Many communities across the country have set for themselves
the ambitious goal of enhancing school readiness. But what does school readiness mean, and how do communities know
whether they have achieved it? This
report is intended to help communities invest wisely in school readiness
initiatives. It begins by summarizing
recommendations from the National Education Goals Panel (NEGP) for defining and
assessing school readiness. The report
then presents a framework for community investments based on an ecological view
of child development. In other words,
this framework considers factors related not only to the child, but also to the
child’s family, early childhood care and education and neighborhood.
The National Education Goals Panel (NEGP) was established in
July 1990 and is comprised of a bipartisan group of federal and state officials
who assess and report on state and national progress toward achieving the eight
National Education Goals set for the nation. The first of these goals states
“by the year 2000, all children in America will start school ready to learn”
(NEGP, 1997a, p. 1). In addressing this
first, important goal, the NEGP identified three components of school
readiness: (1) readiness in the child; (2) schools’ readiness for children; and
(3) family and community supports and services that contribute to children’s
readiness.
Testing is a commonplace feature of American education. Used properly, tests and other assessment
tools can help educators design and deliver the appropriate services for
individual children and can facilitate community-wide or statewide tracking of
children’s status at kindergarten entry and later on. But tests and other assessment tools can also be misused
(Shepard, Taylor, & Kagan, 1996).
For example, they may result in labeling young children prematurely or
inaccurately. They may also lead
communities to focus just on the child’s skills and overlook factors such as
the readiness of schools and the availability of community supports.
Love and colleagues (1994, 1999) also provide detailed
guidance to communities for measuring such community supports for school
readiness as children’s access to high quality and developmentally appropriate
preschool programs. Their
recommendations of specific interview items for inclusion in community surveys
are based on a detailed review of measures being used in national surveys. For example, regarding early childhood care
and education, they recommend items from the National Household Education
Survey and the National Child Care Survey.
An extensive body of research on child development helps
identify the factors that influence children’s readiness for school, beginning
with those closest to the child and moving outward to encompass the family,
early care and education and the neighborhood.
This ecological view of child
development provides a useful framework for understanding where and how
communities can intervene to support and promote healthy child development in
general and school readiness in particular.
This approach may be especially helpful to communities as they set
priorities for investments in school readiness.
There are many programs across the country that may well be
effective in promoting school readiness. In this report, we limit our examples to several programs that
have been rigorously evaluated or for which longitudinal data (with adequate
consideration of background characteristics) are available. A table summarizing findings from the
research literature and their implications for targeted activities to improve
school readiness follows this executive summary; it also appears in Part III of
the report.
Government and private
organizations have experimented with a broad range of approaches to lift
families out of poverty or to address its negative consequences. One set of approaches seeks to raise family
incomes through employment, income supplements or a combination of the
two. Another set of approaches seeks to
address problems associated with poverty through quality early child care,
improved health care and nutrition and parenting education and family
support. Some experimental
interventions for low-income families (including the New Hope Project and the
Minnesota Family Investment Program) have provided wage supplements or earnings
disregards to increase family income and have seen some positive effects on
children’s cognitive and school outcomes (Huston et al., in press; Knox,
Miller, and Gennetian, 2000).
Interventions that reduce or delay
childbearing among young women and men, and/or increase the spacing between
children seem important as well (Bradley et al., 1987; Wood et al., 1993). One particularly promising strategy is to
provide first-time teen mothers with home visits from a public health nurse who
provides important information on prenatal care, child development and family
planning (Olds et al., 1997). Another
promising strategy is to expand outreach to men of all ages for family planning
services.
Despite the fact that kindergarten entry is a critical
period in children’s lives, many schools do not have specific guidelines to
facilitate this transition, nor is there extensive research on best practices
in this area. Nevertheless, the broader
literature on child development and early education does offer some general
guidance for easing the transition to kindergarten:
Promoting children’s emergent literacy skills can take place
both within the family and within early childhood care and education
settings. Children who live in homes
where reading and writing are common and valued usually experience success with
reading as they begin school (Halsall & Green, 1995). But the literacy
environment and activities within a child care setting can also make important
contributions to emergent literacy, especially if a young child does not come
from a home that supports literacy development.
Access to books and print material is important in the home
environment, as is the activity of parent-child book reading. A meta-analysis of the empirical evidence
related to the amount of parent-preschooler book reading found that shared
book-reading is related to language development, emergent literacy and reading
achievement (Ijzendoorn & Pelligrini, 1995). However, many researchers suggest that it is not just the amount of parent-child reading but the quality of reading that is
important. Parents whose children learn
to read more easily do not simply read to their children but ask questions,
help their children relate stories to their own lives and help them learn to
predict events (Halsall & Green, 1995).
Family-based interventions have revealed promising
practices, including providing free children’s books to low-income families
(e.g., Reach Out and Read; High et al., 2000) and teaching parents effective
interactive reading styles to use with their children (sometimes called
“dialogic” reading; Whitehurst & Lonigan, 1998). Other practices that have been studied are home visitation
programs (i.e., Home Instruction Program for Preschool Youngsters; HIPPY) and
family literacy programs (e.g., Even Start; St. Pierre et al., 1993). The available evidence suggests that both
approaches can be effective if
adequate levels of parental access and enthusiasm are ensured.
Access to books and printed material is also important in
the early childhood care and education environment (National Research Council,
1998). Preschool classrooms with play
settings that contain print and where children are encouraged to interact with
it are associated with better emergent literacy (Neuman & Roskos,
1993). Furthermore, the amount and
quality of one-on-one or small group interactions between teachers and children
is highly related to measures of language development (National Research
Council, 2000).
Early childhood care and education interventions suggest
that a combined approach of child-engaging book reading and phonological
training is effective in improving emergent literacy skills (National Research
Council, 1998; Whitehurst et al., 1994; Whitehurst & Lonigan, 1998). Teaching letter-sound knowledge has also
been shown to help children learn to read (National Research Council,
1998).
These findings suggest that interventions focused on aiding
low-income families to relocate to more affluent neighborhoods might improve
young children’s chances of school success.
In the Moving to Opportunity demonstration project, sponsored by the
U.S. Department of Housing and Urban Development, findings from the Baltimore
site indicate that families given housing vouchers restricted to low poverty
areas tend to move to suburbs or low poverty urban areas and, in doing so,
increase their children’s educational opportunities (Ladd & Ludwig, 1997). The alternate strategy of investing in new
businesses and industry in areas with high unemployment or providing
job-training and/or job-placement assistance for unemployed individuals should
also be evaluated for its implications for children.
Interventions related to television usage focus mainly on
parent behavior. Parents and other
adults can monitor the type and amount of television that young children use
and, by doing so, help shape children’s viewing habits and preferences (Hughes
& Hasbrouck, 1996; Truglio, Murphy, Oppenheimer, Huston, & Wright,
1996). Adults can mediate the effects
of television on children’s prosocial, creative and aggressive behaviors by
discussing and interpreting the behavior of characters on the shows children do
view (Huesmann, Eron, Klein, Brice, & Fischer, 1983). However, effective parental behaviors depend
on having accurate information about television programming and its effects on
children. Thus, another intervention
strategy includes educating parents and the general public on the effects of
television programming on children’s development. Research indicates that educational programs such as Sesame Street increase young children’s
letter and number knowledge, vocabulary, and positive attitudes towards school,
while cartoons and adult programming do not (Huston & Wright, 1996; Rice,
Huston, Truglio, & Wright, 1990; Wright & Huston, 1995). Prosocial programs such as Mr. Rogers’ Neighborhood, when combined
with related, reinforcing activities, increase preschoolers’ prosocial behavior
(Huston & Wright, 1998). Research
also finds that watching violent programming is associated with a decrease in
fantasy play among preschoolers and contributes to children’s aggressiveness
(Hughes & Hasbrouck, 1996; Huston-Stein, Fox, Greer, Watkins, &
Whitaker, 1981; Noble, 1970). Although
it is difficult to regulate the amount of violence on television, an
alternative strategy would be increasing the amount of educational and
prosocial television programming available for young children (Kunkel,
1998). This final intervention strategy
would involve legislation on the national level, but would first require an
increase in public awareness and demand for more educational and prosocial
programming for children.
Research on other, newer forms of media such as video and
computers is not as extensive as research on television (Wartella, O’Keefe,
& Scantlin, 2000). Such research
focusing on preschool age children is extremely rare at this time. One review of the literature on video games’
effects on children of all ages found that video games are not directly related
to psychopathology or academic performance (Emes, 1997). Computer use by children under the age of
three is not recommended, based on children’s developmental needs and abilities
(Haugland, 2000). However, some
research shows that computer programs, when combined with activities that
facilitate what the programs are trying to teach, can help 3- to 4-year-olds
develop a range of skills, including long-term memory, manual dexterity and
verbal skills (Haugland, 1992).
Currently, more information is needed about how the medium and content
of newer technologies affect child development and about how best to ensure
that children benefit from these new technologies.
As communities begin to initiate new or augment existing
school readiness efforts, it is essential that decision makers, funders and
other community leaders marshal the full measure of their resources. In particular, they can combine knowledge of
their particular community’s needs, resources and priorities with information
available from research. One important resource is the conceptual work carried
out by the National Education Goals Panel, building on the child development
and early education research. The
NEGP’s work on defining the components of school readiness and the uses and
misuses of readiness assessments (and more recent research building on this
work) are essential background information for any local initiative. The research base also provides a structure
for thinking about where to target community initiatives to strengthen children’s
school readiness (i.e., the child, family, school and/or neighborhood). Finally, the research provides examples of
effective initiatives which helped shape positive early school outcomes, as
well as promising directions for further initiatives. Building on a research base of what works, communities will be
able to put their resources to use more effectively in providing ready schools
and ready students.
In the table below, we summarize in very concrete terms the
most effective “investments” for school readiness, based on the literature reviewed
for this paper. In the left-hand
column, we have noted the level at which one may want to intervene, starting at
the level of the child (e.g., child health) and working outward from an
ecological perspective to the community-level factors and beyond. Activities or components of interventions
that the literature indicates have significant impacts on school readiness are
summarized in the column labeled “What Works.”
The “Mixed Reviews” column contains elements that have been shown to be
effective in some but not all studies, or have been found to be effective for
some groups of children but not for all.
Finally, the “Best Bets” column is for investments that would seem to be
important from a theoretical standpoint, but currently there is no evidence of
their effectiveness based on highly rigorous studies (e.g., experimental
control group studies or longitudinal, multivariate studies).
Indicators work carried out by the Knight Foundation in 26
communities makes clear that members of these communities place a high priority
on supporting the development of young children. As the Knight Foundation works with communities to find ways to
support children’s school readiness, two bodies of knowledge can help guide its
efforts:
·
The work of the National Education Goals Panel Goal 1
working groups, and other researchers, educators and policymakers, in conceptualizing school readiness, and
establishing principles for the
assessment of readiness in young children; and
·
The research on factors
in children’s early years of development that contribute to positive outcomes and diminish the risk of
unfavorable outcomes in the first years of elementary school.
In this background paper, we will refer to the first body of
work as focusing on Assessing school
readiness, and the second body of work as focusing on Investing in school readiness.
We make the assumption that as communities seek to strengthen children’s
school readiness, they will want guidance on these twin topics: Assessments and
Investments.
In Part I of this
background paper, we turn to the Assessment issue. We highlight key conclusions from the work of the National
Education Goals Panel on school readiness and from the related research. We focus on three core conclusions of the
National Education Goals Panel: (1) that school readiness is multifaceted, resting
on social/emotional aspects of children’s development and their health and not
only their cognitive development; (2) that family and community supports for
school readiness are essential to address in an assessment of readiness, along
with characteristics of the individual child; and (3) finally, that school
readiness is a reciprocal phenomenon, involving a process of engagement of the
child and family with the school. As
such, it is important to consider not only the readiness of the child but also
of schools to receive and educate children and especially children with diverse
backgrounds. We also summarize the key principles for appropriate assessment of
young children articulated by the National Education Goals Panel working
groups.
In Part II of
this background paper, we turn to the issue of Investments: how communities can
support children’s school readiness.
The Goal I Technical Planning Group of the National Education Goals
Panel focused on three kinds of community supports: provision of services for
children’s health, supports for parents as children’s first teachers, and
access for all children to high quality and developmentally appropriate early
care and education settings. This work
was groundbreaking in identifying school readiness as a shared responsibility
of the family and community and in articulating that all families and children should have access to basic
supports. Some have urged that we need
now to build on this pioneering work by going beyond this central set of three
kinds of community supports. For
example, it has been noted that diminishing violence in children’s
neighborhoods could also potentially contribute to children’s school readiness
and indeed could be especially salient as a contributing factor in some
communities (Love, Aber & Brooks-Gunn, 1994).
In Part II, we include but go beyond the three core kinds of
community supports considered by the National Education Goals Panel. In particular, we adopt a broader ecological
framework, considering (1) factors related to the child, specifically his/her
physical and mental health; (2) factors related to the family context,
(including parents’ psychological well-being, parent-child relations, family
structure and family socioeconomic status); (3) factors related to children’s
experience of early childhood care and education settings; and (4) factors
related to the neighborhood context.
We present a selective review of the research in Part II,
emphasizing studies that are rigorously implemented experimental evaluations of
interventions, and studies that are longitudinal in design and that use
multivariate analyses to account for background characteristics of families.
Throughout Part II, our aim is to go beyond the broad identification of which
factors appear to be linked to early school outcomes, to the identification of specific strategies (kinds of programs and
activities within these programs) that have been attempted and evaluated,
and for which there is evidence that initiating programs with these activities
has the potential to contribute to improved early school outcomes.
A set of National Education Goals for the United States was
articulated in 1990 by the Governors of all 50 states and the President. The
first of these goals (Goal 1) was that “by the year 2000 all children in
America will start school ready to learn” (Kagan, Moore & Bredekamp, 1995).
There has been much debate about the particular words chosen
in Goal 1 (Lewit & Schuurmann Baker, 1995). For example, a position statement of the National Association for
the Education of Young Children (NAEYC) focused on the appropriateness of the
phrase “ready to learn,” noting that “every child, except in the most severe
instances of abuse, neglect, or disability, enters school ready to learn school
content” (NAEYC, 1998, p. 2). NAEYC
urged that, rather than readiness to learn,
the focus be instead on circumstances that foster children’s readiness for school or that place children at risk for academic failure. Love, Aber and Brooks-Gunn (1999) note a
similar suggestion by the Kauffman Foundation to focus on “the importance of
all children being ready to succeed in the schools they enter” (Love et al.,
1999, p.1). Meisels (1999) emphasizes
the contribution of the school as well as the child in proposing the following
revision: “By the year 2000 all children will have an opportunity to enhance
their skills, knowledge, and abilities by participating in classrooms that are
sensitive to community values, recognize individual differences, reinforce and
extend children’s strengths, and assist them in overcoming their difficulties”
(p. 63).
While the wording of the Goal 1 sentence has been
scrutinized and questioned, few question the progress that has been made
through the subsequent papers produced by the National Education Goals Panel
Goal 1 working groups, and in further research, in conceptualizing school
readiness, and laying out principles for assessing it. The Goal 1 working groups set about their
tasks at an important turning point in the research literature. As summarized by Shepard, Taylor & Kagan
(1996), a 1988 survey undertaken by the National Academy of Sciences and
National Association of State Boards of Education regarding testing practices
for prekindergarten and kindergarten children in all 50 states led to the
conclusion that “many early childhood specialists were well aware that the
testing practices… did not meet professional standards for either sound
measurement or educational decisions” (p. 3).
The survey found IQ-like tests, with a narrow cognitive focus, being
used to make grade placement and retention decisions for young children. There was a lack of evidence of reliability
and validity for some of these tests.
Further, the tests were being used for placement decisions despite
questions about the stability of children’s scores at these young ages. The survey also found inappropriate use of
screening tests (specifically the use of these tests in guiding instruction and
making individual placement decisions, rather than as intended, to make a
decision about the need for in-depth assessment for a possible health,
emotional or learning disability).
Efforts were taken to stop such testing practices for young
children in the late ‘80s, with joint statements by professional organizations,
researchers, and practitioners against the use of testing to deny age-eligible
children entry into kindergarten. A
survey of states regarding their early childhood assessment policies and
practices carried out in 1995-6 found progress away from the use of testing to
determine kindergarten entry and retention, and a diminution in the
inappropriate use of screening instruments (Shepard et al., 1996).
With these steps taken against a very narrow view of school
readiness and against inappropriate assessment practices for young children, it
nevertheless remained to articulate a positive conceptualization for what comprised school readiness in
children, what community supports were central foundations for school readiness
and how to measure these.
