Get documents about "CorterPelletier EarlyDevelopment University of Toronto
Shared by: alicejenny
-
Stats
- views:
- 1
- posted:
- 10/18/2012
- language:
- Unknown
- pages:
- 36
Document Sample
EARLY DEVELOPMENT INSTRUMENT AND INTEGRATED SERVICES
The Early Development Instrument as an evaluation and improvement tool for school-
based, integrated services for young children and parents:
The Toronto First Duty Project
Carl Corter, Sejal Patel, Janette Pelletier, and Jane Bertrand
Atkinson Centre and Institute of Child Study, Department of Human Development and Applied
Psychology, Ontario Institute for Studies in Education, University of Toronto
EARLY DEVELOPMENT INSTRUMENT AND INTEGRATED SERVICES 1
Abstract
Integrated services for young children and families are part of the new policy landscape in early
childhood, but there is limited evidence on the effectiveness of these programs and how they
develop on the ground. This study examined the use of the Early Development Instrument (EDI)
as both a summative program evaluation tool and as a formative program improvement tool
supporting practitioners in Toronto First Duty, an integrated services demonstration project that
combined kindergarten, child care and parenting supports in public schools. Pre-post
comparisons at community demonstration sites and comparisons with matched community sites
using the EDI suggested that the demonstration program was associated with modest
improvements in emotional and social domains of children’s development. Mixed methods and
multiple measures were used to contextualize summative findings in case studies across
demonstration sites. The case studies explored how integration was implemented at different
sites and how dimensions of enacted integration might contribute to positive outcomes for
children and families. A case study of one site showed how an integrated staff team used EDI
school-level profiles, along with formative feedback on program quality, to target and improve
programming. Over the course of implementation, the integrated program environment quality
ratings and EDI scores improved in relevant areas assessing quality of interaction and social-
emotional development. Findings are discussed in terms of the role of the EDI in program
evaluation and in improvement of practice. The potential value of integrated early childhood
services and the challenges of evaluating complex community initiatives is also discussed.
EARLY DEVELOPMENT INSTRUMENT AND INTEGRATED SERVICES 2
In the current societal quest for better ways to support children’s early development and
learning, many variations on traditional services have been examined. These include improving
the quality of preschool child care (Bryant et al., 2003), offering school-based learning in
prekindergarten classes (e.g., Gormley, Gayer, Phillips, & Dawson, 2005), using innovative
instruction in kindergarten (e.g., Pelletier, Reeve, & Halewood, 2006), and involving parents of
preschool children in school-based programs (e.g., Pelletier & Corter, 2005). Another dimension
of innovation occurs when these kinds of services are integrated in school-based hubs, as
advocated by Edward Zigler in the 21C model of Schools for the 21st Century (Desimone, Payne,
Fedoravicius, Henrich, & Finn-Stevenson, 2004; Zigler, 1989).
The study presented in this paper is part of an evaluation of school-based, integrated early
childhood services in the Toronto First Duty (TFD) demonstration project. TFD combined
kindergarten, child care and parenting supports in a seamless integrated early childhood program
based in public schools. The study here describes how the Early Development Instrument (EDI),
a community–level measure of child development, collected periodically in kindergarten across a
Toronto school district, was used in the evaluation and improvement of this innovative and
complex program in five communities. The focus in this paper is on the potential utility of the
EDI as both a formative and summative evaluation tool. In summative evaluations, it measured
possible program effects in pre-post and quasi-experimental comparisons. In formative
evaluations, EDI information was provided to integrated staff teams as one element to be used in
directing and improving programming. In addition, case study analysis of staff teams provided
information about possible processes linking the program to outcomes.
Although ambitious innovations such as early service integration hold real promise, providing
evidence on their effectiveness and fine-tuning them to diverse communities are thorny
EARLY DEVELOPMENT INSTRUMENT AND INTEGRATED SERVICES 3
challenges. Thus, in this project the EDI was one part of a broad-based, mixed methods
evaluation strategy with multiple measures across different levels of analysis and multiple
research designs (Pelletier & Corter, 2006). In fact, evaluation in this project and in other
complex service integration efforts is more akin to the evaluation of complex community
initiatives (CCI’s) than to traditional program evaluations in which there is a circumscribed
treatment regime (Pelletier & Corter, 2006; Connell & Kubisch, 1995). CCI projects are
multidimensional with different program strands operating at different levels - from community,
to institutional, to personal interactions. Furthermore they are “co-constructed” by many
stakeholders and evolve over time (Connell & Kubisch, 1995). In the TFD project, variations in
implementation across sites and parental choice of which services to access meant that there was
no standard “intervention” for each child.
Despite such challenges, this research explored how regular collection of the EDI in a
school district might contribute to program evaluation and improvement of ambitious, integrated
early childhood programming, as well as to regular school feedback and community
improvement (Janus & Offord, 2000). Designing the TFD evaluation as a randomized control
trial was not possible but the effectiveness of the initiative was explored in converging evidence
from less definitive pre-post and quasi-experimental comparisons using EDI data collected by
the school district. In addition to these two designs, which employed the EDI as a summative
program evaluation tool, this study also used a case study design to explore how service
integration worked at one site and how the use of the EDI as a formative improvement tool may
have contributed to patterns of improvement suggested by the summative findings.
None of the three designs employed in this study provides clear cause-and-effect
conclusions by itself. However, taken together they produce converging evidence that suggests
EARLY DEVELOPMENT INSTRUMENT AND INTEGRATED SERVICES 4
working hypotheses about the effectiveness of a complex initiative and about the processes that
may contribute to effectiveness. In the case study, particular attention is paid to the integrated
staff team and the integrated learning environment. Integration of parents and families into the
TFD initiative is another potential avenue of improvement for child outcomes; this possibility is
explored in another study from the larger project (Patel & Corter, 2005; Patel & Corter, 2007).
Children’s everyday interactions with the learning environment supplied by the staff team and
within their own families are likely to be the proximal mediators of program effects. Other
aspects of the process of integrating services, such as governance and access, were also studied
but are not reported here (see Corter et al., 2006 for an overview).
In the literature, a strong conceptual argument for integrated services has been made on
the basis of a social-ecological analysis of how complex social contexts affect child development
and parenting (Lerner, Rothbaum, Boulos, & Castellino, 2002). From an early childhood
practice standpoint, arguments for integration have often been made on the basis of the
importance of “continuity” for fostering children’s security and learning. Continuity, or more
often discontinuity, operates both horizontally as the child and parent move across settings at one
point in time and vertically as the child moves through developmental transitions in time
(Saracho & Spodek, 2003). Continuity may be fostered by service integration (Regional
Educational Laboratories’ Early Childhood Collaboration Network, 1995). For example, child
care integrated with parent education is horizontal integration; child care in schools, for
preschoolers and before and after school care for older children, has the potential to provide
vertical as well as horizontal integration. Integration with more continuity can mean fewer
transitions for the child, more recognition of the child’s individual needs, better and more
consistent programming, and more consistent expectations and support from adults.
