When and Why After-School Programs Support Children’s Development

Deborah Lowe VandellUniversity of Wisconsin

April 30, 2003

Several Factors Have Contributed to an Increased Interest in After-School Programs

• High rates of maternal employment – 69% of married mothers and 71% of single mothers of 6- to

17-year-olds are employed• Concerns about

– negative effects of self-care– youth as victims and perpetrators of crime– lagging academic performance

• Inequities in access to after-school activities and programs

After-School Programs Narrowly Defined and Broadly Defined

• Narrow definition – programs that are offered by schools or other organizations on a daily basis throughout the year

• Broad definition – includes extracurricular activities, sports, clubs, and activities that are offered on a regularly scheduled basis by schools, libraries, and youth organizations

Variations in After-School Programs

• Homework clubs• Preparation for standardized tests• Extension of the school day • Science, math, & computer clubs• Organized sports and recreational sports• Music, drama• Arts and crafts• 4-H, Scouts, YMCA

Current After-School Initiatives• 21st Century Community Learning Centers

– 40M in 1997; $1B in 2002; – 1.2 million students participated in programs located at 3600 schools in 2001– 2,252 applications for funding; 310 awards in 2000– Future funding is uncertain

• Boys and Girls Clubs • State-level initiatives

– California’s After School Learning and Safe Neighborhoods Partnership (30,000 children)

• Local initiatives– L.A.’s Best, The After-School Corp (TASC)

Limitations of Previous Research

• “Evaluation of after-school activities is still limited. Often the information about a program is based on the opinions of experts instead of formal evaluations.” (Working for Children and Families, 2000, p. 7)

• “Circumstances surrounding the type of care provided, the kinds of students who attended the different programs, and what the programs themselves entailed, have rarely been studied in detail.” (Fashola, 1998, p. 3)

• Failure to control for family background and child’s prior functioning

• Failure to consider variations in program quality

Programs are more likely to have beneficial effects….

when program quality is higher

Study of Varying Quality After-School Programs

• Longitudinal study• 150 children recruited in first grade from 30 after-

school programs • Child functioning in 1st – 3rd grades• Program quality assessed at least twice yearly• Include controls for family background and child

prior functioning

Developmental theory and after-school practitioners argue that high quality programs provide students with opportunities to

• Form supportive relationships with adults • Form supportive relationships with peers • Engage in substantive activities that are

meaningful to students

In previous research, my colleagues and I found:• Children report less emotional support when after-school staff are hostile and

negative.

• Children report more positive experiences at programs that offer a greater variety of activities.

• Boys display fewer internalizing and externalizing problems in 1st grade when program staff interact more positively with children.

• Boys obtain lower reading and math grades when program staff are more negative during interactions.

• Frequent negative interactions with peers at the program predict more internalizing and externalizing problems, and poorer social skills, at school.

• Children display better social skills at school when their after-school programs are more flexible.

Measures of After-School Program Quality

• School-Age Care Environment Rating Scale (SACERS)

• Observers’ ratings of program quality

• Child reports: After-School Environment Scale (ASES)

SACERS

• 6 program features (space & furnishings; health and safety; range of activities; interactions between staff and children, parents, classroom teachers, & other staff; program structure; staff development) rated on 7-point scales

• Collected four times in 2nd grade, three times in 3rd grade,

• Settings were typically minimal quality (M = 4.0 in 2nd grade and 4.4 in 3rd grade).

Qualitative Ratings of Program Quality

• Collected 4 times in 2nd grade and 3 times in 1st and 3rd grades

• Ratings were averaged to create annual overall observed quality scores (after reflecting negative regard, negative behavior management, and chaos)

• Good reliability (M alpha = .76; range = .61 - .85)

Qualitative Ratings• Staff positive regard

• Staff negative regard

• Staff uses positive behavior management

• Staff uses negative behavior management

• Programming flexibility

• Age-appropriate activities

• Chaotic setting

Child Report of Program Quality

• After-School Environment Scale (ASES; Rosenthal & Vandell, 1996)

• Collected twice a year • 18 items in 1st and 2nd grade rated on 3-pt scales; 31 items

thereafter rated on 4-pt scales • Overall score (M = 2.4, sd = .4 in 1st and 2nd grades; M =

3.0, sd = .6 in 3rd grade) • Emotional Support (19 items)• Autonomy/Privacy (6 items)• Positive Relations with Peer (6 items)

Cumulative Program Quality

• Mean program quality (Grades 1 & 2) 2nd grade functioning

• Mean program quality (Grades 1 & 2 & 3) 3rd grade functioning

Measures of Child Functioning

Domain Source Measure

Academic grades T Mock Report CardWork habits T (Pierce, Hamm, & Vandell)Social Skills T