The Goal 1 Resource and Technical Planning Groups were asked
to suggest data by which progress towards Goal 1 could be measured. Two kinds of information were recommended:
first, direct indicators of the Goal, reflecting how children are doing and
capturing their welfare and resiliency in the face of risk; and second, direct
indicators of the objectives, reflecting the quality of services afforded young
children. The working groups also subsequently extended the focus to encompass
the issue of the readiness of schools for children.
Below we provide summaries of the progress made by the Goal
1 working groups toward delineating how to measure three different components of
school readiness: (1) readiness in the child; (2) family and community supports
and services that contribute to children’s readiness; and (3) schools’
readiness for children. For the first
and last, the working groups identified the need for conceptual/definitional
work that could lay the groundwork for the development of measures. That is,
the state of research and practice did not yet permit identification of
specific measures, but providing a well-articulated conceptualization would
permit movement toward the selection and development of appropriate measures.
Regarding family and community supports (#2), however, the working groups were
able to suggest and also to provide data for specific measures.
After summarizing the progress made by the Goal 1 working
groups in conceptualizing the three components of readiness, we then turn to an
overview of the guidance provided by the working groups on principles of
assessment in young children. In 1998 the Goal 1 Early Childhood Assessments
Resource Group responded to a specific request by Congress to “create clear
guidelines regarding the nature, functions and uses of early childhood
assessments, including assessment formats that are appropriate for use in
culturally and linguistically diverse communities, based on model elements of
school readiness” (National Education Goals Panel, 1998a, p.3). This document provides specific
recommendations for appropriate assessment practices and for how to guard
against specific abuses that have occurred in the past. With its clear recommendations to
policymakers and educators, this document, too, can provide important
background for the work of the Knight Foundation in specific communities.
The Goal 1 working groups saw their task as laying out a conceptualization for school readiness
in children that could then guide the
selection (and/or development) of specific measures for assessment. They identified five dimensions of readiness
in children’s school readiness, and provided a detailed definition of each of
the dimensions.
This conceptualization has, in fact, subsequently guided
measures development in two major national studies: the Early Childhood
Longitudinal Study-Kindergarten Cohort (a longitudinal study of a large
representative sample of children who entered kindergarten in 1998, following
the children’s development over the early years of schooling), and the Family
and Child Experiences Survey (documenting the experiences and development of
children in a representative sample of Head Start Programs) (see descriptions
in Forgione, 1998; and Love et al., 1999).
This conceptualization has also guided efforts to measure school
readiness at both the state and community level. As one example of efforts at the state level, North Carolina is
working with its legislature on selecting particular items reflective of school
readiness in children that map onto the five dimensions of children’s readiness
identified by the Goal 1 working groups (Maxwell, 2000). At the community level, Love, Aber and
Brooks-Gunn (1994, 1999) provide an extremely helpful and detailed “blueprint”
for selecting measures of each of the five dimensions of children’s school
readiness noted by the NEGP, for use in community monitoring of children’s
readiness.
As we have noted, a fundamental principal underlying the
conceptual work in defining the dimensions of children’s school readiness was
the view that “narrowly constructed, academically-driven definitions of
readiness—heretofore widely accepted—need to be broadened to incorporate
physical, social and emotional well-being” (Kagan et al., 1995, p.1). Beyond this fundamental principle, the
conceptual work was also guided by the following assumptions: (1) that
assessment of young children’s readiness would be intended primarily to inform
public policy regarding the collective well-being of children, rather than to
make decisions regarding particular programs or the educational course of
particular children (see discussion below of the multiple possible purposes of
assessment, and the need to match assessment tools and data collection
strategies to the particular purpose).
As such, it was assumed that data would be collected via a sampling of
all children in a geographical unit of interest; (2) that it would be important
to carry out assessments at multiple points in time (before children entered
school, at the point of school entry, after a period of time in school) rather
than a single time point; (3) that to capture children’s school readiness it
would be important to rely on multiple informants (teachers, parents, the
children themselves); (4) that multiple measurement strategies should be used
rather than any single measure; and (5) that it is inappropriate to develop a
single “threshold” measure of readiness:
Because individual
child performance is multi-dimensional, highly variable across the dimensions,
episodic, and culturally and contextually influenced, the establishment of any
single “readiness” threshold is misleading and dangerous. Rather than advocating any such
developmental threshold or a uni-dimensional approach to assessment, this
effort must be understood as one that attempts to amplify the range of
variables associated with early development and learning (Kagan et al., 1995,
p. 6).
The multifaceted conceptualization of school readiness
developed by the Goal 1 working group, and which continues to guide assessment,
program and policy work, included five dimensions: (1) physical well-being and
motor development; (2) social and emotional development; (3) approaches toward
learning; (5) language development; and (5) cognition and general
knowledge. These dimensions were viewed
as interrelated rather than independent.
Children’s backgrounds, in terms of both culture and exposure to a range
of contexts (such as early childhood care and education settings), were assumed
to affect manifestations of behavior for each dimension.
Below we briefly summarize the rationale for including each
dimension in the conceptualization and note the key components identified for
each, drawing on the document “Reconsidering Children’s Early Development and
Learning: Towards Common Views and Vocabulary” (Kagan et al., 1995).
(1) Physical well-being and motor development. Children who are healthy can more readily
focus on learning. Problems with health or physical development may impede the
learning process by resulting in absences from school, discomfort or the need
to make special accommodations. Children with significant problems with health
or physical development may also face special challenges in terms of their
self-perception in adapting to the school setting and in terms of developing
independence within the school setting.
This dimension encompasses three components, each with
subcomponents:
- I. Physical
development: (A) rate of growth
in terms of height, weight gain, and physical maturation; (B) physical fitness as manifested in stamina, strength,
flexibility, percentage of body fat; and (C) body physiology,
or health/ill-health of the body and organ systems in terms of diseases
(e.g., asthma, ear infections) and disabilities (e.g., hearing loss, poor
vision, neuro-motor disorders).
- II. Physical
abilities: (A) gross motor
skills (e.g., walking, running, jumping and climbing: movements
involving the whole body); (B) fine
motor skills (e.g., activities involving cutting, writing: movements
requiring manual dexterity and precision); (C) sensorimotor skills (integration of sensory information with
movement, requiring vision, hearing, touch and kinethesis); (D) oral motor skills (production of
speech sounds through coordination of breathing and movements e.g., of
mouth, tongue and lips); (E) functional
performance (e.g., feeding and dressing oneself; having independent
mobility and bowel and bladder control).
- III.
Background and contextual conditions of development: (A) perinatal context (exposure to
adverse conditions in utero and around the time of birth, including low
birth weight, preterm birth, exposure to such substances as alcohol, tobacco
and drugs); (B) caregiving
environment (exposure to environmental factors and risks during the
early years, including accident hazards, exposure to toxic substances,
exposure to violence, health practices and health status of parents and
siblings, cohesion of family and social supports available to family); and
(C) health care utilization
(availability of prenatal care, and health promotion and disease
prevention services, including access to and utilization of primary care
services).
(2) Social and emotional
development. Social and
emotional development, while closely related, are distinguished from each other
in that social development reflects the child’s social interactions, for
example with teachers and peers. By
contrast, emotional development pertains to the child’s “feeling states
regarding the self and others” (Kagan et al., 1995, p. 18).
Learning in school occurs through interactions with teachers
and peers. Positive adaptation to
school requires of children such social behaviors as the ability to take turns,
to work cooperatively in a group, to show empathy toward others and
assertiveness (e.g., asking questions, assumption of leadership roles) without
aggressiveness. Positive self-concept
and the ability to interpret one’s own feelings and those of others contribute
to positive interactions and engagement in learning.
- Social
development: The formation of positive relationships with teachers and
peers in the classroom draws on children’s early experiences of
relationships with parents and other family members, yet is also
malleable, influenced by experiences within school. Children’s social competence with peers
is seen as including both the ability to interact and cooperate with
peers, and the ability to form and sustain relationships.
- II.
Emotional development: Components of emotional development that are
important to adjustment and learning in school include self concept
(perceptions of the self), sense of self-efficacy, ability to comprehend
others’ emotions and respond appropriately to them, and ability to
interpret and express one’s own feelings.
(3) Approaches
toward learning. A child may
have skills, knowledge and abilities, but may or may not marshal these to bring
them to bear on the tasks at hand.
Approaches toward learning concern the inclination to use skills,
knowledge and capacities. The Goal 1 working groups saw two key components to
this dimension of school readiness: predisposition (less malleable: either
inborn or developed very early), and learning styles (more malleable:
reflecting how children address the learning process attitudinally).
- Predispositions:
(A) Gender: gender, along with
temperamental characteristics, are assumed to have both a genetic
component and to reflect influences of environment and culture. There is evidence of differences in
parent and teacher expectations of boys and girls. Gender may also predispose children to
interest in particular subjects.
(B) Temperament: temperamental characteristics such as
shyness can influence approach to social situations in the classroom as
well as “the way individuals think, perceive, understand, judge and solve
problems” (Kagan et al., 1995, p. 24).
(C) Cultural patterns and
values: Cultural patterns predispose children to being familiar with,
and comfortable with, certain modes of interaction in the classroom. For example, there are cultural
differences in the degree to which children are expected to listen to, and
receive guidance from, adults as opposed to interacting and
questioning. Similarly, there are differences
in the degree to which children are comfortable with working independently
and with differences in the preferred modality or approach to learning
tasks (e.g., manipulating materials, visual representations, verbal
discussion).
- II.
Learning styles: (A) openness
to and curiosity about new tasks and challenges (e.g., “whether
children are inquisitive or accepting, exuberant or passive,” p. 25); (B) initiative, task persistence and
attentiveness (the ability to plan and undertake a task and then follow
through to its completion despite interruptions or distractions); (C) reflection and interpretation (“the
ability to draw out lessons for future use,” p. 26); (D) imagination and invention (“the
ability to form images of what is not actually present, to extend
conventional thinking beyond the known and to combine previous experiences
to form new ideas,” p. 26); (E) cognitive
styles (e.g., using broad vs. narrow categories of information;
tendency to respond to the first reasonable solution to a problem or to
evaluate alternatives; “tendency to separate details from background and
to analyze information,” p. 26).
(4) Language
Development. Language is
fundamental to the ability to communicate with teachers and peers in the school
setting. It is also critical in
cognitive development in providing a means of conveying concepts. “By
definition, language development is the acquisition of linguistic forms and
procedures and social rules and customs for acts of expression and
interpretation. Such knowledge has
three essential components: content (meaning), form (structure) and use
(function)” (p. 29). The Goal 1 working
groups distinguished between verbal language and language related to emergent
literacy.
·
I. Verbal language: (A) listening (e.g., ability to identify sounds use sounds to form
words, follow directions given orally); (B) speaking
(e.g., ability to produce range of sounds, apply grammatical rules) (C) social uses of language (e.g., use of
social conventions within speech; use of language in a manner appropriate to
context); (D) vocabulary and meaning
(e.g., use of words used and understood); (E) questioning; (F) creative
uses of language (e.g., tell a story).
·
II. Emerging literacy: (A) literature awareness (e.g., recalling familiar stories); (B) print awareness (e.g., understanding
that text remains same across readings, assign sounds to letter combinations);
(C) story sense (e.g., awareness of
beginning, middle, end of a story, and of the permanence of story sequence);
(D) writing process (e.g., producing
ordered scribbling).
(5) Cognition and
general knowledge. This
dimension comes closest to the public perception of the kinds of early learning
important for schooling: mastery of such specific knowledge as names of letters
and shapes. Yet this public perception also restricts cognition and general
knowledge to only one of three aspects identified by Piaget: social
conventional knowledge or mastery of information that rests on conventions of
the particular society (e.g., consonants and vowels in the English
language). Piaget also identified
physical knowledge and logico-mathematical knowledge as important, and forms of
knowledge that the child actively constructs through interaction with the
physical world and others, and these were identified as important aspects of
children’s school readiness.
·
I. Physical knowledge: knowledge about the
properties of objects derived from observation and interaction (e.g., the way
in which an object acts on an inclined plane).
·
II. Logico-mathematical knowledge: knowledge
related not only to physical properties of individual objects but also to
similarities, differences and associations looking across objects, events or
people.
·
III. Social-conventional knowledge: knowledge
that “reflects the agreed-upon conventions of society” (p. 37), for example,
letters of a particular alphabet.
Teacher surveys as
sources of confirmation for a multifaceted view of school readiness. A 1993 survey of kindergarten teachers (the
Kindergarten Teacher Survey on Student Readiness), conducted by the National
Center for Education Statistics, provided evidence that teachers perceive
aspects of school readiness other than Cognition and General Knowledge as
important (Lewit & Schuurmann Baker, 1995; Meisels, 2000; NCES, 1993). In this survey, teachers were most likely to
rate as important aspects of readiness for school that children were physically
healthy, rested and well nourished; able to communicate needs, wants and
thoughts verbally; and that children be enthusiastic and curious in approaching
new activities.
In a more recent survey of 3,600 kindergarten teachers
(Frank Porter Graham Child Development Center, 1999c), substantial proportions
of teachers indicated concern about the skills of children entering
kindergarten. Their focus included but
went beyond academic skills. For example, while 36 percent of the sampled
teachers indicated that about half or more of their class entered kindergarten
with problems relating to academic skills, 46 percent indicated problems with
following directions, 34 percent with working independently, 30 percent with
working in a group, 20 percent with immaturity and 14 percent with
communicating. These concerns “map” onto the school readiness dimensions of
Approaches to Learning, Social and Emotional Skills and Language Development as
well as Cognition and General Knowledge.
Recent Focus on
Social and Emotional Skills Needed for Early School Success. A recent set of two papers (Cavanaugh,
Lippitt, & Moyo, 2000; Huffman, Mehlinger & Kerivan, 2000) and an integrative
monograph based on these, commissioned by the Child Mental Health Foundations
and Agencies Network (FAN), provides a strong focus on the issue of children’s
social and emotional development as bases for early school success. According to the monograph, “children who
are socially and emotionally ready for school generally have improved school
outcomes, better odds of later school and vocational success, better later
social and emotional development and an easier time developing relationships with
their peers” (Peth-Pierce, 2000, p. 2). At the same time, these papers indicate
concern with the number of children entering school without the social and
emotional skills important to positive early school outcomes.
Huffman and colleagues identify risk factors for difficulty
in transitioning to kindergarten. Some
of the risk factors were malleable, and evidence indicated that when changed,
school outcomes for children improved (labeled “causal risk factors”). By contrast, other risk factors either were
not malleable or, when changed, did not alter school outcomes. Among the seven that appeared to be causal,
it is noteworthy that three pertained to the quality of early relationships
(poor parenting practices, difficulties with peer relationships, poor
relationships with teachers), and a further two pertained to social and
emotional issues in the child or parent (early behavior and adjustment
problems, and parental psychological problems). The remaining risk factors
considered causal were cognitive deficits in the child and age at school
entry.
Cavanaugh and colleagues, in reviewing current federal
programs that address risk factors for difficulty in the transition to school,
come to the conclusion that research is not being used sufficiently to direct
federal resources in early childhood policy. For example, while the research
highlights difficulties in early relationships as predictive of problems in the
transition to school, inadequate resources are aimed at addressing developing
and maintaining early positive relationships with adults, particularly for
children in families with fewer socioeoconomic resources. The authors do point out that many states
are engaged in new and innovative efforts to address school readiness and that
some states are moving toward integrated, comprehensive service systems for
young children. While this may result
in fifty different approaches, a comprehensive national strategy is still
missing.
This pair of reviews calls for more research on the social
and emotional skills that foster a positive transition to school, risk and
protective factors for these skills, and for a greater focus on interventions
and policies that can promote such skills in young children.
In addition to articulating Goal 1, the National Education
Goals Panel identified three “Objectives.” These objectives reflect early
experiences that were seen as important supports or foundations for school
readiness (NEGP, 1997b, p. 1):
(1)
All children will have access to high-quality and
developmentally appropriate preschool programs that help prepare children for
school.
(2)
Every parent in the United States will be a child’s first
teacher and devote time each day to helping such parent’s preschool child
learn, and parents will have access to the training and support parents need.
(3)
Children will receive the nutrition, physical activity
experiences and health care needed to arrive at school with healthy minds and
bodies and to maintain the mental alertness necessary to be prepared to learn,
and the number of low birth weight babies will be significantly reduced through
enhanced prenatal health systems.
As Kagan and colleagues (1995) point out, Goal 1 and the 3
objectives, taken together, provide a view of early experiences and school
readiness as linked. They also note
that providing supportive experiences for young children is a responsibility of
both the family and the larger society and further clearly state that all children should have access to
supportive early experiences.
The National Education Goals Panel acknowledged that when
they undertook their work, there were
as yet no direct measures of children’s school readiness (although substantial progress has occurred subsequently).
