The California Census Research Data Center
Data Resources @Berkeley Oct 22, 2012
Census RDCsCensus RDCs
What is an RDC? What is an RDC? What data are available in the What data are available in the
RDC?RDC? What kinds of research can be What kinds of research can be
done with RDC resources?done with RDC resources? What is the process for getting What is the process for getting
access to RDC data?access to RDC data?
What is a Census RDC?
A partnership A set of services, tools
and data A secure & vetted
environment
CCRDCCCRDCCalifornia Census Research Data
CenterBerkeley
The CCRDC is a joint project of the U.S. Bureau of the Census, UC Berkeley, Stanford and UCLA to enable qualified researchers with approved
projects to access confidential, unpublished Census Bureau data CES on the web: http://www.census.gov/ces/index.html
Stanford RDC: https://iriss.stanford.edu/SecuredataCCRDC on the web: http://www.ccrdc.ucla.edu/
RDCs as partnershipsRDCs as partnerships For researchers:For researchers:
Access to huge corpus of non-public use Access to huge corpus of non-public use datadata
For universities:For universities: Support for cutting-edge researchSupport for cutting-edge research Attract and keep data-intensive faculty Attract and keep data-intensive faculty
For Census Bureau:For Census Bureau: Extends pool of expertise on substantive,Extends pool of expertise on substantive, methodological, and statistical issues methodological, and statistical issues
RDCs as partnershipsRDCs as partnerships For researchers:For researchers:
Access to huge corpus of non-public use Access to huge corpus of non-public use datadataMust address topics of interest to the Must address topics of interest to the Census Bureau in developed proposal Census Bureau in developed proposal Must provide working papers and written Must provide working papers and written annual updatesannual updates
Must attempt to provide the benefits Must attempt to provide the benefits promised in proposalpromised in proposal
Must financially support project in most Must financially support project in most casescases
Must adhere to security requirementsMust adhere to security requirements
RDCs as partnershipsRDCs as partnerships For universities:For universities: Support for cutting-edge researchSupport for cutting-edge research Attract and keep data-intensive faculty Attract and keep data-intensive faculty
Finances, provides and maintains secure Finances, provides and maintains secure facility facility Funds full-time Census Bureau Funds full-time Census Bureau administratoradministratorEnters into legal contract delineating Enters into legal contract delineating responsibilitiesresponsibilities
RDCs as partnershipsRDCs as partnerships For Census Bureau:For Census Bureau: Extends pool of expertise on substantive,Extends pool of expertise on substantive, methodological, and statistical issuesmethodological, and statistical issues
Provides and supports administratorProvides and supports administratorProvides feedback on proposals (when Provides feedback on proposals (when submitted)submitted)Provides security infrastructure, Provides security infrastructure, oversightoversightProvides data access, software, Provides data access, software, disclosure avoidance reviewdisclosure avoidance review
Current RDCsCurrent RDCs Washington, DC (1983)Washington, DC (1983)
Center for Economic Studies, U.S. CensusCenter for Economic Studies, U.S. Census Boston, Mass. (1994)Boston, Mass. (1994) UCLA and Berkeley (1999)/(Stanford 2010)UCLA and Berkeley (1999)/(Stanford 2010) Duke (2000) / (RTI-2011)Duke (2000) / (RTI-2011) Chicago, Illinois (2002)Chicago, Illinois (2002) Ann Arbor, Michigan (2002)Ann Arbor, Michigan (2002) Baruch (NYC, 2006) and Ithaca (Cornell, Baruch (NYC, 2006) and Ithaca (Cornell,
2004) 2004) Minnesota (2010)Minnesota (2010) Atlanta (2011) Atlanta (2011) Texas – College Station (2012)Texas – College Station (2012) Seattle (2012)Seattle (2012)
Why RDCs?(Rationale for partnering)
Perceptions of improper use couldPerceptions of improper use could Reduce response ratesReduce response rates Induce Congress to cut Induce Congress to cut funding/programsfunding/programs
Title 13 U.S.C protects confidentialityTitle 13 U.S.C protects confidentiality Identifying microdata cannot be Identifying microdata cannot be releasedreleased Only Census Only Census Employees/temporary staff can Employees/temporary staff can look at individually identifiable datalook at individually identifiable data Access must provide legitimate Access must provide legitimate benefits to benefits to Census Bureau Census Bureau programsprograms
Why use CCRDC data?Why use CCRDC data? Not available elsewhereNot available elsewhere
Establishment level business dataEstablishment level business data Linked data (e.g. worker-firm )Linked data (e.g. worker-firm )
