Sally Obenski and Jim Farber U.S. Census Bureau CNSTAT Panel January 26, 2007 Expanding the Use of Administrative Records: Methods, Applications, and Challenges

2 Overview  Overview and History  Key technical breakthroughs  Decennial and Survey Applications  Medicaid Undercount Study  Value of Integrated Data Sets  Operational and Technical Constraints  Policy Challenges  Conclusions

3 Mandate for Administrative Records Use Title 13, Section 6:  Use administrative records information as extensively as possible in lieu of conducting direct inquires Census Bureau Strategic Plan:  Reduce reporting burden and minimize cost to taxpayer by acquiring and developing high-quality data from sources maintained by other government and commercial entities

4 Legal Guidance and Protections  Title 13, U.S.C., Section 6, 9, and 214  Title 26, U.S.C., Section 6103(j)  Privacy Act of 1974  Paperwork Reduction Act  Government Information Security Reform Act (GISRA)  E-Government Act of 2002, including  Federal Information Security Management Act (FISMA)

5 Safeguarding Administrative Records  Consistent Application of Policies  To ensure that projects have the appropriate legal authorization, comply with existing data agreements, and provide adequate controls to protect confidentiality and privacy  Administrative Controls  Numerous levels of approval  Need-to-know access  Removal of identifiable information  Administrative Records Tracking System  Security and confidentiality training

6 AR Program Evolution Today Program begins mid 1990s July AR Research staff created Survey launched to gather info on potential AR files Early 1990s Statistical uses of AR conference held July /2000 Projects included AREX 2000 and the 1999 StARS prototype 1999 Centralized program emerges Data Stewardship program begins 2001 AR Test for 2000 Census Race Model Addresses Quality Concerns PVS Increases Linking Capacity Infrastructure investments allow new interagency collaborations

7 StARS Provides Technical Infrastructure Person & Address Databases consist of 7 national files: IRS 1040 HUD TRACS IHS SSS IRS 1099 HUD PICMedicare CY2004 recordsPersonsAddresses Raw input894 million767 million StARS308 million152 million

8 Administrative Records Experiment Validated StARS  Local test of AR census models conducted in 5 counties  Coverage issues similar to Census 2000  Validated conformance of StARS to Census 2000 addresses & persons  Improvements to StARS continue, including move to more real-time redesign (E-StARS)

9 NUMIDENT Provides National Reference File  Social Security Administration (SSA) Numerical Identification (Numident) Transaction file with 803 million records  Collapse to 431 million unique SSN records  Usages:  Look-up file that provides demographic data  Social Security Numbers (SSNs) verification/ validation

10 Race and Hispanic Origin Model Rectified Quality Concerns  Initial weakness was dependence on race data from SSA’s SSN transaction file  Census 2000 records matched to SSN transaction file  Model completed missing linkages

11 Person Validation System (PVS) Increases Linking Capacity  Use master file of SSN/name/DOB as reference file  Link addresses with SSN reference file  Match incoming census or survey record using name, address, DOB  Search within address first (high quality match)  Search by name/DOB nationally if address search not successful  Replaces SSN with unique identifier (PIK)

12 Person Validation System (PVS) Increases Linking Capacity  Prepare Numident  Prepare Incoming Data  Run Verification Phase  Run Geokey Search  Run Name Search  Apply PIKs IncomingDataIncomingDataNumidentNumidentVerificationVerification GeokeySearchGeokeySearch NameSearchNameSearch FinalFinal

13 Implementing the ACS Provides Current Long Form Data  Designed to ameliorate constraints of decennial long form data collection  Provides means for timely analyses and estimates at small geographic areas  Provides means to push models based on less granular surveys to smaller geographies

14 AR Integral to Census Bureau Programs  Internal Revenue Service (IRS) 1040  Intercensal Estimates  Small Area Income and Poverty Estimates  CMS Medicare and Medicaid  National Longitudinal Mortality Study  Small Area Income and Poverty Estimates  Small Area Health Insurance Estimates  State Unemployment Insurance Files  Longitudinal Employer-Household Dynamics