However, selected indicators of the 3
objectives were already available in national datasets. Further, uniform and comparable measures of
selected health indicators (though
not measures of parenting practices or participation in early childhood care
and education) were available at the state level as well. The Goals panel
reported annually on a limited set of measures for the objectives and provided
more in-depth reporting in 1997 (with a further in-depth report expected in
2000).
Regarding Objective #3, the health objective, the
1997 report (NEGP, 1997b) provided data on:
- the
point at which mothers first began prenatal care;
- the
percentage of births above and below 5.5 pounds (“low birth weight”) and
3.3 pounds (“very low birth weight”);
- a
Children’s Health Index (percentage of U. S. births with no risk, 1 or
more, 2 or more, 3 or more, with the risk items consisting of third
trimester or no prenatal care, low maternal weight gain, mother smoked
during pregnancy and mother drank alcohol during pregnancy); and
- the
percentage of 2-year-olds with completed basic immunization series for
selected diseases.
These indicators show that in 1995, 4 percent of mothers did
not begin prenatal care until the third trimester or received no prenatal care
at all. There were differences by
race/ethnicity (the figures were 2 percent for white mothers, 7 percent for
Hispanic, and 8 percent for black). The percentage of births below 5.5 pounds
(the standard for low birth weight) was 7 percent, with black infants twice as
likely as those from other racial/ethnic groups to be born at low birth weight
(10 percent vs. 5 percent for white and Hispanic infants). In 1995, 34 percent of all infants born in
the United States had one or more factors on the Children’s Health Index
considered to place them at risk in terms of their long-term health and
educational development. There was a decrease from 1990 to 1995 (from 37
percent to 34 percent) in the percentage of infants with one or more risk
factors. American Indian/Alaskan Native infants and black infants were more
likely to have one or more risks than other racial/ethnic groups. Slightly more than three-quarters of all
2-year-olds had been fully immunized in 1996.
At the national level, black children and low-income children were less
likely to be fully immunized than others.
The availability of state-level data permit examination of
the range in state statistics (NEGP, 1997b).
For example, while in 1995, 81 percent of all mothers in the United
States began prenatal care during the first trimester of pregnancy, this ranged
across states from 60 percent to 90 percent.
Across states, the percentage of infants born at low birth weight ranged
from 5 percent to 13 percent. There was
a substantial range across states in the percentage of children with one or
more health risks, from 24 percent to 42 percent. While 88 percent of 2-year-olds were immunized in the best
states, 64 percent were immunized in the worst states from the point of view of
this statistic. Thus, the state level
data on health indicators reveal substantial variability.
Regarding Objective #2, family-child activities
supportive of school readiness, the report presented indicators of:
- The
percentage of children whose parents read to them every day;
- The
percentage of 3- to 5-year olds whose parents engaged in language and
literacy activities with them regularly (with activities including telling
the child a story 3 or more times in the previous week, and visiting a
library or more times in the previous month); and
- The
percentage of parents of 3- to 5-year-olds who participated in parent
support activities, including attending a parenting class or any support
group to help with parenting since the beginning of the school year; ever
going to a family support center; or receiving more than one home visit
from someone trained to talk about raising children.
National data indicate that in 1995-6, 45 percent of
children between 0 and 2 were read to daily by parents or other family members,
while the parallel figure is 56 percent of children between 3 and 5. These percentages varied by level of
parental education, with parents who had completed high school more likely to
report reading to their preschool-age children. The percentage of 3- to 5-year
old children who were told stories several times a week was 55 percent, while
37 percent visited a library once or more in a month. Both types of literacy activities were reported more frequently
by parents with higher levels of education. The percentage of parents of 3- to
5-year-old children reporting participation in parent support activities was
relatively low, ranging from 13 percent reporting having ever gone to a family
support center, to 8 percent reporting ever receiving a home visit from someone
trained to talk about raising children.
Regarding Objective #1, participation in early childhood
care and education experiences, the 1997 report provided indicators of:
- The
percentage of children enrolled in preschool;
- Markers
of child care quality for preschool centers, including percentage of
teachers/caregivers with child-related training; the percentage of centers
not exceeding the maximum acceptable group size for children; and the
percentage of centers not exceeding the maximum acceptable child/staff
ratios as recommended by the NAEYC; and
- Markers
of child care quality for home-based preschool settings, including
percentage of regulated family day care providers with specific forms of
training and the percentage of home-based settings meeting the standard
for group size for different age mixes of children.
The report notes that the proportion of young children
participating in preschool settings varied with household income, from 79
percent of 3- to 5-year-olds from families with household incomes of more than
$75,000, to 43 percent for children from families with incomes of $10,000 or
less. The report found differences in
the percentage of centers meeting group size and ratio recommendations
according to child age, with more centers likely to meet the recommendations
for 3- to 5-year-olds than infants and toddlers. In 1990, a smaller percentage of caregivers in home-based than in
center-based settings had child-related training and a Child Development
Associate credential. Home-based
settings were more likely to meet recommended group sizes when children were of
similar ages than when they were of mixed ages in the care setting.
The report concludes that “current conditions for young
children are far from ideal. Too many
begin life with avoidable health risks.
Too few are regularly engaged in supportive activities at home with
their families. And far too many do not
have the opportunity to participate in high quality early care and education
programs in safe, caring environments that support their continued development”
(NEGP, 1997b, p. 32). The report also
showed differences in particular indicators by race/ethnicity, parental
education, household income and across states, suggesting limitations in the
extent to which the conditions supporting school readiness are available to all
children.
Shepard and colleagues (1996) report findings regarding state-level reporting on initiatives to
provide the kinds of supports noted in the three objectives and to provide
measures of child and family exposure to the supports identified in the three
objectives. They conclude that
“although many states may have invested in early childhood programs in recent
years, the majority of states do not have initiatives specifically identified
with Goal 1.” Their survey found fewer
than 12 states with reporting mechanisms in place to report on progress towards
the Goal 1 objectives. While the annual
Kids Count report of the Annie E. Casey Foundation does provide state-level
data on child health indicators, the survey conducted by Shephard and
colleagues indicated limited awareness by state early childhood specialists
that health data were linked to Goal 1 and the related objectives. In general, their paper suggests a need to
expand awareness of the three objectives as linked to Goal 1. They also identify the need for work toward
the development of state-level indicators of the objectives.
In recent years, the U.S. Department of Health and Human
Services has provided funding and guidance to a set of states to develop
state-level indicators of school readiness and of child care. There has been substantial progress in these
states, particularly in developing indicators of the use and quality of early
childhood care and education settings (specific indicators being developed and
used by these states are presented in the summary from the project meeting in
Providence, Rhode Island; HHS/ASPE Child Indicator States, 1999). There is some convergence across the states
in terms of what aspects of child care participation and quality are being
measured (Zaslow, 2000). For example,
multiple states are working towards indicators of the proportion of eligible
families receiving child care subsidies, the proportion of children eligible
for early childhood intervention programs who are participating in such
programs, and the proportion of child care centers and family child care homes
that have accreditation.
The Goal 1 working groups of the National Education Goals
Panel also focused on the issue of schools’
readiness for children:
“The Technical
Planning Group strongly believes that child outcomes should not and cannot be
the sole measure of America’s progress toward the first Education Goal. While specifying child outcomes is a
necessary step, it is an insufficient approach. It must be coupled with a commitment to examining social and
institutional readiness to support children’s early development and
learning.
“To that end, the
Technical Planning Group strongly urges that energy be devoted to examining the
readiness and capacity of the nation’s schools to receive young children.”
(NEGP, 1995, p. 41)
In order to address the reciprocal nature of readiness as a
characteristic of both child and school, the National Education Goals Panel
convened a Ready Schools Resource Group.
This group used existing research and knowledge about educational
practices to identify “broad strategies that school and community leaders may
want to consider as they work to strengthen the transition to school and
learning…” (NEGP, 1998b, p. 3). This document does not suggest specific
measures for assessing the readiness of schools, but does begin to identify the
characteristics of ready schools.
Further work is warranted for moving from conceptualization to the
identification of specific measures of schools’ readiness.
The Ready Schools Resource Group identified ten
characteristics of ready schools:
(1) Ready schools smooth the transition between
home and school. For example, they show sensitivity to the cultural gap
that children from low-income and minority families may face in entering
school. They engage in written or
personal outreach practices to help parents and children make the transition to
a new school setting.
(2) Ready schools strive for continuity between
early care and education programs and elementary schools. There may be
substantial differences in level of parent involvement, organization of the
classroom and teaching styles between the elementary school and children’s
early care and education settings. “…Contact with previous caregivers can
facilitate planning for individual students, provide a sense of continuity for
children and parents and allow a better alignment of philosophy, expectations
and curriculum across institutions and the community” (p. 8).
(3) Ready schools help children learn and make
sense of their complex and exciting world through a range of curriculum and
instructional approaches, through high quality instruction, through appropriate
pacing and content of instruction, incentives for learning, effective use of
time and through understanding that learning occurs in the context of
relationships.
(4) Ready schools are committed to the success
of every child by being responsive to the needs of individual children,
providing a setting that is conducive to learning and exploration, showing
awareness of the impact on children of poverty and race, ensuring language
minority children age-appropriate and culturally sensitive instruction and
attempting to meet special needs within the regular classroom, where possible.
(5) Ready schools are committed to the success
of every teacher and every adult who interacts with children during the school
day. They are committed to giving teachers time to develop their skills and
prepare for instruction.
(6) Ready schools introduce of expand approaches
that have been shown to raise achievement. They provide prompt and
supportive intervention to children who are falling behind. They encourage meaningful parent
involvement. They monitor different approaches to classroom organization,
staffing and grouping, and retain and expand approaches that are beneficial to
children.
(7) Ready schools are learning organizations
that alter practices and programs if they do not benefit children, for
example practices concerning retention in kindergarten, redshirting, denying
school entry and accelerating the academic instruction of young children.
(8) Ready schools serve children in communities,
helping to assure access to a range of services and supports in the
community.
(9) Ready schools take responsibility for
results. They carry out assessments to help teachers and parents plan for
the needs of individual students but also for purposes of accountability to the
community. They share data with the
appropriate individuals or groups. “Ready schools organize testing and
screening strategies that are fair and valid and do not, intentionally or
unintentionally, devalue cultural differences” (p. 27).
(10)
Ready schools have
strong leadership. A leader or group of leaders have a clear agenda, have authority to make
decisions and resources to follow through on goals, are visible and accessible.
The national surveys noted above (the Early Childhood
Longitudinal Survey-Kindergarten Cohort, and the Family and Children’s
Experiences Survey) each include items pertaining to the characteristics of the
educational or program setting. These
studies could provide important resources for moving from the conceptualization
of ready schools to the identification of a core set of measures of schools’
readiness that could be used in national, state and community monitoring
efforts.
As we have noted, there is a great deal of activity
currently going on at the state level in developing strategies for monitoring
children’s school readiness. Some
states have chosen to make explicit the view that school readiness involves a reciprocal
relationship between children and schools. North Carolina provides one
example. In working toward a limited
set of school readiness indicators to track over time, current plans in the
state call for measures of schools’ readiness as well as children’s readiness
(Maxwell, 2000; Meisels, 2000). Indeed,
the states’ plan to track school readiness is portrayed visually as a puzzle
with two pieces: condition of children and condition of schools. Measures of schools’ readiness being
considered include kindergarten class size, percentage of kindergarten teachers
with early childhood certification and percentage of elementary school
principals who have completed at least one early childhood course or who have
an early childhood degree.
We have noted that the history of misuse of early
childhood assessment was one of the motivating factors behind the work of the
NEGP Goal 1 working groups. At the
request of Congress, the NEGP worked to articulate a set of principles for the
assessment of young children and to identify safeguards against the misuse of
such assessments in the future. The set
of principles articulated by the Goal 1 Early Childhood Assessments Resource
Group (NEGP, 1998a) is an important resource for the efforts of Knight
Foundation in working with communities on school readiness issues. A further
important resource is an extremely insightful recent review of issues in the
assessment of school readiness by Meisels (2000).
Fundamental to NEGP’s work on this issue is the idea that
there are four different purposes for the assessment of young children. The
four purposes for assessing young children are: (1) assessments to support
learning; (2) assessments for the identification of special needs; (3)
assessments for program evaluation and monitoring trends; and (4) “high-stakes
accountability” assessments, used to make decisions about individual children,
teachers, classrooms or programs. The
NEGP stresses the importance of using assessments for their intended purpose
and suggest great care in attempting to use any one strategy of assessment for
more than one purpose. Further, the
purpose of assessment shapes the content of the assessment, the method of
administration, the technical requirements for the assessment in terms of
reliability and validity, the audience for the information from the assessment
and whether or not it is appropriate to use the information from the assessment
in making decisions about individual children, teachers or programs. Safeguards against misuse also differ by
purpose.
For all four purposes, the distinct aspects of young
children’s development need to be recognized. For example, growth is more rapid
in the period from birth to age eight than at other periods of
development. Development may not occur
at a steady pace, but rather in spurts.
This may make it difficult to capture skills and abilities at any one
point in time. In addition, young children learn and manifest their learning in
different ways than older children. For
example, they tend to learn by doing rather than just listening and may best
manifest what they know in actions rather than in speech or writing. These tendencies make the needs for
assessment of young children very different from those for older children or
adults.
General principles identified by the NEGP are that (1)
“assessments should bring about benefits for children” either individually in
terms of direct services to a child or more collectively in terms of improved
instruction overall; (2) “assessments should be tailored to a specific purpose,
and should be reliable, valid, and fair for that purpose”; (3) “assessment
policies should be designed recognizing that reliability and validity of
assessments increase with children’s age.” Assessments for some purposes should
be postponed until children are older; (4) “assessments should be
age-appropriate in both content and the method of data collection”; (5) “assessments should be linguistically
appropriate, recognizing that to some extent all assessments are measures of
language”; and (6) “parents should be a valued source of assessment
information, as well as an audience for assessment results” (p. 5).
Purpose 1: Assessments to promote children’s learning
and development. Assessments to
shape instruction for individual children occur on an ongoing basis to identify
what children know and can do, and what they should proceed to in their
learning. Assessments for this purpose
draw on and are embedded within the content of the curriculum. They can be collected through direct
observation, collecting samples of children’s work, asking children questions,
and asking questions of informed adults.
Assessment involves observing patterns over time. Technical requirements
for this assessment purpose are the least stringent. While teachers are the
primary “audience,” children and parents themselves can benefit from being
informed regarding this type of assessment.
Also, school administrators can use the information for internal
accountability (to make sure programs are succeeding and that children who need
it are receiving help). The NEGP recommends that all five dimensions of
children’s school readiness be included in such assessments.
In terms of the need for safeguards with respect to this
kind of assessment, the NEGP Goal 1 resource group notes the need to assure
that the learning goals are age appropriate.
They further note that teachers may need tools to assist them in observing
and monitoring the progress of children.
Without such tools, they may use other forms of assessment, especially
screening tools, inappropriately to shape the course of instruction for
individual children. Individual
teachers, schools or districts may lack the resources to develop benchmarks of
progress that are linked with elements of the curricula. Coordination at the state level on curricula
and closely aligned assessments can be extremely helpful.
There have been some interesting recent developments,
described in greater detail below, in the state-level implementation of
assessments for Purpose 1. In
particular, several states have chosen to implement sections of Meisel’s Work
Sampling System, with state-level training of teachers in completing the
procedures and collection and use of the Work Sampling information for every
child in kindergarten and other grades (Meisels 2000).
Purpose 2: Identifying children for health and special
services. Assessment for this
purpose is a first step in the identification
of special problems (such as blindness, deafness, physical disabilities, speech
and language impairment, learning disability, mental retardation) and the need
for special services. Screening
determines if the second step, in-depth assessment by specialists, is needed.
The audiences for such assessments are parents and adults working with
children. Regarding technical
requirements, assessments for this purpose are brief, “fallible,” cost
effective and not administered by a specialist. These characteristics are tied to the fact that screening should
be used to signal the need for follow-up. Serious misuse occurs when such brief
screening measures are themselves used to determine the presence of problems or
need for services without the appropriate follow-up. Misuse also occurs if screening measures are used for
instructional decisions for children (Purpose #1).
Purpose 3:
Monitoring trends and evaluating programs and services. While assessments for purposes 1 and 2 guide
decisions about individual children, assessments for purpose 3 inform aggregate
decisions regarding educational and social programs. Such assessments are used
to monitor trends and evaluate programs and have policymakers as their
audience. According to the Goal 1
resource group, assessments with this purpose should monitor the adequacy of
support services as well as child outcomes (reflecting the five dimensions of
school readiness in tracking child outcomes).
The consequences are for broad programs and policies, not for individual
children, teachers or individual/local programs. There are high standards for technical accuracy, though the
aggregate nature of the data (the fact that they are not used to inform
decisions for individuals) help with the stringency of reliability and validity
requirements.
