parents were implied, ?Having someone you can trust and depend on to care for your child.? It is clear that parents were the subject of this sentence, even though the provider did not mention them explicitly. The priorities for these providers in this group (who specifically mentioned parents or families as a part of quality in family child care) did not reflect the importance 54 of Parents mentioned in the open-ended responses of these providers. Like the sample as a whole, these providers ranked parents as the least important component for quality. The importance of parents was reflected in their FCCERS scores. They offered an adequate level of quality overall and on all subscales with the exception of Parent/Provider (good). In addition, they scored lower than the sample as a whole on the subscales that are more child-centered (Listening/Speaking and Interaction). Teaching and School. Teaching and School (?Teach/School?) was mentioned in eight (50%) of the responses. This theme included words such as teach, school, teachable moment, learning, train, education, and the mention of specific learning activities (e.g. crafts, reading). One example from this group included several of these concepts, ?Someone who can teach children in a fun, informative way. Get them ready for kindergarten, but be flexible enough if a child shows interest in something to take advantage of a teachable moment.? The importance of Teaching/School was reflected in neither the priorities nor the FCCERS performance of those providers who mentioned this theme in their open-ended responses. The providers who mentioned teaching, learning, or school as important for quality did not rank learning activities any higher than the sample overall (fourth). In addition, scores on the Activities and Program Structure subscales were both in the adequate range (same as the total), while unrelated scales were in the good range. Professionalism. Finally, the theme of Professionalism was present in five (31.25%) of the responses. This theme included the concepts of being (a) professional, being more than a babysitter, standards, and any mention of special knowledge (e.g. child 55 development). For example, one provider put it this way, ?Being more professional so I can do my job better with the children.? The providers who mentioned Professionalism in their open-ended responses scored in the good range on the total FCCERS, and in the good or excellent range on all subscales. The quality of care offered by most of these providers was at least one level higher overall than all providers in the FCCP program (on average) across all subscales. Compared to the interviewed sample as a whole, the care they offered was one level higher on all subscales except for Listening/Speaking (both the sample and the Professionalism theme group scored in the good range). They also scored in the good range on Space/Furnishings, Activities, and Program Structure. Their quality was in the excellent range on Personal Care, Interaction, and Parent/Provider. The importance of professionalism was reflected in the priorities of those providers who mentioned it in their open-ended responses. Although (on average) these providers ranked Professional Development fifth out of the seven priorities, relatively speaking, this was higher than the sample as a whole. The FCCERS scores for the providers who mentioned Professionalism also demonstrated consistency with the theme in two ways. First, these providers scored high in all areas of quality as measured by the professional, research-based standard. Second, they also scored at the highest level on the Parent/Provider subscale, which includes items related to professional development. Overall, there was some consistency apparent between the providers? open-ended descriptions of quality, their priorities, and observed quality. All but one of the theme groups (Teaching/School) showed consistency between their descriptions and at least one other area (priorities and/or observed quality). As a group, those providers who 56 mentioned concepts related to the Emotional Climate or Professionalism showed consistency in both areas. Themes and Higher vs. Lower FCCERS Scorers The relationship between observed quality and the providers? own definitions of quality was also examined by comparing the themes mentioned by those whose total FCCERS scores were in the good to excellent range versus those whose scores were in the adequate, minimal, or inadequate range. Dividing the sample in this way, exactly half (eight) of the providers scored in the good to excellent range, while the other half scored in the adequate, minimal, or inadequate range. The average number of themes mentioned by the lower quality group was 3.75, while the average in the higher quality group was 3.13. A similarly high percentage of both lower and higher scorers mentioned the theme of Custodial Care in their responses (87.50% and 75.00%, respectively). In addition, approximately half of both the lower and higher scorers mentioned the themes of Parents/Family and Teaching/School in their responses. This finding indicates that for the providers in this study, these areas did not distinguish those who offered higher or lower quality care. There were also some differences apparent between the groups. Six of the lower scorers mentioned Emotional Climate and Engagement in their responses. In contrast, only half of the higher scorers mentioned Emotional Climate, and only three of them mentioned Engagement. Another striking difference was the Professionalism theme. While half of the higher scorers mentioned this theme, only one of the lower scorers mentioned it in their responses. 57 While the average number of themes mentioned by each provider was 3.44, the number of themes included in the response did not distinguish providers who offered higher or lower observer-reported quality. There was one provider who mentioned two themes and one provider who only mentioned one theme (Professionalism). Both of these providers had an average FCCERS score in the good range. Here is the response from the provider whose open-ended description contained only the theme of Professionalism: Striving for the best possible standards in child care nationally. Whatever the professional, the other home care providers, as well as those who have experience and education in child development have gathered as the standard, that is what we should strive for. The highest standard. In contrast, two providers mentioned five of the six themes and one provider mentioned all six themes. Their FCCERS scores ranged from minimal to excellent, indicating that being able to articulate many aspects of quality care does not necessarily mean that the highest level of quality care is being achieved. Here is the response of the provider who mentioned all six themes and scored in the adequate range on the FCCERS: Meeting the needs of the parents and children in a homelike environment. Going above and beyond the minimum standards. Being compassionate and loving to the children. Address their needs. Like if they get a booboo, reassure them, love them. Teach them to be fair in their play. All my [FCC] kids are just like my grandkids. Hold them, listen to them, read to them. These findings suggest that, although the number of themes mentioned and, specifically, the themes of Custodial Care, Parents/Families, and Teaching/School did not differentiate between higher and lower observer-rated quality, some aspects of the 58 provider perspective on quality did. Lower scorers were more likely to include concepts related to Emotional Climate and Engagement in their responses, while higher scorers were more likely to include Professionalism. Themes, Priorities, and BYTI. One common inconsistency between the providers? open-ended descriptions of quality and their priority rankings was about parents. While the theme Parents/Families was mentioned in half of the open-ended responses as important for quality, all providers ranked it as one of the three least important components of quality. It could be that this inconsistency shows the difference between what providers consider important to quality ?in theory? (based on knowledge of child development, children?s needs, nurturance) and parents as clients (meeting parents? needs). One way to better understand this disparity between providers? descriptions of quality in which the needs of parents are often preeminent and a low ranking of Parents? Needs relative to other components of quality is to look at the item on the BYTI which asked providers how often the children in their care used worksheets. While parents often want to see children coming home with worksheets as evidence of their learning (E. Miller, personal communication, August 6, 2008), the child development literature indicates that this is not the most effective way children learn. For this reason, FCCP mentors advise providers not to use worksheets. Did the providers in this study limit the use of worksheets, putting other important components of quality first (as their priority rankings might suggest)? Or did they continue to use worksheets because, as they reported in their open ended responses, parent satisfaction was an important part of quality? The latter answer is supported by 59 providers in this sample. Twelve of the providers (75.00%) reported using worksheets daily in their program, and only four of the providers (25.00%) said worksheets were not used. Only one of the four providers who did not use worksheets mentioned parents or families in their descriptions of quality (?Do the best job I can taking care of others? children.?) On the other hand, eight of the 12 providers who reported using worksheets often (66.67%) did mention parents and families in their open-ended responses about quality in child care. First Theme Groups While dividing the participants based on all themes mentioned in their response was useful for analyzing the richness and variety of responses offered, the groups were not mutually exclusive; one provider was in all of the groups, and most were in at least two different theme groups. Therefore a simplified approach was used in order to determine if any additional patterns of relationships could be seen. Mutually exclusive groups were created by dividing providers based on the first quality theme mentioned in their open-ended responses. Fourteen of the sixteen providers in the study could be divided into three groups based on the first theme mentioned. Table 6 organizes providers by First Theme group and shows their individual results on caregiving priorities, improving FCCERS status, years in the FCCP program, FCCERS total score (category, score, and