Loneliness C Loneliness and Social Dissatisfaction(Asher, Hymel, & Renshaw)

Depression C Children’s Depression Inventory (Kovacs)

Analytic Plan

• Model 1: Child and family characteristics• Child gender• Prior child functioning (measured in 1st grade)• Firm but responsive parenting (measured cumulatively)• Family income (measured cumulatively)

• Model 2: Cumulative program quality added

• Model 3: Cumulative program quality X child gender added

Academic Grades

Grade 2 Grade 3

N = 105 N = 74

Model 1 (R2) .533*** .454***

Child & family characteristics

Model 2 (change in R2) .037** .051**

Qualitative ratings composite

Model 3 (change in R2) ns ns Ratings composite X gender

Work Habits

Grade 2 Grade 3

N = 105 N = 74

Model 1 (R2) .465*** .353***

Child & family characteristics

Model 2 (change in R2) ns .033 p<.06

Qualitative ratings composite

Model 3 (change in R2) ns .045*

Ratings composite X gender stronger for girls

Social Skills

Grade 2 Grade 3

N = 105 N = 74

Model 1 (R2) .427*** .387***

Child & family characteristics

Model 2 (change in R2) ns ns Qualitative ratings composite

Model 3 (change in R2) ns .061**

Ratings composite X gender stronger for girls

LonelinessGrade 2 Grade 3

N = 119 N = 89

Model 1 (R2) .249*** .493**

Child & family characteristics

Model 2 (change in R2) NA .030*

Child perspective

Model 3 (change in R2) NA .021*

Child perspective X gender stronger for girls

LonelinessGrade 2 Grade 3

N = 119 N = 89

Model 2B (change in R2) .085** .106***

Emotional support (beta) -.314**

Peer affiliation (beta) NA -.345***

Autonomy (beta) NA

DepressionGrade 2 Grade 3

N = 119 N = 90

Model 1 (R2) .372*** .251***

Child & family characteristics

Model 2 (change in R2) .085**

Child perspective

Model 3 (change in R2) ns

Child perspective X gender

NA

NA

Depression

Grade 2 Grade 3

N = 119 N = 90

Model 2B (change in R2) .163***

Emotional support (beta) -.344**

Peer affiliation (beta) NA -.336***

Autonomy (beta) NA

.095***

Programs are more likely to have effects….

when children attend them for more days

Safe Haven Evaluation

– 4 school-based after-school programs serving 152 students grades 3 to 5

– ASES scores (M = 2.7, sd = .3, range 2.6 – 2.9)– Student characteristics

• 77% free or reduced lunch• 71% children of color • 49% single parent households • 47% males

– Program attendance varied from 1 to 163 days (median = 92 days)

Controlling for prior child functioning, children who attended programs for more days

• Had fewer absences from school (-.34*)• Were rated by teachers as

– having better work habits (.27*)– Working well with others (.23+)

TASC Evaluation

• 96 programs• 12,973 very active participants• 17,805 active participants• 8,104 non-active participants• 39,870 non-participants• Math and reading gains, and decreased problem

behaviors for very active participants vs. non-participants

Program Effects Also Are More Evident For

• Low-income children (Grossman; Marshall; Pettit; Posner & Vandell; TASC)

• Children with limited English proficiency (TASC evaluation)

• Low achieving students (TASC)

Why are after-school programs beneficial?

Experience Sampling Study

• 191 low-income 8th grade students • 8 school-based programs in 3 communities • Students wore watches that beeped 35 times

during 1 wk in the fall and 35 times during 1 wk in the spring

• Beeps occurred at random times during the after-school hours, evenings, and weekends

Students Completed Logbooks

• For each signal, students recorded – Who they were with– Where they were – What they were doing– How they were feeling– Levels of effort, concentration, motivation

• On average, students responded to 33 of the 35 signals during the week.

• 12,143 after-school, evening, and weekend experiences were reported.

• 5, 136 of the experiences occurred after school.

Student Activities at Programs and Elsewhere

Activity No program Not at program

At program

Homework 6.3 7.1 10.6***

Enrichment 5.7 6.7 19.2***

Eating 11.0 7.5* 2.6**

TV 19.5 19.2 0.6***

Sports 5.4 4.9 32.4***

Volunteer Service

0.1 0.1 3.5***

Differences in Supervision

No program

Not at program

At Program

Unsupervised peers

16.7 21.0 7.2***

Supervised Peers

26.4 26.0 91.1***

No adults 38.4 38.4 7.5***

Alone 12.9 11.7 0.0

Sib care 5.8 3.5 0.1

Differences in Motivation, Effort, and Feelings

No program Not at program

At program

Motivation 2.7 2.9 3.0***

Effort 1.9 1.9 2.5***

Importance 2.4 2.5 2.9***

Idle 1.4 1.5 1.4***Positivity 2.2 2.3 2.5***

Negativity 1.3 1.2 1.3

The NICHD Study of Early Child Care is well-suited to examining the effects of before/after-

school care• 10 research sites

• Prospective longitudinal design

• Large and diverse sample (n = 933)