15 Current Decennial Census Research  Using AR to “assign” age, race, sex, Hispanic Origin, when a record can be matched  Use AR to identify households with coverage problems  Determine if commercially available & other lists can improve & help build GQ frame

16 Emerging Survey Improvement (1)  Reducing ACS small area variance  Use AR as controls to adjust survey weights right after nonresponse adjustment  Preliminary research highly promising  Obtaining characteristics on nonrespondents  Compared StARS persons to CPS responders to ensure consistency  Used StARS to obtain characteristics of nonresponding households

17 Emerging Survey Improvement (2)  Reacting to disaster and other near-real time requirements  Katrina’s effect on the federal statistical system and our lack of current response data highlighted need  Acquired the USPS National Change of Address File and FEMA’s emergency management and flood insurance files  Developing next generation StARS – near real-time measurements

18 Medicaid Undercount Study (1)  Survey estimates are important to policy research  Examining the large discrepancy between survey estimates and Medicaid enrollment figures  Multi-phased, interagency study, including academia

19 Medicaid Undercount Study (2)  Phase I: Examines quality and characteristics of Medicaid and Medicare files ( )  Phase II: Conducts national match of Medicaid files to Current Population Survey ASEC ( )  Phase III: Conducts selected state matches of Medicaid files to states in CPS ASEC  Phase IV: Conducts national match of Medicaid files to the National Health Interview Survey and compares results

20 Value of Integrated Data Sets  Provides more robust and accurate picture  Builds on strengths of both views while controlling for their weaknesses  Provides better statistics for input into simulations for predictions and funds distribution  As the demand for data increases and budgets decrease data re-use many be the only cost-effective option

21 Operational Constraints  File Acquisition Complexities  Complex Memoranda of Understanding  State by state negotiation  Differences in content definition, quality, and program rules over time  File lag time  MSIS (Federal Medicaid) lags by about 4 years  Most lag for about a year  Many applications require more near real-time response

22 Illustrative Examples  Federal Files—yearly acquisitions  IRS 1040, IRS 1099, CMS Medicare, Medicaid  Federal Files—quarterly or monthly updates  SSA Numident, USPS NCOA  State Files—on as needed basis  MD Food Stamps, MD TANF, MD Child Care Subsidy  State Files—on a quarterly basis  UI Wage and ES 202 (currently from 40 states)  Commercial Files—INFOUSA

23 Technical Constraints  Obtaining the right data in the right format  Varying rates of validation (e.g., Medicare 99%, Medicaid 91%)  Coarseness of administrative data compared to nuances of surveys  Measuring error

24 Policy Challenges  Communicating the benefits vs. privacy concerns  Need for interagency teams to ensure accurate results  Interagency agreements and mission  “Ownership” of the integrated data sets  Growth of possible disclosure risks  Need for longitudinal data bases in order to find an anonomyzed person at an address at a point in time

25 Overcoming the Constraints  Resolving file acquisition issues may require OMB or Congressional assistance  Lag time for general demographics addressed by National Change of Address file—planning move to Enhanced StARS for more near real- time response  Standardized and centralized file acquisition  New files in address search phase and SAS-based matcher increased validation rates  Data Quality Standards team addressing measuring error in integrated data sets  Increasing inter-agency efforts for generating integrated data sets

26 Conclusions  New files and innovations leading to expansion of AR uses  New challenges continue to arise  Regular updating of billions of records to have a near real- time response system  Effectively acquiring state-based records  Understanding integrated data sets  At incipience of a new generation of products, services, and inter-agency opportunities

27 Contact Information Sally M. Obenski Assistant Division Chief for Administrative Records Applications Data Integration Division U.S. Census Bureau Washington D.C Phone: (301) Cell: (301)