rank), and BYTI total score and rank. First theme: Parents. For seven of the providers, parents or families was the first theme mentioned in their descriptions of quality in FCC. This could be done either implicitly or explicitly. For example, one provider started out by saying, ?It means a lot 60 to me to know that I am helping families,? while another provider said, ?Having someone you can trust and depend on to care for your child.? The women in this sample were an average of 54.50 years old, with 11 years of experience in FCC, and approximately five years in the FCCP program (range: 1-7 years). One provider chose not to provide this demographic information. This group was similar to the whole sample on demographic measures on education and income. As a group, these providers had an average FCCERS score in the adequate range on all subscales. Individual FCCERS performance ranged widely, however, with two providers offering inadequate care, two offering adequate care, and three providers who offered care in the good to excellent range. The FCCERS scores did demonstrate consistency with the theme, as the Parent/Provider subscale score was higher than the FCCERS total for all but one of the providers in this group. The priorities of this group differed somewhat from the sample as a whole. For example, Provider G is in this group, the only provider in the sample to rank Professionalism as the most important component of quality. She, however, also ranked the lowest of all providers on both observer-reported and self-reported quality, and she was among those who had not shown improvement. One might speculate that this discrepancy in her stated caregiving priority and her actual quality was because she is one of the newest providers to the FCCP program, and thus, still digesting the changes required to achieve observable change on the FCCERS. Alternatively, it may be that her prioritizing professionalism reflects an aspiration to which she has, as yet, been unwilling or unable to match her behaviors. 61 Also in the Parents first theme group are the only four providers who ranked Healthful Behaviors as the most important component. Three of these four offered good to excellent quality care. The only three providers in the sample who ranked Listening/Talking as third most important (as compared to first or second for all other providers) were also in this group. It may be that even though Parents was ranked last, the way these providers prioritized the other components was a reflection of what parents want. For example, while Listening/Talking is important, few parents would leave their children in an environment where they had a lot of interaction but the basic health and safety needs were not covered. First Theme: Safety. For four of the providers in the sample, safety was the first theme mentioned in their descriptions of quality in FCC. The ideas expressed were either related to children?s safety in general, or the importance of a safe environment. The women in this sample were an average of 47 years old, with a little more than 13 years in FCC, and approximately 6.5 years in the FCCP program. The women in this group were similar to the sample as a whole on measures of education and income. As a group, their average FCCERS total was in the adequate range. Three of the providers in this group scored in the minimal range on the total FCCERS, while one scored in the good range. The priorities of this group were similar to the sample overall, and show both consistency and inconsistency with the first theme mentioned. For example, the only two providers who ranked Space/Furnishings as the first or second most important component of quality were in this group. The Space/Furnishings subscale contains many of the safety items on the FCCERS. In contrast, even though the Personal Care FCCERS subscale and 62 Healthful Behaviors priority contain the other safety-related items, this group ranked Healthful Behaviors third, fourth, or fifth, while the study sample as a whole ranked it second. For the three providers in this group who offered a minimal level of quality, it may be that because they were offering a minimal level of care in the areas on the subscales that included items related to safety (Space/Furnishings and Personal Care), these providers have received a lot of guidance from their mentors in the area of safety. And it may be that because these providers were likely to be hearing so much about safety from their mentors, safety was the first concrete idea they mentioned when answering the open-ended question about quality in child care. First Theme: Professionalism. Three providers first mentioned concepts related to the theme of Professionalism when they described quality in child care. This theme was defined as any mention of the word professional, the idea of being a professional (e.g. ?more than a babysitter?), or standards (e.g., ?striving for the best possible standards in child care?). The average age of the providers in this group was 55 years old with and average of 18 years of experience in FCC and 3.17 years in the FCCP program. The women were similar to the sample as a whole on education and income. Each of the providers in this group performed in the good to excellent range on the total FCCERS and all subscales. The priorities for this group were similar to the sample overall, and demonstrated some consistency with the theme of Professionalism. While the sample as a whole ranked Professional Development as the sixth most important component of quality, the Professionalism first theme group ranked it third, fourth, or fifth. 63 First Theme: Children?s Learning. The provider who first mentioned children?s learning was a 42 year old woman who had been a FCC provider for four years and participating in the FCCP program for one year. That she was relatively new to FCC and FCCP was particularly interesting in light of the level of quality she offered (minimal), and her open-ended response: Someone who cared about the child?s learning. I think it would be someone who loves and cares for the children. I think their education is very important, their minds. When you present something [to the children] you can see them learning? [The provider?s] personality, how they work with the children. The way they get along with them and interact with them. How the children respond: they will show affection. While the first quality theme mentioned in her open-ended response was about children?s learning, she did not include aspects of quality that are scored on the FCCERS, those which the FCCP program tries to impart to the providers who participate (such as child- centered learning, exploration, specific learning activities, and materials). In addition, the FCCP program mentors do not emphasize the provider?s personality or children showing affection toward the provider in response to her care as measures of quality. It is also interesting to note that this provider scored particularly low on the FCCERS subscales related to an optimal learning environment: inadequate on both Activities (2.40) and Program Structure (1.67). To get a high score, providers must have a variety of materials available to children that the children can freely choose to play with, and there must be alternative activities available to children who choose not to participate in group activities. So for this provider, it seems that while children?s learning may be her 64 priority, the way she structures her environment is different from what researchers have found to be an optimal learning environment. First Theme: Warm, Welcoming Home. The provider who first mentioned a warm and welcoming home was a 43 year old woman who had been a provider for 14 years and in the FCCP program for six years. She scored in the excellent range on the FCCERS (6.25) and her open-ended response touched on several different themes of quality in FCC: [A] warm and welcoming home, like a child?s home with school touches. Someone who will be there the whole time the child is a preschooler. Good healthy meals, a lot of reading, a lot of music and fun activities, cooking and craft activities, a lot of love and touching. When [the children] come in the morning they are happy to be here. The ones that are more timid are held, told they will be ok. We sing songs about how they are loved, activities that emphasize love and kindness. Treat them like my grandchildren. In this response, the provider mentioned the environment, activities, interactions with the children, basic needs, love, and treating the children like her grandchildren. She received the highest possible score (7.00) on both the Interaction and Space/Furnishings subscales. Even her lowest scores were in the good range (Program Structure [5.67] and Activities [5.80]), and one category of quality higher than the study sample (adequate on both). Generally speaking, these results indicate that providers? initial theme defining quality tends to show consistency with their priorities and/or observed quality. Although individual contradictions are evident, when looking at the theme groups, most demonstrated consistency in at least one area (priorities or observed quality). These 65 consistencies were not always as expected, however. For example, the Parents priority was typically ranked as the least important component of quality, yet many providers mentioned Parents as the first theme and scored high on the Parent/Provider FCCERS subscale. Consistency was even more evident when comparing themes to observed quality. For example, most of the providers who mentioned Parents as the first theme had Parent/Provider subscale scores that fell into a quality category higher than that of their total FCCERS score. Perhaps more striking, all of the providers who mentioned Professionalism as their first theme offered good to excellent quality, and the average FCCERS score for the Professionalism theme group was also in the good range. In summary, these results demonstrate that components of the provider perspective on quality (in the form of an open-ended description of quality, priority rankings of components of quality, and a self-reported evaluation of quality), as understood through an examination of provider responses grouped by key quality themes, showed some patterns of association. The patterns suggest that providers? stated understanding about what quality means can be seen as generally consistent with their prioritization of its multiple caregiving dimensions and the level of their observer- reported quality. Furthermore, an examination of the ranking of providers according to their observed and self-reported quality in the context of their status as improving versus not improving pointed out the possibility that a provider?s subjective experience of the quality she offers is related to her mentor?s evaluation of her progress. 