EARLY DEVELOPMENT INSTRUMENT AND INTEGRATED SERVICES 5
Despite the rational arguments, empirical evidence on the value of integration is limited
and not always positive, depending on how integration is defined. For example, a study of
integration as coordination of service and referrals through an individual case manager across
otherwise disconnected services, found no benefits (St. Pierre, Layzer, Goodson, & Bernstein,
1999). Different results might be expected when research examines deeper integration of the
kind found in the TFD model where the nexus of integration is the community’s school. In fact,
recent studies have begun to produce some empirical evidence for the benefits of integrated early
childhood services. Helen Penn and colleagues (Penn et al., 2004) reviewed international studies
of integrated early learning and care and found some evidence that these programs benefited
children, particularly those facing multiple risks and particularly when children entered at an
early age. However, these reviewers cautioned that conclusions about “integration” as a general
approach are limited by the different ways and contexts in which early care and education are
combined. In some countries, integration occurs in an established national system and in others it
is a targeted program established as part of a local evaluation. In most cases, evidence on the
process of integration is not brought together with evidence on the impact on children and
families.
The Toronto First Duty (TFD) initiative began in 2002 with broad child development and
parenting support goals and a vision of universally available, integrated services to achieve the
goals. Demonstration sites were established at five schools in Toronto (see Corter et al., 2002,
for details on the project’s context and roll-out; see Corter et al., 2006, for an overview of
findings). Phase 1 of the project was a four-year period with funding for development and
coordination at the five sites. During this period, early childhood service provision across
communities and schools remained reasonably stable in the city and province (Corter et al.,
EARLY DEVELOPMENT INSTRUMENT AND INTEGRATED SERVICES 6
2006), providing a backdrop for evaluating TFD that was not confounded with dramatic changes
for families and children. Evaluation was designed to describe the implementation process and
the effects of the project at three levels: 1) programs, practice and policy; 2) children and
families; and 3) community impacts and awareness. Evaluation included summative reporting on
the project at the end of Phase 1 as well as formative feedback to sites throughout the course of
the project. The EDI was an important part of both the formative feedback and the summative
evaluation.
The rationale for integrating services in TFD was to provide better support for the
development of young children through six years of age, along with support for their parents in
parenting and for accessing quality child care. General design principles were to be implemented
across sites, but in ways that mapped onto local circumstances. At each site, there was to be an
integrated, universally available service core consisting of child care, school kindergarten and
parent education programs. Other community services, such as public health and child and
family mental health also joined, with variations in the mix across sites. Each site was to work
toward deep integration along five dimensions: a common learning environment, a united early
childhood staff team, seamless access, integrated governance, and common family outreach and
involvement strategies. Parents were free to choose which aspects of the integrated service mix
they and their children would take up, although virtually every 4- and 5-year old child at the five
sites attended half-day kindergarten whether or not they participated in other Toronto First Duty
services. Kindergarten is non-mandatory and is typically offered in the Toronto District School
Board as half-day programs for 4-year-olds (junior kindergarten) and for 5-year-olds (senior
kindergarten). Junior kindergarten in the province of Ontario is a universally-available program
that is part of the regular school system. Unlike most U.S. pre-kindergarten programs (Gormley,
EARLY DEVELOPMENT INSTRUMENT AND INTEGRATED SERVICES 7
Gayer, Phillips, & Dawson, 2005), it is fully integrated with senior kindergarten and both levels
of kindergarten share a whole-child curriculum that is not heavily focused on academic
preparation.
Research and evaluation were built in as an integral part of the project by the funders
(Corter et al., 2002). The data gathering and analysis for Phase 1 extended across four years and
across multiple levels of analysis from the children’s experiences in the program to costs and
benefits of the initiative. Mixed research methods and designs were used (e.g., case studies,
quasi-experimental designs, and pre-post community comparisons). These methods included a
variety of qualitative and quantitative techniques for data gathering. However, the most widely
available data on children across project years and sites came from the EDI database. EDI scores
and demographic data were collected periodically on all senior kindergarten children at TFD
sites and at other schools throughout the Toronto District School Board
The EDI has not been widely used for program evaluation and its effectiveness for
assessing community-level reform has not yet been established (Janus & Offord, 2007). In one
evaluation study of school-based readiness centers for parents and preschoolers, the EDI was
used, along with parents’ reports and direct tests of children’s readiness (Pelletier & Corter,
2005). The EDI was correlated with direct child measures of readiness such as the PPVT-R, as
has been reported in other research (Janus & Offord, 2007). However, only direct child tests
showed effects of the school-based readiness centers in comparison with a control group. It
should be noted, however, that the intervention in this study was more modest than the TFD
program and that the EDI was not being used as a district-wide assessment. Other teacher report
measures have been used in studies of early childhood interventions reporting associations
between programs and readiness (e.g., Bryant et al., 2003). According to its originators the EDI
EARLY DEVELOPMENT INSTRUMENT AND INTEGRATED SERVICES 8
is not appropriate as a diagnostic tool for individual children; it is most useful in characterizing
how groups of children are doing (Janus & Offord, 2007). Thus, EDI differences across
communities that are similar in demographics may indicate how well these communities are
supporting the development of young children and their families. Changes within communities
where universal preschool programs have been implemented could point to program effects in
the absence of other community changes. In the TFD project, the EDI was used as part of a
research strategy that included two summative research designs reported in this study. The first
type was a pre-post design in which scores at the TFD sites were compared from early
implementation to full implementation. The second was a quasi-experimental design comparing
TFD sites, after full implementation, to matched community sites without a TFD program. The
EDI was also used as an important formative feedback tool at the sites; this role is described here
as part of a case study of one of the five TFD sites. Multiple qualitative and quantitative data
sources at the site level allowed us to explore possible “process” mediators between the program-
EDI associations found in the summative analyses. The study of process is an under-developed
part of program evaluations. Too often programs are reported to affect outcomes without analysis
of how they are implemented and how they lead to interactions that provide the process link to
effectiveness (Pelletier, & Corter, 2006; Powell, 2005).
This study breaks new ground by examining the EDI as both a summative and formative
evaluation tool within the same project, and by examining formative case studies to elucidate
possible processes mediating program effectiveness. In the following sections of the paper,
Methods and Results are presented separately for each summative evaluation design and for the
case study.
Design 1: Pre-Post Comparisons of EDI in TFD Sites
EARLY DEVELOPMENT INSTRUMENT AND INTEGRATED SERVICES 9
Method
The Early Development Instrument. The EDI is an inventory of school readiness that
kindergarten teachers complete for each child in their class, generally in the spring term after 6
months or so of experience with the class (Offord Centre for Child Studies, 2005a,b). It is
comprised of more than 100 items categorized into five domains.
1. Physical health and well-being - e.g., holding a pencil, running on the playground,
motor coordination, and adequate energy levels for classroom activities.
2. Social knowledge and competence - e.g., knows acceptable behavior in a public place,
self control, cooperation with others, following rules, and ability to play and work with other
children.
