Fccnetwork execsummary[1]
- 1. Herr Research Center for Children and Social Policy at Erikson Institute Report to the Local Initiatives Support Corporation-Chicago Staffed Support Networks and Quality in Family Child Care: Findings from The Family Child Care Network Impact Study Executive Summary Funded by the John D. and Catherine T. MacArthur Foundation
- 2. Staffed Support Networks and Quality in Family Child Care: Findings from The Family Child Care Network Impact Study Executive Summary Report to the Local Initiatives Support Corporation-Chicago Juliet Bromer, Ph.D., Principal Investigator Herr Research Center for Children and Social Policy Erikson Institute Martha Van Haitsma, Ph.D. University of Chicago Survey Lab Kelly Daley, Ph.D. Abt SRBI Kathy Modigliani, Ed.D. Family Child Care Project Funded by the John D. and Catherine T. MacArthur Foundation December 2008
- 3. About the Authors Juliet Bromer, Ph.D., is an assistant research scientist at Kathy Modigliani, Ed.D., directs the Family Child Care the Herr Research Center for Children and Social Policy Project in Arlington, MA. She holds an M.Ed. in child at Erikson Institute in Chicago. She earned her M.S. in study from Tufts University and an Ed.D. in higher education from Bank Street College of Education and and adult continuing education from the University her Ph.D. in human development from the University of of Michigan. She has studied family child care for 20 Chicago. She has conducted research on the family and years; consulted with numerous national organizations, community support roles of child care providers in low- foundations, and governmental agencies; and evaluated income neighborhoods. Her current interests include more than 50 community family child care initiatives. the role of families in child care quality measurement, At Wheelock College, Modigliani developed a provider quality improvement in home-based care, and the role of accreditation system for the National Association for support in child care settings. Family Child Care (NAFCC). Before that she held appointments in the research division and graduate Martha Van Haitsma, Ph.D., directs the Survey Lab at the faculty at Bank Street College of Education. University of Chicago. She earned her M.A. in sociology from the University of Texas at Austin and her Ph.D. Local Initiatives Support also in sociology from the University of Chicago. In both Corporation / Chicago programs she worked on research projects that com- LISC/Chicago’s purpose is to stimulate the redevelop- bined survey, observation, and open-ended interviews. ment of neighborhoods throughout Chicago and to In her work at the Survey Lab, Van Haitsma consults reconnect them to the socioeconomic mainstream in the on research design and directs collection of both qualita- region. Through their wide array of financial products, tive and quantitative data for a wide range of projects. investments, and strategic technical assistance, LISC She taught several methods courses in the college before helps community-based development organizations get starting at the Survey Lab and has been teaching grad- what they need to build vibrant communities—afford- uate-level research methods courses at the university able housing, employment opportunities, commercial since 1999. enterprises, community facilities, and civic engagement. Kelly Daley, Ph.D., is a senior analyst at Abt SRBI. She holds a master’s degree in policy studies from Johns Hopkins University. She earned her doctorate in soci- ology from the University of Chicago. She has had extensive experience working in and researching non- profit organizations. Her research interests include the practical application of social science research methods for policy and social change, political participation, and qualitative research methodology. 1
- 4. Acknowledgements We gratefully thank the 150 family child Other graduate research assistants care providers who permitted us to inter- helped with data analysis. Special thanks view them and observe at length in their goes to Jin Xu who carried out the bulk of homes. Thanks to the network directors the data runs, conducted diagnostics for and coordinators and the leaders of seemingly endless permutations of vari- provider associations who agreed to be ous models, and supplied valuable sta- interviewed and allowed us to review their tistical advice. Others who worked on the records. These generous individuals are data include Julie Berger, Hsi-Yuan Chen, not named to protect their confidentiality. Moira Harden, Leslie Kandaras, Ying Li, We are indebted to the John D. and Basheer Mohammed, Alicia Vandervusse, Catherine T. MacArthur Foundation for and Robert Wyrod. funding this ground-breaking study and to We are also grateful for the insight- the Local Initiatives Support Corporation ful comments and suggestions on ear- (LISC) for sponsoring the study. lier drafts from Lee Kreader, National Ricki Lowitz, senior program officer of Center for Children on Poverty, Columbia LISC/Chicago, offered wise and patient University; Toni Porter, Institute for a Child guidance throughout this study. Care Continuum, Bank Street College Alisú Shoua-Glousberg, consultant, of Education; Donna Bryant, FPG Child provided helpful oversight at several criti- Development Institute, University of North cal stages of this research including the Carolina at Chapel Hill; Linda Gilkerson, project design, final data analysis, and Erikson Institute; Jerome Stermer, Voices report-writing phases of the project. for Illinois Children; Tom Layman, Illinois The late Susan Kontos, from Purdue Action for Children; Laura Pastorelli, North University, served as principal investigator Avenue Day Nursery; Bruce Hershfield, for the design and data collection phases New York City Administration for Children’s of the project, and we wish that she could Services; and Marsha Hawley, Kendall have lived to witness the publication of College. this report. This report is dedicated to her Amy Timm provided editorial review memory. and assistance in producing this report. The study was transferred to Erikson Eboni Howard provided overall guidance Institute in 2007 under the direction of to the principal investigator in producing Juliet Bromer. this report. University of Chicago Survey Lab staff who conducted interviews and in-home For more information about this study observations include and to receive a copy of the full technical report, please contact Virginia Bartot Martha Van Haitsma Juliet Bromer, Ph.D. Kelly Daley Herr Research Center for Children Marcia DaCosta and Social Policy Libby Kaufman Erikson Institute Anne-Maria Makhulu 451 North LaSalle Street Carmen Ochoa Chicago, Illinois 60654-4510 Daniel Steinhelper (312) 893-7127 Elizabeth Tussey jbromer@erikson.edu 2
- 5. Abstract The Family Child Care Network Impact Study is the who participated in a post-baccalaureate certificate pro- first study to examine staffed networks—programs that gram in infant studies customized for family child care provide services and support to family child care (FCC) coordinators), and that offer one or more of the following providers affiliated with the network through at least direct services to providers: one paid staff person. The data were collected between • direct training for providers at the network site; the years 2002 and 2004. The study examines the rela- • visits to FCC homes that focus on helping providers tionship between affiliation with a staffed network and work with children and parents; quality of family child care among affiliated providers • supportive interactions with network staff through in the city of Chicago. The study includes 150 licensed regular meetings, telephone help, and opportunities to FCC providers including network-affiliated providers, give feedback; and a matched control group of unaffiliated providers, and • all of these services combined (training for providers, a third comparison group of providers affiliated with a visits to FCC homes, and supportive interactions). provider-led association. (See the glossary at the end of Providers in these networks have significantly higher this report for definitions of terms.) quality scores than providers in networks that do not The study finds that FCC providers affiliated with have this combination of a specially-trained coordinator staffed networks have significantly higher quality scores and direct service offerings. than unaffiliated providers, confirming the earlier find- The study identifies several other network services ing from Kontos, Howes, Shinn & Galinsky (1995) that that, on their own, have a positive effect on quality. These affiliation