66 V. DISCUSSION This chapter discusses the major findings of the current study in light of what is already known from the literature. Implications for practice and policy are also discussed, as well as the limitations of the study and future directions for research. The primary aim of this study was to explore the provider perspective on quality in family child care (FCC). Providers who participated in the study defined quality in their own words. The themes of custodial care, the emotional climate, engagement, teaching, professionalism, and serving parents and families were common. Additionally, providers prioritized several components of FCC practice identified in the literature as important for quality. Talking with children, healthful behaviors, sensitive discipline practices, and learning activities were the top priorities overall. Relatively speaking, the least important components of quality were space/furnishings, professionalism, and finally, parents? needs. Another primary aim of this study was to identify meaningful patterns of relationship between the provider perspective on quality and quality measured by an observer using the Family Child Care Environment Rating Scale (FCCERS; Harms, Cryer, & Clifford, 2007). In general, there was evidence of consistency between providers? descriptions of quality, their priorities, and/or observed quality. While the number of themes included in the provider response was not related to the level of 67 quality, providers offering lower quality were more likely to mention the Emotional Climate and/or Engagement, while those who offered higher quality were more likely to mention Professionalism. Finally, this study sought to examine patterns of relationship among quality definitions, self-reported quality, and observer-reported quality for those providers who had been identified as improving on the FCCERS and those who were not improving. It was found that those who were improving on the FCCERS typically rated themselves in the upper half of the sample on quality, while those who were not improving typically rated themselves in the lower half of the sample. This was true regardless of the actual observed level of quality. The Provider Perspective on Quality The definitions of quality offered by the providers in this study reflected several of the elements that have been found in previous studies. Like those providers in Nelson?s study (1990), many of the providers in this study incorporated concepts of mothering and a homelike environment into their descriptions of quality. Many providers in this study also embraced their role as professionals, however, a finding that is more consistent with recent studies in which providers discussed FCC in more professional terms (Gable & Hansen, 2001; Harrist, Thompson, & Norris, 2007). In this study, some providers specifically mentioned professionalism, and all providers included at least one component of quality derived from the professional literature without any priming from the interviewer. This finding may be an example of the concept ?professional socialization,? which the FCCP program aims to impart to participating providers through their 68 relationships with trained mentors. However, it is important to keep in mind that providers select themselves into the FCCP quality improvement program and may already embrace the professional part of their role to some extent. The inclusion of both providers? open-ended descriptions of quality and their rankings of components of quality in terms of their relative importance as the ?provider perspective? extended what has been done in other studies. Other studies included only the provider?s words (Nelson, 1990) or only the researcher?s words (Gable & Hansen, 2001; Holloway et al., 2001; Isralowitz & Saad, 1992). In this study, the combination led to a richer understanding of the provider perspective than if only one technique had been used. For example, based on the open-ended response, it was clear that serving parents was important to many of the providers in this study. In contrast, the Parents component of quality was typically ranked last by providers. This apparent inconsistency provides more information about the provider perspective than either measure would have on its own. One explanation, for example, is that the other priorities represent the provider?s attempt to take care of the parents? needs by providing a safe and interactive learning environment for the children. Another possible explanation for the apparent inconsistency may be that it represents an example of the tension between the theory of what is important for quality in child care and the reality of being a FCC provider. Child care providers are entrepreneurs. They are small business owners, dependent on their income from FCC, and they are responsible to their clients (i.e., parents) who may have specific demands. The demands of the clients may at times conflict with what the provider (as a professional) knows is optimal for the children in her care. 69 This tension is illustrated by the high proportion of providers in this study who reported using worksheets on a daily basis. While mentors in the FCCP program are trained to discourage the use of worksheets and instead to help the provider use alternative, more developmentally appropriate, learning activities with children, only 25% of the providers in this study reported that they never used worksheets and the rest reported using them on a daily basis. While providers were not asked to explain their use of worksheets, many commented that the parents wanted to see their children coming home with worksheets. Another possible source for the discrepancy comes from the nature of the task itself. Requiring providers to rank the components of quality relative to each other, rather than asking them to rate how important they were, meant that some of the priorities would fall to the bottom. Most providers shared that the priority task was difficult for them because all of the components on the list were so important. The Provider Perspective and Observed Quality Attitude and Behavior Theory (Bentler & Speckert, 1991) emphasizes the connection between attitudes and behavior, and how understanding one can give insight into the other. One study indirectly examined this relationship with Israeli FCC providers using self-reports (Isralowitz & Saad, 1992). Researchers found that those behaviors FCC providers rated as the most important were also the ones they invested more time in doing. The findings from the current study provide some additional support and extend that research by including observer ratings of quality. In this study, there was evidence of consistency between the provider perspective on quality and observer-reported quality. Overall, the providers were rated especially 70 high by their mentors on the Listening/Speaking and Interaction subscales, those that were most closely related to two of the top three priorities (Listening/Talking and Interaction). After dividing providers into groups based on themes, all but one of the theme groups showed consistency with the priorities and/or observed quality. In addition, the correlation between self-reported quality (BYTI) and observer-reported quality was moderate (r = .63, p < .01), meaning that there was approximately 40% overlap between the measures. Not all of the priorities were reflected in observer-reports of quality. While Healthful Behaviors was ranked second, scores on the Personal Care were in the adequate range. This apparent inconsistency may be due, in part, to the stringent nature of the FCCERS subscale assessing these behaviors. The requirements for the items on the Personal Care subscale are based on recommendations by the Centers for Disease Control. In order to score high on this scale, providers must adhere to very specific standards virtually all of the time the children are in their care. As an example of how difficult it can be to meet the criteria, in order to score in the minimal range, providers must complete all twelve steps in the diapering process. If any one step is not performed correctly in 50% of the diapering observed, the score for the item is automatically a one. Given these strict standards, the fact that providers score in the adequate range on the subscale may actually reflect a high level of attention to Personal Care Routines. On the other end of the priorities scale, while Professionalism and Parents? Needs ranked near the bottom of the priorities, providers (on average) scored in the good range on the Parent/Provider subscale. One possible reason for this apparent discrepancy may be that the subscale is not reflective of the kinds of things providers mean when they talk 71 about parents and professionalism as important. For one thing, the FCCERS Parent/Provider subscale is a combination of the two distinct concepts. There is only one item that is related to the children?s parents, and it primarily addresses information- sharing between the provider and parents. Because parents are clearly an important part of quality for many FCC providers, the measure used may not have adequately assessed relevant behaviors. The most consistent relationship between the provider perspective on quality and observer-rated quality was found among those who mentioned concepts related to Professionalism. The results from providers in the current study are consistent with other studies that have shown participation in professional development activities to be important for quality in FCC (Clarke-Stewart et al., 2002; DeBord & Sawyers, 1996; Fukkink & Lont, 2007; Kontos, Howes, & Galinsky, 1996; Norris, 2001; Pence & Goelman, 1991; Weaver, 2002). While other studies have examined the relationship between involvement in professional development opportunities and quality, this study found that among a group of providers involved in a quality improvement program, those who specifically mentioned professionalism as important for quality offered good to excellent quality in all areas. Quality ranged from inadequate to excellent among those providers who did not specifically mention Professionalism in their descriptions of quality. This finding brings up a question about the relationship between professionalism and quality. Can a provider offer high quality care even if she articulates her definition of quality in a way that is inconsistent with the professional definition of quality? The results from this study suggest that it is possible to do so. For example, Provider F?s 72 description of quality