Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
â3â The Federal-State Cooperative Relationship W ITH ITS MANDATE TO COLLECT DATA on vital events, the Na- tional Center for Health Statistics (NCHS) shares a common chal- lenge with several of its peers in the highly decentralized federal statistical system: functioning as a national-level collector of information on phenomena that are inherently local in nature. Accordingly, mechanisms for cooperation and coordination between federal statistical agencies and state or local authorities are common in the statistical system, ranging from relatively simple awareness-building activities (e.g., the multitude of short- term partnerships that the Census Bureau forges to boost participation in the decennial census) to highly structured contractual and ï¬nancial agreements (e.g., the grants administered by the Bureau of Justice Statistics to support development of criminal history record databases). The Vital Statistics Cooperative Program (VSCP) that has been formed between state and local registration areas and NCHS is, as workshop pre- senter Ed Hunter (Centers for Disease Control and Prevention) noted, a federated system, with the national or federal-level entity of NCHS provid- ing funding, coordination, and standards, but with the individual states and localities retaining signiï¬cant autonomy in their operations. The general structure of the system and the special challenges it facesâsome challenges common to nearly all federal data collection efforts in a time of scarce re- sources, but others unique to the nature of the vital records that are the source of data in the VSCPâwas a recurring theme at the workshop. 35
36 VITAL STATISTICS In this chapter, we summarize some threads of this discussion on the structure of the VSCP Section 3âA describes the role of the states and, more . generally, the challenge of data collection given the civil registration nature of the underlying birth and death certiï¬cates. Section 3âB summarizes the constraints on the vital statistics collection system from NCHSâs perspective as national-level coordinator. Finally, Section 3âC proï¬les selected models of federal-state cooperation elsewhere in the federal statistical system. Workshop presenters and participants received two background papers prepared at the workshop planning committeeâs request: one on the role of the states (prepared by Steven Schwartz, New York City Department of Health and Mental Hygiene) and the second on NCHSâs role. These papers are reprinted in Appendixes A and B, respectively; they have only been min- imally edited for consistency with National Research Council report style. Schwartz gave a presentation at the workshop that closely followed the con- tent of his paper; there was no explicit counterpart presentation of the back- ground paper by NCHS, but several of the paperâs themes were sounded by Jennifer Madans and other NCHS staff (as summarized in Section 3âB). 3âA THE ROLE OF THE STATES Schwartz emphasized the local nature of vital events and the collection of records. At the outset, records of each of the more than 11 million vi- tal eventsâbirths, deaths, marriages, and divorcesâthat occur each year in the United States are processed through one of the more than 6,000 local registrars. These local registrars form a diverse and complex network, and Schwartzâs own local experience offered an interesting perspective on the geographic distribution and the workload of registrars. The state of New York has the most local registrars of any stateâabout 1,500âyet New York City and its more than 8 million inhabitants have only one ofï¬cial registrar. The New York City registrarâs ofï¬ce alone processes about 500 live births and 160 deaths every dayâabout 300,000 vital events annually. Records data funnel through the local registrars to the 57, mainly state- level, registration jurisdictions. As already noted, two citiesâNew York City and Washington, DCâfunction as registration jurisdictions. (Schwartz ob- served that, at one time, registration districts were more city based and, in fact, New York City began as a registration area before New York state.) Each of the 57 registration jurisdictions reports data directly to NCHS through the VSCP Schwartz also noted another centralizing force in the . systemâthe National Association for Public Health Statistics and Informa- tion Systems, the professional association of the state vital records ofï¬ces that was founded in 1933 and works with NCHS and the states on data collection issues.