3. Emotional maturity - e.g., ability to reflect before acting, a balance between being too
fearful and too impulsive, and empathy.
4. Language and cognitive development - e.g., reading awareness, writing and numeracy
skills, proficiency in board games, understanding similarities and differences, and information
recall.
5. General knowledge and communication skills - e.g., socially appropriate
communication of needs and wants, symbolic use of language, story telling, and age-appropriate
knowledge about life and the world.
These domains were established on the basis of factor analyses of completed inventories
of over 16,000 kindergarten children; in a smaller scale study the EDI was found to have
adequate psychometric properties in terms of some forms of reliability and validity, such as
convergence between EDI teacher reports and parents’ reports on the five domains (Janus &
Offord, 2007). Individual items on the inventory are generally rated on five-point scales or
EARLY DEVELOPMENT INSTRUMENT AND INTEGRATED SERVICES 10
answered as yes or no items (a copy of the EDI is available from the website at the Offord Centre
for Child Studies, 2005a). An example of a yes/no item on the inventory is “Would you say that
this child can’t sit still, is restless?” (Emotional maturity); an example of a rated item is “How
would you rate the child’s level of energy throughout the day?” (Physical health and well-being).
Most items include scoring criteria that help to define the rating scale points or “yes/no”
distinctions (Offord Centre for Child Studies, 2005b). The EDI also includes demographic items
such as age, gender, language status (English Language Learner or English First Language), and
special needs status.
EDI data were collected by the Toronto District School Board (TDSB) at several time
points over Phase 1 of TFD and were made available to the research team. These included both
board-wide collection and samples of smaller numbers of schools at some points. Teachers were
given training in the administration of the EDI, which typically consisted of one workshop
session with a TDSB early years curriculum specialist and were also provided with the Teachers
Guide prepared by the Offord Centre at
http://www.offordcentre.com/readiness/files/2005.EDI.ENG.Guide.pdf). Reliability of individual
teacher ratings was not assessed by the board. Furthermore, teachers at TFD sites were not
“blind” with respect to being part of the special TFD program. Nevertheless, the research team
conducted analyses of the stability of the EDI at the community level across years prior to full
implementation of TFD. These analyses showed that scores were stable in TFD sites through
early implementation. In particular, scores from TFD school sites had been collected in either
2001 or 2002 and these were compared with scores at the same schools in 2003 during the early
implementation of TFD. One-way analyses of variance for each school assessed mean
differences between years of baseline testing for total scores and for each domain. As reported
EARLY DEVELOPMENT INSTRUMENT AND INTEGRATED SERVICES 11
earlier (Toronto First Duty Research Team, 2003), there were no significant differences for any
school site between the 2003 scores and earlier EDI data, for total scores or for any of the five
domains. Thus, the scores in 2003 represented a reasonable community baseline for evaluating
changes in sites across implementation. The results also suggest that teacher EDI reports at TFD
sites were not biased by participation in the TFD project during the first two years of the project.
In the spring of 2005, at the end of the implementation period, the EDI data were used to
examine the community-level impact of TFD by looking at change over time in TFD school sites
between early implementation in 2003 and full implementation in 2005. Despite the stability in
the years immediately preceding implementation, cohort effects could still confound
comparisons of EDI between 2003 and 2005 in the TFD sites. Thus, potentially confounding
differences for kindergarten children in the EDI database were investigated prior to comparing
the 2003 and 2005 EDI scores. The samples for the comparison were all senior kindergarten
children at TFD sites in 2003 (N=361) and 2005 (N=319). Declining enrolments at the two
school sites accounted for the lower number of children rated in 2005. The variables examined
were language status, special needs status, age, and gender. Univariate F tests were used to
examine the effect of Year (2003 vs. 2005) on each of these variables. Age at time of
administration of the EDI was the only variable that differed significantly between TFD sites
across the two years, F(1, 710)=10.544, p=.001. In 2003 the mean age was 5.76 and in 2005 it
was 5.68. This difference reflected the fact that in 2003 the EDI was collected somewhat later in
the spring term than in 2005. Since age plays a significant role in the EDI ratings, this factor was
statistically controlled in comparing 2005 and 2003 by adding it as a covariate to univariate F
tests of the effect of Year on EDI scores. Aggregated scores for sites were thus compared across
years for total EDI score and each of five domains.
EARLY DEVELOPMENT INSTRUMENT AND INTEGRATED SERVICES 12
Results
Results of the pre-post comparison showed that Total EDI scores did not change between
early and full implementation. However, there was a significant increase in Emotional Maturity,
F(1,666)=6.32, p=.012, and a marginal increase in Social Knowledge and Competence,
F(1,673)=3.691, p=.055, from 2003 to 2005. The mean Emotional maturity score was 8.03 (SD,
1.61) in 2003 and 8.30 (SD, 1.28) in 2005. The mean Social knowledge and competence score
was 8.22 (SD, 1.84) in 2003 and 8.44 (SD, 1.66) in 2005. Effect sizes for these changes, as
represented by Cohen’s D, were .198 for Emotional maturity and .149 for Social knowledge and
competence. These are small effects as gauged by the rule of thumb of small, medium and large
effect sizes being approximately .2, .5, and .8 respectively (Cohen, 1992).
Significant changes were not seen in Physical health and well-being, Communication
skills and general knowledge, or Language and cognitive development. Physical development
was generally not a target in programming at the sites. Scores in Communication skills and
general knowledge were strongly and negatively associated with minority language status; the
fact that roughly 58% of the children in TFD programs were learning English as an additional
language (Corter et al., 2006) may have moderated improvements in this area. In any case, the
improvements on the EDI in social and emotional domains, along with the complementary
observation that several sites had explicit aims for programming and child improvement in
social-emotional development, suggest that there may be causal connections between TFD
programs and improved outcomes for child development. These possibilities are examined
further in the next set of results comparing TFD sites with matched control communities and in a
case study examination of how one site used the EDI as a tool for designing programming to fit
EARLY DEVELOPMENT INSTRUMENT AND INTEGRATED SERVICES 13
how children in the community were doing, with subsequent change in program environments
and EDI scores in that site.
Design 2: Quasi-Experimental Comparison of TFD and Matched Community Sites
Method
Another way of assessing the association between EDI scores and the TFD program was
to compare TFD school sites, at the end of implementation in 2005, with matched community
school sites that did not have the TFD program. Four TFD schools were matched with four
comparable schools, on the basis of physical proximity to the TFD schools and on the basis of
risk status. These same schools were also used as a basis for comparing parent involvement
between schools with and without the integrated preschool services in Toronto First Duty (Patel
& Corter, 2005; Patel & Corter, 2007). Senior kindergarten students comprised the samples for
the TFD schools (N=222) and the comparison schools (N=184).
