Child care_ early learning_ and by pengxiuhui


									Child Care, Early Learning, and
   Children’s Development

       Cathryn Booth-LaForce, PhD, FAPS
  Charles & Gerda Spence Professor of Nursing
          Conference on Early Learning
              September 24, 2007
Media Reports
    Some Recent Headlines
   “Poor Behavior is Linked to Time in Day Care”
                             ---New York Times

   “Study Links Child Care to Acting Out”
                             ---Associated Press

   “How nurseries „still breed aggression‟”
                              ---London Times
   “Day Care is Linked to Behavior Lasting Through
    6th Grade”
                         ---The Wall Street Journal

   “Few Effects of Poor Daycare Last Past Age 11”
   “Quality of Early Child Care Makes a Difference
    But Good Parenting Matters More, US Study
                          ---Medical News Today

   “Day Care News: Parents, You Count Most”
“A day care center, the sort of place in which
 bullies are bred, according to a new study”
                      ---New York Times, 2001
Some statistics….
National Maternal Employment
      Statistics (2005)

   60% of married mothers of preschool
    children are in the work force

   53.5% of married mothers of infants
    are in the work force

                   ---Monthly Labor Review, Feb, 2007
        Child-Care Statistics
   Each year, parents + government
    spend roughly $50 billion on child
                                 ---Univ. MD, 2006

   About 12 million children are in child
    care in the U.S. (9.8 million under 5
    years in 40+ hours)
                         ---U.S. Census Bureau, 1999
Why are young children spending
  so much time in child care?

       National trends—maternal
       Families need two incomes
       Work preference
Costs to Women of the ―Off Ramp‖

   Short (1-2 year) time out—lose 18%
    of earning power (28% in business
    and banking/finance)

   3 or more years—lose 37% of
    earning power

                       ---Center for Work-Life Policy, 2005
 What are the effects of
childcare on children and
 NICHD Study of Early Child
Care and Youth Development
NICHD Early Child Care Research Network
Virginia Allhusen     University of                 Lyz Jaeger           St. Joseph’s University
                        California: Irvine          Deborah Johnson      Michigan State University
Mark Appelbaum        University of                 Jean Kelly           University of Washington
                        California: San Diego       Bonnie Knoke         Research Triangle Institute
Jay Belsky            Birkbeck College,             Nancy Marshall       Wellesley College
                        University of London        Kathleen McCartney   Harvard University
Cathryn Booth-LaForce University of Washington      Fred Morrison        Loyola University Chicago
Robert Bradley        University of Arkansas        Phil Nader           University of
                        at Little Rock                                      California: San Diego
Celia Brownell        University of Pittsburgh      Marion O’Brien       University of Kansas
Peg Burchinal         University of North           Margaret Owen        University of Texas-Dallas
                        Carolina at Chapel Hill     Ross Parke           University of
Bettye Caldwell       University of Arkansas                                California: Riverside
                        at Little Rock              Chris Payne          University of North
Susan Campbell        University of Pittsburgh                              Carolina at Greensboro
Alison Clarke-Stewart University of                 Deborah Phillips     Georgetown University
                        California: Irvine          Robert Pianta        University of Virginia
Martha Cox            University of North           Suzanne Randolph     University of Maryland,
                        Carolina at Chapel Hill                             College Park
Sarah L. Friedman     NICHD                         Wendy Robeson        Wellesley College
Willard Hartup        University of Minnesota       Susan Spieker        University of Washington
Ty Hartwell           Research Triangle Institute   Deborah Vandell      University of Wisconsin-
Kathryn Hirsh-Pasek   Temple University                                     Madison
Aletha Huston         University of Texas-Austin    Marsha Weinraub      Temple University
   NICHD Grant #HD25447
   Susan J. Spieker, Co-PI
   Jean F. Kelly, Co-PI
   Sumi Hayashi, Site Coordinator
   Participating families
…and a small army of data collectors
   Purpose of the NICHD

