Transcript CSC Associate's Presentation Oct05

Evaluating Early Childhood
Mental Health Consultation
Evaluation and Expansion in Connecticut
Walter S. Gilliam, PhD
October 27, 2010
The Edward Zigler Center in Child Development and Social Policy
Child Study Center
Yale University School of Medicine
The Rationale for ECMHC in CT
Expulsion Rates (per 1,000)
(1 in 36)
(1 in 149)
(1 in 476)
Gilliam, WS & Shahar, G (2006). Preschool and child care expulsion and suspension: Rates and
predictors in one state. Infants and Young Children, 19, 228-245.
Gilliam, WS (2005). Prekindergarteners left behind: Expulsion Rates in state prekindergarten programs.
FCD Policy Brief, Series No. 3. Available: www.fcd-us.org/resources/resources_show.htm?doc_id=464280
2
State PreK Expulsion Rates
Connecticut
16% of Classrooms
12 Expulsions per 1,000 Enrolled
(1 in Every 81 Preschoolers)
7th Highest State in the Nation
3
Access to Support Associated with
Decreased Expulsion Rates
% Classrooms Expelling
14.3%
10.3%
8.0%
No Access
On-Call Access
On-Site Access
Access to Behavioral Support Staff
Gilliam, WS (2005). Prekindergarteners left behind: Expulsion Rates in state
prekindergarten programs. FCD Policy Brief, Series No. 3.
4
Early Childhood Consultation Partnership
• Created in 2002 – Funded by DCF & SDE
• Availability: All CT child care & ece programs
serving 0- to 5-year olds
• Referral-source: child care directors (teachers &
parents)
• Services: Child/classroom-focused consultation,
Teacher training, Home-based component
• Consultants: 10-20 MA-level throughout state
• Dosage: Brief (3 months), intense (6-8 hrs/wk)
• Supervision: Group, Individual, Agency-based
5
Evaluated in 3 Statewide RCTs
• Study #1 (2005-2007)
– Preschool (3-4 yo) programs in CT
• Study #2 (2008-2010, finishing)
– Preschool (3-4 yo) programs in CT
• Study #3 (2008-2010, finishing)
– Infant/Toddler (0-2 yo) programs in CT
• Inclusion: Requesting ECCP; Consenting
• Exclusion: Prior ECCP treatment
6
Sample Sizes
Classes
Target
Children
Random
Peers
Treatment
43
75
NA
Control
42
69
NA
TOTAL
85
144
NA
Treatment
44
88
88
Control
44
88
88
TOTAL
88
176
176
Treatment
17
17
34
Control
18
18
36
TOTAL
35
35
69
Study 1 (P)
Study 2 (P)
Study 3 (I/T)
7
Sample Characteristics
Study 1
(P)
Study 2
(P)
Study 3
(I/T)
Child Care
82%
86%
100%
Head Start
13%
6%
0%
5%
8%
0%
63%
57%
20%
M=16.9
M=16.8
M=9.2
Male
73%
79%
72%
Female
27%
21%
28%
Setting
Public School
Teacher BA+
Class Size
Target Child Gender
Target Child Race/Ethnicity
White
56%
74%
67%
Black
14%
4%
22%
Latino
20%
14%
0%
Other/Multi
10%
8%
11%
8
Study 1 Evaluation Measures
• Classroom Quality & Interactions
– Early Childhood Environment Rating Scale (ECERS-R)
– Arnett Caregiver Interaction Scale (CIS)
• Teacher Beliefs and Feelings
– Parental Modernity Scale
– Child Care Worker Job Stress Inventory
– Teacher Depression (CES-D)
• Child Behavior Problems
– Conners Teacher Rating Scale – Long Form (CTRS-LF)
– Social Skills Rating System (SSRS)
9
Study 1 Results (Target Children)
CTRS Oppositional
CTRS Hyperactivity
75
75
70
70
65
65
60
60
55
55
ECCP
Control
50
ECCP
Control
50
Pretest
Posttest
Pretest
(F = 10.68**; d = 0.57)
Posttest
(F = 10.62**; d = 0.51)
12
CTRS Restless-Impulsive
75
10
70
8
65
6
60
4
55
2
ECCP
Control
SSRS Externalizing
ECCP
Control
0
50
Pretest
Posttest
(F = 4.05*; d = 0.34)
Pretest
Posttest
(F = 4.33*; d = 0.39)
10
Study 1 Results (Summary)
• Positive Effects on
– Teacher-rated Externalizing Behaviors
• No Effects on
– Non-Externalizing Child Behaviors
• Internalizing Behavior Problems
• Prosocial Behaviors
– Classroom Environment
• ECERS-R (all domains)
• Arnett Caregiver Interaction Scale (all domains)
– Teacher Beliefs and Feelings
• Authoritarian Beliefs
• Job Stress, Control, and Satisfaction