Matching variables. The initial matching variable was a summary index of demographic
risk (“Learning Opportunities Index” or LOI) used by the Toronto District School Board (TDSB)
to allocate resources and for other administrative purposes. The LOI combines information from
school records, and census data on the average and median income, parental education,
proportion of lone-parent families, recent immigration, housing type, and mobility of families
with school-aged children (Toronto District School Board, n.d.). LOI scores are percentiles
representing a school’s standing among 389 elementary schools in the district, with higher scores
reflecting a ‘higher risk’ school population. The percentile scores for the TFD integrated services
school sites averaged .79; for the control sites the average was .75. Because the LOI was
developed as an administrative tool, there is limited information on its measurement properties.
For the TDSB as a whole, the index percentile predicts school performance on standardized
EARLY DEVELOPMENT INSTRUMENT AND INTEGRATED SERVICES 14
provincial tests measuring academic achievement in third grade reading, writing and math; for
example, the correlation between LOI and third grade writing scores in 2002 was r=-.55
(Pearson’s correlation, p<.01) with higher risk percentiles being associated with lower
achievement scores. In contrast, there was no association of LOI with total EDI scores within the
narrow range of LOI percentiles represented by schools within the TFD and control school
samples. Nevertheless, we went beyond preliminary matching on this index to compare key
demographic factors that were also available in the EDI database in order to test the equivalence
of the samples of kindergarten children from TFD and control sites. Demographic comparisons
of the two samples showed that there were no significant differences in age, gender, and special
needs status. However, the difference in English-only versus English as an additional language
approached significance (p=.059, two-tailed, Fisher’s Exact Test). In the TFD sample, 52.9% of
families spoke a language other than English in the home, whereas in the control sample, 62.4%
spoke another language. Although this difference was not statistically significant, language status
was included as a factor in the ANOVAs comparing TFD and control schools.
Results
Mean differences in EDI scores between TFD sites and control sites were not significant
for Total score, Physical health and well-being, Language and cognitive development, and
General knowledge and communication skills. The difference for Emotional maturity was
significant, F(1, 405)=4.88, p=.028, and the difference for Social knowledge and competence
approached significance, F(1, 405)=3.413, p=.065. The mean Emotional maturity score was 8.18
(SD., 1.35) in TFD sites and 7.87 (SD, 1.46) in control sites. The mean Social knowledge and
competence score was 8.40 (SD, 1.71) in TFD sites and 8.08 (SD, 1.77) in control sites. Thus,
children in the TFD cohorts had made greater gains in emotional maturity and in social
EARLY DEVELOPMENT INSTRUMENT AND INTEGRATED SERVICES 15
knowledge. Effect sizes for these differences, as represented by Cohen’s d, were .220 for
Emotional maturity and .183 for Social knowledge and competence. Again these are small
effects on the order of those found in the pre-post community comparisons.
Because the site type samples differed marginally in the demographic factor of language
background, the site comparisons were repeated with a 2 x 2 ANOVA with site type (TFD vs.
Control schools) and language status (English-first language vs. English as an additional
language) as factors. English language status did not interact with site type in any EDI scale
analyses but there were significant main effects of language status on General knowledge and
communication skills, F(1, 417)=71.16, p=.001. The mean score for English-only children was
7.99 (SD=2.47) and for ESL children was 6.24 (SD=2.92).
Design 3: Case Study of One Site
Method
Mixed methods were used to collect the information to describe and analyze the
development, implementation, and potential effects of the TFD program at each of the five sites.
The data collection included meeting notes and other records from site agencies; participant
observation in organizational meetings at the site; key informant interviews with site leaders;
focus groups and interviews with program staff, parents, and children; and surveys with parents
and staff. Direct observation of programs and small-scale sampling of children with direct
observations and testing were also carried out. Additional data were provided by a City of
Toronto client information system that provided detailed intake and tracking information on
participant demographics and on use of each element of the integrated services array so that
questions of who used the services and how often, could be answered.
EARLY DEVELOPMENT INSTRUMENT AND INTEGRATED SERVICES 16
From the beginning of the project there was continual qualitative monitoring of
organizational and programming development. A member of the research team generally
attended each site’s monthly management meetings and wrote field notes and collected meeting
minutes and handouts, as well as other documents produced at the site. In addition, a research
team member attended some of the integrated staff team meetings that focused on use of
information from the EDI and improvement of program quality with monitoring via the ECERS-
R. Thus the key elements of how the staff teams worked were directly documented in meeting
notes and observations. The research team wrote an in-depth case study for each site at the
beginning of implementation and filed periodic site progress reports as part of overall project
reporting (see reports at http://www.toronto.ca/firstduty/reports.htm). Draft versions of the site
reports were shared with the sites to check the accuracy of facts and interpretations. We also
attempted to reduce subjectivity of the reports by making use of different perspectives within the
research team. Each case site report was written by a different member of the research team and
was reviewed by a second team member who had also attended meetings at the site.
At all the sites, the research team reported back to the sites on general findings from the
progress reports, and on site-specific findings on child measures, including EDI patterns, and
program environment assessments on the Early Childhood Environmental Rating Scale-Revised
(ECERS-R). The ECERS-R is a standardized early childhood environment rating scale that
measures seven areas of environment quality (Harms, Clifford, & Cryer, 1998). Trained graduate
student researchers who were not part of the research team and were “blind” to any preliminary
findings carried out environment observations using the ECERS-R in 2003 and again in 2005.
The school district provided detailed reports on EDI patterns to the individual schools, including
comparisons to school board “norms” in years when board-wide testing was done. Although EDI
EARLY DEVELOPMENT INSTRUMENT AND INTEGRATED SERVICES 17
results were available to all schools in the board, the ECERS-R was provided only to TFD sites.
Furthermore, attention to both the EDI and ECERS-R went beyond school staff and kindergarten
classrooms at TFD sites, where the integrated staff teams considered implications of EDI profiles
for programming and areas which could be improved based on feedback on the ECERS-R.
Site selection. One site, the York Early Years-Wilcox site (YEY-W), was selected for
analysis in this paper for several reasons. First, the pattern of EDI results matched the general
findings from all sites combined. Second, the case analysis showed the clearest picture of an
integrated staff team using EDI along with ECERS-R in a knowledge-building approach to
program improvement. Third, the site was the first to begin implementation of the TFD approach
and was the “Early Leader” among the five sites. Fourth, the site represented a clear test of
whether the TFD programs would have “universal” appeal since the school sits on a
demographic boundary between a middle-class area and an area with high rates of family
poverty, recent immigration, etc. Thus, the case study is not “typical” of the other sites. In fact,
the five sites varied in many dimensions since the TFD approach was not a rigid program to be
implemented in exactly the same way across sites, so no site was typical. For example, the
overall design principles for integration were actually achieved more fully at another site and
stronger results for ECERS-R and EDI improvement were found for a third site.
