2/26/16  

1  

Connecting the Dots…  

Using Big Data to Improve Educational & Well-Being Outcomes for Children in Silicon Valley

Co-Directors, Silicon Valley Regional Data Trust: Dr. Rodney T. Ogawa

Marcy Lauck Dr. Bill Erlendson

Patty Ponzini

2013 NSF Grant Awarded ���to UC Santa Cruz

“Building Community Capacity for ���Data-Intensive Research in Education”

u  Technical Infrastructure

u  Building trust among partners

u  Model data-sharing agreements

u  Defining research and analytic needs

Regional Scan All 58 agencies contacted positively responded and requested help with:

2/26/16  

2  

u  39% Free/Reduced meals

u  44% gap between ethnic groups meeting UC/CSU requirements

u  Only 33% of Hispanic youth score Proficient or Advanced in ELA on the state test compared to 73% of White students.

u  Student Population 412,486

Regional Context by the Numbers

VISION ���Silicon Valley Regional Data Trust

The SVRDT is a university, agency, and community collaborative focused on the integration and application of data to inform programs and policies for children and their families.

Establish  &  Maintain  

Data Security

Data Privacy

Confidentiality

3  Levels  of  Protec:on  

Technical

Legal

Governance

Trust

Building Institutional Trust

2/26/16  

3  

Exercise

Jonathan’s Story

u Began building the DataZone data warehouse u Developed FosterVision collaboratively with DFCS & JPD

2012 SCCOE Initiatives

SCCOE DataZone Vision

u  Low-cost data warehousing

u  Dashboards for decision making

u  Secure partitions

u  Support LCAP goals

u  Aggregate multiple data sources (SIS, Assessments, HR, Finance)

u  Improved data quality

2/26/16  

4  

How DataZone Works

SIS

Assessment

Finance

Other Data

Sources

Data Warehouse

Analysisand

ReportingTools

Reporting and AnalysisServicesTurning data into usefulinformation

TrainingLearning how to use data tomake informed decisions.

State and FederalReportingMeeting reporting compliance

DisseminationSharing data with thecommunity (ie: report cards)

Schools InteroperabilityFramework

US US DOEDOE’’s s Vision of Data Flow for a DVision of Data Flow for a D33M SystemM SystemData Sources

Students,  Subgroups,  Cohorts  

Teacher  

CELDT,  SBAC  

Forma:ve  Assessments  Student  &  Teacher  AIendance  

Discipline  

Programs  

Schools,  Grade  Levels  

Years  Teaching,  Professional  Development  

2011-2012 2012-2013 2013-2014 2014-2015 2015-2016

Courses,  Grades,  GPA  

Data for Instructional Coherence

SCCOE DataZone

FosterVision  

District  Data  District  Data  

District  Data  

Other  Agency  Data  

Social  Services  

Juvenile  Proba:on  

2/26/16  

5  

FosterVision 2015

Converging Vision

Silicon  Valley  Regional  Data  Trust    MOU  Established  2015  

DataZone & FosterVision UCSC NSF Grant

Secure Information Sharing Environment

3 County Offices of Education:���Santa Clara COE - 31 Districts Santa Cruz COE - 11 Districts San Mateo COE - 24 Districts

University of California Santa Cruz

Social Services

Mental Health & Public Health SCCOE

DataZone

Juvenile Justice

Governance

Ensures data security, privacy & confidentiality Supports data-intensive research

Informs school & agency programs, policies and practices

Educational Technology Solutions

Silicon Valley Regional Data Trust

2/26/16  

6  

Why SVRDT?

u  Eliminate the lag time in providing and coordinating critical services

u  Allow service providers to be “on the same page” u  Identify the most critical factors impacting student

outcomes u  Identify the most at-risk children and families u  Determine best practice interventions u  Solutions are a community wide responsibility

SVRDT Theory of Change

SVRDT = Big Data

u  Data will be of sufficient scale, diversity, and complexity that it requires new architecture, techniques, algorithms, and analytics to manage it and extract valuable (and sometimes hidden) knowledge

u  UCSC’s Data Science team is among the best in the world

2/26/16  

7  

Building Blocks of the SVRDT

u  Innovative Analytics to produce timely and relevant information

u  Innovative Data Visualization to synthesize data across multiple sources, make actionable connections and identify solutions.

u  Build Partnerships among researchers, community and family members working directly with children and families across the Valley.

u  Continuous Development to integrate new tools, new data and better methods as we grow.

A Sister Lighthouse Department of Health Services, Pittsburgh, PA u  Built a comprehensive data warehouse with data sharing among

Probation, Housing Authority, Welfare, Pittsburgh Public Schools & Carnegie Mellon University

u  Predicting which children are most likely to be chronically absent •  Most predictive variable is within school-year move

u  Understanding which schools have the highest concentrations of chronically absent students who are on probation and receiving public benefits

u  Working with school districts to determine most effective interventions

u  Using predictive analytics to prevent child abuse

u  Data and GIS mapping help determine placement of treatment centers

Creating a Learning Community

We can create collective impact by: u  Facilitating cross agency data sharing

u  Connecting practitioners and researchers

u  Informing practice and local and regional efforts

u  Providing economies of scale

2/26/16  

8  

There is a regional support for this collaboration

Interagency networks & cross- sector coalitions are possible

Applied research & predictive analysis from UCSC

Students & families benefit through alignment of community resources

This is the Tipping Point

Questions & Answers