with a provider support group is associated effects on quality are not as large as those described with higher quality care in FCC homes. The study above, but may point to secondary pathways for improv- further finds that providers in networks with specially- ing quality among network-affiliated providers. Providers trained coordinators (defined as a coordinator who par- in networks that offer any of the following services—use ticipated in a post-baccalaureate certificate program in of a formal quality assessment tool, frequent visits to FCC infant studies customized for family child care coordina- homes to help providers work with children and parents tors) have significantly higher quality scores than pro- (10 times within 6 months or possibly fewer if coordina- viders affiliated with associations. This study also yields tor has specialized training), initial training for newly- new information about staffed networks as a particular licensed providers, or on-going training for providers at type of support organization, and the effectiveness of the network site—have higher quality scores than provid- services offered by networks to affiliated FCC homes. ers in networks that do not offer any of these services. This study finds that multiple factors must come In addition, the study finds that services such as together at once for strong network effects on quality. referrals to external trainings, peer mentoring, free The study identifies specific network characteristics and materials, and business help do not have a significant services that are correlated with significantly higher- relationship to higher quality care. These are all ser- quality care among affiliated providers, once individual vices that do not focus on network staff-provider interac- provider characteristics such as relevant education are tions around care of children and quality improvement. controlled. The following paragraph describes the larg- The study does find, however, that materials and busi- est effects found on quality of care in this study. ness services may help providers maintain financial sta- First, the study finds that networks with staff that bility and retain network affiliation. have regular and supportive communication with pro- Finally, the study compares staffed networks to volun- viders through meetings, telephone help, and formal tary, provider-led associations and identifies significant feedback channels have a greater effect on the quality of differences between these organizations in the types of care offered by affiliated providers than networks that services and supports offered to affiliated providers. The lack these multiple avenues for interaction. Second, the study concludes with policy and program recommenda- study finds a large effect on the quality of care offered tions regarding staffed networks as a quality improve- by providers affiliated with networks that are staffed by ment strategy for family child care. a specially-trained coordinator (defined as a coordinator 3
- 6. Overview Family Child Care (FCC) Quality providers including visits to FCC homes, training, and and Staffed Networks business assistance. Yet few standards or regulations exist in Chicago concerning the quality or nature of sup- The quality of care young children receive in early care port/oversight that staffed networks supply to affiliated and education settings is crucial to their later develop- providers, thus resulting in great variation among these ment and school success (Shonkoff & Phillips, 2000). networks in the frequency, type, and quality of services Estimates vary on how many children are in family delivered to providers. child care arrangements. National surveys find that Several changes in the local policy environment from 1 in 10 children to nearly a quarter of children immediately preceding the study period prompted spend some time in family child care homes, and a concerns about if and how staffed networks in Chicago majority of young children from low-income families are could live up to their promise of improving the quality of cared for in home-based settings while their parents family child care among their member providers. When work (Morrissey, 2007). Aside from offering a develop- the Chicago Department of Human Services (CDHS) mental support to young children, FCC providers have received a contract for Early Head Start (EHS) in 1998, the potential to support low-income working parents they called upon all of their regular center-based Head and communities (Bromer & Henly, 2004; Gilman, Start grantees to launch family child care networks 2001). to oversee administration of these EHS slots in FCC Yet, a growing body of evidence also suggests a crisis homes. This resulted in a sudden expansion of the in the quality of child care young children receive, with number of staffed networks, many run by groups with family child care settings rated as adequate to poor (see limited knowledge of family child care or how to run a Morrissey, 2007 for a review; Kontos, Howes, Shinn staffed network. & Galinsky, 1995). Moreover, FCC providers serving This set of local conditions, along with the wider mostly subsidized, low-income children have been found absence of research about the effect of staffed networks to offer lower quality care than their middle-income on quality of care in FCC homes, prompted the Local counterparts (Kontos et al. , 1995; Raikes, Raikes & Initiatives Support Corporation (LISC) to sponsor and Wilcox, 2005). Studies find that the negative effects the John D. and Catherine T. MacArthur Foundation of low-quality child care may be worse for low-income to fund the research reported here. In the environ- children than for children from higher-income families ment described above, an assumption that network (Shonkoff & Phillips, 2000; Votruba-Drzal, Coley & membership would automatically yield higher quality Chase-Lansdale, 2004). Such studies point to the need care seemed unwarranted. Providers could choose a for more research on how to improve quality in family network that offered just a few services, or they could child care settings, especially those serving predomi- drop network membership altogether to avoid monitor- nantly low-income families and communities. ing. In this environment, networks were allowed to offer Networks, also referred to in some states as systems, minimal services simply because it was economically hubs, or satellites, that offer on-going support to FCC beneficial. providers are one strategy for quality improvement in In their landmark study of quality in family child family child care (Hershfield, Moeller, Cohen & the Mills care, Kontos et al. (1995) found that support group affil- Consulting Group, 2005). Such networks exist in many iation, in general, was an important correlate of quality. states, although there are currently few mandated stan- Yet no studies to date have looked systematically at the dards for these services. Staffed networks in the city of particular characteristics of staffed networks that are Chicago, the site of this study, provide services to FCC related to quality child care. Clearly, more information 4
- 7. was needed about what networks offer that increases Description of Staffed Networks, quality of care in member provider homes. This study is Provider-led Associations, and the first to take a detailed look at staffed networks in a Providers large urban community and to examine the particular characteristics and services of networks that are associ- Staffed Networks ated with quality child care. A total of 35 staffed networks were included in this study although only 26 networks had providers who Study Design and Methods were eligible to participate. In order to participate in the study, network-affiliated providers had to be licensed The Family Child Care Network Impact Study took by the state of Illinois, had to be affiliated with these place during the years 2002 to 2004 and includes 150 networks for at least six months, had to be affiliated licensed FCC providers in the city of Chicago. The study with only one network, and had to hold no affiliation used a matched control design wherein 80 staffed net- within the past year with other support groups such as work-affiliated providers were matched on key provider provider-led associations. characteristics (age, gender, race/ethnicity, experience, education, and neighborhood type) to a control group of Network coordinators 40 unaffiliated providers. The matched control group Each of the networks in this study had a coordinator was designed to isolate the effect of network affiliation on staff to work directly with FCC providers at