Results
Overview of implementation. At the York Early Years-Wilcox site, community partners
joined with JR Wilcox Public School, which was already the site of a community child care
centre and a parenting and family literacy centre, to establish YEY-W. Community partners
included several child and family service agencies that were eager to take on implementation of
the TFD model and applied to a joint call for proposals from the funders. Funding provided for a
EARLY DEVELOPMENT INSTRUMENT AND INTEGRATED SERVICES 18
site coordinator, as well as a community outreach worker. The coordinator provided ongoing
leadership in building an integrated and cohesive front-line staff team, with an increasing focus
on program improvement and child outcomes as described briefly below. More details on the full
implementation of the TFD program at the YEY-W site are available in the final project report
(Corter et al., 2007).
During implementation many barriers were overcome and others remained. Some
barriers were “people issues” such as blending different front-line staff from different
professional backgrounds (Corter et al., 2006). Here YEY-W made real progress. Early on there
were tangible tensions as revealed in staff interviews and surveys. Some kindergarten teachers
felt “roped in” and resisted joint work with child care professionals Some front-line staff and key
informants reported that change happened too fast and without sufficient planning participation
by front-line staff (Ioannone & Corter, 2006). Other barriers were physical and financial. For
example, there were challenges around the use of facilities and space at all five TFD sites since
the schools were embedded in a large school system adjusting to the aftermath of amalgamation
and continuing budget cuts.
In measuring the success of the YEY-W site against the aims of TFD project funders, the
aims of the site management and professionals, and the needs of families, there were a number of
positive accomplishments to highlight. There was a new array of flexible care options and
parenting-child programming, which included weekends and summer. During the 2004-05
school year, core early learning activities were provided to 96 children in four classes across
junior and senior kindergarten, including the 39 children forming the EDI sample of 5-year olds
in senior kindergarten. Additional programs included full-time and flexible part-time care, a
family literacy centre and weekend literacy program, parent education programs, and a summer
EARLY DEVELOPMENT INSTRUMENT AND INTEGRATED SERVICES 19
readiness program. Parent survey results indicate good client satisfaction and increasing feelings
of being consulted about the services (Patel & Corter, 2007).
Intake and tracking data indicated that attending families represented the diverse
demographics of the school catchment area, in keeping with the universal, not targeted, service
aims of TFD. As one index of diversity at the site, maternal education of participating TFD
families ranged from 31% with high school or less, 30% with more advanced training such as
community college, and 34% with university degrees (and 5% missing data). Another indication
of effective outreach was that the intake and tracking data for the year 2004-05 suggested that
most children and families in the EDI cohort at YEY-W had made use of early childhood
services in addition to kindergarten. Thus, the number of children of senior kindergarten age
(and/or their parents) who used early childhood services other than kindergarten was 36, close to
the total number of children in senior kindergarten (N=39). It should be noted, however, that
TFD programs were open to all preschool children in the community, including children who
eventually attend a Catholic school system, so a few of the 36 families using TFD were not from
the EDI cohort. In terms of particular services used by the 36 families, 24 children has used half-
day or flexible part-time chilld care, 15 had attended parent-child programs with a parent, and 6
of their parents had attended parent workshops.
Another important point to note is that use of TFD programs other than kindergarten was
even greater by children/parents before children entered junior kindergarten at age 4. In 2004-
2005, 165 children under 4 and/or parents used TFD services at the Wilcox school. In terms of
particular services used by the 165 families, 39 children had used full-time or flexible part-time
child care, 131 had attended parent-child programs with a parent, and 50 of their parents had
attended parent workshops. These data show that TFD provided cumulative experience via
EARLY DEVELOPMENT INSTRUMENT AND INTEGRATED SERVICES 20
additional early childhood programming for most families before kindergarten entry. Thus the
effect of the intervention was to provide more access to early childhood services for more
children and families. Efforts to improve the quality of those services at the YEY-W site was
also a focus as described below.
EDI as an improvement tool at the YEY-W site. A major theme in the YEY-W case study
was the TFD emphasis on improving quality through integrated staffing and programming with a
“knowledge-building” approach. The site improved quality by utilizing data on program quality
and on needs of children. With this information in mind, the integrated staff team worked
together to make changes in the “learning environment” that would “make a difference for
children”. The learning environment was a focus of weekly discussions by the integrated staff
team over the life of the project. Initially a major emphasis was to “integrate” the existing
curriculum requirements for the provincial kindergarten with the programming approaches of
child care and parenting programs that fell into different provincial auspices. As the project
developed, the staff team focused more and more on the goals and quality of the learning
environment and its effects on children, particularly in the joint kindergarten and child care
programming. For example, integrative programming to foster social-emotional development
became an area for special attention at YEY-W during 2003. This was partly spurred by findings
from the 2003 EDI results for senior kindergarten children at the YEY-W school, which showed
that their rates of “low readiness” were nearly double the board norms in the two EDI assessment
areas of Emotional maturity and Social knowledge and competence. These findings were
conveyed to the site via Figure 1. The figure showed the percentage of children at the site
deemed to be “low readiness” on each dimension of the EDI. Low readiness was defined as the
lowest tenth percentile for scores across the entire school board.
EARLY DEVELOPMENT INSTRUMENT AND INTEGRATED SERVICES 21
Insert Figure 1 here
The integrated staff team explored various ways of addressing this need. Some of the
ongoing weekly integrated staff efforts in joining the kindergarten and child care curriculum
approaches supported social-emotional outcomes, since there was major emphasis on this area in
the provincial Kindergarten Program used in YEY-W junior and senior kindergarten classes. In
addition, the ECERS-R reports on program quality were workshopped and discussed with ideas
of how to improve the quality of social and language interaction for children. Eventually a school
board-sponsored Second Step program (Joseph & Strain, 2003) promoting healthy social
development and effective behavior management was adopted across YEY-W programs in 2004.
The school principal arranged for joint professional development for all staff to be trained in the
common approach. The program included activities to develop children’s social skills and
mutual respect. Key informant interviews and field notes suggest that the program was well
implemented in various settings with the possible exception of the parenting and family literacy
centre, where the activities differed from ongoing programming.
There were other forms of less formal programming for social emotional development as
well. Staff had established a Friendship Club to recognize and celebrate friendly actions by
children and it was observed in action during the second round of ECERS-R observations. Case
study notes showed that the integrated staff, including child care staff as well as kindergarten
teachers, also worked to implement the provincial kindergarten curriculum, which includes
social emotional development as a major focus. Thus the improvement in outcomes at this site
cannot be linked to the Second Step program alone. In fact, other TFD sites where socio-
emotional gains occurred did not implement the Second Step program through the integrated
staff team.
EARLY DEVELOPMENT INSTRUMENT AND INTEGRATED SERVICES 22
In parallel with the integrated programming efforts, objective evidence from the ECERS-
R observations showed improvements in the quality of the integrated child care and kindergarten
environment. In particular, the environmental dimension of “Interaction”, which should reflect
an emphasis on improved social relations, showed an increase in quality between early
implementation and full implementation in 2005, as did several other dimensions. Items
comprising the Interaction scale rate the quality of the interaction between teachers and between
children, as well as teachers’ management approaches and support for children’s development of
social and concept skills. In 2005, the mean score for items on this subscale was 7, the maximum
possible, and an increase from the score of 6.2 observed two years earlier.