THE FEDERAL-STATE COOPERATIVE RELATIONSHIP 37 Schwartz observed that the registration jurisdictions vary considerably in their capacities and individual procedures, their staff size and expertise, and their level of system automation and electronic database implementation. He said it is important to bear in mind that the state and local vital records ofï¬ces must fulï¬ll three basicâand sometimes competingâroles. Two of these roles have historically dominated the work of the ofï¬ces: 1. Civil registration of vital events: The most basic function of the vi- tal records ofï¬ces, the civil registration function of vital events, has the important implication of making the ofï¬ces huge customer service operations. Schwartz said that walk-in customers have pressing legal needs for records and certiï¬ed copies and may require corrections or amendments to existing legal documents; as custodians of the records, the vital records ofï¬ces also have a responsibility to be prompt and responsive. Schwartz noted that his New York City registrarâs ofï¬ce ï¬elds on the order of 700 walk-in customers a day, seeking copies of records or other services; the approximately 800,000 paid copies of birth and death certiï¬cates that the ofï¬ce issues each year accounts for about $12 million a year in revenue. 2. Public health statistics collection: It is the processing of records dataâ the information on the birth and death certiï¬catesâthat ultimately populates the vital statistics data ï¬les. Schwartz noted that there is a constant tension in resource allocation between the statistical role and the civil registration and customer service role and that the statistical side must take second place. Yet a third and no less important role has arisen and been made explicit in law in recent years: vital records ofï¬ces are also front lines in national security efforts. Both Schwartz and Hunter noted that birth certiï¬cates have become particularly sensitive because they are breeder documents that are the basis for many other important documents and legal statuses. Birth cer- tiï¬cates can constitute proof of U.S. citizenship; because each vital records reporting jurisdiction maintains a contract with the Social Security Adminis- tration (SSA), birth certiï¬cates also trigger issuance of Social Security num- bers and eligibility for beneï¬ts. Birth certiï¬cates are also used to obtain state driverâs licenses and federal passports, key means of establishing identity. The Intelligence Reform and Terrorism Protection Act, which became law in December 2004, was a partial implementation of the recommen- dations of the National Commission on Terrorist Attacks Upon the United States (2004). Among its provisions was a set of minimum standards meant to secure birth certiï¬cates. Hunter argued that the changes in the 2004 act were not really newâa 1996 immigration act passed by Congress contained many similar provisions. However, the 2004 act was substantively differ- ent in important ways and carried particular urgency given that several of
38 VITAL STATISTICS the terrorists in the attacks of September 11, 2001, had obtained passports and identity documents using fraudulently obtained birth certiï¬cates. Rec- ognizing the federated nature of the vital registration systemânot national or federal per se, and so lacking the ability to directly effect changesâthe 2004 bill simply stipulated what the federal government would accept as a valid birth certiï¬cate. This left implicitâand up to the states and localitiesâ what changes to the system were needed to meet those standards in order for certiï¬cates to be valid for federal purposes. As Hunter summarized, and Schwartz echoed, the demands of this new national security role for vital records ofï¬ces are considerable: ⢠The 2004 act sought to reduce the hundreds of different variations of birth certiï¬cates, including commemorative or ceremonial ones issued by the states, that were previously allowed. The act deï¬ned standard, recognizable paper certiï¬cates, printed on a certain type of security paper, as well as other provisions for basic structure. ⢠In addition to securing the physical document, the 2004 actâs provi- sions are intended to secure the system by which they are issued. One part of this revised system is a requirement for rapid ascertainment of death certiï¬cation and a direct matching of death to birth records to preclude fraudulent use of birth certiï¬cates of the deceased. Like other background check systems, the critical requirement of this matching is that it needs to cross all jurisdictions and it needs to be fast. However, although the act was signed in December 2004, Hunterâs workshop presentation described âregulations for secure systemsâ as âpendingâ and still in progress; NCHS put draft regulations together, but these are still circulating through the federal system. ⢠Although the actâs text only explicitly speaks to standards for birth certiï¬cates, the requirement of matching of birth and death records tacitly also implies standards for death certiï¬cates. Schwartz noted that the new national security role is one that is resource intensive for localities and, again, one that can blur the statesâ ability to fo- cus on the public health data collection role. In discussion, Kenneth Prewitt (Columbia University) noted another way in which the security role poten- tially clashes with the data collection role. He worried that, to the extent that electronic vital registration systems become portrayed as a homeland security tool or even a type of law enforcement mechanism for detecting fraud, complications may arise for statistics. That is, security may become so tight and participation sufï¬ciently strained that it may be more difï¬cult to move the system into the kind of social, public health surveillance system that is needed for detection of early disease or other health incidents. In terms of the functioning of the VSCP Schwartz expressed strong sup- , port for the basic distributed nature of the system. The locally distributed
THE FEDERAL-STATE COOPERATIVE RELATIONSHIP 39 system of collecting vital event data gives state and local registrars maximum leverage to work closely and directly with their source data providers: hos- pitals and physicians, funeral homes, nursing homes and clinics, and so forth who contribute the information that populates vital records. This approach satisï¬es local authoritiesâ need to effectively be the master of their own data streamâto best know their own data and their own data providers. However, he also agreed that it is not a perfect system, and noted several particular challenges: ⢠At the national level, the compiled vital statistics data can only be as timely (and as high quality) as the weakest state. Reporting lags and data quality or consistency issues of an individual state, large or small, can impair the national system. ⢠A related challenge is that the distributed nature of the VSCP makes training, educating, and querying of the source data providers a key aspect of improving and maintaining end data quality, particularly the case for cause-of-death reporting, for which consistency in approach is critically important. However, such training and education efforts are probably the ï¬rst thing to suffer in a competition for scarce resources. ⢠The individual registration jurisdictions are sensitive to the amount of funds that NCHS provides through the VSCP and uncertainty over , NCHSâs resources can have signiï¬cant local effects. In discussion, Schwartz commented that New York Cityâs use of NCHS-provided VSCP funds is principally to pay staff. He noted the example of NCHSâs discontinuation of abortion reporting in 1995, the result of which was that New York City lost more than $75,000 in funding and had to forgo a staff position and active surveillance of abortion providers. Hence, in recent reports, the city has had to attribute drops in the number of abortions to the cessation of active monitoringâ rather than a real declineâbecause its ability to accurately measure activity has been impaired. ⢠A further complication in the distributed nature of the system was sug- gested by thenâCensus Bureau director Steve Murdock in his luncheon remarks at the workshop, drawing on his experience as state demogra- pher of Texas. In states with extensive rural populations, such as Texas, the county clerks responsible for processing birth and death certiï¬cates (not to mention other government documents) may be part-time po- sitions and, hence, data collection and processing (and furtherance of the national vital statistics collection efforts) might not be a high pri- ority. He said that the part-time nature of these jobs contributes to the growing pains that occur as paper-and-pencil registration systems become computerized systems and to the lags that result in some areas.