the net- on quality of care among affiliated providers. The study work. Of these coordinators, 70% held a college degree also includes a third comparison group of 30 providers or higher and most had some kind of relevant educa- affiliated exclusively with a provider-led association. tion or training in early care and education. Although The comparison group of association-affiliated providers the study did not specifically ask coordinators about in this study is not representative of association provid- the source of their relevant education, 10 coordinators ers in Chicago, but rather a sub-set of providers who are from 10 networks (38%) with providers in this sample affiliated exclusively with an association. reported that they participated in a post-baccalaureate Interviews were conducted with network staff, asso- certificate program in infant studies customized for ciation leaders, and affiliated providers in the study network coordinators. The program took place at a local sample. In-person interviews with network staff and institution of higher education1 and was designed at association leaders focused on organizational history the request of the local department of youth services as and goals as well as services offered to member provid- ers. Telephone interviews with providers gathered pro- 1This certificate program in infant studies for network coor- vider reports about network and association services. dinators was offered by Erikson Institute in Chicago. Some Two observational measures of child care quality were coordinators reported this specialized training during the inter- used in provider homes: The Family Day Care Rating views, but it was coded only as “relevant education,” along with Scale (FDCRS), which assesses global quality of the other relevant education coordinators might have received. family child care home including the environment, rou- Once this study was transferred to Erikson Institute in 2007, we were able to obtain a list of network coordinators in the tines, learning activities, and provider-child interactions study who attended the infant studies certificate program to (Harms & Clifford, 1989), and the Arnett Caregiver break out those who had received this training from those who Interaction Scale (Arnett CIS), which measures provider had general education in the area of child development. We sensitivity to children in care (Arnett, 1989). then matched these names to coordinators we had interviewed at networks with providers in the study. 5
- 8. Table 1. Characteristics of Post-Baccalaureate Certificate Table 2. Five Dimensions of Staffed Network Program for Network Coordinators Services to Providers Program Component Description Type of Service Description Course work and Graduate-level, academic sequence Visits to FCC homes • Monitor quality supervised internship of four semester-long courses • Check for licensing violations followed by a supervised internship. • Observe and work with children • Talk to providers about their Focus on Course work focuses on working work with children and parents infant-toddler care with providers who care for infants • Meet with parents and toddlers and their families. Course work also covers knowledge Education/ training • Knowledge of child development base specific to infant/toddler • Training for providers at the development and care. network site • Referrals to off-site training Adapted for network Curriculum focuses on how to and education coordinators support FCC providers in their work • Tuition reimbursement programs with children and families.* Professional and • Regular provider meetings Relationship-based Program emphasizes supportive supportive relationships • Telephone help curriculum relationships between coordinators • Opportunities for feedback and between instructors and coor- to the network dinators. Modeling of relationship- • Peer mentoring programs building helps coordinators develop supportive and effective relationships Material resources • Lending libraries with providers in their networks and • Free toys, books, equipment ultimately helps providers develop Business services • Recruitment and enrollment supportive relationships with children of families and families in care (Gilkerson & • Payment of fees Kopel, 2004; Stott & Gilkerson, 1998). • Administration of subsidies Funding and endorsement Coordinators were fully funded to • Help with taxes participate in the certificate program by local and federal government entities, and sponsored by their network agencies. *The 18-credit certificate program offered by Erikson Institute was also modified to a 15-credit program, with three internship credits waived for coordinators’ prior experience in the field. 6
- 9. part of their work to implement a new Early Head Start Provider-led Associations grant. Participation in this program by network coordi- The 12 provider-led associations in this study are nators (specially-trained coordinators) turns out to be a qualitatively distinct from staffed networks. Provider- key predictor of higher quality among affiliated provid- led associations are groups of providers who volunteer ers in this study. their time to participate in association activities, are Although the study did not involve an evaluation independent of any sponsoring agency, and do not have of the certificate program, conversations with the cer- paid staff. Associations offer peer support, mentoring, tificate program director and instructor (which took and professional encouragement to provider members place after the data collection period) helped illuminate and depend on the leadership of individual providers. five unique aspects of the program (see Table 1). They All but two association leaders in the sample had an include: Graduate-level course work and a supervised associate’s degree or higher and most had professional internship; a focus on infant-toddler development and training in child development or early childhood educa- care; a curriculum intentionally adapted for FCC net- tion. By and large, associations are not deeply involved work coordinators; a relationship-based curriculum; and in the day-to-day operations of member providers, full funding and endorsement by local and federal gov- although many offer training or referrals to external ernment entities. training, regular association meetings, and telephone help-lines. A few even offer occasional home visits to Network services member providers. Staffed networks in Chicago vary in the type and fre- quency of services offered to affiliated providers. Based on reports from network staff and affiliated providers, the study conceptualizes five dimensions of services that are offered by networks (although not uniformly or consistently) to affiliated providers (see Table 2): Visits to FCC homes; education and/or training for pro- viders; supportive professional relationships; material resources; and business services. Analyses of network services and quality in this study rely on provider reports of services received or available. Provider reports are a more reliable measure of network services because network leaders are motivated to put their best foot forward, while providers have no incen- tive either to over or under report the services their net- works provided. 7
- 10. Table 3. Demographic Characteristics of FCC Providers by Affiliation Status Matched Characteristics Staffed Network Control Provider-led N=80 N=40 Association N=30 Race and Ethnicity Black or African American 65% 65% 90% Latina or Hispanic 31% 27% 0 White 1% 8% 10% Asian 2% 0 0 Age Mean age 46 47 45 Years of Experience Mean numbers of years in child care 5.6 5.9 7.2 Highest Education Level Less than high school 16% 15% 3% High School or GED 13% 15% 20% Some college but no degree 43% 38% 47% A.A. degree 19% 18% 17% B.A. degree or higher 10% 15% 13% Neighborhood Poverty Average proportion of persons below poverty from Census 2000 23% 20% 19% Unmatched Characteristics Highest Relevant Education* None 25% 36% 43% Some college 33% 26% 43% CDA credential 30% 25% 7% A.A. degree 14% 8% 7% B.A. degree or higher 1% 3% 0% Average Household Monthly Income* $3,447 $3,041 $3,194 *Staffed network providers had higher levels of post-secondary relevant education in child development or early childhood education and higher incomes than the control group providers most likely due to higher rates of CDA completion and higher paying Early Head Start slots. 8