Replication of summative findings at the site level. EDI scores also increased in the two
areas - social and emotional - where the integrated team had worked together for improvement.
As in the pre-post analyses of EDI reported earlier for all TFD sites, age at time of completion
was covaried in a univariate F test of the effects of year of data collection for the YEY-W site.
Differences from early implementation (2003) to full implementation (2005) were significant for
both Emotional maturity, F (1,71)=8.138, p=.006; and for Social knowledge and competence,
F(1,71)=5.743, p=.022. Parallel to the finding for all sites, the effect size, as well as significance
level, was higher for Emotional maturity than for Social knowledge (Cohen d’s of .60 and .50,
respectively).
Emotional development was at higher levels in the YEY-W site in 2005 than in the
closest comparator school among the control sites. A quasi-experimental comparison was carried
out with univariate F tests comparing the two schools, which were similar in LOI, (.74 and .75)
and school size (39 and 40 children, in two senior kindergarten classes at each school). Again the
difference favoring the single TFD site was significant for Emotional maturity, F (1,77)=8.038,
EARLY DEVELOPMENT INSTRUMENT AND INTEGRATED SERVICES 23
p=.006, but did not reach significance for Social knowledge, F(1,77)=2.758, p=.101, with
corresponding medium and small effect sizes (Cohen d’s of .61 and .32, respectively).
These quantitative results are supported by converging findings from other lines of
evidence at the site level. For example, surveys of all front-line staff working in the project
showed that they perceived that children were benefiting in readiness for school, both socially-
emotionally and academically, from participation in the project, a finding that also held across
other sites (Ioannone & Corter, 2006). Specifically, among the kindergarten teachers, child care
staff, and family support staff surveyed at YEY-W in 2005 at the point of full implementation,
eight of ten agreed or strongly agreed that the program helped “children get ready for school
socially” or “academically.”
Discussion
This study provides summative evidence linking the introduction of an innovative
program of integrated early childhood services at school sites to improved EDI scores in social
and emotional domains several years later. The evidence also supports a complementary
methodological point about the EDI as an evaluation tool: the population-based administration of
the EDI across a school district effectively supports evaluation of early childhood programs in
particular communities. The case study evidence from one integrated service site shows that
formative use of the EDI can contribute to evidence-based practice and team building among
professionals, which in turn may lead to improved program environments, and better child
outcomes. Nevertheless, the EDI is only one part of a multi-method approach to evaluating and
understanding the kind of complex community initiative that Toronto First Duty represents.
The summative evidence for the effectiveness of the Toronto First Duty initiative comes
from a pre-post comparison within sites and from a quasi-experimental comparison to matched
EARLY DEVELOPMENT INSTRUMENT AND INTEGRATED SERVICES 24
community sites. In both kinds of comparisons, evidence for the effectiveness of the initiative
was found in the social and emotional domains of the EDI. Changes were not seen in Physical
health and well-being, General knowledge and communication skills, or Language and cognitive
development. Physical development was generally not a target in programming at the sites. The
domain of General knowledge and communication skills is strongly and negatively associated
with minority language status; the fact that over half of the children in TFD programs have
English as an additional language may have moderated improvements in this area.
Language development was an indirect focus across sites in the sense that early literacy
was emphasized in the development of TFD integrated learning environments; yet the EDI data
did not show an impact on language. There are a number of possible explanations for the lack of
impact. Early literacy is already a focus for separate services of child care, junior and senior
kindergarten, and family support and literacy programs, so integrated TFD programming may
not have added much to children’s day-to-day language learning. Alternatively, the EDI may not
have picked up program effects that would be revealed by direct language measures such as the
PPVT-R (e.g., Pelletier & Corter, 2005). In fact, direct pre/post testing of small convenience
samples of children at three TFD sites with the PPVT-R showed positive changes (Corter et al.,
2006).
The summative findings on social-emotional development are encouraging in terms of
the potential value of the integrated TFD program. Social-emotional regulation is an area for
increasing attention in the conceptualization of readiness (Blair, 2002) as well as in programming
and assessment. The literature suggests that better social emotional adjustment and self-
regulation at kindergarten are important contributors to better transitions and later success. For
example, Rimm-Kaufman, Pianta and Cox (2000) studied kindergarten teachers’ reports of why
EARLY DEVELOPMENT INSTRUMENT AND INTEGRATED SERVICES 25
children experienced difficulty in transition to school. The majority of problems were a
combination of difficulty in following directions (46%) and lack of social skill (21%). Another
study looked at teacher ratings of social and self-regulation- “learning-related skills”- in
kindergarten and demonstrated links to children’s math and reading development through grade
6 (McClelland, Acock, & Morrison, 2006).
Evidence for the effectiveness of early childhood programs in producing social emotional
readiness gains is mixed for large-scale programs such as Head Start (Anderson et al., 2003) but
is stronger for some focused and “manualized” interventions such as the Incredible Years
(Joseph & Strain, 2003). Although the social-emotional advantage found in TFD kindergartens
across all 5 sites was relatively small, the larger effect sizes found within the case study site
presented here suggest that the integrated model can be implemented in particular ways that
boost practically significant gains for children.
The suggestive summative findings from the EDI are also supported by converging
findings from other lines of evidence in this study. For example, as reported in the case results,
surveys of all front-line staff working in the project showed that they perceived that children
were benefiting in readiness for school, both socially and academically from participation in the
project (Corter et al., 2006). Furthermore, the findings from the case study presented here show
explicit efforts by front-line staff to program for improving emotional maturity and social
emotional competence. EDI data provided a baseline pointing to need in these domains and also
provided post implementation evidence that programming was succeeding. Parallel findings on
objective ratings of improvement in program quality via the ECERS-R supported the
interpretation that programming efforts were successful in improving the quality of children’s
EARLY DEVELOPMENT INSTRUMENT AND INTEGRATED SERVICES 26
social interactions. Thus, the case study evidence suggests that staff teamwork and program
improvements may have contributed to better social-emotional outcomes at the site.
An important question is whether the changes in programming and possible effects on
children could have happened without integration. This case study material suggests not. Child
care staff would have not been trained in the Second Step program without the intervention of
the school principal. Delivery of an integrated curriculum including the social-emotional goals
or the provincial kindergarten curriculum would not have extended beyond the kindergarten
teacher and half day kindergarten program into child care time. Consideration of feedback on
program environment quality on the ECERS-R would have been difficult to orchestrate across
separate programs and professional groups.
Nevertheless, there are many possible “processes” intervening between the introduction
of a complex integration program and improvements in child outcome so definitive causal claims
cannot be made. The work of the integrated staff team in programming is emphasized here. As
an example of another potentially important process, involving parents in preschool services and
thereby increasing their capacity to become involved in learning and the school may have
contributed to better child outcomes in the TFD project. In another study in this project, we
found that parent involvement in kindergarten and parents’ confidence in helping their children
learn in TFD schools were greater than among parents at matched schools without integrated
preschool services (Patel & Corter, 2007; Patel & Corter, 2005).