A pitfall here is the use of assessments collected for the
purpose of monitoring instead to inform decisions about individual children,
teachers, classrooms or local programs, despite the fact that technical
requirements for reliability and validity, having aimed these for aggregate
level data, would not make this appropriate.
A safeguard is using sampling strategies, so that data are not available
for every child. Matrix sampling
(sampling so that individual children receive only parts of the assessment) is
a further strong safeguard against misuse of data collected for this
purpose.
There are interesting recent developments with respect to
assessments with purpose 3. Earlier we noted state efforts to strengthen the
collection of indicators on such issues as young children’s participation in
early childhood care and education settings, and the quality of such care
(HHS/ASPE Child Indicator States, 1999).
In addition, individual states are working to select a limited number of
measures of children’s school readiness, to report at the aggregate level, and
to track over time. For example, in
North Carolina they are working to select a measure or a small set of measures
for each of the five dimensions of school readiness identified by the Goal 1
working groups. As suggested in the
NEGP assessment document, these data would be collected through a
representative sampling of children, with matrix sampling to prevent misuse of
the data for making educational decisions for individual children. The measures
would rely on multiple informants, including direct assessment, maternal report
and teacher ratings (Maxwell, 2000).
Love and colleagues (1994, 1999) also provide detailed
guidance to communities on how to measure children’s readiness and community
supports for readiness. Their recommendations of specific interview items and
child assessments is based on a detailed review of measures being used in
national surveys, evaluations of early childhood programs, and child
development research. For example, regarding the community support of
children’s access to high quality and developmentally appropriate preschool
programs, they guide communities to use specific interview items from the
National Household Education Survey, the National Child Care Survey, and the
Interactional and Developmental Processes study questionnaire. For measuring children’s Social and
Emotional Development, Love and colleagues recommend specific scales from the
Social Skills Rating System, and the Behavior Problems Index (see Love et al.,
1994 for sources for each of these surveys and measures).
Purpose 4: Assessing academic achievement to hold individual
students, teachers and schools accountable (“high stakes assessments”). These assessments are reported for individual
students, classrooms, teachers or schools. There are serious consequences of
assessments with this purpose, for example, retaining a child in a grade or
awarding a teacher merit pay. Technical
standards for assessments with this purpose are stringent, as scores for
individual students must have sufficient reliability and validity to provide
the basis for decisions at this level. Departures from appropriate testing
procedures (for example providing more time than allowed or teaching to the
test) can have serious implications for the results of such assessments. Misuse
here involves using assessments for purpose 4 without adequate reliability or
validity to make decisions at the individual level. Given low reliability and validity for young children, the Goal 1
resource group recommends the use of assessments for purpose 4 beginning only
at the end of third grade.
The Goal 1 resource group suggests great caution in using
any assessment for more than one purpose.
For example,
Assessments could not
at the same time be flexible and informal enough to be useful to teachers in
day-to-day teaching and learning and still meet the technical requirements of
reliability, standardization, comparability, validity and fairness that must be
satisfied for accountability reporting. (NEGP, 1998a, p. 33)
With care that the technical requirements for each are being
met and that the content is appropriate, however, the resource group noted that
there could be coordination of assessments for purposes 1 and 3 (assessment to
shape individual instruction, and assessment to monitor at an aggregate level).
In an extremely thoughtful review of the history of school
readiness assessment, Meisels (1999) identifies four approaches to the
assessment of children’s readiness, each reflecting a different underlying
theory. His review leads him to support the fourth, or Interactionist view of
school readiness and to propose the use of assessments that reflect this view.
The first approach to school readiness, Idealist/Nativist,
sees readiness as a maturational issue.
In this view, readiness is not seen as something that can be accelerated
in a child, but emerges according to an internal timetable. The Gesell School
Readiness Test grew from this tradition (and was used extensively to make
placement decisions for children, e.g., whether they were ready to begin
kindergarten). The second approach, Empiricist/Environmental, by contrast, sees
readiness as externally driven.
Assessments in this tradition seek evidence of knowledge and skills
provided by the environment. For
example, they seek evidence of knowledge of such specific information as
colors, shapes and one’s address. In this perspective, readiness is absolute:
it is either present or absent in the child. The major achievement tests grow
out of this tradition. A third approach, Social Constructivist, sees mastery as
a relative rather than absolute concept. Communities each provide their meaning
for readiness. Children who are ready in one community may not be in
another. Finally, the Interactionist
view focuses jointly on the child’s status and that of the educational
setting. Assessments are embedded in
the curriculum and are carried out over time and in the educational setting
rather than removed from the everyday setting. They are used for individual
instructional planning as well as evaluation.
The Work Sampling System, developed by Meisels, follows this latter
approach. It relies on checklists
completed by the child’s teacher, portfolios of the child’s work and summary
reports. Teachers are trained in the observation and recording of students in
actual classroom situations and are provided benchmarks to assist them in
describing the children’s behavior and performance.
In an important recent development, a number of states are
implementing work sampling approaches (with a number specifically using
Meisel’s methodology), both for the purpose of shaping children’s individual
instruction (NEGP Purpose #1) and also to aggregate data from individual
students to monitor student progress in larger geographical units (NEGP purpose
#3). These states are in the early
stages of training teachers, collecting and examining pilot data and implementing
the procedures.
The use of Work Sampling Systems clearly addresses the
recommendations of the NEGP Goal 1 resource group to strengthen assessment for
purpose 1 by using resources at the state level to train teachers in
observation of student skills and development, to provide tools for guiding the
documentation and rating of student progress and for developing
curriculum-linked benchmarks of progress that are coordinated across a state.
Yet, as recommended by the NEGP resource group, the use of these assessments
for multiple purposes will need to be carefully monitored. Some issues that
will need to be addressed over time include (1) the use of parts of the full
Work Sampling System rather than the entire system in a number of states. Reliability and validity have been assessed
for the full system, leaving open the question of how well the particular
segments of the system are functioning.
This is particularly important given the more stringent reliability and
validity requirements for assessments used for purpose 3. (2) Given that all children will be assessed for purpose 1 using work sampling
approaches in these states, it is clearly not possible to employ the sampling
strategies suggested to protect from misuse of assessments when they are used
also for purpose 3. It will be important to follow implementation and assure
that this approach is not being used to make placement decisions for individual
children or to provide feedback, rewards, etc. to individual teachers (i.e.,
for high stakes assessments). (3) The
NEGP working groups recommended that assessments be based on multiple
informants. Work sampling approaches
rely very heavily on teacher report, guided by training in the observation of
students and their work. Despite the
provision of training and benchmarks for describing progress in the Work
Sampling System, the heavy reliance on a single observer means that it will be
important to examine the possibility of any biases associated with this
particular informant. Biases could
potentially be exacerbated by teacher awareness that the assessments will be
used for monitoring as well as for purposes of informing instruction. Other possible biases could concern ratings
of minority/language minority children or children of lower socioeconomic status.
There has been substantial progress over the past decade in
terms of the conceptualization of school readiness and in terms of establishing
positive practices for the assessment of young children. The conceptualizations of school readiness
that have emerged and the delineation of principles for assessment have
implications for Knight Foundation’s work within selected communities.
One of the goals of the work within communities will be to
determine how each community is defining school readiness and seeking to
support children’s readiness. In
addition to documenting the approach in each community, a further goal of this
work is to determine how well community approaches map onto the definitions and
approaches suggested in the research on school readiness. Our review of the “Definitions and
Assessments” components of the literature in Part I of this review suggest a
number of specific questions that we will want to address in carrying out site
visits within communities. In
particular, we will want to ask:
·
Are communities thinking in terms of all three
components of school readiness identified in the literature review: children’s
readiness, schools’ readiness, and family and community supports?
·
Do the Knight communities want to come up with
community-specific definitions of school readiness or work toward close
correspondence with definitions used at the national level?
·
Focusing specifically on children’s readiness, are
communities defining children’s readiness in term of all five dimensions
delineated by the National Education Goals Panel Goal 1 working groups, or are
they more narrowly focused on the cognitive aspects of readiness?
·
What assessment strategies do the communities currently
pursue to measure children’s school readiness? Is the assessment (are the
assessments) clearly organized around a particular assessment purpose
(instruction, screening, progress at an aggregate level, individual
accountability)? Are the strategies for
collecting the data and technical requirements appropriate to the purpose? Are all five dimensions of readiness
encompassed in the assessment?
·
Are the communities attempting to strengthen schools’
readiness for children? Which of the
strategies identified by the Goal 1 working groups are they pursuing? Are there any attempts within the
communities to measure schools’ readiness for children? What funding streams are supporting these
efforts?
·
Are there community efforts to strengthen family and
community supports for children’s school readiness? If so, do they focus (as the three objectives articulated by the
Goal 1 working groups do) on health, early childhood care and education and
parenting support and practices? Do
they go beyond this set to further supports?
Are there any attempts to track these kinds of supports through
indicators at the community level?
·
Is the particular community part of any state-level
efforts pertaining to children’s school readiness, the readiness of schools or
supports for school readiness? To what
extent did the community have input into this initiative? What is the community reaction to the
initiative?
·
Are there special efforts within the community to
support the school readiness of children from diverse backgrounds, in terms of
language, race or ethnicity? What are
these efforts? Are they being evaluated
in any way?
Now that we have addressed definitions of school readiness
and assessments of school readiness, it is time to turn to investments in school readiness.
What can communities do to support children’s school readiness? The Goal
I Technical Planning Group of the National Education Goals Panel focused on
three kinds of community supports: provision of services for children’s health,
supports for parents as children’s first teachers, and access for all children
to high quality and developmentally appropriate early care and education
settings. In Part II, we include but go
beyond the three core kinds of community supports considered by the NEGP. In particular, we adopt a broader ecological
framework, considering (1) factors related to the child, specifically his/her
physical and mental health; (2) factors related to the family context,
(including parents’ psychological well-being, parent-child relations, family
structure, and family socioeconomic status); (3) factors related to children’s
experience of early childhood care and education settings; and (4) factors
related to the neighborhood context.
These reflect “layers” of influence on children’s school readiness,
moving from the child, to the family, to care settings beyond the family, and
to the broader social context. This ecological view of child development
provides a useful framework for understanding where and how communities can
intervene to support and promote healthy child development in general and
school readiness in particular. This
approach may be especially helpful to communities as they set priorities for
investments in school readiness.
An extensive body of research on child development helps
identify the factors that influence children’s readiness for school, beginning
with those closest to the child and moving outward to encompass the family,
early care and education, schools and the neighborhood. In Part II of this report, we present a
selective review of the research pertaining to each layer of the ecological
model. We focus as much as possible on
studies that use measures of children’s early school progress (academic and/or
behavioral adjustment in school) as outcomes.
We emphasize (1) studies that are rigorously implemented experimental
evaluations of interventions, in which aspects of the environment are manipulated
and early school outcomes examined; and
(2) studies that are longitudinal, involving the examination of aspects
of the environment as predictors of early school outcomes but that use
multivariate analyses taking background characteristics of the families into
account. We have emphasized these types
of studies for several reasons.
Experimental-control group studies are better at controlling for
selection bias than other designs (as long as there is not extensive attrition
in the sample over time) and they also can address causal relationships more
clearly than other designs. In
addition, longitudinal studies that contain adequate consideration of
background characteristics can address change over time and predictive validity
better than cross-sectional studies. We
also highlight studies that have been replicated with similar results across
different populations and geographic regions, because successfully replicated
interventions appear to have a better chance of being replicated in additional
locations than do studies that have been carried out in a single place, at one
single point in time. For some issues,
especially in the area of health, where the linkages to child school outcomes
have been well established (such as child exposure to lead), we rely on studies
reporting basic epidemiological/descriptive research.
Throughout Part II, our aim is to go beyond the broad
identification of which factors appear to be linked to early school outcomes,
to the identification of specific
strategies (kinds of programs and activities within these programs) that
have been attempted and evaluated, and/or for which there is evidence that
initiating programs with these activities has the potential to contribute to
improved early school outcomes. Due to the criteria we set for our selected
literature review, we may not have identified all programs and activities across the country that may be
effective in promoting school readiness.
Over the past century, the United States has been successful
in giving children a healthy start on life.
This includes lower infant and child death rates and lower chronic
health conditions (Brown et al., 1999).
However, serious physical and mental conditions that can affect a
child’s ability to learn in school still persist. Antonia Novello, former Surgeon General for the United States,
started the “Healthy Children Ready to Learn Initiative” with the assumption
that a child is ready to learn if he or she arrives at school physically and
mentally healthy (Novello, 1991). The
National Education Goals Panel agreed with this initiative when it deemed
physical well-being, motor development and socioemotional well-being to be
important criteria for school readiness.
This section will focus on the health of the child as it relates to
school readiness. We will cover the
topics of prenatal and infant health, appropriate immunizations, nutrition,
unintentional injuries, lead exposure, dental health and child emotional and
behavioral problems.
Infant health is extremely important in regard to the
cognitive and social components of school readiness, because problems at this
age may lead to problems in the school years.
For instance, a child’s birthweight can have significant effects on his
or her cognitive abilities and behavioral competence (e.g., Gross, Brooks-Gunn,
Spiker, 1992; Liaw & Brooks-Gunn, 1993).
The Infant Health and Development Study (IHDP) has been one major
investigation studying the effects of poor infant health.
The IHDP was a three-year, multisite, randomized trial of an
intervention for low birth weight, preterm infants from diverse ethnic
backgrounds. The intervention included
pediatric monitoring, referral, follow-ups, home visits, attendance of the
infant at a child development center and support group meetings for the
parents. Several reports have been
written, but two are presented here for illustrative purposes. From the study data, it was found that lower
child neonatal health status is a strong predictor of lower cognitive abilities
tested at 12, 24 and 36 months (Liaw
& Brooks-Gunn, 1993). Neonatal
status remained a strong predictor even after the effects of maternal
environment and the treatment were taken into account. It was found, though, that the intervention
was effective at 36 months in regard to gains in receptive language and
visual-motor and spatial skills (Brooks-Gunn, Liaw, & Klebanov, 1992). In fact, cognitive development scores were
7.2 points higher for the intervention group as compared to the control
(McCormick, McCarton, Tonascia, & Brooks-Gunn, 1993). Moreover, the effects were most positive for
the most at-risk families (i.e., children whose parents had a high school
education or less and who were of ethnic minority status; Brooks-Gunn, Gross, Kraemer,
Spiker, & Shapiro, 1992). A later
study, examining the same sample of children at eight years of age, supported
the earlier findings by showing that a subsample that was heavier at birth
(though still low birthweight) had higher scores on several cognitive tests
(i.e., reading, math, overall IQ) than a lower, low birthweight cohort
(McCarton et al., 1997). However, it
should be noted that the cognitive test scores of the low birthweight cohort
were still within a normal range, based on a nationally representative
sample. Other studies have found
similar results (e.g., Holmes, Reigh, & Rieff, 1988). A reanalysis of IHDP was conducted by
investigators independent from the IHDP research consortium (Baumeister &
Bacharach, 1996). The researchers found
that birthweight, either directly or interacting with the intervention, was a
significant predictor of cognitive outcomes.
However, only seven percent of the variance of cognitive outcomes was
accounted for by the intervention model.
Other factors such as maternal intelligence and the home environment
were considered to be important mediating factors for the entire sample. The results from the reanalysis also suggest
that cognitive deficits in early childhood (i.e., at three years of age) may
regress toward the mean as the child ages, even within two years of the
program’s completion (i.e., when the child is five years old). This finding suggests that cognitive
deficits may disappear without the need for an intervention.
The impact of parenting skills on infant health was
supported by an evaluation of an at-home video course for parents of preterm
infants (Brown, Yando, & Maxwell, 2000).
The video course, produced by My
Baby U consists of a set of videos and short books that teach mothers about
an infant’s state of awareness, individual temperament and cognitive, motor,
perceptual and emotional capabilities and needs. Each video in the set is for a
different age (i.e., one-month-old, three-months-old, five-months-old,
seven-months-old, ten-months-old, and one year old) and is sent to the parent
when the child reaches the age in the specific video. Two hundred fifty-one expectant mothers were randomly assigned to
the video intervention (n=116) or a non-intervention control group (n=135). Mothers in the intervention group had
children that had fewer severe illnesses over the course of the first year as
compared to a non-intervention control group.
There is general consensus that immunizing children is
beneficial. Fewer severe diseases
result, which enables children to attend school in a healthy state on a regular
basis and results in fewer financial and psychological burdens on
families. According to the National
Immunization Survey (NIS), initiated by the Centers for Disease Control and
Prevention, vaccination rates for the recommended round of immunizations for
children age 19 to 35 months is the highest ever recorded (Centers for Disease
Control and Prevention, 1998). This finding, based on 1997 data, showed a
national immunization level of 83.3 percent, ranging across states from 69 to
91 percent. This rate, based on surveys
of 32,652 household representing 33,064 children, is an increase of 1.5 percent
over the 1996 level. It should be noted
that immunization levels for Hepatitis B and Pertussis have not reached the same
high rates for vaccination as for other diseases (Teitelbaum & Edmunds,
1999).