More detail than anywhere elseMore detail than anywhere else Detailed geo-spatial variablesDetailed geo-spatial variables Virtually no top or bottom coding Virtually no top or bottom coding Possible to link to other non-Census Possible to link to other non-Census
datadata Bigger SamplesBigger Samples High Quality Sampling FramesHigh Quality Sampling Frames ExtensibilityExtensibility
Access and Disclosure Access and Disclosure IssuesIssues
All researchers must be Census Bureau All researchers must be Census Bureau employees or have Special Sworn Statusemployees or have Special Sworn Status Fingerprints, security forms, penaltiesFingerprints, security forms, penalties
Projects must showProjects must show Benefits to BureauBenefits to Bureau Scientific MeritScientific Merit FeasibilityFeasibility Need for non-Public use DataNeed for non-Public use Data Minimal Risk of DisclosureMinimal Risk of Disclosure
All output goes through disclosure All output goes through disclosure review process (Interim and Final review process (Interim and Final Outputs)Outputs) Statistical output: YesStatistical output: Yes Tabular Output: NoTabular Output: No
• Demographic Data• Economic Data• Health Data• Trade Data• Crime Victimization
Data in the RDC’s
Key Demographic Surveys& Censuses
Decennial Census of Population and Housing (1970-2000)
American Community Survey (1996-2009)
Current Population Survey (1967-2010) Survey of Income and Program
Participation (1984-2008) American Housing Survey (1984-2009) National Longitudinal Survey (1966-
1999)
Decennial Census of Population and Housing
Flagship Data Collection of Census Bureau
Includes both universe and sample data
Public Use products include Summary Files
Pre-tabulated counts, multiple geographic summary levels
Public Use Microdata Individual/Household level data, PUMAs
Decennial Census Decennial Census 1970, 1980, 1990 & 20001970, 1980, 1990 & 2000
vs. Public Use Microdatavs. Public Use Microdata Lowest level of Lowest level of geographygeography
available in the PUMS is an area available in the PUMS is an area that contains 100,000 people that contains 100,000 people (PUMA)(PUMA)
RDC version includes more RDC version includes more detailed geographic informationdetailed geographic information
current residencecurrent residence place of workplace of work prior place of residenceprior place of residence
Decennial CensusDecennial Census vs. Public Use Microdatavs. Public Use Microdata
Larger sample sizeLarger sample size 100% of short form respondents100% of short form respondents One in six answered long formOne in six answered long form
PUMS has 5% of populationPUMS has 5% of population Improves analysis of small Improves analysis of small
populations/sample sizespopulations/sample sizes Less top-codingLess top-coding
Continuous variables, such as income, Continuous variables, such as income, are top-coded at a higher levelare top-coded at a higher level
More detailed codes (race, education, More detailed codes (race, education, multi-race, e.g. type of native American)multi-race, e.g. type of native American)
Documentation at ICPSR(but NOT the data)
What can you do with it? Analyses of Segregation School Choice Preferences Impacts of Indian Casinos Patterns of Migration Impacts of Subsidized Childcare Residential and Work Enclaves Spatial Mismatch Impacts of Vietnam DraftLook for yourself (CES Discussion Paper
Series)
American Community American Community SurveySurvey
All surveys with all information All surveys with all information collected on surveycollected on survey Household or person-level dataHousehold or person-level data Detailed geography (census block)Detailed geography (census block) No top or bottom codingNo top or bottom coding
1996 through 2009 currently 1996 through 2009 currently availableavailable
Can be linked to other data sources, Can be linked to other data sources, where feasible and permissiblewhere feasible and permissible
Confidential Versions of Confidential Versions of Your Favorite Public Use Your Favorite Public Use
DatasetsDatasets Survey of Income and Program Survey of Income and Program
Participation (SIPP) Participation (SIPP) National Longitudinal SurveyNational Longitudinal Survey Current Population Survey Current Population Survey
(March)(March) American Housing SurveyAmerican Housing Survey
Economic datasets: Economic Census
Economic datasets: Firms
Economic datasets: Establishments
Economic datasets: Transactions
Economic datasets: BR
Longitudinal Business Database
Longitudinally linked Business Longitudinally linked Business Censuses Censuses All non-farm establishments with All non-farm establishments with
paid employees in (almost) all paid employees in (almost) all industries industries 24 million unique establishments24 million unique establishments Excludes airlines, agriculture, RRExcludes airlines, agriculture, RR
Longitudinal Business Database
LBD includes LBD includes PayrollPayroll EmploymentEmployment OwnershipOwnership Detailed geographic informationDetailed geographic information Industry at 6-digit NAICS (more Industry at 6-digit NAICS (more