Part of the reason integrated preschool programs may be effective is their potential to set
in motion multiple processes that may mediate between programs and positive outcomes. Thus
more parents and children may be pulled into a service array offering more varied programming.
Better relationships among professionals working together in an integrated approach may lead to
EARLY DEVELOPMENT INSTRUMENT AND INTEGRATED SERVICES 27
better relationships with parents and better client satisfaction, as well as to increasing horizontal
and vertical continuity. These processes may also contribute to parents’ capacity to be involved
as children enter kindergarten. Some of these elements were found in an evaluation of Zigler’s
21C integrated service model Desimone et al., 2004); child outcomes were not reported in this
study but the model’s potential for building staff teams and relationships with parents was
shown.
In terms of methodological considerations, the EDI was a useful part of the evaluation
strategy in this project, but further exploration of its effectiveness in program evaluation and
community reform efforts is warranted. Other studies have raised issues of bias in teacher
reporting; for example, halo effects based on teacher-child relationships may color reporting of
other areas (e.g., Pianta, Steinberg, & Rollins, 1995). Bias of teachers in the TFD sites could
have been a problem if positive expectations about the project inflated teacher reporting.
However, no changes were found during early implementation of TFD as compared to earlier
baseline data. Furthermore the gains in EDI were limited to social and emotional development
and other evidence such as relevant improvement in program environments and parent reports
provided support for the convergent validity of the measure. Since “blind” observers using the
ECERS-R rated program environments, the finding of increases in the dimension of program
“interaction” quality provides evidence for the convergent validity of the teachers’ ratings of
increases in the social and emotional dimensions of the EDI.
Beyond the issue of potential bias as a threat to validity of the measure, differential
sensitivity for different domains of the EDI in detecting real differences is also a concern. For
example, in this study the EDI revealed differences in emotional and social domains
corresponding to programming efforts but did not reveal differences in language, despite
EARLY DEVELOPMENT INSTRUMENT AND INTEGRATED SERVICES 28
attention in programming. Was this due to less effective programming, “ceiling effects” in
language environments (early childhood staff already providing optimal, literacy-rich
programming, as noted earlier), or is the language dimension of the EDI a less sensitive measure
than other domains? Put another way, do the day-to-day interactions of children and kindergarten
teachers provide better sampling of social and emotional functioning than of their language?
Correlations have been reported between PPVT-R scores and the EDI language scale (Janus,
2001) but the possibility that associations are mediated by socio-economic status has not been
ruled out. A clear strength of the EDI is that it is based on a very broad sample of social-
emotional functioning, which is impossible to obtain in brief, direct observations or tests of
children.
Another advantage of the EDI is that the community and district level of administration
provides a common platform for engaging multiple stakeholders in the findings on early child
development, at the same time that it permits the evaluation of program effects at the school
level. In fact, the case study evidence from this study shows that practitioners took EDI evidence
seriously in their professional work, and other findings show that policy makers have also paid
attention. For example, at a provincial and national level, the Toronto First Duty model is
highlighted in a re-issue of an influential early years policy document, the Early Years Study 2
(McCain, Mustard, & Shanker, 2007) as an example of integrated early childhood programming
central to an integrated system. It uses the TFD experience to propose specific public policies
necessary to move from fragmented program delivery to a coherent early years system in Ontario
and elsewhere in Canada. This is a key finding from the current evaluation. The EDI contributed
to evidence-based practice and program improvement and thus to attention in policy
development.
EARLY DEVELOPMENT INSTRUMENT AND INTEGRATED SERVICES 29
A caution to close this discussion is Penn and her colleagues’ (Penn et al., 2004)
reminder that findings on the integration of early childhood services will be limited by the
context in which they are collected. Thus, the lessons on “integration” from this study are
working hypotheses that will not apply across different cultural contexts or even to different
neighborhoods in Toronto in exactly the same way. Part of this uncertainty is the inevitable
variation in how complex change initiatives are taken up in different communities. However, the
approach and value of local knowledge building and formative evaluation, in which tools like the
EDI can play a part, is a universal message.
EARLY DEVELOPMENT INSTRUMENT AND INTEGRATED SERVICES 30
References
Anderson, L.M., Shinn, C., Fullilove, M.T., Scrimshaw, S.C., Fielding, J.E., Normand, J., et al.
(2003). The effectiveness of early childhood development programs: A systematic
review. American Journal of Preventive Medicine, 24, 32-46
Blair, C. (2002). School readiness as propensity for engagement: Integrating cognition and
emotion in a neurobiological conceptualization of child functioning at school entry.
American Psychologist, 57, 111-127.
Bryant, D., Maxwell, K., Taylor, K., Poe, M., Peisner-Feinberg, E., & Bernier, K. (2003). Smart
Start and preschool child care quality in North Carolina: Change over time and relation
to children's readiness. Chapel Hill, N.C: FPG Child Development Institute. (ERIC
Document Reproduction Service No. ED473699). Cohen, J. (1992). A power primer.
Psychological Bulletin, 112, 155-159.
Connell, J.P, & Kubisch, A.C. (1995). Applying a theory of change approach to the evaluation of
comprehensive community initiatives: Progress, prospects, and problems. In A.C.
Kubisch, K. Fulbright-Anderson, & J.P. Connell (Eds.), New approaches to evaluating
community initiatives: Vol. 2. Theory, measurement and analysis. Washington,
DC.:Aspen Institute.
Corter, C., Bertrand, J., Griffin, T., Endler, M., Pelletier, J., & McKay, D. (2002, May). Toronto
First Duty Starting Gate Report: Implementing Integrated Foundations for Early
Childhood. Retrieved November 30, 2006, from the City of Toronto Web site:
http://www.toronto.ca/firstduty/sg_report.pdf
Corter, C., Bertrand, J., Pelletier, J., Griffin, T., McKay, D., Patel, S., et al. (2006, June 8).
Toronto First Duty Phase 1 Summary Report: Evidence-based understanding of
EARLY DEVELOPMENT INSTRUMENT AND INTEGRATED SERVICES 31
integrated foundations for early childhood. Retrieved October 6, 2006, from the City of
Toronto Web site: http://www.toronto.ca/firstduty/TFD_Summary_Report_June06.pdf.
Corter, C., Bertrand, J., Pelletier, J., Griffin, T., McKay, D., Patel, S., & Ioannone, P. (2007,
September). Toronto First Duty Phase 1 final report. City of Toronto.
Desimone, L., Payne, B., Fedoravicius, N., Henrich, C.C., & Finn-Stevenson, M. (2004).
Comprehensive school reform: An implementation study of preschool programs in
elementary schools. Elementary School Journal, 104(5), 369–389.
Gormley W.T.J., Gayer T., Phillips D., & Dawson, B. (2005). The effects of universal pre-K on
cognitive development. Developmental Psychology, 41(6), 872-884.
Harms, T., Clifford, R., & Cryer, D. (1998). The Early Childhood Environment Rating Scale –
Revised Edition (ECERS-R). Vermont: Teacher’s College Press.