emphasized custodial care and mothering, yet the level of quality she offered as observed by her mentor was in the good range overall: Having someone you can trust and depend on to care for your child. Making sure that they are comfortable, fed, given proper nutrition, any other needs are being met. That the kids are happy and being loved like a mother would if she were home to do it. Being talked to and played with like a mother would. This provider?s priorities and FCCERS scores were also largely consistent with her description of quality, regardless of the apparent lack of ?professional? concepts. Improving versus Not Improving on Observed Quality One of the original aims of the study was to compare those providers who were improving on the FCCERS with those who were not improving. Those comparisons generally did not yield many differences. One reason for this may be that the ?not improving? group as originally defined was not a single cohesive group, but rather, made up of those who scored high and those who scored low on the FCCERS. Those providers who scored high on the FCCERS did not have a lot of room for improvement, and it is possible that a ceiling effect was responsible for their lack of improvement. Nevertheless, the qualitative examination of improving status, the BYTI, and the FCCERS did reveal a noteworthy pattern. Seven of the eight providers who were not improving on the FCCERS (regardless of the observed level of quality they offered) rated themselves in the bottom half relative to the other providers in the study, whereas seven of the eight providers who were improving on the FCCERS rated themselves in the top half relative to the other providers in the study (regardless of the observed level or quality they offered). This finding may be an example of the effect of the provider?s subjective 73 experience on her self-report. All of the providers in this study have been participating in a quality improvement program in which they receive regular feedback from a mentor about their progress. Based on these results, it seems that the providers are aware of their progress and that this awareness effects how they evaluate their own quality. This effect should be kept in mind when using self-reports with FCC providers who are regularly evaluated on quality. While the BYTI may reflect the provider?s subjective experience, this finding does bring up the question of the validity of the BYTI itself as a measure of quality. In this study, the BYTI was moderately correlated with the FCCERS, and there was approximately 40% overlap between the two measures (r = .63; r2 = .40). That means that 60% of the score on the BYTI is not predicted by the provider?s score on the FCCERS. The BYTI was originally developed based on the FDCRS, and while the FDCRS is similar to the FCCERS in many ways, the differences make the BYTI less like the FCCERS than its predecessor. One of these differences is the concept of ?accessibility.? Accessibility on the FCCERS means that children have unrestricted access to that item or feature for at least an hour a day. The concept of accessibility is part of the scoring for more than one-third of the items on the FCCERS, yet is absent on the BYTI. Because of this discrepancy, a provider could accurately rate herself high on an item on the BYTI (for example, that she has many books for children to read), yet receive a 1 from an observer on the FCCERS (if the children are not given access to those many books for an hour a day). Another difference between the BYTI and the FCCERS that contributes to the lack of overlap between the two measures of quality is the actual items. Six of the 25 74 items on the BYTI are not a part of the FCCERS. For example, the BYTI asks providers how satisfied they are with the space they have available for child care. Half of the items on the FCCERS are not mentioned in any form on the BYTI, neither as individual items nor combined with other items. Most of these items come from the Personal Care Routines, Program Structure, and Parent/Provider subscales. This means that BYTI and FCCERS are more similar in some areas then others. For example, the Space/Furnishings subscale on the FCCERS is well represented on the BYTI. Does that mean that the BYTI is not a useful measure of quality? Not necessarily. While the BYTI should not be used as a substitute for an objective measure of quality, in this study it offers insight into the provider?s subject experience of the quality of care she offers in relation to her mentor?s assessment of her quality. This information cannot be obtained by observer ratings of quality and (as described below in the section on implications for practice) can be used by researchers to better understand how the provider?s subjective experience is related to the quality of care she offers and her receptivity to training and quality improvement. Because of the potential benefits gained from the information in the BYTI, researchers should work to revise it to better reflect the information obtained by the FCCERS; this could be accomplished by including more items from the FCCERS, dropping items not measured by the FCCERS, and incorporating important concepts from the FCCERS (e.g. accessibility). This would allow for more direct comparison between the measures and a better understanding of how providers and observers rate the same behaviors. 75 Implications for Future Research and Theory This study gave insight into the provider perspective on quality of sixteen FCC providers participating in a quality improvement program. The results revealed that providers included components of quality from research-based perspective in their own descriptions of quality. Future research should explore the provider perspective using larger, more representative samples of FCC providers. Doing so will allow us to answer several important questions. For example, do providers who are not involved in a quality improvement program define quality in FCC using the same terms? Would the concept of professionalism emerge as a theme, and if so, would it be related to higher quality among those providers? This study highlighted the importance of parents in the minds of these FCC providers. However, measures based on the professional definition only narrowly address the provider-parent relationship. Future research should explore further the provider perspective on this relationship. Results would have implications for measures, as they may need to be revised to reflect the complex and central role that parents? needs play for FCC providers. This area also represented an apparent inconsistency between provider descriptions of quality (in which parents were often mentioned first) and priorities (in which parents? needs were often ranked last). Future research should address these (and other) discrepancies by identifying and exploring them further, perhaps by asking providers themselves to explain the apparent inconsistencies. Another important direction for future research would be to study these questions longitudinally. Does the provider perspective on quality change over time? What conditions seem to induce change? Is there a causal relationship between the provider 76 perspective on quality and observed quality? In this study, it was found that those providers who were improving on observer-rated quality rated their own quality in the top half relative to the sample, while those who were not improving on observer-reported quality rated themselves in the lower half. A longitudinal study of these measures would be able to determine whether or not this relationship is actually a causal relationship. In addition, it would be interesting to see how self- and observer-rated quality are related in a sample of providers who do not regularly receive feedback on their quality. Researchers should also explore how a provider?s subjective experience of her quality influences other important factors (e.g., motivation, receptivity to training, satisfaction). Finally, while the results of this study overall revealed consistency between the provider perspective on quality and observer-reported quality, many inconsistencies were also apparent at the provider level. Because of these differences, Attitude and Behavior theory is insufficient to explain the relationship between the provider perspective on quality and her quality related-behaviors. There are certainly factors other than her attitude that influence her behaviors. A theoretical approach that could guide future research into what those factors may be would be to frame research questions using an Ecological Systems theory (Brofenbrenner, 1992). An ecological framework directs us to look at different levels of the environment or context to see how these factors influence the outcome of interest. Multiple influences exist for FCCP providers at the proximal, microsystem level, such as relationships with other providers, parents, their mentor, and the children in their care. How do the characteristics of each of these relationships influence the provider?s quality related- behavior? How does the provider navigate conflicting demands among those in her 77 microsystem? At the mesosystem level, researcher might ask how parent-child and parent-parent relationships influence provider quality behaviors. At the macrosystem level, researchers might ask, how do different states? policy approaches to FCC impact, encourage, or limit certain attitudes and quality-related behaviors for FCC providers? What combinations of factors at different levels seem to foster higher (or lower) quality care? How do providers incorporate conflicting demands at different levels (e.g. parents and policies)? Implications for Practice and the FCCP Program The results from this study are useful to FCC providers and other professionals in the field of FCC. One important implication is that, while professionalism is related to quality, it should be noted that some providers in this study who offered high quality care did not incorporate professional verbiage in their definitions of quality. Nevertheless, these providers also showed that they could offer high quality care even if their working definition of quality seemed to contain elements in conflict with a professional definition (e.g., ?treat them like my grandchildren?). Given this finding, and in light of the fact that one of the most common complaints FCC providers have about training is its irrelevance to their daily challenges (Taylor, Dunster, & Pollard, 1999), understanding the provider perspective is essential in designing and marketing professional development opportunities to FCC providers. Rather than relying solely on terminology from the research