detail in some cases)detail in some cases) Other variables available (e.g. sales) Other variables available (e.g. sales)
but coverage varies across sectorsbut coverage varies across sectors
Employer-Employee Linked Employer-Employee Linked DatasetsDatasets
LEHD: Longitudinal Employer LEHD: Longitudinal Employer –– Household DynamicsHousehold Dynamics
Quarterly data on employment and Quarterly data on employment and wages from state unemployment wages from state unemployment insurance agenciesinsurance agencies
Contains basic demographic data for all Contains basic demographic data for all employeesemployees
Establishments linked to the LBDEstablishments linked to the LBD 49/50 states are currently participating 49/50 states are currently participating
(not all available in RDC setting)(not all available in RDC setting)
Synthetic products
Synthetic products
National Center for Health National Center for Health StatisticsStatistics
We are now hosting research using confidential NCHS and AHRQ We are now hosting research using confidential NCHS and AHRQ data in the CCRDC data in the CCRDC
Rules for access and disclosure the same as those in their Rules for access and disclosure the same as those in their enclavesenclaves http://www.cdc.gov/nchs/r&d/rdc.htm http://www.meps.ahrq.gov http://www.ciser.cornell.edu/NYCRDC/documents/http://www.ciser.cornell.edu/NYCRDC/documents/
NCHS_RDC_Data.pdfNCHS_RDC_Data.pdf No requirement to demonstrate Census benefitNo requirement to demonstrate Census benefit..
Long list of datasets – including NHIS, NHANES, NSFG, Long list of datasets – including NHIS, NHANES, NSFG, LSOA….LSOA….
http://www.ciser.cornell.edu/NYCRDC/documents/NCHS_RDC_Data.pdfhttp://www.ciser.cornell.edu/NYCRDC/documents/NCHS_RDC_Data.pdf
New DataNew DataNational Center for Health National Center for Health
StatisticsStatistics http://www.cdc.gov/rdc/http://www.cdc.gov/rdc/
National Health and Nutrition Examination Survey (NHANES)NHANES combines interviews and physical examinations to NHANES combines interviews and physical examinations to assess the health and nutritional status of adults and children assess the health and nutritional status of adults and children in the United States. in the United States.
National Health Care Surveys (NHCS) National Health Care Surveys (NHCS) A family of provider-based surveys that provide reliable A family of provider-based surveys that provide reliable information about health care providers, services, and patients. information about health care providers, services, and patients. NN
National Health Interview Survey (NHIS) The NHIS collects data on a broad range of health topics The NHIS collects data on a broad range of health topics through personal health interviews conducted in the home. through personal health interviews conducted in the home.
National Vital Statistics System (NVSS) National Vital Statistics System (NVSS) NVSS works with state vital registration systems to compile NVSS works with state vital registration systems to compile data on births, deaths, marriages, divorces, and fetal deaths. data on births, deaths, marriages, divorces, and fetal deaths.
New DataNew DataNational Center for Health National Center for Health
StatisticsStatisticsNational Health Care Surveys (NHCS)
National Ambulatory Medical Care Survey (NAMCS) National Hospital Ambulatory Medical Care Survey (NHAM
CS) National Survey of Ambulatory Surgery (NSAS) National Hospital Discharge Survey (NHDS) National Nursing Home Survey (NNHS) National Home and Hospice Care Survey (NHHCS) National Survey of Residential Care Facilities (NSRCF)
NHIS: Health Topics Demographics and SES Health status and disability Injury and poisonings Health insurance coverage Access to care Health services utilization Immunization Chronic conditions Health behaviors Height & Weight
New DataNew DataNational Center for Health National Center for Health
StatisticsStatisticsNational Vital Statistics System (NVSS) National Vital Statistics System (NVSS)
Births (Natality) Deaths (Mortality) Fetal Death (Fetal Mortality) Linked Birth/Infant Death (Linked Fetal Mortality) Marriages and Divorces (Marital Status) National Maternal and Infant Health Survey (NMIHS) National Maternal and Infant Health Survey (NMIHS) National Mortality Followback Survey (NMFS) National Mortality Followback Survey (NMFS)
New DataNew DataNational Center for Health National Center for Health
StatisticsStatisticsOther NCHS Data SourcesOther NCHS Data Sources
Longitudinal Studies of Aging (LSOA) Longitudinal Studies of Aging (LSOA) The LSOA follows two cohorts of people 70 years of age and over to The LSOA follows two cohorts of people 70 years of age and over to measure changes in their health, functional status, and health measure changes in their health, functional status, and health service use. service use.