Ioannone, P., & Corter, C. (2006). TFD Frontline Survey Report 2004-05. Unpublished
manuscript, University of Toronto, Ontario, Canada.
Janus, M. (2001, May). Validation of a teacher measure of school readiness with parent and
child-care provider reports. Poster presented at the 13th Annual Research Day,
Department of Psychiatry and Behavioural Neurosciences, McMaster University,
Hamilton, Ontario.
Janus, M., & Offord, D. (2000). Readiness to learn at school. ISUMA, 1(2), 71-75.
Janus, M., & Offord, D. (2007). Psychometric properties of the Early Development Instrument
(EDI): A teacher-completed measure of children’s readiness to learn at school entry.
Canadian Journal of Behavioural Science, 39, 1-22.
EARLY DEVELOPMENT INSTRUMENT AND INTEGRATED SERVICES 32
Joseph, G.E., & Strain, P.S. (2003). Comprehensive evidence-based social-emotional curricula
for young children: An analysis of efficacious adoption potential. Topics in Early
Childhood Special Education, 23(2), 65-76.
Kagan, S., & Neuman, M. (2003). Integrating early care and education. Educational Leadership,
60(7), 58-63.
Lerner, R.M., Rothbaum, F., Boulos, S., & Castellino, D.R. (2002). Developmental systems
perspective on parenting. In M. H. Bornstein (Ed.), Handbook of parenting (2nd ed., Vol.
2, pp. 315-344). Mahwah, NJ: Erlbaum.
McCain, M.N., Mustard, J.F., & Shanker, S. (2007) Early Years Study 2: Putting Science into
Action. Toronto, Ontario: Council of Early Child Development.
McClelland, M. M., Acock, A. C., & Morrison, F. J. (2006). The impact of kindergarten
learning-related skills on academic trajectories at the end of elementary school. Early
Childhood Research Quarterly, 21(4), 471-490.
Offord Centre for Child Studies. (2005a). Early Development Instrument: A population-based
measure for communities 2004/2005. Retrieved November 30, 2006, from the Offord
Centre Web site:
http://www.offordcentre.com/readiness/files/2005.EDI.ENG.Instrument.pdf
Offord Centre for Child Studies. (2005b). Early Development Instrument Guide. Retrieved
November 30, 2006, from the Offord Centre Web site:
http://www.offordcentre.com/readiness/files/2005.EDI.ENG.Guide.pdf
Patel, S., & Corter, C. (2005, April). Parents, Preschool Services and Engagement with Schools.
Poster session presented at the biennial meeting of the Society for Research in Child
Development, Atlanta, GA.
EARLY DEVELOPMENT INSTRUMENT AND INTEGRATED SERVICES 33
Patel, S., & Corter, C. (2007). Building capacity for parent involvement through school-based
preschool services. Submitted.
Pelletier, J., & Corter, C. (2005). Design, implementation and outcomes of a school readiness
program for diverse families. The School Community Journal, 15(1), 89-116.
Pelletier, J., & Corter, C. (2006). Integration, innovation, and evaluation in school based early
childhood services. In B. Spodek & O. Sarracho (Eds.), Handbook of research on the
education of young children (2nd ed., pp. 477-496). Mahwah, NJ: Lawrence Erlbaum
Associates Publishers.
Pelletier, J., Reeve, R., & Halewood, C. (2006). Young children’s knowledge building and
literacy development through Knowledge Forum. Early Education and Development,
17(3), 323-346.
Penn, H., Barreau, S., Butterworth, L., Lloyd, E., Moyles, J. Potter, S., et al. (2004). What is the
impact of integrated out-of-home settings on children aged 0-6 and their parents? In:
Research Evidence in Education Library. London: EPPI-Centre, Social Science Research
Unit, Institute of Education, University of London.
Pianta, R.C., Steinberg, M.S., & Rollins, K.B. (1995). The first two years of school: Teacher-
child relationships and deflections in children's classroom adjustment. Development and
Psychopathology,7(2), 295-312.
Powell, D.R. (2005). Searches for what works in parenting interventions. In T. Luster & L.
Okagaki (Eds.), Parenting: An ecological perspective (2nd Ed., pp. 343-373). Mahwah,
NJ: Lawrence Erlbaum Associates Publishers.
Regional Educational Laboratories’ Early Childhood Collaboration Network. (1995). Continuity
in Early Childhood: A Framework for Home, School, and Community Linkages.
EARLY DEVELOPMENT INSTRUMENT AND INTEGRATED SERVICES 34
Retrieved December 1, 2006, from Southwest Educational Development Laboratory Web
site: http://www.sedl.org/pubs/catalog/items/fam17.html
Rimm-Kaufman, S., Pianta, R., & Cox, M. (2000). Teachers’ judgments of problems in the
transition to Kindergarten. Early Childhood Research Quarterly, 15(2), 147-166.
Saracho, O., & Spodek, B. (2003). Recent trends and innovations in the early childhood
education curriculum. Early Child Development and Care, 173(2-3), 175-183.
St. Pierre, R., Layzer, J., Goodson, B., & Bernstein, L. (1999). The effectiveness of
comprehensive, case management interventions: Evidence from the National Evaluation
of the Comprehensive Child Development Program. American Journal of Evaluation,
20(1), 15-34.
Toronto District School Board. (n.d.). TDSB Learning Opportunities Index: An Overview
(DRAFT). Toronto, Ontario: Author.
Toronto First Duty Research Team. (2003). Toronto First Duty July 2003 Progress Report.
Retrieved November 30, 2006, from the City of Toronto Web site:
http://www.toronto.ca/firstduty/FDJune03Rep.pdf
Zigler, E.F. (1989). Addressing the nation’s child-care crisis: The School of the Twenty-First
Century. American Journal of Orthopsychiatry, 59(4), 484-491.
EARLY DEVELOPMENT INSTRUMENT AND INTEGRATED SERVICES 35
Author note
We thank the Atkinson Charitable Foundation, the City of Toronto, and the Government
of Canada’s Social Development Partnerships Program for funding this research. The opinions
and interpretations in this publication are those of the research team and do not necessarily
reflect those of the funders. We are grateful to all participants and partners who contributed data
to the project research, with special thanks to Fidelia Torres who coordinated the YEY-W site, to
Maria Yau and Jill Worthy, who helped to provide data from the Toronto District School Board
and to Petr Varmuza and Joseph Lee, who provided data from the City of Toronto. We would
like to acknowledge the important contributions of other members of the Toronto First Duty
Research Team: Dan Keating, Theresa Griffin, Donald McKay, and Palmina Ioannone, as well
as the administrative support of Christine Davidson in the Laidlaw Centre, Institute of Child
Study.
Correspondence concerning this article should be addressed to Carl Corter, Atkinson
Centre, Institute of Child Study, 45 Walmer Road, Toronto, Ontario M5R 2X2, Canada; email:
ccorter@oise.utoronto.ca.