literature to present and organize training sessions, practitioners may be better served by incorporating the words providers use in order to connect with them and better communicate the relevance of training experiences. Based on the results of this study and others, it is clear that a desire 78 to serve families and concern for children are at the heart of quality for many FCC providers. Those concepts are generally not at all in conflict with the goals of most training opportunities: improving quality in order to benefit the children. By using the words providers use to describe these concepts, rather than professional terminology, practitioners may be able to bridge the perceived gap between the provider perspective on quality and the research-base perspective. This study also offers a model for engaging with providers to understand their perspective on quality in FCC. This is important for quality improvement programs like the FCCP, in which mentors are considered quality partners and not experts. By involving the provider?s own ideas in a discussion of quality, it creates a platform for building a professional relationship. Connecting with providers at that level may help mentors and other quality improvement professionals to tailor training in ways that are more meaningful to the provider. It also reveals areas where positive feedback might be readily accepted, constructive feedback should be carefully given, and providers may need more knowledge to expand their definitions to include other important components of quality. For example, positive feedback about provider-child interactions may be especially well received by a provider who believes such interactions are at the heart of quality. At the same time, because it is important to the provider, care should be taken when constructive feedback is needed. A mentor could point out a strength that the provider has in one context (e.g., conversation with children upon arrival) and help the provider to understand how that same skill could be applied in another context (e.g., conversation with children during meal time). 79 In addition, knowledge of the provider?s definition of quality may reveal areas in which she needs extra support. For example, the high level of daily worksheet use in this sample shows that these providers might benefit from training that addresses how to approach parents whose desires for child care conflict with what is known about developmentally appropriate practice. Providers may need help increasing their knowledge and ability to talk about developmentally appropriate practices, as well as training in how to talk with parents effectively. Additionally, the results for the Improving and Not Improving groups on the BYTI suggest that feeling ?stuck? may negatively influence providers? subjective experience of their own quality. Mentors and other professionals should be sensitive to this tendency and be careful to continue to recognize areas of growth, even with providers who are already performing at a high level. This may be especially true for providers who are in an ongoing quality improvement program and are evaluated on a regular basis. Implications for Policy The results from this study also offer implications for policy. First, it will be important for policymakers and researchers to work together to develop and implement policies that support quality in FCC. The current study and others have clearly shown that there is a relationship between professionalism and quality in FCC. Policies should focus on elevating the status of FCC as a profession through regulation, standards, and other supports. Requiring involvement in professional development opportunities will be an important component of increasing professionalism among FCC providers, but policies should include supports designed to eliminate the obstacles that many providers face. 80 Funding should be set aside to support accessible, high-quality training opportunities that are relevant to FCC providers. Limitations There are several limitations of this study to consider when reviewing the results in the context of research, practice, and future directions. One area of limitation is related to the sample. Because of the small sample size, it was not possible to run quantitative analyses to establish (statistically) the relationship between study variables. In addition, the nature of the sample limits generalizability of the study results. Prior to study recruitment, two pools of FCC providers were identified from all providers who were currently participating in the FCCP program. These providers were all operating in the state of Alabama and voluntarily participating in a mentor-based quality improvement program. While attempts were made to randomly recruit participants from the pools of Improving and Not Improving providers, self-selection into the study certainly influenced the final sample. The providers who participated in the study differed from all current FCCP providers on race, being primarily Caucasian, and provided higher quality care as observed by their mentors in the areas of Listening/Speaking and Activities. Other limitations were measurement-related. In an effort to increase participation, the interview was limited to 30 minutes. This time restriction did not allow for an in- depth interview to better understand the provider perspective on quality. While the responses were informative, they were not as rich as one would have wanted. In addition, interpretation of the discrepancies between caregiving priorities and the FCCERS were hindered because the priorities do not directly translate to FCCERS subscales or items. While they are similar, it is not possible to determine whether or not the providers? 81 interpretation of the priority warrants direct comparison to the corresponding FCCERS subscale. It is also important to note that, although FCCP mentors represent outside observers, they are not independent observers. Mentors have a relationship with their FCCP providers and are invested in their improvement. This investment may influence their objectivity in scoring providers on the FCCERS. However, the relationship between the provider and mentor, developed over an extended period of time, could also be seen as affording the mentor-observer in-depth knowledge of a provider?s caregiving that may reflect the provider?s quality no less accurately, if somewhat differently, than a stranger?s assessment over the course of one afternoon. Despite these limitations, this was the first known study with FCC providers that examined the relationship between providers? priorities and both self-reported and observer-reported quality behaviors. The qualitative results provide an initial view of the provider?s way of thinking about what quality caregiving involves. Conclusion Research has typically paid little attention to the provider perspective on quality in FCC. This study revealed that providers describe quality in different ways, some more consistent with the professional definition than others. At the heart of quality for many of these providers is an awareness of the importance of their service to families. While all providers in the study were participating in a mentor-based quality improvement program, the value of professionalism was explicitly communicated and endorsed by a minority of the providers. Those providers who explicitly mentioned professionalism as an important part of quality offered high quality care. There were other providers, 82 however, whose perspectives did not explicitly mention professionalism, yet they also offered high quality care that reflected a professional approach. It is important to continue to explore the provider perspective on quality in FCC, because it may hold important keys to increasing participation in professional development, increasing the efficacy of training programs, and ultimately better quality caregiving in FCC. 83 REFERENCES Abell, E., Miller, E. B., Keiley, M., & Ma, Y. (Under review.) The Family Child Care Partnerships program: Mentoring as a quality improvement strategy. Arnett, J. (1989). Caregivers in day care centers: Does training matter? Journal of Applied Developmental Psychology, 10, 541-552. Austin, A. M., Lindauer, S. L. K., & Rodriguez, A. (1997). Conditions of caregiving, provider nurturance, and quality care. Early Child Development and Care, 135, 21-22. Bailey, S., & Osborne, S. (1994). Provider perspectives on the content and delivery of training for family day care. Child and Youth Care Forum, 23, 329-338. Bandura, A. (1986). Social foundations of thought and action: A social cognitive theory. Englewood Cliffs, NJ: Prentice-Hall, Inc. Bentler, P., & Speckart, G. (1981). Attitudes ?cause? behaviors: A structural equation analysis. Journal of Personality and Social Psychology, 40, 607-627. Bradley, R. H., & Caldwell, B. M. (1984). The relation of infants? home environments to achievement test performance in first grade: A follow-up study. Child Development, 55, 803-809. Bromer, J., & Henly, J. R. (2004). Child care as family support: Caregiving practices across child care providers. Children and Youth Services Review, 26, 941-964. 84 Bronfenbrenner, U. (1988). Interacting systems in human development: Research paradigms: Present and future. In N. Bolger et al., (Ed.), Persons in context: Developmental processes (pp. 25-49). New York: Cambridge University Press. Buell, M. J., Pfister, I., & Gamel-McCormick, M. (2002). Caring for the caregiver: Early Head Start/Family child care partnerships. Infant Mental Health Journal, 23, 213- 230. Caldwell, B. M., & Bradley, R. H. (1984). Home observation for measurement of the environment. Little Rock, AR: University of Arkansas at Little Rock. Campbell, P. H., & Milbourne, S. A. (2005). Improving the quality of infant-toddler care through professional development. Topics in Early Childhood Special Education, 25, 3-14. Clarke-Stewart, K. A., Vandell, D. L., Burchinal, M., O?Brien, M., & McCartney, K. (2002). Do regulable features of child-care homes affect children?s development? Early Childhood Research Quarterly, 17, 52-86. Cryer, D. (2003). Defining program quality. In D. Cryer & R. M. Clifford (Eds.), Early childhood education and care in the United States (pp. 31-46). Baltimore: Brookes. Deery-Schmitt, D. M., & Todd, C. M. (1995). A conceptual model for studying turnover among family child care providers. Early Childhood Research Quarterly, 10, 121- 143. Debord, K., & Sawyers, J. (1996). The effects of training on the quality of family child care for those associated with and not associated with professional child care organizations. Child & Youth Care Forum, 25, 7-15. 