- 11. Table 4. Characteristics of FCC Programs by Affiliation Status Staffed Network Control Provider-led Association Characteristics N=80 N=40 N=30 Mean number of children enrolled 6.8 7.2 7.1 (includes part-time) Infants (under age 1) enrolled 55% 58% 60% Homes with 1 or more assistants 75% 77% 80% Percent with any Head Start 53% 0% 0% or Early Head Start slots Percent with any private fee-paying families 39% 60% 70% Providers not matched to either the network or unaffiliated group. Table 3 shows the demographic characteristics of the Association providers in this sample were almost all 3 groups of providers in this study. The similarity of African-American, almost exclusively established pro- the distribution of most characteristics for the first two viders, and had higher levels of college education than groups—network-affiliated providers and unaffiliated network providers but lower levels of relevant education (control group) providers—is due to the matched control in child development or early childhood education. group design of the study. Providers were matched on Although providers were not matched on program race, age, experience, education, and neighborhood type. characteristics, Table 4 shows that FCC programs are Providers were not matched on relevant education and similar across the 3 groups except for source of payment. household income. The 30 association-affiliated provid- ers in this study were selected as an additional com- parison group to network-affiliated providers but were 9
- 12. Findings Network Affiliation and Quality and providers affiliated with any network, regardless of coordinator qualifications. The following section details The study finds that affiliation with a staffed network is a specific features of networks that have a positive effect on strong predictor of global quality in FCC homes in a low- quality among affiliated providers as well as those aspects income urban context. Network affiliation has a signifi- of networks that do not impact quality in this study. cant and positive association with higher global quality scores when comparing network-affiliated providers with unaffiliated providers, even after controlling for Network Services, Coordinator other provider and program characteristics associated Qualifications, and Quality with quality, such as provider’s relevant education, Large Effects on Quality household income of the provider, and ages of children The study finds networks that emphasize supportive inter- in the family child care program. actions between staff and providers, and networks that The average global quality (FDCRS) score for the have a specially-trained coordinator who delivers services 150 providers observed for this study is 3.80,2 which is to providers, have the largest positive effects on quality considered “adequate” but not “good.” Despite low overall among affiliated providers. These findings suggest that scores, the average global quality score for network- coordinators are central to the effectiveness of network affiliated providers is 3.99, or “adequate” as compared services for providers. Table 5 summarizes the effects to the average score for unaffiliated providers, which is of particular network services and network coordina- 3.38 or “minimally adequate.” Moreover, 10% of network- tor qualifications on global quality (as measured by the affiliated providers score a 5 or above indicating “good” FDCRS) among the 80 network-affiliated providers. quality, whereas none of the unaffiliated providers score Two pathways towards large network effects on quality in the “good” range. Similarly, 11% of network-affiliated are found, controlling for individual provider character- providers score a 2, indicating “inadequate” or “poor” istics such as relevant education, Early Head Start par- quality, compared to 40% of unaffiliated providers scor- ticipation, and ages of children in care.3 First, the study ing a 2. In other words, although most network-affiliated finds a large and positive effect on global quality among providers have quality scores that indicate “adequate” providers affiliated with networks that offer regular but not “good” quality care, very few offer poor, inad- opportunities for supportive interactions between net- equate care. By contrast, none of the unaffiliated provid- work staff and providers. As Table 5 shows, providers ers offer “good” care, and nearly half offer care that may in networks that offer the combination of regular meet- be considered harmful to children. ings, telephone help, and opportunities for providers to The study goes on to find that providers affiliated with a give feedback to network staff, have significantly higher network that has a specially-trained coordinator (participated global quality scores than providers in networks that do in a post-baccalaureate certificate program in infant studies not offer this group of opportunities. customized for coordinators) have significantly higher global quality and provider sensitivity scores than unaffiliated pro- 2 The FDCRS rates providers on 32 standards with scores rang- viders. Moreover, providers affiliated with a network that has ing from a low of 1—a score that designates “inadequate” care, a specially-trained coordinator also have significantly higher to a high of 7—a score that designates “excellent” care. global quality scores than providers affiliated with provider- 3Although the study does not control for self-selection into led associations. This finding underscores the central find- certain staffed networks by providers, interviews with provid- ing in this study that networks with qualified staff have ers suggest that providers did not join networks based on the a unique opportunity to improve quality in family child particular services offered or qualifications of network staff. care homes. Indeed, the study finds no significant quality Providers often joined networks in order to receive the higher Early Head Start subsidy rate, free materials and/or business difference between providers affiliated with associations help—services that are not associated with quality in this study. 10
- 13. Table 5. Effects of Network Coordinator Qualifications and Network Services on Global Quality (FDCRS) Among Network-affiliated Providers (n=80 providers) Network Coordinator Qualifications Large Effects on Qualitya Modest Effects on Qualityb No Effects on Qualityc Experience • Prior experience working with children Education • Participation in specialized • Coordinator level of general certificate program education • Coordinator has non-certificate relevant education/training Network services Visits to FCC homes • Use of formal quality assessment • Check for licensing violations • High frequency visits (10 times • Discuss health/safety information in 6 months) focused on working • Monthly visits with a child Education/ training • Training for providers at • Referrals to external training network site • Tuition reimbursement • Training for newly licensed providers Professional and supportive Combination of supportive • Peer mentoring relationships interactions: • regular meetings • telephone help, and • opportunity to give feedback to network Material resources • Lending libraries and business services • Free toys, books, equipment • Recruitment of families • Administration of fees/subsidies • Help with taxes Network Coordinator Qualifications AND Network Services Specially-trained coordinator d Specially-trained coordinator AND any of the following AND any of the following: (in order of increasing effect size): • Lending libraries • Training for providers at • Free toys, books, equipment network site or • Recruitment of families • Visits to FCC homes focus • Administration of fees/subsidies on child / parent; or • Help with taxes • Combination of supportive interactions (regular meetings, telephone help and opportunity to give feedback to network); or • Combination of all 3 services (training, visits, interactions) a Large effects are defined as more than half a point higher score on the FDCRS (as determined by ordinary least-squares regression analyses). b odest M effects are defined as less than half a point higher score on the FDCRS (as determined by ordinary least-squares regression analyses). c No effects on quality are defined as no statistically significant positive effect on FDCRS scores. d articipated in a post-baccalaureate certificate program in infant studies customized for coordinators working with FCC providers P 11