Unfortunately, certain populations have not had as much
success with increasing immunization rates.
For instance, in a study conducted in low-income areas of Los Angeles, only
53 percent and 70 percent of African American and Latino children,
respectively, were up-to-date on their immunizations at 3 months of age (Wood
et al., 1995). The rate decreased to 42
percent for Latino children and 26 percent for African American children at 24
months. Though the reasons for the
lower rates were not conclusive, there are obviously barriers that must be
overcome. For instance, there may be an
unintentional bias to detect low rates, and subsequently try to increase those
rates, in Caucasian American and higher income neighborhoods. A recent report by the Institute of Medicine
(IOM) Committee on Immunization Finance Policies and Practice (Institute of
Medicine, 2000) states that the current methods for tracking immunization rates
through private insurers and providers excludes certain high-risk populations,
such as children on Medicaid. The
report concludes that non-detection of low vaccination coverage in distinct
pockets may provide a “reservoir” for disease outbreaks. Therefore, the IOM committee recommends an
increase in public and private spending of $1.5 billion over the next five
years to fund an effective immunization tracking system.
One study attempted to overcome low rates of immunization in
a medium-sized Midwestern city (Yokley & Glenwick, 1984). Families of 1,133 immunization-deficient
children were randomly assigned to six groups: group 1 received a general
letter that indicated the types and timing of childhood immunizations; group 2
received a client-specific letter that contained immunization information about
the particular child; group 3 received a specific letter and increased public
health access; group 4 received a specific letter and a monetary incentive;
group 5 received a control letter; and group 6 received no letter. A monetary incentive had the largest effect
on increasing immunization rates, followed by the increased access group and
the specific letter group. The
researchers concluded that although monetary incentives were the most
effective, using client-specific prompts, such as letters, might be the most
cost-effective.
The Committee on Nutrition of the American Association of
Pediatrics has developed a series of recommendations for infant feeding during
the first six months of life (Committee on Nutrition, 1998). These suggestions include breastfeeding,
using infant formula if not breastfeeding, not using cow’s milk and not giving
solid foods in the first four months.
For the fifth and sixth months, solid foods can be introduced if the
infant is able to support himself or herself in a seated position.
The effects of proper nutrition can be seen in Korenman,
Miller and Sjaastad’s (1995) study using data from the National Longitudinal
Survey of Youth. They found that
nutritional deficits contributed unique and significant variance to lower
cognitive test scores when the home environment and maternal characteristics
were taken into account.
Grantham-McGregor (1995) conducted an in-depth review of the
literature on cognitive and behavioral deficits of nutritionally deprived
children. From case controlled studies,
it was concluded that, using epidemiological principles, there is a causal link
between nutrition and cognitive and behavioral outcomes. However, there is still a question about
whether acute episodes are more or less detrimental than smaller but persistent
nutritional deficits.
The Special Supplement Nutrition Program for Women, Infants,
and Children (WIC) and the Food Stamp program are both designed to help parents
provide proper nutrition for their children.
Several studies have been designed to examine the effects of these two
programs. Compared to giving only
additional income to families, both WIC and Food Stamps have been found to
increase the nutritional intake of children; WIC was found to increase the
intake of 10 nutrients while Food Stamps increased the intake of five nutrients
(Rose, Habicht, & Devaney, 1998).
Gordon and Nelson (1995) performed a separate evaluation of WIC, comparing
participants to income-eligible non-participants. Using data from the 1988 National Maternal Health Survey, the
investigators found that WIC participants were more likely to use infant
formula over breastfeeding than income-eligible non-participants, thus going
against the recommendations of the Committee on Nutrition (1993). WIC participants were also more likely to
use infant formula over cow’s milk at five and six months. Although WIC and Food Stamps are not
completely effective in changing feeding behaviors, it appears that they are effective
programs for improving the nutritional intake of children.
There were over 13,000 child deaths from unintentional
injury in 1996; over six times that number of children had non-fatal injuries
(CDC, 1999). The injuries can result
from a plethora of causes, such as bicycle accidents, car crashes, drowning or
being caught in a fire. Aside from
pain, suffering and the emotional and economic cost to the family of an injured
child, long-term deficits of cognitive, behavioral, and motor functioning of
the child may occur. This could lead to
a reduction in the quality of life of the child (Deal, Gomby, Zippiroli, &
Behrman, 2000).
Two types of strategies have been implemented for preventing
unintentional injuries: interventions that target the individual, and those
that target the community. These
strategies have been reviewed in the most recent edition of Future of Children (DiGuiseppi &
Roberts, 2000; Klassen, MacKay, Moher, Walker, & Jones, 2000). DiGuiseppi and Roberts (2000) reviewed
randomized controlled studies of individual-level programs. The studies were restricted to those
examining children under 20 years of age, having interventions delivered in a
clinical setting and having randomly assigned experimental and control groups. Interventions for motor vehicle restraint
use that only included educating the parents did not produce a significant
decrease in injuries compared to the control groups. However, when education was paired with subsidized devices (i.e.,
lending car seats) or when education was paired with positively reinforcing the
safety behavior, there was a strong effect of the education on improved safe
behavior and decreased injury.
Parent education on its own appeared to be significantly
effective for using safe hot tap water temperatures in order to prevent hot
water scald burns. Similar effects were
found for smoke alarm ownership.
However, interventions for childproofing homes were not as
successful, with little effect shown for even intensive education-based
interventions. Reasons for this include
the interventions possibly needing to be more
intense, the families being confused by the vast number of safety practices or
the inability to acquire the various safety devices. Overall, interventions seem to be effective for vehicle safety
and using safe hot tap water if they are implemented in a straight-forward
manner and the instructions for the safety procedures are easy for the families
to follow.
Community-level interventions generally focus on changing
safety behaviors by altering social norms, either through legislation or
community change. Interventions were
reviewed if they included a control group, had a target population between 0
and 19 years of age and the study reported changes in injury rates or injury
reducing behavior. For bicycle helmet
use, Klassen and colleagues (2000) concluded that the most effective strategies
were those that used multiple pathways, such as passing legislation requiring
helmet use paired with an educational campaign. This strategy proved to be effective in a Mid-Atlantic state,
with 47 percent of the legislation-plus-education group using helmets compared
with 19 percent of the education-only group and 4 percent of a control.
Similar multiple strategies were again effective for
increasing use of motor vehicle restraints.
However, there was an interaction between the intervention strategies
and income level, with the interventions having a stronger effect on low-income
as opposed to high-income families.
Pedestrian and general injury prevention programs that use
parent and community education, television programs, school-based education
and/or various community programs have not had very robust effects. A possible problem with these findings is
that the evaluations compared samples
that were too close in proximity and therefore the control groups may have been
affected by the programs. Overall,
there is support for using some types of community-based strategies for
preventing unintentional injuries.
Using multiple community-wide methods, such as legislation combined with
an educational campaign, appears to be most effective.
The percentage of children with elevated levels of lead in
their blood has declined sharply since the 1980s. Between 1991 and 1994, 4.4 percent of children ages one to five
had elevated blood lead levels, compared to 88.2 percent between 1976 and
1980. However, 8 percent of poor and
minority children have elevated levels of lead in their blood (U.S. Department
of Health and Human Services, 1999).
Exposure to lead can lead to serious problems, especially for young
children, as it can interfere with the development of the brain and other organ
systems. Although high levels of lead
exposure are now rare in the U.S., the more common chronic, low-level exposure
can lead to problems such as lowered IQ, short attention span, hyperactivity,
stunted growth and reading and learning disabilities. There is no medical treatment for lead poisoning, and the only
way to prevent it is by removing the source of exposure (Alliance to End
Childhood Lead Poisoning, 2000).
The most common way to be exposed to lead is from lead-based
paint. Even though lead in paint was
banned through federal regulations in 1978, two-thirds of all housing contains
some leaded paint. This only becomes
dangerous when lead dust is created by deteriorating paint (a problem in
poorly-maintained housing) or when repainting and remodeling is done without
proper safeguards. Two other possible
points of childhood exposure to lead are soil (contaminated by flaking exterior
paint) and water (contaminated by lead from old pipes). The best way to prevent lead exposure in
children is good home maintenance, keeping indoor areas where children play
dust-free and clean and having children wash their hands after playing outside
and before meals (Alliance to End Childhood Lead Poisoning, 2000). Parental education in such practices and
careful building maintenance may be the keys to seeing the levels of lead
exposure continue to decline, especially among families living in poverty.
Tooth decay, also called dental caries, can be a source of
pain and suffering for children and can permanently damage the functioning of
their teeth. Oral health is important
to child development. At the most
fundamental level, a child with dental caries may not be able to eat properly,
so physical growth may be diminished.
Tooth loss can also impair speech development among young children and,
later, can result in an inability to concentrate or increased absences from
preschool (Platt & Cabezas, 2000).
Dental caries can be caused by a lack of access to
preventive care, a lack of information about proper dental hygiene, poor
feeding habits or poor nutrition. The
prevalence of dental caries is highest among children living in poverty—the
rate is 29.7 percent for poor children between the ages of two and five,
compared to 8.6 percent of non-poor children (U.S. Department of Health and
Human Services, 1999).
Even though dental disease is the most common disease among
children, dental services are usually not coordinated with other health and
social welfare services in communities.
An exception to this is Head Start, which has a federally mandated
dental component that includes screening, preventive services and education of
parents. Platt and Cabezas (2000)
suggest that oral health screenings and parent education in oral health
practices should be a part of other health screenings. Alternately, referrals for dental health
services and parental training in proper feeding and dental hygiene (which is
especially important as children’s permanent teeth begin growing in) should
occur at points of entry into other services such as WIC, child care, home
visits and during immunizations.
Finally, increasing access to dental care providers would be a necessary
step in decreasing the percentage of children with dental caries.
In comparison with non-depressed children, depressed
children, which comprise between seven to 14 percent of the child population,
have been found to have more maladaptive social behaviors, such as poorer
problem-solving and conflict-negotiation skills and affect-regulation deficits
(Rudolph, Hammen, & Burge, 1994).
Children of depressed mothers are at an increased risk of
developing depression (Hammen, Adrian, Gordon, Burge, & Jaenicke, 1987;
Radke-Yarrow, Nottelmann, Martinez, Fox, & Belmont, 1992). However, other factors such as family stress,
should be considered as well.
Radke-Yarrow and colleagues (1992) observed children (ages 1.5-3.5) of
depressed (i.e., unipolar), manic-depressive (i.e., bipolar), and non-depressed
mothers, along with their older siblings (ages 5 to 8), at baseline and at a
three year follow-up to determine how maternal depression affects
children. Findings supported their
hypothesis that children of unipolar depressed mothers would exhibit some form
of psychopathology (i.e., either depression, anxiety, or both). The question of whether long-term maternal
depression or present maternal emotional state are more predictive of child
depression was examined in a small study of 8- to 16-year-old predominantly
Caucasian American children and their mothers.
The investigators found that the current emotional state of the mother
was most predictive of child behavioral problems, including child
depression. Maternal depression,
however, is possibly part of an additive model in which parenting behaviors are
included (Burge & Hammen, 1991).
However, a more recent review of the depression literature has
conflicting findings, concluding that chronic depression has the most negative
impact on children’s mental health (Beardsley, Versage, & Wright,
1997). Although more research is needed
before a definitive answer can be given about the chronicity of maternal
depression, the research is rather definitive that maternal depression,
regardless of length of illness, predicts child depression. Intervening with the mother, therefore,
could be one avenue for preventing emotional and behavioral problems in
children.
Based on the current state of the evidence, which varies in
its completeness and quality, a number of implications can be suggested for
communities seeking to improve child health.
For child emotional well-being, studies have found that maternal
depression is predictive of child depression, although the effect of the length
of maternal depression on children has not been definitively concluded. This finding implies that addressing current
parental psychological problems may have immediate benefits for children (or,
two-generation interventions, where both parent and child depression are
simultaneously addressed may work well, too).
Again, parent-child interactions and parenting strategies may be the
link between maternal and child depression.
To boost immunization rates, it appears that client-specific
prompts accompanied by monetary incentives work best, although a more
cost-effective option is client-specific (i.e., personalized) prompts
accompanied by increased access to public health institutions.
In contrast, giving money to parents does not necessarily
increase the likelihood of them buying nutritious food for their children and
properly feeding infants (e.g., giving the infant breast milk or formula
instead of cow’s milk). However,
government programs such as WIC and Food Stamps, which provide vouchers for
food, do increase the nutritional intake of children. Increasing the availability and use of these programs to
income-eligible families seems to be a somewhat effective strategy for insuring
a good diet for infants and young children.
Parent education on the harmful effects of lead exposure and
improper dental care may encourage better hygiene practices in the family. Increasing access to dental care providers
and coordinating their services with other social services has also been
suggested as a strategy to improve children’s dental health.
Finally, in terms of avoiding unintentional injuries, it
appears that parent education is most effective when accompanied by subsidized
equipment, such as vehicle safety, or by positive reinforcement of the safety
behavior. However, community-wide
efforts accompanied by local legislation may be another strategy for
effectively preventing childhood injuries.
Various domains of child mental and physical health have
been described. From this review, one
can see that there are enough data on the predictors of infant health and child
mental health with which to implement programs. Focusing on the education and parenting quality of the guardian
appears to be an important part of any program, such as treating maternal
depression or teaching the parent how to care for the physical and mental
well-being of his or her child.
Supplying the family with basic resources such as proper food and
medical care also appears to be essential.
These recommendations hold true for nutrition, immunizations, lead
exposure, dental health and unintentional injury, although the findings are
less conclusive.
Currently nearly one in five children in the United States
live at or below the poverty level. The
rate increased from 17 percent in 1975 to 22 percent in 1993, and then lessened
to 19 percent in 1997 (U.S. Department of Health and Human Services,
1999). However, the rate for children
under age five is even higher, at 23 percent (Annie E. Casey Foundation,
2000). The large number of young
children in poverty is cause for concern because it has been found that poverty
during the first five years of life is more detrimental than poverty
experienced at any other point during childhood or adolescence. In particular, poverty before age five is
associated with fewer total years of schooling, so it would appear that such
children are set on a trajectory for school failure and dropout (McLoyd, 1998).
The effects of poverty on childhood development are
numerous. Poor children have worse
nutrition than nonpoor children, and more physical health problems such as low
birthweight, growth stunting, and more hospitalizations. Lower scores on standardized tests for
verbal ability have been found as early as age two (Brooks-Gunn, Britto, &
Brady, 1999). Negative effects of
poverty have also been found by age five on cognitive skills, including reading
readiness, number skills, problem solving, creativity and memory (Stipek &
Ryan, 1997). Furthermore, poverty is
associated with an increase in emotional and behavioral problems (McLoyd,
1998).
It is clear that poverty has a strong association with
children’s school readiness, and there are many possible pathways through which
it can influence developmental outcomes.
As proposed by Brooks-Gunn, Britto, and Brady (1999), one pathway is
nutrition and health, often beginning with mothers’ lack of prenatal care. Another is a home environment unsuitable for
early learning and development, and even presenting physical dangers for
children. Parental mental health and
parent-child interactions are other factors to consider (see later section on
the home environment). The stress
associated with poverty or the loss of a job can lead to problems in maternal
mental health which, in turn, can lead to poor parenting. Parents in poverty are more often punitive
and coercive, and often deal with their children in a way that lacks support,
involvement and consistency. Another
possible pathway is through community effects as poor families often have no
choice but to live in undesirable neighborhoods (see later section on
neighborhoods). Furthermore,
low-quality child care can have detrimental effects on child development
(Peisner-Feinberg & Burchinal, 1997).
Here we discuss several experimental studies that highlight some of
these pathways.
Many approaches have been employed to address the problem of
poverty and school readiness. Trying to
teach parents work skills and raise families’ incomes above the poverty level
is one option. Another is to address
the problems that are associated with child poverty by doing such things as
providing quality early child care, improving nutrition and health care,
teaching parents effective parenting skills and educational activities to carry
out with their children, or improving the social network and support for
stressed poor mothers. The main focus
of the three studies that will be summarized here was the alleviation of
poverty itself.
The aim of the New Hope Project was to raise poor families’
incomes above the poverty level and observe the effects on children (Huston et
al., in press). Participating parents
were older than 18, had one or more child between ages one and ten, had incomes
at or below 150 percent of the poverty level and were willing to work 30 or
more hours per week (those in the experimental group without jobs were given a
community service job that paid minimum wage).
Families were randomly assigned to the experimental group (366 families)
or the control group (379 families). In
addition to help finding jobs, experimental group families were given wage
supplements, subsidies for childcare, and health insurance. Child outcomes including school performance
and motivation, social behavior, and psychological well-being were
investigated. Data were collected two
years after the initiation of the project.