85 Fukkink, R. G., & Lont, A. (2007). Does training matter? A meta-analysis and review of caregiver training studies. Early Childhood Research Quarterly, 22, 294-311. Gable, S., & Halliburton, A. (2003). Barriers to child care providers? professional development. Child and Youth Care Forum, 32, 175-193. Gable, S., & Hansen, J. (2001). Child care provider perspectives on the role of education and training for quality caregiving. Early Child Development and Care, 166, 39- 52. Harms, T., & Clifford, R. M. (1989). Family Day Care Rating Scale. New York: Teachers College Press. Harms, T., Cryer, D., & Clifford, R. M. (2007). Family Child Care Environment Rating Scale ? Revised Edition. New York: Teachers College Press. Harrist, A. W., Thompson, S. D., & Norris, D. J. (2007). Defining quality child care: Multiple stakeholder perspectives. Early Education and Development, 18, 305- 336. Holloway, S. D., Kagan, S. L., Fuller, B., Tsou, L., & Carroll, J. (2001). Assessing child- care quality with a telephone interview. Early Childhood Research Quarterly, 16, 165-189. Isralowitz, R. E., & Saad, I. A. (1992). Self-report measures of attitudes and behaviors among Israeli family child care providers. Early Child Development and Care, 81, 101-107. Johnson, J. O. (2005). Who?s minding the kids? Child care arrangements: Winter 2002. Washington, DC: U. S. Census Bureau. 86 Kontos, S., Howes, C., & Galinsky, E. (1996). Does training make a difference to quality in family day care? Early Childhood Research Quarterly, 11, 427-445. Kontos, S., Howes, C., Shinn, M. & Galinsky, E. (1997). Children?s experiences in family child care and relative care as a function of family income and ethnicity. Merrill-Palmer Quarterly, 43, 386-403. Maccoby, E., & Martin, J. (1983). Socialization in the context of the family. In P. Mussen (Ed.), Handbook of child development (Vol. 4, pp. 1-101). New York: Wiley. Morrissey, T. W. (2007). Family child care in the United States. Child Care and Early Education Research Connections: Reviews of Research Series, 1-28. National Institute of Child Health and Human Development Early Child Care Research Network. (1996). Characteristics of infant care: Factors contributing to positive caregiving. Early Childhood Research Quarterly, 11, 269-306. Nelson, M. K. (1990). Mothering others? children: The experience of family day-care providers. Signs, 15, 186-605. Norris, D. J. (2001). Quality of care offered by providers with differential patterns of workshop participation. Child and Youth Care Forum, 31, 111-121. Pence, A., & Goelman, H. (1991). The relationship of regulation, training, and motivation to quality of care in family day care. Child and Youth Care Forum, 20, 83-101. Rosenthal, M. K. (1991). Behaviors and beliefs of caregivers in family day care: The effects of background and work environment. Early Childhood Research Quarterly, 6, 263-283. Rusby, J. C. (2002). Training needs and challenges of family child care providers. Child and Youth Care Forum, 31, 281-293. 87 Shlay, A. B., Tran, H., Weinraub, M, & Harmon, M. (2005). Teasing apart the child care conundrum: A factorial survey analysis of perceptions of child care quality, fair market price, and willingness to pay by low-income, African American parents. Early Childhood Research Quarterly, 20, 393-416. Strauss, A., & Corbin, J. (1990). The basics of qualitative research: Grounded theories procedures and techniques. Newbury Park, CA: SAGE publications, Inc. Taylor, A. R., Dunster, L., & Pollard, J. (1999). ? And this helps me how?: Family child care providers discuss training. Early Childhood Research Quarterly, 14, 285- 312. Tout, K., Zaslow, M., & Berry, D. (2006). Quality and qualifications between professional development and quality in early care and education settings. In M. Zaslow & I. Martinez-Beck (Eds.), Critical issues in early childhood professional development (pp. 77-110). Baltimore: Paul H. Brookes Publishing Co. Weaver, R. H. (2002). Predictors of quality and commitment in family child care: Provider education, personal resources, and support. Early Education and Development, 13, 265-282. 88 Tab le 1 Ta ble of Me asu res of Qu ali ty i n Ch ild Ca re Wh at is Me asu red Pro vid er- ch ild in tera cti on s: po siti ve , pe rm issiv e, pu nit ive , a nd de tac he d Ex pe rien ces o f o ne ch ild in th e en viro nm en t: c are giv er resp on siv en ess , a ffe cti ve sta tes , a nd intera cti on s Emo tio na l & Ve rba l Re spo nsiv ity ; Ac cep tan ce of the Ch ild ; Org an iza tio n o f th e E nv iro nm en t; Play M ateria ls; Inv olv em en t; Va riety Sp ace & fu rnis hin gs fo r c are an d lea rnin g; Ba sic Ca re; Lan gu ag e a nd Re aso nin g; Lea rni ng Ac tiv ities ; So cial De ve lop me nt; Ad ult Ne ed s Sp ace & Fu rnish ing s; P ers on al Ca re Ro uti ne s; L isten ing & Talk ing ; Ac tiv ities ; In tera cti on ; P rog ram Stru ctu re; Pa ren ts an d P rov ide r Pro gra m Size ; S pa ce & Fu rnish ing s; Ba sic Ca re Ro uti ne s; L an gu ag e- Re aso nin g Ex pe rien ces; L earn ing Ac tiv ities ; P are nts & Staff Me tho d o f S co rin g Ob serv er rec ord s freq ue nc y o f oc cu rre nc e o f d iffere nt intera cti on s Ob serv er rec ord s freq ue nc y o f b eh av iors an d q ua lity of in tera cti on s Ob serv er rec ord s p rese nc e o r a bse nc e o f eac h it em Ob serv er ass ign s it em s sc ore of 1-7 (hig he r sc ore s rep rese nt hig he r q ua lity ) Item sc ore ba sed on pre sen ce/ab sen ce of ma teria ls, fe atu res in th e e nv iro nm en t, qu ali ty of care giv er- ch ild in tera cti on s Su bsc ale an d to tal sc ore s ca lcu late d b y av era gin g it em sc ore s Sa me as fo r th e F DCRS Se lf- rep ort; m ult iple ch oice ; h igh er sco res rep rese nt hig he r q ua lity ; s om e ite ms rev erse -sc ore d Qu ali ty Ty pe Pro cess Pro cess Glo ba l Glo ba l Glo ba l Glo ba l Ye ar Pu blish ed 19 89 19 96 19 84 19 89 20 07 20 01 Au tho r(s) Arn ett NICH D ECCRN Ca ldwe ll & Bra dley Ha rm s & Cli ffo rd Ha rm s, Cry er, & Cli ffo rd Ho llo wa y, Ka ga n, Fu lle r, Tso u, & Ca rro ll Titl e o f M easu re Ca reg ive r In tera cti on Sc ale (CIS ) Ob serv ati on al Re co rd of the Ca reg ivi ng En vir on me nt (ORCE) Ch ild Ca re Ho me Ob serv ati on fo r Me asu rem en t o f th e En vir on me nt (CC -HO ME) Fa mil y Da y Ca re Ra tin g Sc ale (F DCRS) Fa mil y Ch ild Ca re En vir on me nt Ra tin g S cale (FCCERS ) Be rke ley -Ya le T elep ho ne Interv iew (BYTI) 89 Table 2 Subscales on the FCCERS and Other Measures of Quality Measures of Quality FCCERS Alabama DHR ORCE CC-HOME Space and Furnishings Health, Safety, and Universal Precautions Physical Environment Personal Care Routines Health, Safety, and Universal Precautions Listening and Talking Language Development Language Stimulation Activities Child Development Stimulation of Cognitive Development Academic Stimulation Learning Materials Variety of Activities Interaction Positive Discipline and Guidance Sensitivity and Responsiveness Responsivity Positive Regard Acceptance Detachment Involvement Flat Affect Modeling Intrusiveness Program structure Quality Child Care and Licensing Fostering Exploration Organization Parents and Provider The Provider and the Family 90 Table 3 Subscales on the FCCERS and Quality in the Scientific Literature Source of Quality Categories FCCERS Nelson (1990) Gable & Hansen (2001) Harrist et al. (2007) Isralowitz & Saad (1992)a Space and Furnishings Child feels at home Physical space planning Personal Care Routines Health, safety, and nutrition Caregiver practices Emergency planning Listening and Talking Activities Developmentally appropriate Developmentally appropriate Creative activity planning Education enrichment planning Interaction Like a mother Communication and rapport Child care skills Program Structure Needs of special children Parents and Provider Administration Visibility and involvement Parent relations Personal attributes Finances and resources Budget management Professionalism Professionalism Relations with other providers Level of education Staff characteristics Work efficiency a Items not relevant to FCCP providers were omitted. 91 Table 4 Demographic Characteristics by Group, T-test and Chi-Square Analysis Characteristic All FCCP Study Sample ?2 Age in years N = 167 N = 14 Mean (SD) 48.64 (10.42) 50.79 (10.27) Range 25-76 35-76 Years experience N = 175 N = 15 Mean (SD) 12.52 (7.26) 11.97 (5.89) Range 2.50-40 2-20 Ethnicity N = 175 N = 16 4.10* White 48.60% 75.00% Minorities 51.40% 25.00% Education N = 174 N = 15 1.36 High school or GED 41.40% 26.70% Some college, no degree 34.50% 46.70% Post-secondary degree 24.10% 26.70% Household Income N = 164 N = 14 2.87 Less than $20,000 23.20% 14.30% $20,001-$40,000 28.00% 14.30% $40,001-$60,000 31.10% 50.00% Greater than $60,000 17.70% 21.40% FCC Income N = 168 N = 14 0.62 Less than $20,000 52.40% 42.90% $20,001-$40,000 39.30% 50.00% Greater than $40,000 8.30% 7.10% *p < .05 92 Table 5 Mean FCCERS Total and Subscale Scores by Group FCCERS All FCCP (N = 197) Study Sample (N = 16) DifferenceA Total M (SD) 4.39 (1.16) 4.75 (1.54) 0.36 Quality Category Adequate Adequate Space and Furnishings M (SD) 4.44 (1.30) 4.78 (1.81) 0.34 Quality Category Adequate Adequate Personal Care and Routines M (SD) 4.48 (1.35) 4.74 (1.77) 0.26 Quality Category Adequate Adequate Listening and Speaking M (SD) 4.31 (1.68) 5.10B (1.64) 0.79 Quality Category Adequate Good C Activities M (SD) 3.56 (1.35) 4.11 (1.62) 0.55 Quality Category Minimal Adequate C Interaction M (SD) 5.56 (1.50) 5.63 (1.63) 0.07 Quality Category Good Good Program Structure M (SD) 4.75 (1.83) 4.75 (2.07) 0.00 Quality Category Adequate Adequate Parents and Provider M (SD) 5.06 (1.14) 5.22 (1.36) 0.16 Quality Category Good Good A Difference between FCCERS mean for All Current FCCP providers and the study sample B p = .07 C Observable change of one category and/or one point on the FCCERS; Kontos et al., 1996) 93 Tab le 6 Ind ivid ua l Re sult s f or Se lec t S tud y M easu res BYTI total sco re (Ra nk )E 55 .25 67 (1 ) 52 (1 1) 57 (6 ) 59 (3 ) 53 (1 0) 52 (1 1) 44 (1 6) 57 (6 ) 56 (8 ) 50 (1 5) 52 (11 ) 61 (2 ) 58 (5 ) 56 (8 ) 59 (3 ) 51 (1 4) FCCERS categ ory - t otal sco re (Ra nk )D Ad q - 4. 75 Ex c ? 6. 59 (1 ) Ad q - 4 .86 (9 ) Ad q ? 4. 49 (1 0) Ina d ? 2. 70 (1 5) Ex c ? 6. 22 (3 ) Go od ? 5.5 9 ( 7) Ina d ? 1. 14 (1 6) Go od ? 5.6 8 ( 5) Min ? 3.5 1 (1 4) Min ? 3.7 3 (1 2) Min ? 3.5 7 (1 3) Ex c ? 6. 22 (3 ) Go od ? 5.3 2 ( 8) Ex c - 6. 08 (5 ) Ex c ? 6. 25 (2 ) Min ? 