- 14. Second, the study finds a large and positive effect on coordinators to be effective, they must have reasonable global quality among providers affiliated with networks caseloads of providers. Furthermore, the certificate that have a specially-trained coordinator and that offer program’s particular focus on infant-toddler care (most one or more of the following direct services to providers: FCC providers care for very young children) and the • direct training for affiliated providers at the customization of the curriculum for FCC coordinators network site; may partially explain the impact of this program. Other • visits to FCC homes focused on helping providers relevant education/training in child development or early work with children or parents, and; childhood education among coordinators does not have • opportunities for supportive interactions with network a significant association with higher quality care among staff through the combination of regular meetings, affiliated providers. telephone help, and opportunities to give the network regular feedback. Modest Effects on Quality The study also finds that the use of a formal quality assess- Providers in networks with a specially-trained coordi- ment tool by networks, frequent visits to FCC homes, training nator and that offer all of the above services combined for providers at the network, and coordinator prior experi- (training for providers, visits to FCC homes, and sup- ence in child care and specialized coordinator training each portive staff-provider interactions) have FDCRS quality on their own have a significant yet more modest effect on scores of 5.07, on average, which is considered “good” quality, and point to additional ways in which networks may care for children. Although there are few providers from impact the quality of care among affiliated providers. our sample in these networks, this particular combina- These modest but significant effects of network ser- tion has the greatest effect on global quality in the study vices and coordinator qualifications on quality are detailed (close to a point higher FDCRS score). below. In addition, two combinations of a specially-trained • Providers in networks that use a formal quality coordinator and a package of direct services have a assessment tool during visits to FCC homes have significant and positive effect on provider sensitivity to significantly higher global quality scores, on average, children in care: than providers in networks that do not offer formal • Networks that have a specially-trained coordinator and assessments of quality. supportive staff-provider interactions through meet- • Providers in networks that conduct visits to FCC ings, telephone help, and feedback opportunities; and homes to help providers work with children or parents • Networks that have a specially-trained coordinator at least 10 times within a 6 month period (or possibly and the combined package of all three services—train- fewer if coordinator has specialized training5) have ing for providers at the network, visits to FCC homes, and supportive staff-provider interactions. 4 Five networks with providers in the study had multiple coor- Several factors may help explain the large effect of dinators who worked with providers. Only one coordinator was specially-trained network coordinators on quality of care interviewed at each network and the coordinators that were offered by affiliated FCC providers. The 10 coordinators interviewed in each of these five networks did not participate in who participated in the specialized certificate program the certificate program. However, it is possible that other coor- dinators in those five networks, who were not interviewed and were the only coordinators at their network, and thus it thus could not be identified, did attend the program. is likely that providers in these networks were receiving 5 Although high-frequency visits have a significant relationship services directly from these coordinators.4 Moreover, 80% to higher quality care, the study also shows that specially- of networks with specially-trained coordinators also had trained coordinators who work with children during visits to coordinator-to-provider ratios of less than 1 to 12, which FCC homes have a greater effect on quality than frequency of supports the idea that in order for specially-trained visits alone. Although some frequency of visits is obviously nec- essary for coordinators to have an impact on provider practices, the study cannot identify an ideal number of visits. 12
- 15. significantly higher global quality scores and more sen- sitive interactions with children, on average, than pro- No Effects on Quality viders in networks that do not conduct this intensity of The study goes on to find that some network services, includ- visits to FCC homes. A majority of providers (78%) who ing monitoring FCC homes, external training for providers, report high frequency visits to their homes also belong peer mentoring, and material and business resources, are to networks that have what could be considered an NOT significantly associated with higher quality care among optimal coordinator to provider ratio of no more than affiliated providers. None of these four network service 12 providers per coordinator. areas involve direct interaction between network staff • Providers in networks that offer education or training and providers around care of children, as described below: for providers at the network site have higher global • visits to FCC homes focused on monitoring for licensing, quality scores and more sensitive interactions with health, and safety violations (versus visits focused on children, on average, than providers in networks that helping providers work with children and parents); do not offer education or training for providers at the • referrals to external training or tuition reimburse- network site. ments for providers (versus on-site training for pro- • Providers in networks that offer introductory training viders at the network); for both licensed providers just joining the network • peer mentoring programs, which offer supportive peer or for those just becoming licensed also have higher relationships, (versus opportunities for professional global quality scores than providers in networks that support between network staff and providers); do not offer introductory training. • material resources and business services (versus ser- • Providers in networks with a coordinator who has pro- vices that involve interactions and support between fessional experience working with children, either as an network coordinators and providers). FCC provider or as a center-based teacher, have higher Although material resources and business services do global quality scores than providers in networks that do not impact the quality of care in this sample of network- not have coordinators with child care experience. affiliated providers, these services may help providers • Providers in networks with a specially-trained coordi- run better businesses. Providers report that the follow- nator have higher global quality scores than provid- ing material resources and business services helped ers in networks that do not have a specially-trained them improve their businesses (percentages of network- coordinator. affiliated providers reporting this are shown): Two network services, on their own, have a differen- • help with recruitment of families (60%) tial effect for newly-licensed versus experienced pro- • free supplies (55%) viders. For more experienced providers, visits to FCC • help with payments (38%) homes that focus on working with a child, or talking to • business skills (37%) a provider about a child, have a greater effect on global Most of these services are delivered at the agency quality than they do for newly licensed providers. For level rather than by network coordinators and, in addi- newly licensed providers, use of a quality assessment tion to improving providers’ business practices, may tool during visits has a greater effect on global quality be important factors in attracting providers to join and than it does for experienced providers. remain affiliated with networks. 13