To examine how variations in child-rearing
contexts (childcare, home, school, etc.) are
related to children’s social, emotional,
cognitive, and language development and
Ecological Model: Concurrent and
     Longitudinal Relations
 Family and maternal                             Demographic
   characteristics                              characteristics

       Home                               Childcare and school
    environment                               environments


                      Social, emotional,
                  cognitive, language, health
              Families in the Study
                            Recruited in these locations
   1,364 eligible births
    occurring during 1991
   Sampling designed to
    assure adequate
    representation of
    major socio-
    demographic niches
   Ten data collection
   24 hospitals
          Exclusion Criteria
   Mother younger than 18 years
   Family planned to move
   Multiple birth
   Infant had disability
   Infant stayed in hospital > 7 days
   Substance abuse—mother
   Mother did not speak English
   > 1 hour from lab site
   Extremely unsafe neighborhood
Current Locations of the
  1009 Study Families
   Who are the Families in the
                           Income-to-Needs (1 m)
Maternal Education
                           13% poverty
                           18% near-poverty
10%   no HS degree
                           69% non-poor
21%   HS degree or GED
33%   some college
21%   college degree
15%   postgrad education   Marital Status (1 m)

                           14% single
      Race/Ethnicity (%)
                      Study   USA
White, non-Hispanic    75      65
Black, non-Hispanic    13      16
Hispanic                6      15
Asian                   1       3
Native American         1       1
Other                   4      NA
 Work/School Plans at Birth

Employment/School Plans--child‟s first year:

          Yes            79%
          No             18%
          Don’t know      3%
Planned Hours of Work/School

      < 10      4%
      10-29    29%
      30+      67%
Early Entry into Many Hours of

   Median age at entry—3 months
   Mean hrs per week (1-54 months)—31 hrs
   “Informal” care arrangements most
    common during first year
Study Assessment Ages

    1, 6, 15, 24, 36, 54 months
    K, Grades 1, 2, 3, 4, 5, 6
    Age 15
Types of Assessments

     Interviews
     Questionnaires
     Observations
     Direct Assessments
     Records
   Mothers
   Fathers/Partners
   Child-care Providers
   After-school Care Providers
   Teachers
   Principals
   Children
   Best Friends
   Mothers and Teachers of
    Best Friends
   Nurse Practitioners

   Home
   Child-care arrangements
   School
   After-school settings
   Neighborhood
         Number of Variables
          (birth to Grade 6)

   Raw data—70,000 variables
   Analysis data sets—8,700 variables
   Data are available to other investigators
What did we measure in
the child-care setting?
         Child-Care Measures
   Quantity : Hours/week
   Observed Quality
   Type :
      Relative/ In Home
      Child-care Home

      Child-care Center
        Quality of Care
(Observational Record of the Caregiving Environment)
    6, 15, 24, 36, 54 months of age
          ORCE Behaviors
•   Shared positive affect   •   Stimulates cognitive
•   Positive physical            development/teaches
    contact                      academic skill
•   Responds to              •   Facilitates behavior
    talk                     •   Mutual exchange
•   Speaks positively to     •   Negative/restricting
    child                        actions (reversed)
•   Asks questions of        •   Speaks negatively to
    child                        child (reversed)
•   Other talk to child
            ORCE Ratings

•   Stimulation
•   Sensitivity/responsiveness
•   Positive regard
•   Detachment/disengagement
•   Flat affect
•   Intrusiveness (at 36, 54 months)
•   Fosters exploration (at 36, 54 months)
       ORCE is related to…

   Quality of the physical environment
   ―Regulable‖ features of child care
             Type of Care:
           Child Care Centers
   Larger groups of
   More toys
   More structured
   More children per
   Children grouped
    by age
Child-Care Homes
           More informal care
           More time in free
           Varying ages of
            children; often
           Activities are more
      Relative/In-Home Care
   Most informal
   Care provider
    follows usual
    routine and
    incorporates child
   Little structure
   May be just
    caregiver and child
    or other related
…including Fathers
      Quality of Care by Type

   Full range of quality in every type of care.
   Especially during the first two years, the
    average quality of care was higher in less
    formal care with fewer children.