3.9 7 (1 1) Ye ars in FCCP 5.0 7.0 1.5 7.0 7.0 6.5 7.0 1.0 7.0 6.5 6.5 6.5 5.5 1.0 3.0 6.0 1.0 Im pro vin gC , NI -H, NI -L N/A Im pro vin g Im pro vin g Im pro vin g Im pro vin g NI -H NI -H NI -L Im pro vin g Im pro vin g NI -L NI -L Im pro vin g Im pro vin g NI -H NI -H NI -L Prio rities B Pa r Pa r Ac t Pa r Ac t Sp ace Pa r Pa r Pro Pa r Pa r Sp ace Pa r Pa r Pa r Sp ace Pa r Pro Ac t Pa r Sp ace Pa r Pa r Pro Sp ace Pa r Pro Sp ace Pa r He alt h Sp ace Sp ace Pro Sp ace Sp ace Pro Pro Pro Sp ace Pro Ds pln Ac t Ds pln He alt h Pro Pro Pro He alt h Ac t Pa r Pro Ac t Sp ace He alt h Ac t Pro Ds pln Sp ace Ds pln Ac t Ds pln Ac t He alt h Sp ace Ds pln Pro Ds pln He alt h Ds pln Ds pln Sp ace He alt h Talk Talk Ac t Talk He alt h Ac t He alt h Ac t Ds pln Pro Ds pln Ac t Ds pln He alt h Talk Ds pln Ds pln Ds pln Ac t Talk He alt h Talk Sp ace Ds pln Talk Ac t Talk He alt h He alt h Ac t Talk He alt h Talk Talk He alt h He alt h He alt h Pro Sp ace Talk Talk Ds pln Talk Ac t Talk Talk Talk First Th em eA N/A Pa ren ts Pa ren ts Pa ren ts Pa ren ts Pa ren ts Pa ren ts Pa ren ts Sa fety Sa fety Sa fety Sa fety Pro fess ion ali sm Pro fess ion ali sm Pro fess ion ali sm We lco mi ng ho me Lea rnin g Pro vid er Sa mp le A B C D E F G H I J K L M N O P 94 Tab le 6 Ind ivid ual Resu lts fo r Select Stu dy Me asu res (co ntin ued ) A First co ncr ete them e m en tio ned in op en -en ded resp on ses to th e q uesti on ?Wh at do es q ua lity in FC C mea n to yo u?? B Co mp on en ts of Qu ality as p rio ritized by FC C pro vid ers fr om m ost to least im po rtan t C FC CE RS (Har ms et al. , 2 00 7) statu s o ver tim e ( Im pro vin g, No t Im pro vin g Hig h Sco re [NI -H] , No t Im pro vin g L ow Sco re [NI -L ]) D Lev el of qu ality (ca teg orical ), to tal sco re, an d r an k ( relativ e to th e s am ple) on th e Fam ily C hild C are En vir on men t Ratin g Sc ale E To tal sco re an d r an k ( relativ e to th e s am ple) on th e B erk eley -Yale Telep ho ne Inte rview (Ho llo way et al. , 2 00 1) 95 Table 7 Most Common Key Wordsfrom Open-Ended Responses Key Words Percent of ResponsesA (Number of providers) Child, children, kids 100% (16) Love, loving, loved 43.75% (7) Safe, safety, safe environment 43.75% (7) Teach, teaching, taught, train, teachable moment 43.75% (7) Care, caring, cared for, take care of 37.50% (6) Listing specific basic care needs (e.g., feeding or diapering) 37.50% (6) Talking to, listening to, communicating with [the children] 37.50% (6) Parents, families 37.50% (6) Meeting needs 31.25% (5) Like my own (grand)children, like a mother would 31.25% (5) Morals, right and wrong 31.25% (5) School 25.00% (4) Homelike environment 25.00% (4) Activities (including specific learning activities, e.g. crafts) 25.00% (4) Professional, professional standards, national standards 25.00% (4) A Percent of open-ended responses to the question ?What does quality in FCC mean to you??in which the key words were found (out of 16 providers) 96 Ta ble 8 Th eme Gr ou ps an d Ot he r S tud y M ea sure s Prio rity Ra nk ing s (m ost t o lea st i mp ortan t) Pa r Pa r Pa r Pa r Pa r Sp ace Pa r Pro Pro Sp ace Pro Pro Pa r Sp ace Sp ace Ac t Pro Sp ace Sp ace Pro Pro Ac t Sp ace He alt h Ac t Ac t Ac t Ac t Ds pln Ds pln Ac t Ds pln Ds pln Ds pln He alt h He alt h He alt h Ds pln He alt h He alt h He alt h Ds pln Talk Talk Talk Talk Talk Talk Talk Prio rities Co rre spo nd ing to Th em e N/A Sp ace /Fu rnish ing s He alt hfu l Be ha vio r Re spo nsiv en ess / Disc ipli ne Liste nin g/T alk ing Pa ren ts/F am ilies Ac tiv ities Pro fess ion al De v?t Lev el of Qu ali ty on FCCERS Su bsc ale Co rre spo nd ing to Th em e N/A Ad q (S pa ce) Ad q (P ers Ca re) Go od (In tera cti on ) Ad q (Li ste n/S pe ak ) Go od (Pa ren t/P rov ide r) Ad q (Ac tiv ities ) Ad q (P rg Stru ctu re) Ex c (Pa ren t/P rov ide r) Me an FCCERS To tal Ad q 4.7 5 Ad q 4.6 5 Ad q 4.5 7 Ad q 4.5 3 Ad q 4.6 0 Ad q 4.6 9 Go od 5.8 1 Me an Ye ars in FCCP 5.0 0 5.5 8 4.6 0 5.4 4 5.4 4 4.5 6 3.6 0 Me an Ye ars as FCC Prov ide r 11 .97 12 .96 11 .94 10 .00 11 .69 13 .00 11 .40 Th em e Gro up Sa mp le (n = 1 6) Cu sto dial Ca re (n = 1 3) Emo tio na l Cli ma te (n = 1 0) En ga ge me nt (n = 9 ) Pa ren ts & Fa mil ies (n = 9 ) Tea ch ing & Sc ho ol (n = 8 ) Pro fess ion al (n = 5 ) i i`eure I . ~t urd ('loud 01 Open-Fndcd Responses (All Words) aMAM 0% compassionateVU play learning `--i wa,m professional interactionI r," quality, n~ea` e, dependabil ity ~ ~- t i v i t i eUe4. d-, , wing w1 ing comfortable listen helping happy f~,iOthe~- iaf7 rmlnr inJilnig pl a c e allprt~nn 4 nvrturiny ""'t attention n res "s E,-:a c ood nutrition ~~ mannershealthy p a re ntS development available holding Iftschool besth o trainingbasics talk education standardsmaterials reading hguce ?. Word C'Iuud ol'Open-I';nded Responses (Words Said at least Twice) holding development ID to 0activities APIMV& reading music interaction available k thealthy e s talkcomfortable prof esslonal mother dependability helpin~ ntitritir~r) standards wUV d I" ta,~manners f o o d quality 0"", ho"% V I A0% ff-% - W-" A--M n place training educationb a s is materials listen school full --- - teachlearning play, happy r 99 APPENDIX A FCCERS Items by Subscale 100 FCCERS Items by Subscale Space and Furnishings 1. Indoor space used for child care 2. Furniture for routine care, play, and learning 3. Provision for relaxation and comfort 4. Arrangement of indoor space for child care 5. Display for children 6. Space for privacy Personal Care Routines 7. Greeting/departing 8. Nap/rest 9. Meals/snacks 10. Diapering/toileting 11. Health practices 12. Safety practices Listening and Talking 13. Helping children understand language 14. Helping children use language 15. Using books Activities 16. Fine motor 17. Art 18. Music and movement 19. Blocks 20. Dramatic play 21. Math/number 22. Nature/science 23. Sand and water play 24. Promoting acceptance of diversity 25. Use of TV, video, and/or computer 26. Active physical play Interaction 27. Supervision of play and learning 28. Provider-child interaction 29. Discipline 30. Interactions among children Program Structure 31. Schedule 32. Free Play 33. Group time 34. Provisions for children with disabilities Parents and Provider 35. Provisions for parents 36. Balancing personal and caregiving responsibilities 37. Opportunities for professional growth 38. Provisions for professional needs 101 APPENDIX B Berkeley-Yale Telephone Interview 102 Berkeley-Yale Telephone Interview PROGRAM SIZE 1. On a typical morning, that is, between 9 am and noon, how many children are present in your setting? _______ children 2. On a typical morning, including yourself, how many people work with you in your setting? _______ workers SPACE AND FURNISHINGS 3. Family child-care settings vary as to the amount of space they have available to post child-related pictures an artwork. Which describes your child-care setting? a. There is no space available to display child-related pictures, mobiles, or children?s artwork. b. There is some children?s artwork displayed and you have some store-bought or adult-made pictures for children to look at. c. There is much children?s work displayed, at least two items per child enrolled. Some of it is down low at the child?s eye level. d. There are many items of interest to children displayed where the children can see them. The display is changed at least monthly to match the children?s activities and interests. 4. Do you have any areas in your setting that are specifically set up just for one type of play, like a block area or a dress-up area? a. Yes b. No 5. Which best describes how you prevent children from breaking fragile objects like flower vases? a. You teach children not to touch them. b. You remove them from the areas used by children. 6. How satisfied are you with the amount of space you have for children? a. Somewhat satisfied b. Moderately satisfied c. Very satisfied 7. We are interested in learning about the availability of items for active play, for example, tricycles. Which best describes your child care setting? a. Little active play equipment is available at this time b. You have some equipment in good condition, but there is not a lot of variety. c. The room has a wide variety of equipment in good condition. 103 d. The room has many different kinds of equipment in good condition. The equipment stimulates skills on different levels. For example: tricycles with and without pedals. BASIC CARE ROUTINES 8. We?re interested in how things go when children arrive in the morning. Which of the following is most like your child-care setting? a. You are often too busy to greet children individually. b. Most of the children and parents will be greeted as they arrive. With so many families coming and going, however, some children may arrive without being greeted. c. You greet each child and parent upon arrival. d. You have a conversation with each child and each parent upon arrival. You also use this time to talk informally with the parents or to help a child become involved in an activity. 9. How often do you have a chance to sit with the children while they are eating? a. Never b. Sometimes c. Often d. Always LANGUAGE-REASONING EXPERIENCES 10. Sometimes budgets don?t allow child-care providers to purchase all the toys and materials they would like. The next question refers specifically to the amount of education materials relating to language development, including books as well as music tapes and picture card games. Which best describes your program? a. There are fewer than 6 children books and no other materials available. b. There are at least 10 children?s books and some other materials that you use at least 3 times a week. c. There are at least 20 children?s books and various other materials for the children. Yu have at least one daily planned activity, such as reading or saying nursery rhymes. d. You check out materials from the library once a month or add to the material in other ways and use them in daily activities. 11. On an average day, how many minutes per day does someone read aloud to the children? _______ minutes 12. How often do you ask children specific questions about the story when you read aloud? a. Every day b. Most of the time c. Sometimes 104 d. Rarely 13. Which best describes the type of informal conversation that takes place in your setting? a. You talk with the children primarily while managing routines like toileting, or to correct a child?s behavior. b. You have time for short, social conversations with most of the children. c. You have many conversations with children and try to make comments that build on ideas presented by them. d. You make sure to have a conversation with each child every day and often ask questions to encourage them to talk more. LEARNING ACTIVITIES 14. When it comes to materials involving hand-eye coordination, such as pegboards and puzzles, which best describes your setting? a. At this time, I have no hand-eye coordination materials. b. There are some hand-eye materials available for children to use independently. c. There is a variety of hand-eye materials that are rotated to maintain interest. They are also organized and labeled to encourage self-help. 15. When it comes to art activities and materials, which best describes your setting? a. There are no art materials available for use by children. b. There are some materials, including drawing, at least twice a week. c. There are crayons and paper, or other drawing materials available daily. Art materials needing supervision are planned at least 3 times a week, such as cutting and pasting, or painting. d. There are at least 2 different activities offered daily. Activities include at least one 3-dimensional material per week, such as clay or carpentry. 