- 16. Table 6. Characteristics of Staffed Networks and Provider-led Associations Staffed Networks Provider-led Associations Organizational Characteristics • Part of an established social service • Independent group of providers umbrella organization • Funded by member dues, occasional • Funded by external agencies one-time grants (e.g. Early Head Start) Staff and Leaders • Paid staff to work directly with providers • No formal, paid staff to work with providers • Staff are not providers • Voluntary and fluctuating leadership • Some network coordinators have • Leaders are also FCC providers specialized training to work with FCC providers Services for Providers • Services focus on initial training for • Services focus on professional beginning providers and raising quality development of experienced providers of care for children • Benefits often include social activities • Benefits often include access to other for providers outside hours of caring for umbrella group services or facilities for children children or their families Differences Between Staffed Networks networks have paid staff who may be specially trained and Provider-led Associations to work directly with providers. Networks also function under the funding and organizational capacity of an This study finds that staffed networks offer a different menu umbrella agency. Provider-led associations do not have of services than provider-led associations. Many networks paid staff or stable funding sources, and their existence focus their services on new providers, child care quality, and depends on an individual association leader’s commit- child and family well-being. Provider-led associations focus ment and availability (see Table 6). their efforts on seasoned providers, professional advocacy, Providers in networks are more likely than association and peer support for providers. Although some associa- providers to report that network affiliation helps their tions offer occasional home visits and direct training for business and financial stability through reduced-cost or providers, these services are dependent on unpredict- free supplies, marketing of programs to parents, and reli- able funding and on the time constraints and circum- able and regular administration of subsidy payments. stances of individual association leaders. Moreover, 14
- 17. Policy Implications Findings from this study regarding specific character- findings regarding network services associated with istics and services of staffed networks have concrete higher quality care in this study may be categorized implications for policy-makers and administrators seek- as structural and process features of networks (Philips ing to improve the quality of family child care. & Howes, 1987; Kontos et al. , 1995). Conceptualizing network services in terms of structural and process fea- Government and other stakeholders should consider invest- tures of quality may facilitate the development of stan- ing in staffed networks as a potentially effective quality dards for networks. improvement strategy for family child care in low-income, As shown in Table 7, in this study structural fea- urban communities. This study finds that affiliation with tures of networks refer to components that can be easily a staffed network is significantly associated with higher regulated, such as specialized training for coordinators, quality care. Prior research suggests that children from frequency of visits to FCC homes, low coordinator to low-income families are more likely to be cared for in provider ratios, use of a formal quality assessment tool family child care homes and may benefit from high- during visits, and training and educational workshops quality child care settings. Thus, improving quality in for providers at the network site. Process features of family child care may be one way to improve outcomes networks in this study refer to components that are not for low-income children and families. easily regulated but are observable, such as visits to Government and other stakeholders should consider invest- FCC homes that help providers work effectively with ing in specialized graduate-level training for network coor- children and parents, strong coordinator-provider rela- dinators who work directly with FCC providers. This study tionships that are responsive and respectful of provider finds that the combination of a specially-trained coordina- needs, and opportunities for providers to give feedback tor and direct services to providers focused on working to and have a voice within the network. Prior research with children is a key component of staffed networks that has found that structural and process aspects of qual- have higher quality providers. An approach to quality ity in child care facilities and programs are linked to improvement that includes specially-trained staff, who child outcomes (Vandell & Wolfe, 2000). Yet despite the deliver training and technical assistance to providers, is findings from this study regarding quality outcomes, a more effective strategy than support services without the research reported here points to the need for future a specially-trained coordinator. Although the study did research to examine the relationship between network not observe the network coordinator certificate program quality and child outcomes. in this study, key aspects of this program may contribute Government and other stakeholders should consider creat- to its effectiveness in helping coordinators work with and ing mandatory standards based on structural aspects of support providers: graduate-level academic course work staffed networks that are associated with higher quality FCC. and supervised internship; a focus on infant-toddler care; Additional investments in the process aspects of staffed a curriculum adapted for FCC network coordinators; a network quality should also be considered such as assuring relationship-based curriculum; and funding and endorse- the content of visits to FCC homes is focused on helping ment by local and federal government. providers work with children and parents, and implementing Government and other stakeholders should consider creat- programs and practices that lead to strong network-provider ing a set of quality standards for staffed networks. With the relationships. Early Head Start may be a promising exception of Early Head Start standards, networks at sponsor of networks, as some of the services associated the time of this study (2002–2004) had few standards with quality in this study are mandated by Head Start to follow, which resulted in a range of network services. standards, including coordinator to provider ratios and Borrowing terms from child care quality measurement, frequency of visits to FCC homes. However, Early Head 15
- 18. Start standards alone may not be enough to ensure Table 7 Structural and Process Features of Staffed Network . quality outcomes. Indeed, the study finds no differences Services Associated with Higher Quality Care in quality between providers who have Early Head Start Structural Features of Networks slots and those who do not.6 Network Coordinator • Participated in a post-baccalaureate Government and other stakeholders should encourage Qualifications certificate program focused on collaborations between staffed networks and other orga- providing support to FCC providers nizations that serve FCC providers, including provider-led who care for very young children and associations, unions that represent providers, and resource their families. and referral agencies. Increased collaboration and part- • Prior experience working with children either in FCC or center- nerships between support organizations could reduce based setting. redundancies in support systems and maximize the potential of different support groups to help providers. Visits to FCC Homes • High-frequency visits to FCC homes (at least 10 times within 6 months Services such as lending libraries and business help, or possibly fewer if coordinator has for example, are not directly related to quality of care specialized training) for children, but may be important for improving busi- • Low coordinator to provider ratio ness and other provider-focused outcomes. Such services (no more than 12 providers per may be better delivered by organizations that focus on coordinator) provider advocacy, peer support, and business stability. • Use of a formal quality assessment tool in visits to FCC homes Packaging different types of services through collabora- tions between networks and associations, for example, Education/ Training • Training and education at the may make it easier for providers to access the types of network site for affiliated providers. • Introductory training for new services they need for both quality improvement and providers business support. Government and other stakeholders should consider finan- Process Features of Networks cial incentives for FCC providers to join staffed networks and Visits to FCC Homes • Content of visits to FCC homes is improve their quality of care. In the current study, higher- focused on helping the provider work paying Early Head Start slots for children attracted pro- with children and parents. viders to join networks. Further collaboration between Professional and • Opportunities for professional Early Head Start and family child care may be one way Supportive Relationships and supportive relationships to bring providers into networks and consequently raise between coordinators and the quality of care offered in these FCC homes. Other providers: Combination of regular state-level incentives might include tiered reimburse- meetings, telephone help-lines, and opportunities to give feedback ment rates with network providers receiving higher reimbursements than non-network providers. 6It is important to note that all networks in the study had some Early Head Start or Head Start slots. However, not all providers affiliated with the networks were allocated one or more of these slots. 16