---NICHD ECCRN, 1996
What did we measure in
  the home setting?
Parenting Quality
Quality of Mother-child
Ratings from 15 minute videotaped
 structured play interactions:

   Sensitivity to distress
   Sensitivity to nondistress
   Detachment
   Intrusiveness
   Cognitive stimulation
   Positive regard
   Negative regard
   Flat affect
Home Observation for Measurement of
     the Environment (HOME)

   Checklist of quantity and
    quality of support and
    stimulation available to
    the child in the home
    (e.g., books, age-
    appropriate play
    materials, appropriate
    responses to child, affection)

   Based on interview of
    mother with child
Is early, extensive participation
  in childcare a risk factor for
     insecure attachment?
    A 30-sec introduction to attachment…

   Secure attachment—”comfortable sense of trust
    in the primary caregiver”
   Security is predicted by
    warm, sensitive, responsive parenting from
    primary caregiver
   Insecurity is predicted by
    detached, uninvolved, unresponsive, intrusive
   Insecurity is a risk factor for subsequent
    behavior problems, problems with
    peers, relationships, poor social competence
          Attachment Results

   Security/insecurity was related to the
    quality of parenting.
   Security/insecurity was not related to the
    quality, quantity, or age of entry into
   Dual-risk effect: Very insensitive parenting
    plus poor quality childcare, or many hours
    in childcare—greater proportion of
    insecure infants.

---NICHD ECCRN, 1997
   Child Care and Child
Outcomes: More Questions
 Is child care related to child cognitive,
  language and social outcomes at
  4.5 years?
 If so, how? What are the specific effects
  of quality, quantity of care, and type of
  care on child outcomes?
 How big are these effects?
       Statistical Controls

Site, gender, ethnicity, maternal education,
proportion time mother had partner in
household, maternal depression, income,
maternal sensitivity
I. Quality of Early Child

   For preschoolers, higher quality
    care over the first 4.5 years is
    associated with

     better   pre-academic skills

     better   language skills
     Differences in Child Care Quality vs.
      Differences in Parenting Quality:
            Language Competence
             105                                     105              m=102.5
Language                                Language
Competence                              Competence
             100               m=97.6                100
             95                                      95

             90                                      90

             85                                      85

              0                                       0
                     Low        High                         Low        High

                   Child Care Quality                      Parenting Quality
                         d = .29                                d = .87
II. Type of Child Care

   More experience in child care
    centers is associated with

     betterlanguage skills
     more problem behaviors
III. Quantity of Child Care
     (all types of care)
 More hours of child care over the
 first 4.5 years is associated with

    more problem behaviors
     (aggression, disobedience)
   All types of care
   Not a function of quality
   No threshold
   Not just assertive behavior
   Not clinical levels of aggression
   Differences in Amount of Child Care
   vs Differences in Parenting Quality:
    Behavior Problems at 54 Months
         55                                        55
Caregiver                              Caregiver
Reported                    m=51.7
                                       Reported         m=51.0
Behavior                               Behavior
Problems 50                            Problems    50               m=48.8

         45                                        45

         0                                         0
              <10 hrs/wk   30>hrs/wk                     Low         High
               Quantity of Care                         Parenting Quality
                   d = .38                                   d = .23
Grade 5/6 Results
       Statistical Controls
Site, gender, ethnicity, maternal education,
proportion time mother had partner in
household, maternal depression, income,
maternal sensitivity, classroom quality,
after-school hours.
I. Quality of Early Child

   Higher quality care over the first
    4.5 years is associated with:

     higher   vocabulary scores in Grade 5
II. Type of Child Care

   More experience in child care
    centers is associated with

     morebehavior problems in
     Grade 6
Comparison of Effect Sizes

     Vocabulary:
         Child-care quality--.06
         Parenting quality--.25-.33
     Behavior Problems
         Prop. center care--.08-.12
         Parenting quality--.11-.19
Behavior-Problem Mean Scores