Overall, the impacts were greater for boys than they were
for girls. The authors posit that the differences between boys and girls were
due to the fact that girls were already at higher levels than boys on the
measures they used. Nevertheless, the
effects for boys were significant and noteworthy. New Hope boys scored higher than control group boys on measures
of educational progress and motivation.
They also received better teacher ratings of classroom behavior as they
were seen as less disruptive, aggressive and hyperactive. It was also found
that parents expressed feeling less stress, more social support and greater
goal-expectations for themselves. One
aspect of the research that needs careful consideration is the fact that New
Hope children spent more time than control group children in preschool programs
and after school child care. Since the
researchers did not separate out the effects of each component of the
intervention, this raises the possibility that increasing families’ incomes
above the poverty level might not be necessary, but that child care subsidies
alone might be enough to improve developmental and school outcomes.
A similar program with similar results was the Minnesota
Family Investment Program (MFIP; Knox, Miller, & Gennetian, 2000). A large pilot study of MFIP as a replacement
program for the traditional Aid to Families with Dependent Children (AFDC) was
begun in 1994. MFIP is a program of
financial rewards for work and mandatory participation by those who have been
receiving long-term welfare in employment-focused services. Thus, families’ welfare receipt does not
decline and their incomes are raised above the poverty level. Fourteen thousand families were randomly
assigned to either MFIP or AFDC.
Researchers found that participation in MFIP led to large increases in
employment and earnings, a decline in domestic abuse and a small increase in
the marriage rate. Perhaps most
importantly as far as school readiness is concerned, children in MFIP exhibited
fewer behavior problems, performed better in school, and were more engaged in
school. It is noteworthy that the program
included no sort of education for parents in parenting skills, so it would
appear that raising their incomes was enough to produce positive results for
children and families. However, as with
the New Hope Project, it must be taken into account that participating children
took part in more stable child care.
Because of participation in child care, even though it seems that
raising families’ incomes above the poverty level has many positive effects,
the findings cannot be attributed to increased income alone.
A third program related to poverty, and to the alleviation
of welfare receipt in particular, is the Job Opportunities and Basic Skills
Training Program (JOBS). As part of the
Family Support Act of 1988, JOBS was designed to provide either job training
and education or the quick acquisition of a job for those receiving
welfare. JOBS was replaced by the
Personal Responsibility and Work Opportunity Reconciliation Act (PRWORA) in
1996, shifting the emphasis from job training to job placement (Brooks-Gunn,
Britto, & Brady, 1999). However, an
evaluation of JOBS provides valuable lessons on the efficacy of parental
education and job training in alleviating poverty and altering child
development.
The Child Outcomes Study involved over 3,000 families in three
cities (Atlanta, GA, Grand Rapids, MI, and Riverside, CA). Mothers of preschool-age children were
randomly assigned to one of three programs between 1991 and 1994. The first was a JOBS program that emphasized
the rapid acquisition of employment; the second was a JOBS program that
emphasized longer-term training and education; and the third was the
traditional AFDC (control group). Two
years later, child outcomes were assessed, including cognitive development, academic
achievement, behavior, emotional development, health and safety. Assessments consisted of cognitive testing
of the children and mother reports (Zaslow, McGroder, & Moore, 2000).
Participation in JOBS had few and small impacts on child
development but the pattern of impacts is important when considering school
readiness. The effects on behavioral
and emotional development were mixed, but, in general, the effects on cognitive
development and academic achievement were positive. The job training and education program in particular produced some
favorable impacts on cognitive school readiness tests (Zaslow, McGroder, &
Moore, 2000). Thus, it can be surmised
that maternal participation in job training and education classes can
positively influence poor children’s school readiness (at least in the
cognitive domain), though effects were small.
Summary: Family Economic Risk
As far as the best interventions for improving the school
readiness of children in poverty are concerned, Brooks-Gunn (1995) suggests
that “two-generation programs” might be the most effective. Such programs would provide education and
work training for poor parents, quality child care, parenting classes and
referrals for services. To that list should
be added wage supplements to bring incomes above the poverty level, the effectiveness
of which has been tentatively shown by the New Hope Project and MFIP.
Rates of out-of-wedlock childbearing, divorce, cohabitation
and remarriage in the United States have risen substantially over the past
several decades. Changes in living
arrangements, marriage and divorce can cause financial and emotional strain on
families with children, and these stressors may have a direct and/or indirect
effect on children’s well-being. In particular,
the experience of multiple transitions in family situations (including
frequency of moves) during childhood is associated with an increased
probability of repeating a grade (Wood, Halfon, Scarlata, Newacheck, &
Nessim, 1993) and lower educational attainment (Aquilino, 1996). Even when living arrangements are relatively
stable, children may fare differently depending on how many children are in the
family, whether the child’s birth was intended and wanted by the parents,
whether the parents were younger or older, or whether both parents reside in
the home.
This section discusses the effects of family composition on
children’s school readiness skills. We
will cover several topics, including unwanted and unintended pregnancy, close
birth spacing, teen parenthood, single- vs. two-parent families, nonresident
parents, cohabiting couples and step- and reconstituted families. To date, there are few rigorous,
longitudinal studies that specifically address the effects of different family
configurations on children’s school readiness.
Nevertheless, the available evidence indicates that family structure at
the time of a child’s birth and during the first several years of life will
have long-lasting effects on children’s development in general, and
specifically on children’s adjustment to school. Where relevant studies are available (e.g., studies relating
family structure characteristics to educational achievement), they will be
mentioned below.
Unwanted and Unintended Pregnancy
Women who have unplanned pregnancies are often in poorer
health before the pregnancy and are less likely to avail themselves of prenatal
care (Brown & Eisenberg, 1995); these circumstances can have negative
consequences for life outcomes of the child.
In fact, unintended pregnancy has been found to be associated with low birthweight,
infant mortality, a lowered likelihood of obtaining immunizations during early
childhood, and poor child health (Moore, Manlove et al., 1997). Children who are the result of an unintended
pregnancy also have been found to have lower levels of positive emotional
expression at age two, and delayed verbal abilities during the preschool years
(Moore, Manlove et al., 1997). These
children are also at a greater risk for abuse and neglect (Zuravin, 1991). An analysis of a nationally representative sample
of families (the National Survey of Families and Households, NSFH) found that
mothers with unwanted births spent less time with their children and were more
likely to spank or slap their children than mothers with wanted births (Barber,
Axinn, & Thornton, 1999).
Furthermore, the same researchers found that, among a second national
sample of mother-child pairs followed over 30 years (the Intergenerational
Panel Study of Mothers and Children), mothers with unwanted births had lower
quality relationships not only with the unwanted child, but also with the rest
of the children in the family (Barber et al., 1999). Other researchers estimating sibling fixed-effects models find
that most of the negative outcomes reflect characteristics of the women who have
unintended births (Joyce, Kaestner, & Korenman, 1997).
There are several populations that deserve targeting in an
effort to reduce unwanted and/or unintended pregnancies. One group is teenage women. Current estimates are that more than 80
percent of pregnancies to teen mothers are unplanned. Another group is unmarried couples. Current estimates are that 43 percent of births to unmarried
couples are unintended, and 15 percent are unwanted (Abma, Chandra, Mosher,
Peterson, & Piccinino, 1997). A
third group is men—of all ages. The
intendedness of fatherhood may be related to the stability of the marital or
cohabiting union between the parents (Furstenberg, 1995), the amount and
quality of interaction between father and child (Halle, Moore, Greene, & LeMenestrel,
1998), and whether a non-resident father pays child support (Working Group on
Male Fertility and Family Formation, 1997).
However, the question of causality is an issue: Is having an unstable
union the cause or effect of having an unintended birth? More research is needed in this area to
illuminate the answers to these questions.
What we do know is, to date, most public- and privately-funded programs
to delay unintended pregnancy target women and the few that do extend their
services to men tend to focus on adolescent males only (Halle et al.,
1998).
The Number and Spacing of Children
Family size has been noted as a strong predictor of
children’s cognitive and educational status (Blake, 1989). In general, children with fewer siblings
tend to have higher cognitive and educational outcomes. In an analysis of six large-scale national
studies, Blake (1989) found that the number of siblings was the second most
important predictor of educational attainment among white males, after father’s
schooling. It is thought that
advantages exist for children in smaller families due to the extra time and
financial resources that parents can devote to each child. A study by Gottfried and Gottfried (1984)
found that children between the ages of one and four from small families,
compared to their peers in larger families, scored higher on standardized tests
of ability and had more enriching home environments, as assessed by the HOME
scale.
A recent study has examined the origins of the long-standing
association between family size and child cognitive outcomes. Guo and VanWey (1999) used sibling data from
the National Longitudinal Survey of Youth – Child Supplement (NLSY-CS) to
control for unobserved selection factors and found that family size is no
longer a significant predictor of cognitive outcomes. These findings are quite controversial (Downey, Powell, Steelman,
& Pribesh, 1999; Phillips, 1999) and additional research can be
anticipated. A recent analysis by
Rodgers, Cleveland, van den Oord and Rowe (2000) also used within-family (i.e.,
sibling) data from the NLSY-CS and also found that family size per se is not a significant determinant
of children’s cognitive abilities.
Rather, the authors suggest that it is the strong correlation between
maternal and child IQ that may account for the effects attributed to family
size in past research, along with the fact that mothers with many children tend
also to have lower IQ’s than mothers with fewer children. They conclude that low-IQ parents make large
families, but large families do not make low-IQ children. While the controversy over the linkages
between family size and cognitive ability are important from a scientific and
methodological perspective, it should be noted that much more research is
needed in this area. For example, as
yet research has not examined critical non-cognitive outcomes for children such
as health, safety, behavior problems, and emotional well-being and their
relation to family size.
Beyond the number of children in the family, the spacing
of children is also cited as being important.
Close child spacing is related to poorer infant health (Spratley &
Taffel, 1981) that in turn may make children more susceptible for learning
disabilities and language delays (Bradley, Caldwell, Rock, Casey, &
Newlson, 1987). An intervention
originally designed to prevent maternal and child health problems has been
found to be effective in reducing unwanted pregnancies and delaying second
births among first-time, low-income teen mothers. The intervention consists of having a registered public health
nurse visit the homes of first-time mothers at regular intervals during
pregnancy, and up through the child’s second birthday. During the home visits, the nurses cover
topics ranging from personal health habits (e.g., reducing cigarette and drug
use, adequate diet) to parenting behaviors (e.g., emphasizing sensitivity and
responsivity) to home safety. A 15-year follow-up on the original,
randomized sample of 400 woman (89% white) in the semirural county of Elmira,
NY found that women who received nurse home visitation during pregnancy and
infancy had fewer subsequent births (1.3 vs. 1.6) and longer intervals between
the first and second birth (65 vs. 37 months) than women in the comparison
group who only received free developmental screenings for their children during
infancy (Olds et al., 1997). A second
randomized trial conducted in a more urban setting (Memphis, TN) with a
primarily African American sample and under less controlled circumstances
produced remarkably similar results (Olds et al., 1998). It appears that having nurses rather than
paraprofessionals conducting the home visits is an important ingredient for
program success. A replication of the
original Elmira study in Denver, CO compared families visited by nurses and
paraprofessionals who were following the same program model. Researchers found that there were
significant differences between nurses and paraprofessionals in the length of
visits, topics covered during visits and number of program dropouts
(Korfmacher, O’Brien, Hiatt, & Olds, 1999). Effects of nurses vs. paraprofessionals on mother and child
outcomes are still not available at this time, however.
Timing of Parenthood
The age at which adults become parents turns out to be a
contributing factor in child well-being.
Several studies have documented that children of teenage mothers
compared to children of older mothers have lower levels of cognitive and
educational attainment, lower levels of academic achievement, and higher levels
of behavioral problems (Baldwin & Cain, 1981; Broman, 1981; Furstenberg,
Brooks-Gunn, & Morgan, 1987; Hofferth, 1987; Maynard, 1997; Moore,
Morrison, & Greene, 1997). Other
factors (such as human, economic, and social resources) account for much of the
variance attributed to teen parenthood (Cooksey, 1997) but not all of it
(Maynard, 1997; Moore et al., 1993).
The age of becoming a father also has implications for
children’s life outcomes. Due to low
rates of marriage and high rates of divorce among teenage parents, adolescent
fathers have less contact with their children than do older fathers. Several national studies have found that
nearly half of young unwed fathers visit their infants at least once a week up
through the age of two, but their level of visitation and other forms of
involvement slow as their children became preschoolers and diminish
precipitously once children enter the elementary school years to the point that
only 20 percent of unwed fathers report visiting their school-aged children at
least once a year (Lerman, 1993; McLanahan, Garfinkel, Brooks-Gunn, & Zhao,
1998; Mott, 1990). The lower levels of
involvement of teenage fathers may be due to their lack of preparedness for the
fathering role in terms of both financial and emotional responsibility
(Furstenberg, Brooks-Gunn, & Chase-Lansdale, 1989). Older fathers (i.e., men who become fathers
in their late 30s or more) have been found to be more responsive and
affectionate with their children, more helpful with child care and housework
and more satisfied with the parenting role than are younger fathers (Coltrane
& Ishii-Kuntz, 1992; Cooney, Pederson, Indelicato, & Palkovitz, 1993;
Volling & Belsky, 1991).
Furthermore, children born to older couples are more likely to be
planned and wanted (Brown & Eisenberg, 1995).
To our knowledge, there have been no studies that have
looked at the impact of the timing of parenthood on school readiness. However, in the near future we will have
several new datasets available that will meet our data needs for this
topic. The Early Head Start (EHS) Study
and the Early Childhood Longitudinal Study – Birth Cohort (ECLS-B) Study are in
the process of collecting data on nationally-representative samples of young
children and their families, and following these children through school
entry. In addition to gathering data on
children’s cognitive, social and emotional development, the two datasets will
collect information on family history and household composition as well as data
on aspects of children’s home and child care environments.
Single- vs. Two-Parent Households
Twenty-seven percent of children in the United States live
in single-parent households while 68 percent live in two-parent households
(U.S. Department of Health and Human Services, 1999). Children living in
single-parent households face more challenges with regard to school success
compared to children living in two-parent families. For example, children from two-parent households are more likely
to get better grades and to have fewer behavior problems than children from
single-parent households (McLanahan & Sandefur, 1994; Thompson, Entwisle,
Alexander, & Sundius, 1992).
Furthermore, Aquilino (1996) found that children born to unmarried
mothers who lived exclusively in single-parent households were particularly
likely to have low educational attainment.
We were unable to locate any experimental or longitudinal, multivariate
studies that specifically focused on the impact of single parenthood on school
readiness.
The challenges faced by children in single-parent households
stem, in part, from economic disadvantage (see section on family economic risk,
above). Children growing up in families
headed by a single mother are five times more likely than children in
two-parent families to be poor (U.S. Department of Health and Human Services,
1999). Indeed, the vast majority of
families receiving welfare are headed by a never-married, divorced, or
separated mother. Nevertheless, some
researchers have found that even after controlling for family socioeconomic
status, race/ethnicity, and other background characteristics, children from
divorced or never-married households are less likely to complete high school
and more likely, as adults, to be in low-wage jobs than children from
two-parent households (Amato, 1993; Amato & Booth, 1997). Still, the
differences in school outcomes for children growing up with one parent vs. two
are relatively small (Zill, Morrison, & Coiro, 1993) and therefore the
research should not be construed to suggest that children in single-parent
households are doomed for school failure.
Similarly, it should not be assumed that all children in two-parent
families are better off than children in single-parent families. Children exposed to high levels of marital
conflict exhibit higher levels of behavior problems than children who
experience divorce or separation (Morrison & Coiro, 1999).
Cohabiting Households
Nearly one-third of all births in the United States occur to
unmarried women (Curtin & Martin, 2000), yet not all of these women are
raising their children alone.
Thirty-nine percent of nonmarital births occur to cohabiting couples
(Bumpass & Lu, 2000). Research on
cohabiting couples is limited, and rarer still is research that looks at the
effects of cohabitation on children in the family. Indeed, we were unable to find any studies that examined the
impact of living in a cohabiting household on children’s school readiness
skills. What we do know from research
is that, from an economic standpoint, children living in cohabiting households
are better-off than children living in single-parent households without another
adult, but they are still not as economically secure as children living in
two-parent families (Manning & Lichter, 1996). Because cohabiting relationships
in the United States are less stable than legal marriages, and marriages that
are preceded by cohabitation are less likely to last than marriages not
preceded by cohabitation (Axinn & Thornton, 1992; Lillard, Brien, &
Waite, 1995), it is likely that children of cohabiting couples will experience
more physical, social, and economic turbulence than other children in
two-parent families.