• Depression
11
Studies 2 & 3 Measures
• Classroom Quality & Interactions
– Classroom Assessment Scoring System (CLASS)
– Preschool Mental Health Climate Scale (PMHCS)
• Target Children [+Random Peers]
– Conners Teacher Rating Scale – Long Form (CTRS-LF) or
Infant-Toddler Social Emotional Assessment (ITSEA)
– Social Skills Rating System (SSRS) [+Random Peers] or Brief
Infant-Toddler Social Emotional Assessment (BITSEA)
– Preschool Social Behavior Scale (PSBS)
– Preschool Expulsion Risk Measure (PERM) [+Random Peers]
– Time Sampling (Structured Activities & Free Play)
• (Relational Aggression, Oppositionality, Disruptive Behavior)
– Home-School Collaboration
• Family Involvement Questionnaire
• Parent Satisfaction with Educational Experiences
12
Studies 2 & 3 Measures (more details)
• Time Sampling
– Each -- 30 segments (10 sec on, 10 sec off)
– Relational Aggression (Free Play)
• Any verbal or nonverbal behavior that (a) excluded others
from play or encouraged others to exclude a child or (b)
threatened to exclude or ignore.
– Oppositionality (Structured Activity)
• Includes defiance or refusal to follow adult directions,
arguing with adults, and temper outbursts in response to
adult directives.
– Disruptive Behavior (Structured Activity)
• Includes behaviors that are not necessarily in response to
adult directions, but are either impulsive, show difficulty
sustaining attention or waiting turn, disturbs other children,
or displays an emotional or behavioral outburst.
• Condition-Blinded Objective Raters
13
Studies 2 & 3 Measures (more details)
• Home-School Collaboration
– 11 items, 4-point Likert Scale (rarely-always)
– Alpha = .81
– Example items:
• “The parent talks to the teacher about how the
child gets along with his/her classmates at
school.”
• “The parent talks to the child’s teacher about
the child’s accomplishments.”
• “The parent talks with the child’s teacher about
personal or family matters.”
– (Family Involvement Questionnaire; Fantuzzo, Tighe, & Childs,
2000)
14
Effect Sizes for Studies 1-3
PRELIMINARY
Study 1 (P)
Study 2 (P)
Study 3 (I/T)
TARGET CHILDREN
Teacher Ratings
CTRS/ITSEA
.57
.37
SSRS-Ext
.39
.52
.66
Direct Observation
Relational Aggression
-.02
-.30
Oppositionality
-.25
.58
Disruptive Behavior
-.13
1.89
.44
.58
-.23
.29
Home-School Collaboration
RANDOM PEERS
SSRS-Ext/BITSEA
Effect Sizes: Trivial, Small, Medium, Large
Effect estimates for Studies 2 and 3 are preliminary and based on
incomplete data.
15
Conclusions (so far)
• Positive or Likely Positive Effects
– Teacher-rated externalizing behaviors in targets (P & I/T)
– Observed oppositionality & disruptive behaviors (I/T)
– Home-School Collaboration (P & I/T)
• Null or Likely Null Effects
– Teacher depression and job stress (Study 1 only)
– ECERS-R, CLASS (P & I/T)
• Too Close to Call at This Time
– Teacher-rated externalizing behaviors in random peers (P & I/T)
• Not Yet Examined
– Preschool Mental Health Climate Scale (PMHCS)
– Preschool Expulsion Risk Measure (PERM)
– All parent-rated forms
16
The ECCP Evaluations in Sum
• What we have going for us
– ECCP is brief and well-defined
– ECCP is popular (built-in waitlist controls)
– ECCP is stable and well-supported
– ECCP is a willing evaluation participant
– State support &funding for evaluation
• What we have as challenges
– Fluid programs (especially infant/toddler)
– Need more/better measures (fidelity)
– Evaluation funds still shoestring (for RCTs)
17
Walter S. Gilliam, PhD
The Edward Zigler Center in Child Development and Social Policy
Child Study Center
Yale University School of Medicine
230 South Frontage Road
PO Box 207900
New Haven, CT 06520-7900
Phone: 203-785-3384
Email: [email protected]
For reports, please visit: www.ziglercenter.yale.edu
18