- 19. Program Recommendations Findings from this study have several implications for for providers to give the network formal feedback about agencies that sponsor staffed networks in large, urban the program services. Such feedback may help providers communities. feel they have a professional voice in the network and foster positive and trusting relationships between staff Staffed networks should invest resources into hiring coordi- and providers. Provider feedback also offers a source of nators with a bachelor’s degree and encourage coordinators program and service development for network directors to enroll in graduate-level training focused on working with and coordinators that is directly responsive to the needs providers, very young children, and families. This study of providers. finds that specially-trained coordinators, who attended a post-baccalaureate certificate program in infant studies, Staffed networks should focus their resources on develop- enhance the effectiveness of direct services to providers ing training programs for providers at the network rather including training for providers, visits to FCC homes, than making referrals to off-site programs or offering tuition and staff-provider interactions. This coordinator certifi- reimbursement programs for providers. On-site training cate program is not a professional-development train- for providers at the network may enable coordinators ing, but rather a coordinated academic, credit-granting to customize trainings for providers in the network and program in infant studies offered at an institution of offer opportunities for providers to develop professional higher education. relationships with other providers as well as with net- work staff. Staffed networks should hire coordinators who have prior experience working with children either in family child care Staffed networks should invest their resources in visits to or center-based settings. Direct experience working with FCC homes. In order to carry out quality-focused visits, children helps coordinators understand the work of staffed networks should commit to limiting provider casel- FCC providers and may enable them to develop trust- oads for coordinators to no more than 12 providers per coor- ing and supportive relationships with providers in their dinator, in order to assure adequate frequency and intensity networks. of visits. This study finds that the following characteris- tics of visits to FCC homes have a significant relation- Staffed networks should find ways to develop supportive ship to higher quality care among affiliated providers: interactions between network staff and providers through • network uses a formal quality assessment tool in regular meetings for providers, telephone help, and oppor- FCC homes; tunities for providers to give network staff feedback. This • specially-trained coordinator works with children study finds that networks that offer this combination of during visits to FCC homes; opportunities for staff-provider interactions have some • specially-trained coordinator talks to providers about of the greatest effects on quality. Regular meetings for children during visits to FCC homes; providers should focus on topics identified by providers • specially-trained coordinator talks to providers about or focus on training topics related to working with young parents during visits to FCC homes; children and families. Networks should also provide • network staff make regular and frequent visits to some mode of regular communication between coordina- FCC homes (at least 10 times in 6 months or possibly tors and providers in addition to scheduled visits to FCC fewer if coordinator has specialized training) to help homes. Providers should have regular telephone access provider work with children and parents. to someone at the network for technical assistance. Finally, networks should have in place some procedure 17
- 20. Staffed networks should differentiate their services depend- Finally, staffed networks may encourage more experienced ing on providers’ experience levels. Individualized services, providers to join or form their own associations. Provider- such as visits to FCC homes focused on working with led associations may be an additional support for pro- children, may be more effective for experienced provid- viders in networks and dual affiliation may be beneficial ers. Services that help providers understand quality, to many providers. This study finds that associations such as use of a formal quality assessment tool during a offer different kinds of supports to providers than net- visit, may be most effective for newly-licensed providers. works do—mostly in the areas of advocacy and peer net- working. Association involvement may be a particularly Staffed networks may consider offering business services effective quality improvement step for more experienced and/or material goods to providers as an incentive for provid- providers in a network. ers to join the network. Such services, however, should not replace quality-focused services such as visits to FCC homes, direct training for providers, or opportunities for staff-provider interaction. 18
- 21. Future Research Findings from this study point to the potential for devel- children in care (Bromer, 2006). Networks have the oping standards and best practice models for staffed potential to support and enhance these relationships. networks. Future studies may include piloting these Future studies might look at how networks interact models and examining the impact of network services on with and support parents and how networks help pro- quality of care over time. In particular, the study finds viders work effectively with parents. that specialized coordinator training is a key predic- Future studies may also examine the effectiveness tor of higher quality networks and providers. However, of networks for different groups of providers including limited information is known about how this coordinator family, friend, and neighbor providers and license- certificate program helps coordinators work effectively exempt providers serving low-income families. The with providers. Future research may involve examining current study finds that some network services are the processes by which training of network coordinators more effective for newly licensed providers, while other impacts providers, children, and families. The child care services are more effective for more experienced provid- field is in need of detailed information about profes- ers. Future research could examine the different ways sional development processes in order to replicate effec- networks support quality across provider types and lev- tive quality improvement programs. els of experience and licensing status. Given the large The current study was designed to understand the numbers of low-income children who are cared for in relationship between network affiliation and quality of license-exempt homes, understanding how to support care, as measured by standard assessments of the fam- quality in these settings seems an important goal for ily child care environment and provider-child interac- future studies. tions. Yet other outcomes in addition to program quality Another area for future investigation is the com- may yield further information about the impact of net- munity and neighborhood role of networks. Networks works. The child care field is currently in the midst of have the potential to support neighborhood-based FCC reassessing and conceptualizing current approaches to providers and to help providers develop a positive pres- measuring and defining quality child care that include ence in their local communities. Some research has better alignment between assessments of child care examined the community-building roles of FCC provid- quality and child outcomes (Child Trends, 2006). Future ers, documenting the neighborhood-watch function that studies could examine the impact of network affilia- many providers in low-income neighborhoods perform tion on child outcomes in addition to program quality (Bromer, 2006; 2002). Networks that support provid- outcomes. ers have the potential to increase recognition and vis- Parent perspectives and parent outcomes may be ibility of providers in neighborhoods, and to enhance another area to examine in future studies of networks. the impact providers have on children, families, and Parents are central players in young children’s develop- communities. ment and experiences in child care. Many FCC provid- ers often develop close relationships with parents of 19
