ENHANCE Update Research Underway on the Validity of the Child Outcomes Summary (COS) Process ECO Center Advisory Board Meeting March 8, 2012 Arlington, VA ENHANCE is funded by grant R324A090171 from the the U.S. Dept. of Education, Institute for Educational Sciences
ENHANCE Update. Research Underway on the Validity of the Child Outcomes Summary (COS) Process . ECO Center Advisory Board Meeting March 8, 2012 Arlington, VA. ENHANCE is funded by grant R324A090171 from the the U.S. Dept. of Education, Institute for Educational Sciences. Topics. - PowerPoint PPT Presentation
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
ENHANCE UpdateResearch Underway on the Validity of the Child Outcomes Summary (COS) Process
ECO Center Advisory Board Meeting
March 8, 2012Arlington, VA
ENHANCE is funded by grant R324A090171 from the the U.S. Dept. of Education, Institute for Educational Sciences
Topics• Project design - review• Data collection progress• What we are learning
– Implementation– Child assessment study – State data study
• Next steps
Project Design
1. Conduct a program of research to examine the validity of ratings generated by COS and identify conditions that lessen validity.
2. Revise the COS and supporting materials based on study findings.
3. Identify a series of validity analyses that can feasibly be conducted in states to allow each state to examine the validity of its own COS data on an ongoing basis.
ENHANCE Project Objectives
Validity – What Are We Trying to Demonstrate?
• Validity is NOT a characteristic of an assessment or measurement device.
• Validity is a characteristic of the data produced by the tool and how these data are used.
• Are data valid for the purpose of…..• Implications:
– State A’s COS data could be valid; – State B’s COS data could not be.
Standards for Educational and Psychological Testing(1999) by American Educational Research Association, American Psychological Association, National Council on Measurement in Education
If data were valid for this use, then we would see….
– Collect evidence to examine each of those propositions
Examples of propositions in the COS Validity Argument
3. Children differ from one another with regard to level of functioning in the 3 outcome areas as reflected in COS ratings.
7. Functioning (COS ratings) in an outcome area at time 1 is related to functioning in that area at a later point in time.
9. COS ratings will be related to the nature and severity of the child’s disability.
Design
Real World
Not controlled Conditions
Design:37 Project Data Collection Sites
19 Programs Part C • Illinois• Maine• Minnesota• New Mexico• North Carolina• Texas • Virginia
18 Districts Part B Preschool • Illinois• Maine• Minnesota• New Mexico • South Carolina• Texas
9
Studies and Data Collection Progress
ENHANCE Studies
• Provider Survey• Team Decision-Making• Comparison with
Child Assessments • State Data Study
Provider SurveyGoals• Learn about COS implementation – processes in use• Identify providers’ knowledge and training experiences • Describe perceptions about if COS produces an accurate
rating and influences on that• Understand impact of COS on practice• Process & Sample
• Online survey responses• All providers in program who participate in COSStatus• Survey underway, continues through April
Team Decision-Making Study
Goals• Examine understanding and
application of outcomes and rating criteria
• Describe team process • Identify if ratings are consistent with
evidence discussedProcess & Sample• Video teams discussing COS ratings
• 210 children’s teamsStatus• Data collection underway• Code videos this summer & fall
What Are We Learning?
Considerable variability across states and even across programs, within a state Training Ongoing staff support and quality assurance Teaming (not just for COS) Parent involvement Timing and Process
Implication: Results will tell us about COS validity under
real-world conditions
Implementation
Number of Providers in COS Ratings - Preliminary
1 2 3 4 5 or more0.00
0.05
0.10
0.15
0.20
0.25
0.30
0.35
0.40
0.45
EI % (n = 99)ECSE % (n = 61)
Perc
enta
ge o
f COS
form
s
Number of providers
Goals• Compare entry and exit COS ratings
to BDI-2 and Vineland-II scores• Compare conclusions from COS
and assessments
Process & Sample• Longitudinal, external assessor at
program entry & program exit • 216 children
Study Status• Local, trained assessors in place• Recruiting families since Aug. 2010• Sample shows expected variability,
including initial COS ratings, tool scores
Comparison with Child Assessments Study
Comparison with Child Assessments Study – Preliminary…
3. There is variability in children’s functioning in the three outcome areas and that variability is reflected in the COS ratings.
Validity argument claims
Distributions of Preliminary COS Ratings (1-7)
EI(n=71)
ECSE(n=49)
Ratings
3. There is variability in children’s functioning in the three outcome areas and that variability is reflected in the COS ratings.
10. COS ratings in the corresponding outcome areas are moderately correlated with:
o the social-emotional (Outcome 1), o cognitive (Outcome 2), o communication (Outcome 2), and o adaptive (Outcome 3) domain scores of assessment tools.
Validity argument claims
• Methods table
Methods
• Preliminary correlations between COS Ratings and assessment tools
• What expect to see?
Methods
.70
.42
.00
1.0
Preliminary Correlations: COS Ratings & Assessment Scores
ECSE larger COS-Assessment Correlations than EI
COS with relevant assessment tool domains EI Mean (range)
ECSE Mean (range)
Positive Social Relationships 0.33 (.20-.52) 0.61 (.51-.68)
Acquiring and Using Knowledge and Skills 0.25 (.22-.28) 0.70 (.67-.75)Taking Appropriate Action to Meet Needs 0.49 (.36-.64) 0.47 (.40-.57)
Outcome 2: Acquiring and Using Knowledge and Skills
EI(n=71)
ECSE(n=49)
Outcome 3: Taking Appropriate Action to Meet Needs
Summary of preliminary findings
Patterns• Means for groups generally follow expected directions
on assessment tools• Group comparisons showed expected differences• Effect sizes were nearly all larger for ECSE than EI
For COS – assessment tool comparisonsFor comparisons between assessment tools by the
same external assessor
More data are needed for final conclusions
State Data Study
Goals• Examine characteristics of COS data
and relationships to other variables• Look for consistency in patterns
across states to test claimsSample• All valid COS data within the state for a reporting year • 15-18 states conducting all analyses• Additional states sharing select analysesStatus• Piloted procedures with 3 Part C, 3 Part B Preschool states• Now working with 4 Part C, 6 Part B Preschool states• Recruiting more states, requesting data
7. Functioning, as reflected in the COS rating, in an outcome area at time 1 is related to functioning in that area at a later point in time.
Validity argument claims
Year Taking Appropriate Action to Meet Needs
State A 08-09 0.47State A 09-10 0.52State A 10-11 0.50State B 08-09 0.61State B 09-10 0.61State C 10-11 0.59
Correlations:Entry and Exit Ratings Part B 619
Preliminary state data
3. There is variability in children’s functioning in the three outcome areas and that variability is reflected in the COS ratings.
14. COS rating distributions at entry will be related to the disability-related characteristics of the population served by states.
Validity argument claims
Part C 08-09 entry ratings across states Taking appropriate action to meet needs
0.0%
5.0%
10.0%
15.0%
20.0%
25.0%
30.0%
35.0%
State AState BState D
1 2 3 4 5 6 7
COS Ratings
Next Steps and Reactions
Next steps
• Gather more state data• Complete data collection
involving local programs/districts• Analyze provider survey results• Code videos
• Questions? Reactions?• Implications for the national
data? • Implications for ECO?
Questions? Reactions?
Find out more
• ENHANCE Website– http://ENHANCE.sri.com
• ECO Center Website– http://www.the-ECO-center.org