Involvement of the Non-Resident Parent
Because of high rates of nonmarital births, divorce and
separation, many children currently reside with only one parent. For the most part, non-resident parents are
men. In general, non-resident fathers are less
involved in their children’s lives than are resident fathers. The research portrays mixed results
regarding whether contact between non-resident fathers and their children has a
positive, negative or null effect on child well-being (e.g., Furstenberg,
Morgan, & Allison, 1987; King, 1994; Peterson & Zill, 1986). Some of the discrepancies in findings may be
due to the inadequacy of most measures to distinguish between positive and
negative contact (Halle et al., 1998).
However, the research is in general agreement that financial
contributions from non-resident fathers are associated with better behavioral,
emotional and academic outcomes among children (Furstenberg, Morgan, &
Allison, 1987; King, 1994; Knox & Bane, 1994).
Stepfamilies, Reconstituted Families, and Other Family Configurations
Despite the prevalence of divorce and remarriage, there is
relatively little research on the impact of stepfamilies and reconstituted
families on child well-being. There is
even less research that specifically relates these to children’s school
readiness. One difficulty in conducting
this research is tracking the fluidity of changing family configurations. There is some information on the effects of
children living with their grandparents; nevertheless, the findings are
inconclusive. For children of unmarried
mothers, the impact of living with grandparents is mixed. Aquilino (1996) found that in some cases,
children of unmarried mothers who also lived with their grandparents had lower
school achievement, while others seemed to benefit from a more stable
environment that included extra adult monitors and supports.
Summary: Family Structure Risk
The research suggests that children who are the result of a
planned pregnancy and who reside in a low-conflict, two biological parent
family have, on average, more optimal school outcomes. For those children who reside with one
parent, having the non-resident parent remain actively involved in the child’s
life may be advantageous. Although the
research is mixed, it may be fair to hypothesize that children would benefit
from contact with their non-resident parent, provided that contact was of a positive
nature. It is clearly important that
non-resident parents continue to contribute monetarily to their child’s
upbringing, as financial support from non-resident parents has been found to be
a factor in children’s school success.
Since non-resident fathers tend to diminish their involvement with their
children as they reach school age, it may be worth exploring ways to keep men
involved (in terms of spending time, having a positive relationship with their
children, and providing financial support) at this critical point of their
children’s development.
Interventions that reduce or delay childbearing among young
women and men, and/or increase the spacing between children seem important as
well. One particularly promising
strategy is to provide first-time teen mothers with home visits from a public
health nurse who will provide important information on prenatal care, child
development and family planning.
Another promising strategy would be to expand outreach to men of all
ages for family planning services.
There is no clear evidence for the effect of cohabitation or other
family configurations on children’s school readiness at this time, though these
circumstances are related to poorer outcomes on other measures.
The child’s home environment is comprised of the interaction
that the child has with his or her parents and the cognitive and social
stimulation that is present in the household.
The stimulation may come from learning materials such as books and
educational games or from toys with which the child plays. The parents’ characteristics may also have
an effect on the child. For instance, a
parent with a high level of education and a strong work ethic may have
expectations for higher educational and occupational attainment for the
child. A depressed parent may model
more negative behaviors. The present
section on the child’s home environment will discuss those studies that have
examined the effects of the home environment on children’s cognitive and
socioemotional development.
Parental Depression
The depression of a parent, especially long-term, can have
adverse effects on a child’s behavior and mental health. Hammen and colleagues (1987) conducted a
longitudinal study to determine if mother’s diagnosis (i.e., unipolar, bipolar,
medically ill, and normal) predicted child psychological outcomes of 8- and
16-year-olds. At a six-month follow-up,
behavior problems were found to be predicted by maternal depression and the
chronic strain that depression can place on the family. It should be noted that, although
significant, the effect of depression on child behavior was relatively
small. Possibly a longer interval
between initial assessment and follow-up would have resulted in stronger
associations (Beardsley et al., 1998).
Similar results were obtained over a longer period in a study comparing
children of bipolar, unipolar and normal mothers from a baseline at 1.5 to 3.5
years to a follow-up at 5 to 6 years of age (Radke-Yarrow et al., 1992). In that study, when mother and psychiatrist
reports were combined, maternal depression, regardless of whether it was
unipolar or bipolar, was related to disruptive behavior and depression in the
child at the follow-up. Although these
studies paint a picture of a direct effect between maternal depression and
child outcomes, a broad review of the parent depression literature identified
parent-child interactions as a possible mediator of child developmental
outcomes (Downey & Coyne, 1990). As
mentioned in the child health section of this report, interventions targeting
the parent’s depression (or a two-generation approach in which both the parent
and child are treated) may be effective in reducing behavioral problems and
depression in the child.
Parent-Child Interactions
The parent-child relationship is the first important,
socializing relationship that a child may encounter. Many researchers and policymakers believe that, if changes are
sought in child outcomes, interventions should focus on the parent (Dyer
Harnish, Dodge, & Valente, 1995). Several interventions directed at parents
have been implemented and evaluated. One program, Parents as Teachers (PAT),
has been implemented and evaluated in several regions in California. Through home visits from parent educators
beginning during pregnancy, PAT seeks to teach parents to be knowledgeable and
well-adjusted parents. Wagner and
Clayton (1999) reported on two evaluations of PAT. Both evaluations used a randomized experimental design. In the first evaluation, assessing a program
serving a predominantly Latino population in Northern California, an
experimental group (n = 298) received PAT services, while a control group (n =
199) was untreated. The families in the
experimental group were visited an average of 20 times over a three year period. Visits normally lasted 45 to 60 minutes and
usually involved the mother and the children in the home. During the visit, the parent educator
conducted periodic screenings of the children’s health and developmental
trajectory and taught the parent basic parenting skills. In the other evaluation, serving adolescent
parents in Southern California, participants received one of four services: PAT
program services alone (n = 177); case management services alone (n = 174); PAT
plus case management services (n = 175); or no treatment (control group; n =
178). The program services group was
offered monthly home visits through the child’s second birthday. The services during these visits, which
averaged 10 over the two year period, mirrored those in the other program. Case management services took place separate
from the home visits. Case managers
provided referrals for physical and psychological functioning, as well s for
education, nutrition, and environmental risks.
The results, overall, produced small and inconsistent gains in child
outcomes, such as positive behaviors and cognitive development. However, children in a Latino
Spanish-speaking subsample in the PAT groups showed significant gains in
cognitive, communication, and self-help development. The most gains in the Latino Spanish-speaking group were from the
combine PAT and case management group.
In addition to the Parent as Teachers programs, several
other interventions that place paraprofessionals and professionals in the homes
of new mothers have been conducted throughout the country, such as Healthy
Families America; the Family, Infant, Preschool Program (FIPP); and the Home
Instruction Program for Preschool Youngsters (HIPPY). Overall, randomized experimental and quasi-experimental design
evaluations of these programs demonstrate improved parent-child interactions,
reduced maternal depression and increased cognitive abilities of the children
(see Powell, 1993 and the Spring/Summer 1999 edition of The Future of Children, for an overview of these projects).
One of the most successful interventions for improving both
maternal and child outcomes is the Nurse Home Visitation Program, designed by
Dr. David Olds. This program was
studied in a series of randomized, control-group studies in three different
locations: one semirural county (Elmira, NY), and two urban areas (Memphis, TN
and Denver, CO). The first randomized
trial was begun in 1977 in Elmira, New York.
Four hundred women who were pregnant with their first child were
recruited for the study (89 percent were white). Eighty-five percent of the women had one or more of the following
risk factors at study entry: low-income, single parent, or younger than age
19. The women were randomly assigned to
one of four treatment groups. Families
in Treatment 1 were given free developmental screenings of the child at age 12
and 24 months. Families in Treatment 2
received these screenings, plus free transportation for prenatal and well-baby
care through the child’s second birthday.
Families in Treatment 3 were provided the screening and transportation
services, and in addition were provided with a nurse home visitor during the
pregnancy. Treatment 4 families
received all services provided to families in Treatment 3, but the nurses
continued to visit the home through the child’s second birthday. During the home visits, the nurses covered
topics ranging from personal health habits (e.g., reducing cigarette and drug
use, adequate diet), to parenting behaviors (e.g., emphasizing sensitivity and
responsivity), to home safety. Results
of the study indicated significant differences in the level of child
maltreatment and child cognitive developmental outcomes based on receiving a
nurse home visitation intervention. Reviews of randomized trials of other home
visitation programs indicate that, to produce comparable effects, the
interventions must include all the elements that are part of this particular
model, including the use of nurses rather than paraprofessionals (Korfmacher et
al., 1999; Olds & Kitzman, 1993).
Additional longitudinal research suggests that the emotional
quality of the parent-child relationship is an important factor in child
outcomes. While longitudinally
examining the role of maternal depression on the development of child emotional
and behavioral problems in a sample of representative first-grade boys and
girls, Dyer Harnish et al. (1995) found that the effect of maternal emotional
states on child outcomes were partially mediated by the quality of mother-child
interactions. Zaslow, Hair, Dion, Ahluwalia, and Sargent (2000), longitudinally
studying the effects of maternal depression on child cognitive and social
development, came to similar conclusions.
A sample of 351 mothers participated in a welfare-to-work program, and
their 3- and 4-year-old children were observed in a two-year follow-up to
determine the effects of the program on child and maternal outcomes. In regard to the child outcomes, the
investigators found that the effects of maternal depression on child outcomes
were mediated by supportive parent-child relationships.
Parenting practices have also been found to moderate
parental depression. For instance, the
NICHD Early Child Care Research Network (1999) analyzed, longitudinally,
maternal sensitivity to child needs as a moderator
of the effects of maternal depression.
Researchers obtained maternal reports of depression from time of the
birth of the woman’s child until the child was 36 months of age. Observational data of the mother-child
interaction was collected from the sample of 1,215 women and their
infants. Results supported a moderating
effect, showing that high maternal sensitivity predicted better outcomes for
those children who had depressed mothers.
Further evidence of the effects of parenting practices can
be seen in the child antisocial behavior literature. Parent-child relationships have been demonstrated to have both a
direct and indirect effect on child behavioral problems (e.g., Patterson, 1986; Patterson, DeGarmo,
& Knutson, 2000). To explain this
phenomenon, Patterson (1986) has developed a performance model for antisocial
boys, based on data from a predominantly Caucasian American sample. He describes how the interaction of poor
parenting practices such as yelling at the child (what Patterson considers
“inept discipline”) with coercive behavior by the child will result in a
downward spiraling of both poor parenting strategies and child coercion. The parenting strategies thus mediate the
effects between the child’s behavior and later antisocial behaviors for boys.
Antisocial behaviors subsequently predict a decline in peer acceptance, child
self-esteem, and academic performance.
A more recent longitudinal analysis by Patterson, DeGarmo, and Knutson
(2000), based on a similar sample, sought to discriminate between hyperactivity
and antisocial behaviors in children.
Although the initial analysis found hyperactivity to have a strong
relationship to boys’ antisocial behavior, the relationship disappeared once
parent discipline was entered into the equation. The parent’s antisocial behavior was also a strong predictor of
later antisocial actions of and hyperactive outbursts by the child. The long term effect of child antisocial behavior
is demonstrated by the finding that boys’ antisocial behaviors have a very
strong relationship to the early onset of delinquency.
Analyses of the National Household Education Survey (NHES, a
nationally representative sample of households in the U.S.) has found that
fathers’ involvement both at home and at school is significantly related to
children’s school success, even after accounting for mother involvement (Nord,
Brimhall, & West, 1997).
Furthermore, observational studies have noted that paternal praise (as
opposed to harsh words or indifference) is associated with higher educational
goals, achievement and better classroom behavior on the part of the child
(Feldman & Wentzel, 1990; Radin, 1986; Smith, 1989; Wentzel & Feldman,
1993). Although these studies report
findings for children already in school (indeed, many studies focus on
adolescents), their findings suggest that similar levels of paternal
involvement would foster children’s transition to schooling.
The Physical and Social Environment (The HOME)
The quality of the physical environment along with the
social and emotional interaction with the parent in the home is also important
to school readiness (Caughy, 1996; Hu, 1987; Klebanov, Brooks-Gunn, McCarton,
& McCormick, 1998; Korenman et al., 1995; Liaw & Brooks-Gunn, 1993;
Molfese, DiLalla, & Bunce, 1997; Molfese, DiLalla, & Lovelace, 1996;
Teo, Carlson, Mathieu, Egeland, & Sroufe, 1996). In order to assess both the physical and social qualities of the
home, an index, the Home Observation for Measurement of the Environment (HOME)
was developed (Bradley & Caldwell, 1984).
The index measures eight variables within the home that are thought to
affect child development: learning materials, stimulation of communicative
competence, physical environment, parental nurturance and responsivity,
academic stimulation, modeling and encouragement of social maturity, variety in
experience and acceptance of the child.
This psychometrically sound measure has been used in numerous national,
regional and local surveys and studies. A shortened version, HOME-SF, has been used
in a number of large-scale surveys with similar findings.
Higher scores on the HOME have been found to predict
positive cognitive and social development among children. For instance, a child’s HOME score was the
single most important predictor of intelligence among Caucasian American
children from three to six years of age (Molfese, DiLalla, & Bunce,
1997). Liaw and Brooks-Gunn (1993)
conducted an analysis to examine the cognitive development of African American
and Caucasian American low birthweight, preterm infants from 12 to 36 months of
age. The analysis supported HOME
scores, along with neonatal health, and a parenting education and health
screening intervention, as variables that discriminated among cognitive development
groups, with more enriching environments predicting better cognitive outcomes
(e.g., high and stable cognitive development being better than low and stable
cognitive development or high cognitive development declining to average). HOME scores have also been used in assessing
preschool cognitive and verbal abilities of Caucasian American infants
(Molfese, DiLalla, & Lovelace, 1996).
HOME scores measured at three years accounted for significant and unique
variance in predicting cognitive and verbal outcomes at four years in a sample
of 41 boys and 53 girls. However, when
broken down into subscales, only academic and language stimulation were found
to be uniquely predictive.
HOME scores have also been shown to mediate and moderate the
effects of other variables in predicting positive outcomes of school
readiness. Klebanov and colleagues
(1998) found that the HOME mediated the effects of family income and family
risk scores on cognitive development at three years of age in preterm low
birthweight children. Research has also
supported the home environment as a moderator of cognitive outcomes. Academic readiness of school-age children
was assessed using National Longitudinal Survey of Youth data to examine the
effects of perinatal and post-natal health, family income, maternal education
and the quality of the home environment (Caughy, 1996). HOME-SF scores and maternal education,
considered as a composite score of the home environment, added unique variance
to the model. HOME scores also
moderated the effects of poor infant health on cognitive outcomes, with a
better home environment tempering the negative effects of poor health. Similar results were found when early
cognitive and motor developmental status and home environment (in this case,
maternal responsivity to and involvement with the child) at 12 months were
examined as predictors of cognitive performance at 24, 36, and 48 months of age
(Hu, 1987). High scores on
developmental and home variables each predicted general cognitive ability individually
and together. When combined with
parent-child attachment and child’s psychological adjustment in a comprehensive
index, HOME scores at 30 months of age predicted math and reading achievement
in first, third, and sixth grades (Teo et al., 1996).
Summary: The Home Environment
As can be seen, there are different components of the home
environment that can affect child outcomes.
Maternal depression, parent-child interactions, and the physical
environment all have been found to be related to children’s developing
cognitive and socioemotional competencies. Results across multiple studies seem to
suggest that parenting practices and parent-child interactions might be the
best places to direct interventions.
Parenting practices mediate the direct effect between maternal
depression and cognitive and social outcomes for children. A coercive parent-child relationship
encourages deviant child behavior. Even
the physical environment in the home may be mediated by parenting
practices. For instance, a competent
parent may make sure to provide sufficient stimulation and opportunities for
the child to learn. This would be in
line with one of the three main goals of the National Education Goals
Panel: “Every parent in America will be
a child’s first teacher and devote time each day helping his or her preschool
child learn; parents will have access to the training and support they need”
(NEGP, 1997a, p. 1). Given the effect
that parent-child interactions and parenting practices can have on a child, it
appears to be worth investing in giving parents the training and support they
need to be their child’s first teacher.
Family-level factors, including family economic status,
household composition, and the physical, emotional, and psychological
components of the home, are all important for the child’s social and cognitive
development. Family planning behaviors
that increase birth spacing also benefit children’s cognitive and physical
outcomes. Education-based interventions
using well-qualified educators (e.g., public health nurses) to increase the
parenting knowledge of mothers have been used effectively in specific
populations. Further evidence with more
diverse populations, however, is needed.
The relatively sparse data on resident fathers precludes us from making
recommendations concerning males in the household at this time. However, data suggest that positive contact
between non-resident parents (usually fathers) and children may benefit
children’s cognitive and social development.