- 22. Glossary Family Child Care (FCC) is paid, place children with the providers in their groups in order to be able to apply for non-parental child care offered in a pro- network and administer the payments to grants. Associations have no paid vider’s home. This study included only the providers under purchase-of-service staff and no regular income aside from FCC providers licensed by the state of contracts. In Chicago, the staffed net- member dues. Illinois (through the Illinois Department work agencies employ a coordinator to Affiliated indicates membership in of Children and Family Services). Some oversee and deliver services to member a staffed network or provider-led people use the term to include unregu- providers. Larger networks may have association. lated, non-parental home-based care additional staff. by relatives or neighbors, and in some Unaffiliated designates providers who Network directors are the program or states the provider is not required to are not affiliated with any support orga- agency directors who oversee a staffed live within the FCC home. In Illinois, a nization. That is, they do not belong to network. Some of the organizations that provider without an assistant is licensed either a staffed network or provider-led run staffed networks are very small so to care for up to eight children under association. Because the unaffiliated that the overall organization director also the age of 12, no more than five under providers in this study were matched directs the network. Other organizations the age of five and no more than three to the network providers on several are large and complex so that the net- under the age of two. With an assistant dimensions and serve as a primary com- work director is not the head of the entire the rules allow for more young children parison group, they are also referred to agency but directs the network and or up to 12 children if some of them are as matched control providers. These some other set of programs. school-age and are cared for before and providers were selected based on demo- after school only. Network coordinators are the staff per- graphic characteristics that matched the sons who deliver the network program network providers in the study. These Support organization refers to any services for providers. The network coor- providers serve as a comparison group organization that offers formal or informal dinator works for (or with) the network for the network providers. supports to family child care providers. director defined above. The coordinator is Quality in this study refers to quality Staffed network or network refers to a the person responsible for visits to FCC of family child care care as measured family child care support network with homes; setting up and running meetings by either the Family Day Care Rating paid staff attached to a pre-existing social and trainings for the providers; interact- Scale (FDCRS) or the Arnett Caregiver service organization. The staffed network ing directly with providers by phone, mail, Interaction Scale (CIS). provides oversight, direct education/ser- email and/or in-person; getting providers vices and/or links to education and ser- referrals or direct services, etc. Global quality in this study refers to vices for family child care providers. The quality of family child care as measured Specially-trained network coordina- providers who belong to staffed networks by the Family Day Care Rating Scale tors are coordinators who participated in in this study are independent contractors (FDCRS). a post-baccalaureate certificate program and are not employed by the network. in infant studies, customized for coor- Provider sensitivity to children or sen- Typically, staffed networks screen and dinators working with family child care sitive interactions with children in this register children and their families for providers. study refers to quality of family child care federal or state child care programs such as measured by the Arnett Caregiver as Head Start/Early Head Start or vouch- Provider-led association refers to a Interaction Scale (CIS). ers for Transition Assistance for Needy group of providers who come together Families (TANF). Agencies that sponsor voluntarily to form a mutual support or staffed networks generally target low- professional group. Some associations income families. Most staffed networks are organized as 501(3)c non-profit 20
- 23. References Arnett, J. (1989). Caregivers in day-care centers: Does training Morrissey, T. W. (2007). Family child care in the United States. matter? Journal of Applied Developmental Psychology, 10(4), (Reviews of research literature review). New York: Child Care 541–552. & Early Education Research Connections. Bromer, J. (2002). Extended care: Family child care, family Philips, D., & Howes, C. (1987). Indicators of quality in child support, and community development in low-income neighbor- care: Review of the research. In D. Phillips (Ed.), Quality in hoods. Zero to Three, 23(2), 33–37. child care: What does research tell us? Research Monograph of the National Association for the Education of Young Children Bromer, J. (2006). Beyond child care: The informal family sup- (NAEYC), Vol. 1, 1–19. Washington, DC: NAEYC. port and community roles of African-American child care pro- viders. Dissertation. University of Chicago, Chicago, IL. Raikes, H. A., Raikes, H. H., & Wilcox, B. (2005) Regulation, subsidy receipt, and provider characteristics: What pre- Bromer, J., & Henly, J. R. (2004). Child care as family support? dicts quality in child care homes? Early Childhood Research Caregiving practices across child care providers. Children and Quarterly, 20(2), 164–184. Youth Services Review, 26, 941–964. Shonkoff, J. P., & Phillips, D.A. (Eds.) (2000). From neurons Child Trends (2006). Measuring quality in early childhood to neighborhoods: The science of early childhood development. and school-age settings: At the junction of research, policy, and Washington, DC: National Academy Press. practice. Executive Summary. Washington, DC: Department of Health and Human Services. Stott, F., & Gilkerson, L. (1998). Taking the long view: Supporting higher education on behalf of young children. Gilkerson, L., & Kopel, C. C. (2004). Relationship-based sys- Zero to Three, 18(5), 27–35. tems change: Illinois’ model for promoting social-emotional development in Part C early intervention (Occasional Paper Vandell, D. L., & Wolfe, B. (2000). Child care quality: Does it No. 5). Chicago: Erikson Institute, Herr Research Center for matter and does it need to be improved? (Special report No. 78). Children and Social Policy. University of Wisconsin-Madison Institute for Research on Poverty. Gilman, A. R. (2001). Strengthening family child care in low- income communities. New York: Surdna Foundation. Votruba-Drzal, E., Coley, R. L., & Chase-Lansdale, P. L. (2004). Child care quality and low-income children’s development: Harms, T., & Clifford, R.M. (1989). Family Day Care Rating Direct and moderated effects. Child Development, 75, 296–312. Scale. New York: Teachers College Press, Columbia University. Hershfield, B., Moeller, A., Cohen, A.J., & the Mills Consulting Group (2005). Family child care networks/systems: A model for expanding community resources. New York: Child Welfare League of America. Kontos, S., Howes, C., Shinn, M., & Galinsky, E. (1995). Quality in family child care and relative care. New York: Teachers College Press.
- 24. About the Center The Herr Research Center for Children and Social Policy informs, guides, and supports effective early childhood policy in the Great Lakes region. Unique in its regional approach, the center brings together perspectives from policy and research to promote the well-being of young children from birth to age 8 and their families. This center builds on the work of an applied research center established at Erikson in 1997 with a generous gift from the Jeffrey Herr family. The center expanded its mission in 2005 with additional funding from the Herr family and from the McCormick, Joyce, and Spencer Foundations and the Children’s Initiative, a project of the Pritzker Family Foundation. Herr Research Center for Children and Social Policy at Erikson Institute 451 North LaSalle Street Chicago, IL 60654-4510 www.erikson.edu/hrc