• Extensive measures of family background, early child care, and child prior functioning

Study Participants Recruitment Sample

N = 1,364

52% boys

24% children of color

11% moms not high school graduates

14% single mothers

1st Grade Sample

N = 933

50% boys

20% children of color

10% moms not high school graduates

11% single mothers

Maternal Reports of Before/After-School Arrangements

• 4 telephone interviews (K fall & spring; 1st grade fall and spring)

• Time spent each weekday between 7 a.m. and 7 p.m. in each of 11 nonmaternal care arrangements

Five Types of Before/After-School Arrangements

• Before/after-school programs

• Extracurricular activities

• Sitter care

• Father care

• Nonadult care

Cumulative Participation Scores were Created

• Mean hours in the care arrangements were skewed, so dichotomous (yes/no) participation scores at each of the 4 interviews were created

• Cumulative participation scores were based on the proportion of interviews (2 in K, 2 in 1st grade) in which each type of care was used – Never– Sometimes– Consistently

Percentage of Children (n = 933) Who Never, Sometimes, and Consistently Participated in Different Types

of Out-of-School Arrangements (Children Can Have Multiple Arrangements)

Never Sometimes Consistently

Programs 60 25 15

Extracurricular 20 53 27

Sitter 14 58 28

Father 25 52 23

Nonadult 72 25 3

Family Predictors • Maternal employment

hours

• Parenting– Observed – Endorsed strategies

• Demographic factors– Family income– Household structure– Race and ethnicity– Maternal education

• Full day vs half-day kindergarten

• Early child care– M hours 3-54 months– % center care – % child care homes– % father care– M quality 6-54 months– Exclusive maternal care 3-

54 months (yes = 1)

Child Predictors

• Gender

• Behavior problems – 54 months– CBCL

• Language competence – 54 months– Preschool Language Scale

Examination of Relations between Before/After-School Care and Child

Developmental Outcomes

• Multivariate analyses of covariance (covariates were all family factors in previous analyses, matching child outcome variables at 54 months, & child gender), followed by ANCOVAs and pairwise t-tests

• Time (never, sometimes, consistently) in 5 types of care entered simultaneously

Child Developmental Outcomes: First Grade

• Academic outcomes

– Woodcock-Johnson: Letter-word identification

– Woodcock-Johnson: Applied problems

– Academic grades

– Teacher-reported work habits

• Social outcomes – teacher report

– Behavior problems

– Social skills

• Social outcomes - mother report

– Behavior problems

– Social skills

Participation in Extracurricular Activities was Associated with Academic

Outcomes

• MANCOVA F (8, 1656) = 2.23, p < .05

• ANCOVAs– Letter-word identification (p < .05)– Applied problems (p < .001)

Academic Achievement of Children who Never, Sometimes, and Consistently

Participated in Extracurricular Activities

Letter-word Applied problems

Never 111.1a 112.0a

Sometimes 112.4a 111.8a

Consistently 115.5b 116.1b

Number and Duration of Extracurricular Activities in a Week

• Children who participated in extracurricular activities typically had a single activity each week. Very few children had more than 2 activities.

• Children who participated in extracurricular activities typically spent between 1 and 3 hrs a week in the activities

Types of Extracurricular Activities• Team sports (21-34%)

• Individual sports (18-27%)

• Dance & music lessons (17-32%)

• Youth organizations (7-18%)

• Tutoring (0-1%)

• Academic enrichment (2-4%)

Father Care was Associated with Teacher-Reported Social Outcomes

• MANCOVA – F (6,1680) = 2.36, p = .03

• ANCOVA– Less externalizing behavior (p < .05)

Externalizing Behaviors (T scores) of Children Who Never, Sometimes, and Consistently Received Father Care

• Never 52.2a

• Sometimes 51.6a

• Consistent 50.0b

Participation in other types of before/after-

school care was not associated with child

functioning in first grade

Conclusions

• Consistent participation in extracurricular activities during kindergarten and first grade was associated with children’s academic achievement.

• Voluntary structured activities during nonschool hours may have beneficial effects on student performance at school.

Next Steps• Relations between After-School Experiences and

Students’ Academic, Social, and Behavioral Functioning

• Study of Promising Programs

• Longitudinal Study of After-School Arrangements - SECCYD

• Academic vs. extracurricular activities

dvandell@wisc.edu

for more information