Furthermore, fathers’ involvement with children at home and in their
children’s schools boosts children’s school performance and enthusiasm for
school, above and beyond mother involvement.
Services related to early childhood care and education are
important in preparing children to be ready for school. It is estimated that 53 percent of all 3-
and 4-year-old children are enrolled in early childhood programs including Head
Start, day-care centers, and preschools.
However, there is a difference in enrollment rates by socioeconomic
status—58 percent of families above the poverty threshold enroll their children
in programs, compared to 41 percent of those below the poverty threshold (U.S.
Department of Health and Human Services, 1999). This is cause for concern because research suggests that poor
children particularly may benefit from such programs, and enrollment might even
be a necessary precursor for them to begin school ready to learn (Barnett,
1995). Also of concern is the fact that
many programs lack quality. Low-quality
programs often just meet the routine care needs of children but provide few
opportunities for meaningful learning activities or language stimulation
(Peisner-Feinberg & Burchinal, 1997).
Quality early child care and education programs can enhance
cognitive, emotional and social development, especially among economically
disadvantaged preschoolers (Kagan & Neuman, 1997). Participation in programs can lead to
short-term gains in IQ scores, better kindergarten achievement, lower rates of
grade retention and special education placement and higher rates of high school
graduation (Barnett, 1995). In the
following section, several studies of early childhood care and education will
be summarized and the best practices found in each study will be highlighted.
The High/Scope Perry Preschool Project was conducted in
Ypsilanti, Michigan and started in the late 1960s. It included a sample of 123 African-American children born in
poverty who were divided into two samples at ages 3 or 4. The control group received no intervention
while the experimental group was placed in a high-quality preschool (High/Scope
Educational Research Foundation, 2000).
The program encouraged active
learning among children by allowing them to initiate activities and control
their environment. Children learned in
an environment rich with materials and then reported back to their teachers on
what they had achieved. Teachers
encouraged children’s experiences in the areas of initiative, social relations,
creative representation, music and movement, language and literacy, and logic
and mathematics. The program also
included weekly home visits by teachers in order to discuss and practice
activities for parents to carry out with their children. Teachers received curriculum training and
supervision, and only five or six students were assigned to each teacher
(Schweinhart & Weikart, 1993).
Data were collected on program participants and control
group children every year between ages 3 and 11, and then at ages 14, 15, 19
and 27. At the latest assessment,
information was gathered from 95 percent of the original participants through
interviews, data from schools, social services, and arrest records. The results are striking—71 percent of
program participants had graduated from high school compared to 54 percent from
the control group. Fifty-nine percent
of program participants had received welfare assistance as adults, compared to
80 percent of controls. Fifty-seven
percent of preschool participants had out of wedlock births, compared to 83
percent of controls (High/Scope Educational Research Foundation, 2000). Thus, participation in a high-quality
preschool program that included a great deal of contact between the home and
school produced significant, long-lasting results.
The goal of the Carolina Abecedarian Project was to give
high-risk children educational experiences early in life so that they could
achieve school success. Between 1972
and 1977, four yearly cohorts were randomly assigned to receive either
high-quality child care (a total of 57 children) or no treatment (a total of 54
children). Ninety-eight percent of the
children were African American and born to a single mother who had not
graduated from high school at the time of the child’s birth. The intervention was begun by age 3 months,
and children in the treatment group received child care for 6 to 8 hours per
day, 5 days per week through kindergarten entry at age 5 (Horacek, Ramey,
Campbell, Hoffman, & Fletcher, 1987).
Some services, such as nutritional supplements, social work services and
medical care were provided to control group families in order to ensure that
those were not the only factors accounting for different outcomes between the
two groups (Ramey & Campbell, 1991).
The caregiver-to-infant ratio in the child care center was
1:3 initially, and increased to 1:6 as children got older (Campbell &
Ramey, 1995). The activities that
teachers carried out targeted four areas: cognitive and fine motor skills,
social and self-help skills, language, and gross motor skills. Activities were individualized for infants
and children based on readiness. As the
children reached age 3 or 4, the center became a preschool program with centers
for a variety of activities. Activities
were designed to seem playful and spontaneous to the children. Teachers were trained to place a special
emphasis on language development by speaking with children about their daily
lives in a developmentally appropriate way (Ramey & Campbell, 1991).
As early as age 18 months, children receiving the
intervention had higher scores on IQ tests, and the difference persisted when
children were tested at ages 12, 15, and 21 years. Furthermore, children who received treatment had higher scores on
reading and math tests from elementary school through early adulthood. At age 21, 35 percent of experimental
participants had graduated from or were enrolled in a four-year college,
compared to 14 percent of control participants (Frank Porter Graham Child
Development Center, 1999b). Regarding
grade retention, by age 15, 31.2 percent of the preschool subjects had been
retained, compared to 54.5 percent of the control group. The findings for special education were
similar—24.5 percent of the experimental group were placed in special education
compared to 47.7 percent of the control group (Campbell & Ramey,
1995). Thus, as with the Perry
Preschool Project, the long-term findings indicated long-term positive impacts.
Unlike the preceding studies, the Cost, Quality, and
Outcomes Study did not utilize an experimental, random-assignment design. Instead, 401 preschools in four different
regions were assessed for quality beginning in 1993, and then the progress of a
total of 826 children was tracked through the second grade (Frank Porter Graham
Child Development Center, 1999a). The
quality of child care centers was assessed by examining the physical
environment, the use of teaching practices related to motor, language,
cognitive and social development, teacher sensitivity and responsiveness and
teaching style (child-centered vs. didactic) (Peisner-Feinberg & Burchinal,
1997).
Quality of child care was correlated with assessments of
cognitive skills and language development during the preschool year. Ratings of the relationship between the
teacher and individual children were correlated with socioemotional
development. Results were found after
controlling for familial factors such as poverty and maternal education that
might explain differences in outcomes.
The strongest effects were found for children at risk for school failure
(Peisner-Feinberg & Burchinal, 1997).
These effects were seen through the second grade as children who had
received higher quality care had better language and math skills, and fewer
problem behaviors. Children who had
closer relationships with their preschool teachers exhibited better classroom
social and thinking skills and language ability at the beginning of elementary
school (Frank Porter Graham Child Development Center, 1999a). Thus the investigators found that two
aspects of the preschool setting are important for young children as they make
the transition into school—classroom practices (materials and activities) and
relationships with caregivers.
Perhaps the most well-known early childhood education
program is Head Start, a large-scale and publicly funded project serving 3- and
4-year-old children from families below the poverty line. The idea behind its creation in 1965 was to
narrow the gap between children who are economically disadvantaged and those
who are not in terms of school readiness and long-term academic
achievement. Head Start includes many
services such as early childhood education and nutrition, and medical, dental
and mental health (Lee, Brooks-Gunn, & Schnur, 1988). Collaboration with other service providers
and community programs is sought.
Another important component to Head Start is parent involvement. Teachers make home visits, and parents are
encouraged to volunteer in classrooms, attend parent education classes and
enroll in adult education (Mallory & Goldsmith, 1991). There is flexibility in the way that Head
Start is implemented—providers can vary services as long as they meet minimum
requirements. For example, they can
choose their own curriculum and decide to how best meet the specific needs of
particular communities (Advisory Committee on Head Start Research and
Evaluation, 1999).
Many varied studies have been conducted on Head Start
programs around the country. Early
gains on cognitive test scores among those who attend Head Start compared to
those who do not attend preschool or other programs have been found (Lee,
Brooks-Gunn, & Schnur, 1988).
However, several studies suggest that such short-term gains fade with
time, but that long-term gains such as higher rates of high school graduation,
fewer grade repetitions and lower rates of special education placements accrue
to Head Start attendees. Thus,
advantages persist in ways that are measured differently than simple cognitive
gains, and perhaps initial boosts in test scores set children on a trajectory
for better school achievement (Lee, Brooks-Gunn, Schnur, & Liaw,
1990). Currently, proposals are being
reviewed by the U.S. Department of Health and Human Services for an
experimental impact study of Head Start.
As a result of this study, in years to come, more definitive evidence of
the influence of Head Start on school readiness will be available.
What can be learned from such studies about the best
practices for early childhood care and education? It is very clear that quality matters. Children need environments in which they are not just provided
with basic care but are stimulated to learn.
Successful programs emphasize a number of different areas of development
including cognitive, language, social and emotional. The studies cited
emphasize the fact that small student/teacher ratios are beneficial. Teachers should also work to foster close,
caring relationships with their students, an important element in positive
socioemotional development.
Furthermore, contact between the program and children’s homes appears to
be important—parent involvement should be sought and encouraged so that parents
know what their children are learning and are able to extend early education
into their homes.
The Head Start model emphasizes that, along with
collaborations between schools and homes, collaborations with community
programs and service providers are also important so that children can receive
all of the services they need. Finally,
it is clear that quality early childhood care and education is most needed for
children in poverty—they are the children who benefit the most from it yet they
are currently the least likely to receive it.
For five-year-old children, the transition from preschool or
home to kindergarten can be an extremely stressful time. Children face new expectations for
independence and responsibility, goals that are more formal than those in
preschool and larger class sizes.
Children begin to have interactions with teachers that focus on their
academic progress. They must also learn
to negotiate more formalized routines (Rimm-Kaufman & Pianta, under
review). The transition to kindergarten
is also important when considering how poverty and minority group membership
affect schooling. Stratification begins
early and can affect the entire course of a student’s education. Many poor and minority students enter school
lacking some of the competencies required by the school, creating a mismatch
between what children know and what the school expects. This sets certain students on a trajectory
for school failure (Entwisle & Alexander, 1993). Also, despite the fact that the transition to kindergarten is a
critical period in children’s lives, many schools do not have specific
guidelines in place. Furthermore,
little research has been done on what practices are best under what
circumstances. In the following
section, recommended transition practices will be summarized, along with the
supporting evidence that is available.
The transition to kindergarten involves connections between
many different spheres including individual children, families, schools and
communities. In order to facilitate a
successful transition, the connections between the spheres must be fostered
(Rimm-Kaufman & Pianta, under review).
To begin with individual children, if a child possesses certain
characteristics and experiences, the transition into elementary school is
likely to be easier. Social skills,
including play behavior, peer group entry strategies and communication skills,
have been found to be important. The
development of such skills can be facilitated by interactions with peers before
school entry, either through parents’ creation of chances for their children to
interact with other children or through preschool experience (Maxwell &
Eller, 1994). Preschool can also have
an impact on cognitive development (see section on early child care and
education). Entwisle (1995) suggests
that cognitive gains can aid in the transition to school by helping children be
placed in higher ability groups and creating greater teacher expectations,
setting children on the track for continued success.
The active role that schools take when children are
transitioning, and even before the transition begins, is essential. One recommended practice is contact between
kindergartens and preschools (Smolkin, 1999).
Kindergarten teachers should visit the preschools their students will be
coming from in order to understand the curriculum and to get to know their
future students through interactions with them and through preschool teachers’
reports. Kindergarten teachers can then
begin planning for individual students before they enter their classrooms,
helping to ease the transition (NEGP, 1998b).
Another way to build connections between preschool and kindergartens is
by arranging for preschool children to visit the kindergartens they will be
attending. If children know what to
expect, they might feel less anxiety early in kindergarten. The effectiveness of this practice was
demonstrated by the Head Start Transition Project (Kagan & Neuman, 1998,
details below).
If connections between kindergartens and preschools are
important, perhaps connections between schools and homes are even more so,
especially because not every child attends preschool. The Goal 1 Ready Schools Resource Group suggests that schools
should reach out to families well before children start school (NEGP,
1998b). Practices can include written
or personal communications to parents about the school and steps they could be
taking with their children, such as reading to them. Other options are home visits by teachers or principals or
orientation sessions before the school year begins. It is recommended that practices should continue after school
starts with family visits to schools, the solicitation of parent volunteers in
classrooms and teacher visits to homes (Smolkin, 1999). When the connections between schools and
homes are strong, parents know what their children are learning and can support
the development of that new knowledge in their homes (Melton, Limber, &
Teague, 1999).
Several studies have demonstrated the effectiveness of
school-home connections during the transition to kindergarten. The Head Start Transition Study, found that
the more frequently kindergarten teachers used certain transition practices,
the lower child stress was during the first month of school. Practices included training parents to use
their rights and responsibilities in schools, creating support groups for
parents and distributing book lists and recommending activities for parents to
carry out with their children during the summer before kindergarten (Kagan
& Neuman, 1998). Another program
called On Board Early is used in Baltimore County, Maryland, and serves
families during the transition to both preschool and kindergarten. Teachers maintain communication with parents
through conferences and logs that pass back and forth. Trained, specialized staff members are hired
as liaisons, and they make home visits and suggest activities to parents to
carry out with their children. Although
the direct effects on children as they transition have not been reported,
investigators have found that On Board Early has led to greater involvement by
parents in their children’s early literacy and closer parent-teacher
relationships during the transition into school (Swick et al., 1997).
Connections between schools and community resources should
be added to the interplay between children, homes and schools. If schools are connected to health and
social services, teachers can respond to the individual needs of their
students, helping to ease students’ transitions into kindergarten (NEGP,
1998b). Such connections can also serve
as a useful means for educators to find and contact the parents of young children
who are not enrolled in preschool as a way of reaching out to them before they
enter kindergarten. Furthermore,
connections with community organizations would allow educators to help more
easily children in transient or homeless situations receive the services they
need. Through connections between
children, parents, schools and community organizations, the transition into
kindergarten can be made easier, setting children on a trajectory for
continuing school achievement and positive outcomes.
Another important issue in the transition to kindergarten is
age of entry. There are certain
researchers and educators who believe that school readiness is based on
biology, and that children have inner clocks for development. Based on this reasoning, if children are too
immature for school at age five, they simply need to be held out for a year—a
practice called redshirting. Redshirting has been quite common, even
among parents who hold their children out of kindergarten for a year simply to
give them an edge over other children.
The practice has been found to lead to the acceleration of expectations
in kindergarten, creating classrooms that are developmentally inappropriate for
the 5-year-olds they are supposed to serve (May & Kundert, 1997). The necessity of redshirting is not
supported by the evidence. Gullo and
Burton (1992) compared older and younger kindergarten students and found that
there is no difference in children’s development or readiness for first grade
at the end of kindergarten based on age.
The finding suggests that any differences between children that are
present at the beginning of kindergarten can be alleviated by the end of the
year.
May and Kundert (1997) espouse the view more commonly held
today that development is not based solely on biology, but that learning
depends on experience and the learning environment. Therefore, the school must accommodate to the child, no matter
what his or her developmental stage. One approach used by some schools has been
to place children who are found to be unready for kindergarten or first grade
in modified classes. This creates an
extra year of schooling for children as it becomes either an extra year of
pre-kindergarten or a half step between kindergarten and first grade. However, the evidence does not support the
practice (both in terms of academic skills and achievement and social,
emotional and behavioral outcomes) and suggests that developmentally immature
children benefit the most from being in regular classrooms (May & Kundert,
1997). This shows that the definition
of school readiness should be broad enough to include children at varying
stages of development. Because all
children are ready for school, except for those with extreme developmental
disabilities, it is necessary for schools to have in-depth, careful screening
programs and individualized services for children with serious problems. This raises another pressing concern about
school transition practices—assessment.
As discussed earlier, assessment at school entry can serve
four purposes: to plan teaching by finding out what children already know and
what they need more help in; to identify children with special needs; to
evaluate programs and monitor trends; and to hold schools and programs
accountable (NEGP, 1998a). Problems
arise when tests intended for one purpose are used for another. Furthermore, tests are often used
inappropriately for identifying children to hold out of kindergarten and
tracking children into ability groups.
Such practices can be dangerous because tests are often not accurate
with young children as they develop and change at such a rapid pace. For example, Pianta and McCoy (1997) found
that assessments at kindergarten entry were not very predictive of outcomes
such as grade retention and standardized test scores in the third grade. Another danger of using assessments to make
important decisions about children is that low scores are most often found for
poor, minority children (Ellwein, Walsh, Eads, & Miller, 1991). Such children are the ones who benefit the
most from a stimulating classroom experience and are not helped by being held
out of kindergarten for a year or by being placed in low-ability groups or
classes.
The Goal 1 Early Childhood Assessments Resource Group (NEGP,
1998a) suggests that assessments should always be beneficial for children or
for programs. They should be used for a
specific purpose, be age-appropriate and be linguistically appropriate. Parents and teachers should be used as
sources of information. Educators who
use assessments to make decisions about individual children should recognize
that tests might not be reliable until children are in third grade or
older. Assessments should be used as a
first step to identify students who might not be ready for school, but then
further testing should be carried out so that informed decisions can be
made. Then individualized services,
rather than simply holding children out of kindergarten, should be designed.
Final Report to the Knight Foundation