      Average score               50.0
      ―At-risk‖ score             60.0
      No time in child care:      49.6
      1-2 years in center care:   50.0
      3 years in center care:     51.4
       (10% of sample)
      4 years in center care:     52.0
       (5% of sample)
Effect Size Considerations…

     Length of time between
     Parenting—genetic and
      environmental influences?
     Childcare—cumulative societal
Classroom effects?
    Are the effects of child care

   Dmitrieva, Steinberg, & Belsky (in press)
   3440 children in 282 Kindergarten
    classrooms—ECLS-K study
   Beginning and end of Kindergarten year
   Externalizing problems (teacher report)
   Achievement (teacher report + testing)

What are classroom composition
effects, beyond demographic
variables and effects of the child’s
individual child-care history on
her/his development?
For children with limited child-
      care experience…..

 Those in classrooms with many
 children with extensive center-based
 child-care experience had better
 academic growth than did children in
 classrooms with few children with
 extensive center-based experience.
For children with limited child-
      care experience…..

 Those in classrooms with many
 children with extensive child-care
 experience were more aggressive
 and disobedient than were children
 in classrooms with few children with
 extensive child-care experience.
Back to the NICHD Study…
What are the limitations of the

   It is not nationally representative
   We did not include the most high-risk
    disadvantaged families
   It is not an ―experiment‖
    Therefore, we cannot claim that child
    care causes child outcomes.
    What are the strengths of the
   Largest, longest-term study of child care in
    relation to child development
   Prospective study
   Began at birth
   ―Ecological‖ model – included data about
    the family, home, school, neighborhood
   Multiple aspects of child development &
   Exceptionally high-quality data
   Diversity of investigators’ views
Summing up: What do these
     results mean?
Parent(s) Matter!
     Parents are spending more
      time with their children
   1965--mothers spent 10.2 hrs per week
    tending to their children
   2003--mothers spent 14.1 hrs per week
   1965—fathers spent 2.5 hrs per week
   2003—fathers spent 7 hrs per week
   2003 paid work + time spent with child =
    65 hrs/wk for mothers, 64 for fathers.
                          ---Bianchi et al., 2006
In our study at 7 months of age…
Compared families in which the infant was
in 30 or more hours of childcare vs. at
home with mom since birth.
 Time interacting with mother—only 12
  hours difference per week.
 Time not related to quality of mother-
  infant interaction or child outcomes.

                        ---Booth et al., 2002
  A conservative politician…

Our study ―…proves what has long been
obvious, that kids do better if nurtured by
their own parents.‖
 Is there a developmental
advantage to staying home
        with mom?
―Thank goodness I’m a stay-at-
        home mom!‖
     No evidence that exclusive
       maternal care is best

   Only 52 children with exclusive maternal
    care, birth to 60 months
   Not different from child-care children in
    cognitive, language, social development
   High-quality child care > exclusive
    maternal care > low-quality child care—
    cognitive and language development.
Are childcare quality and
  quantity important?
    Results of other studies…

   Importance of child-care quality for
    lower-income and at-risk children
    (correlational and experimental
   Quantity findings consistent with
    those from ECLS and from a study in
Consider child-care
quantity and quality from
the perspective of the
child‟s everyday
Quality of Care in the U.S.

     Poor         8%
     Fair        53%
     Good        30%
     Excellent    9%

                 ---Booth et al., 1999
―Ira’s Discount Day Care‖
       My Recommendations
   Let’s stop making parents feel guilty and
    focus on supporting them

   Let’s stop thinking that staying at home
    with mom is the ―gold standard‖
    More recommendations…
   Family leave policies
   Flexible employment; re-entry strategies
   Improve child-care quality and choices
   Educate parents about their importance;
    practical strategies
   Pay attention to what is going on in child
            More science…
   What is the role of stress?
   Given the small effect sizes for childcare
    quality, weighing of costs and benefits of
    specific improvements
   More data about ―daily life‖
   Classroom effects
   Age 15 results
   Relationship study—Age 17.5

To top