National Immunization Survey (NIS) National Immunization Survey (NIS) The NIS monitors immunization coverage of children between 19 The NIS monitors immunization coverage of children between 19 and 35 months of age with a telephone survey and provider records. and 35 months of age with a telephone survey and provider records.
National Survey of Family Growth (NSFG) National Survey of Family Growth (NSFG) Collects information on family life, marriage and divorce, Collects information on family life, marriage and divorce, pregnancy, infertility, use of contraception, and men's and women's pregnancy, infertility, use of contraception, and men's and women's health. health.
State and Local Area Integrated Telephone Survey (SLAITS) State and Local Area Integrated Telephone Survey (SLAITS) Collects health care information at the state and local levels to Collects health care information at the state and local levels to facilitate state and local area estimates to meet varied program and facilitate state and local area estimates to meet varied program and policy needs. policy needs.
NCHS Data Linkage Activities (Linkage)NCHS Data Linkage Activities (Linkage)To enhance research value, NCHS links records from its population To enhance research value, NCHS links records from its population based surveys with other sources including Death Certificates based surveys with other sources including Death Certificates (NDI), Medicare Claims (CMS), Social Security Benefits (SSA), Air (NDI), Medicare Claims (CMS), Social Security Benefits (SSA), Air Monitoring Data (EPA). Monitoring Data (EPA).
RDC Research RDC Research EnvironmentEnvironment
““Thin Client” computing. Thin Client” computing. Servers in Maryland, accessed via remote terminalsServers in Maryland, accessed via remote terminals Standard statistical software (SAS, Stata, Guass, Standard statistical software (SAS, Stata, Guass,
Matlab, etc.)Matlab, etc.) Standard Datasets kept on serversStandard Datasets kept on servers Other software/data coordinated by Other software/data coordinated by
Administrator/CES staffAdministrator/CES staff
Secure Environment Secure Environment Restricted and monitored keycard accessRestricted and monitored keycard access No VisitorsNo Visitors No Laptops, internetNo Laptops, internet Printing limited, RDC AdministratorPrinting limited, RDC Administrator
Virtual RDC at Cornell Virtual RDC at Cornell (Synthetic Data, Zero Obs files)(Synthetic Data, Zero Obs files)
http://www.vrdc.cornell.edu/news/http://www.vrdc.cornell.edu/news/
FeesFees
$15,000 Standard Annual Project Fee$15,000 Standard Annual Project Fee Waivers may be available for Graduate Waivers may be available for Graduate
Student ResearchersStudent Researchers Additional Fees for complex matching Additional Fees for complex matching
requiring CES staffrequiring CES staff Additional Fees for NCHS data – initial file Additional Fees for NCHS data – initial file
creation and processing, creation and processing, but much but much smaller annual feesmaller annual fee
NEWNEW
Newly “recovered” historical household/population surveys and Newly “recovered” historical household/population surveys and business/economic surveys. business/economic surveys. Expedited access for evaluation purposesExpedited access for evaluation purposes Non-March CPS supplements, economic Censuses, ASMs…Non-March CPS supplements, economic Censuses, ASMs…Write for details if you have questions.Write for details if you have questions.
Kauffman Firm Survey Data Extension- Data MatchingKauffman Firm Survey Data Extension- Data Matching(http://www.kauffman.org/kfs/Travel-Grants-Program/Call-for-(http://www.kauffman.org/kfs/Travel-Grants-Program/Call-for-Proposals-%E2%80%93-KFS-Data-Extension-%E2%80%93-Data-Proposals-%E2%80%93-KFS-Data-Extension-%E2%80%93-Data-Mat.aspx)Mat.aspx)
National Crime Victimization Survey, 2008 - 2009National Crime Victimization Survey, 2008 - 2009
OtherOther
CES Mentorship programCES Mentorship program(US citizens only)(US citizens only)
Virtual RDCVirtual RDChttp://www.vrdc.cornell.edu/news/synthetic-http://www.vrdc.cornell.edu/news/synthetic-
data-server/data-server/
INFO 7470: Spring Semester 2011INFO 7470: Spring Semester 2011http://www.vrdc.cornell.edu/info747/http://www.vrdc.cornell.edu/info747/
course_outline.htmlcourse_outline.html
Contact InformationContact Information
RDC web site: RDC web site: http://www.ccrdc.ucla.edu/
email: email: [email protected] RDC phone: (510) 643-2262RDC phone: (510) 643-2262 RDC administrator: Angela AndrusRDC administrator: Angela Andrus RDC executive director: Jon StilesRDC executive director: Jon Stiles CES: CES:
https://www.census.gov/ces/index.htmlhttps://www.census.gov/ces/index.html