16. Family child-care homes vary greatly on the amount of space and resources available to provide sand and water play. Do you have provisions for sand play (or a similar material like rice) indoors? a. Yes b. No 17. What about sand play outdoors? a. Yes b. No 18. What about water play indoors? a. Yes b. No 19. What about water play outdoors? 105 a. Yes b. No 20. I am interested in the resources available for dress up or dramatic play activities. Which best describes your child-care setting? a. There are not special materials available for dramatic play. b. There are some props available for dramatic play, mostly to play house. c. There is a variety of dramatic play props and they involve at least two themes. For example, house keeping and work. d. There is a variety of props involving two themes. The props are arranged in their own space and include child-sized play furniture, like a small stove or a baby stroller. 21. How often do the children have access to the television or videos? a. Everyday b. A few times a week c. A few times a month d. A few times a year or never 22. How often do you talk with the children about what they are watching on the television or VCR? a. Always b. Often c. Sometimes d. Rarely or never 23. How often do the children in your setting use worksheets to learn a skill? By this we mean exercises to learn their ABC?s or practice numbers, not drawing or art. a. Everyday b. A few times a week c. A few times a month d. A few times a year or never PARENTS AND STAFF 24. Do you have a regularly scheduled parent conference? a. Yes b. No 25. I am interested in knowing how you are able to balance personal and caregiving responsibilities. Which description best describes you? a. Many housekeeping duties and family errands come up throughout the day. b. You make some changes in your own schedule of housekeeping and family errands on a day-to-day basis to meet caregiving responsibilities. 106 c. You make plans so that family responsibilities and caregiving seldom interfere with one another. You have a substitute available as an emergency backup. 107 APPENDIX C Transcript of Open-Ended Responses 108 Open-Ended Responses Provider A Being available when the parent needs you. Having accessible materials for children at their ages and stages. Listening to and communicating with the children. Knowing the developmental stages and what to offer the children to help them accomplish the? Bringing out the qualities of the child at the ages and stages they are in as they develop. Reading is the most important to me because the more words you read to the children, the more their vocabulary increases. Provider B Meeting the needs of the parents and the children in a homelike environment. [P] Going above and beyond the minimum standards. Being compassionate and loving to the children. [P] Address their needs, like if they get a booboo. Reassure them, love them, and teach them to be fair in their play. All my kids are just like my grandkids. Hold them, listen to them, read to them. Provider C Dependability and consistency for parents. Parents never having to worry about where they are, what kind of care they are getting or food they are eating. A closer atmosphere than a commercial center. All of the children I care for are like my own children. Provider D Something that you can depend on. Spending quality time with the children, having resources available for them to use. [P ? resources] Safe home environment. Don?t have any vicious animals that would cause the child harm. Make sure that all of the safety equipment is there that they need. Safety to me plays a big part in quality. Provider E Giving the best care that you can. Meeting the parents? needs, talk with them. [P ? best care] Making sure that their needs are met as far as teaching them moral values, personal hygiene, manners, meeting their nutritional, physical, and emotional needs. 109 Provider F Having someone you can trust and depend on to care for your child. [P ? care for] Making sure that they are comfortable, fed, given proper nutrition, any other needs are being met. That the kids are happy and being loved like a mother would if she were home to do it. Being talked to and played with like a mother would. Provider G It means a lot to me to know that I am helping families. [P ?helping families] Know the child is taken care of and safe. That the child is getting what they need: love, care, good food, taught right from wrong. Train them: how to use the spoon, how to use the bathroom and be potty trained. I train them the basics: right, wrong, how to talk, ABC?s. What they need is someone to change their diaper, keep their tummies full, and sit with them in a rocking chair and rock them to sleep. Provider H Provide children with a safe and loving environment. [P] Safety issues are already covered when children come in. When children come in, they feel they are free to interact with the materials you have in the day care setting. Provider I Safe, nurturing place for kids to come. A bright cheerful place with toys they can play with and activities they can do. Inviting. That they feel safe and loved. Playing with them, involved. Not just putting them in a room and letting them play all day, ignoring them. Provider J First and foremost, it is a safe and clean environment. It provides adult interaction with the children. It teaches the children different areas of things that they need to know for school. Not necessarily sit down type activities, but it also teaches them manners, discipline, and how to socialize and interact with other children. Has the feel of a home environment so they get that experience because they may be at daycare more than at home. 110 Provider K Do the best job I can taking care of others? children, making sure they are safe and helping them learn. Make sure they are played with and loved. [P-loved] Treat them like I treat my own children when they were little. I know they are happy when they are smiling and looking in my face, so I do things that make them do that. I talk to them as much as I can. Provider L Being more professional so I can do my job better with the children. There are so many resources that help us teach the children so they are ready for school. Help the children be all that they can be. [P-?be all they can be?] Children can turn around and remind me of things they?ve learned. Repeating back to me, and the parents telling me they are happy with what the child is doing. Provider M Striving for the best possible standards in child care nationally. Whatever the professional, the other home care providers as well as those who have experience in education in child development have gathered as the standard, that is what we should strive for. The highest standard. Provider N Being more than just a babysitter. Being someone who can provide a safe, loving, structured environment for the children. Someone who can teach children in a fun, informative way. Get them ready for Kindergarten, but be flexible enough if a child shows interest in something to take advantage of a good teachable moment. Provider O Warm and welcoming home, like a child?s home with school touches. Someone who will be there the whole time the child is a preschooler. Good healthy meals, a lot of reading, a lot of music and fun activities, cooking and craft activities, a lot of love and touching. [P ? love] When they come in the morning they are happy to be here. The ones that are more timid are held, told they will be ok, sing songs about how they are loved, activities that emphasize love and kindness. Treat them like my grandchildren. 111 Provider P Someone who cared about the child?s learning. I think it would be someone who loves and cares for the children. I think their education is very important, their minds. When you present something to them, you can see them learning. [P- loves and cares] You can tell if you are around someone. [P- how can you tell?] Their personality, how they work with the children. The way they get along, treat them, and interact with them. How the children respond, they will show affection [toward the provider]. 112 APPENDIX D Open-Ended Response Coding Key 113 Open-Ended Responses Coding Key Coding Key for Themes Present anywhere in the open-ended response Providers can have more than one theme present in their responses If a response has any of the components of a theme, that counts Coding First Themes * Denotes use for coding ?first themes? (parents/families, professionalism, & safety) Code based on the first concrete theme mentioned in the response Some responses may not fall into any of the 3 themes CUSTODIAL CARE ? Any form of the general term ?basic needs? ? Referring to any specific basic needs (e.g. toileting, nutrition, etc.) ? Any form of the word ?safety? ? Referring to the ?environment? or how the environment is organized (e.g. accessibility) EMOTIONAL CLIMATE ? Any reference to emotional needs ? Any form of the words love, warm, nurture ? Any form of ?care? or ?close? when they are emotional, not in a physical or basic needs sense ? Soothing physical touch ENGAGEMENT ? Listening to, talking to, communicating with ? Playing with ? Quality time TEACHING AND SCHOOL ? Any form of the word ?teach? as it refers to academic-related teaching (e.g. ABC?s, not manners) ? Referring to school or getting children ready for school ? Any form of the word ?learning? ? Mentioning specific learning activities (e.g. reading, crafts, etc.) *PARENTS AND FAMILIES ? Any form of the words ?parents? or ?families? 114 ? You can tell parents are ?on her mind? (OK if it is implied: idea that parents matter, serving & helping them, dependability, taking care of others? children, etc.) *PROFESSIONALISM ? Mentions standards ? Any form of the word professional ? The idea of being professional (e.g. ?more than a babysitter?) *SAFETY (use for ?First Themes? ONLY) ? Any form of the word ?safe?