40 VITAL STATISTICS Like Hunter, Schwartz commended the development of electronic sys- tems for the automated processing and veriï¬cation of birth and death records. The system has made great strides in moving away from having the data on paper forms key-entered by local ofï¬ces; web-enabled systems that permit hospitals and doctors to enter and certify data electronically have the potential for improving data quality and timeliness. The catch, he noted, is that these automated systems carry extremely high start-up costsâon the order of $1 million per death or registration systemâthat state and local registration ofï¬ces have had difï¬culty obtaining from their parent govern- ments. The resultâas noted by other workshop presentersâis a lack of uni- form implementation. Schwartzâs paper in Appendix A provides additional details on speciï¬c electronic systems that have been developed or proposed, including the national State and Territorial Exchange of Vital Events that is proposed to permit matching of birth and death records across all vital statistics ofï¬ces. Going forward, Schwartz suggested that it is important for the stake- holders in the VSCP to consider ways to look at the systemâs return on investment. He argued that the system is not broken but that there is no clear measure of how good or how bad it is: some estimates of the return on investment of resources at the national and state levels would be valuable for building support for the system among policy makers. He suggested the example of the SSAâs Enumeration at Birth Program (EAB; initiated in 1990) that assigns Social Security numbers to newborns, with parental approval, when a birth certiï¬cate record is processed. The SSA Ofï¬ce of Inspector General audited the program (using SSAâs own data) and estimated the av- erage cost of the traditional processâindividual parents walking into local Social Security ofï¬ces to obtain a number for their newborn childrenâto be $18.70 per record process; in comparison, the audit suggested that the total cost of processing records under the EAB system (including the fee that goes to the states for processing) is only $3.74. As Schwartz noted, the resulting estimate of about $60 million in savings each year (about $15 per record, multiplied by about 4 million births per year) is a clear and compelling assess- ment of the return on investment of the EAB program. Noting that the EAB program might be a relatively easy-to-measure case as a pure administrative system, Schwartz argued that these kinds of ï¬gures are worth considering in relation to vital statistics. Measuring the value of vital statistics is consider- ably harderâtrying to quantify things such as the value to research of the data ï¬les and the returns of the use of locally held vital records data to pop- ulate immunization or lead-exposure registries or to target newborn home visits by nurses. Still, such measures are important to consider in making the overall VSCP better and demonstrating its unique value.
THE FEDERAL-STATE COOPERATIVE RELATIONSHIP 41 3âB CHALLENGES AND LIMITATIONS AT THE NATIONAL CENTER FOR HEALTH STATISTICS The largest problem facing the current VSCP from NCHSâs perspective was ï¬rst raised very early in the workshop. Committee on National Statis- tics director Constance Citro observed that the workshop was planned with three Cs as themes: to celebrate the many and growing needs served by vital statistics, to critique the program to identify strengths and weaknesses, and to contribute to a better understanding by policy makers of the value of vital statistics. NCHS director Ed Sondik countered that the third of these Cs could readily be simpliï¬ed to âcostsââgrappling with the continuing chal- lenge of obtaining high-quality data through a cooperative program when both federal and state resources are tight. He commented that in order for the center to balance the agencyâs budget, NCHS does not have sufï¬cient resources to ï¬ll all of the vital statistics needs. Indeed, he commented that NCHS has been meeting with its Board of Scientiï¬c Counselors speciï¬cally to discuss options for the programs should NCHSâs generally ï¬at funding continue. The presentation by Jennifer Madans (NCHS) in the ï¬nal session of the workshop echoed the concern about costs. She observed that the desire to build the vital statistics system and revamp itâincluding further promotion of electronic systemsâis coming at a time when both NCHS and state gov- ernments are facing tight funding constraints. In its planning, NCHS has had to generally assume a ï¬at budget going forward and simultaneously wrestle with the data collection costs for information collections in all other areas in health. Accordingly, she noted a certain level of frustration by all vital statistics stakeholders: there is a strong need for the data and great pride in the vital statistics system, but not a great deal of latitude for massive im- provement in one single program without resources. The sensation is one of striving to meet todayâs problems by putting off tomorrowâs problems, which are investments in future capacities: the problem is that sooner or later the VSCP is going to get to tomorrow. From NCHSâs perspective, a constant challenge given the cooperative nature of the VSCP is determining fair shares of costs. As mentioned above, the states and localities must deal with the civil registration aspect of vital events, and the staff and resource allocation to keep up with customer ser- vice is a very signiï¬cant local administrative focus. The key questions are: What is the cost of gleaning the data from the records and the value of the information that the states and localities possess (both in local totals and compiled national data)? How do those values lead to a determination of who is responsible for what part of the overall costs? In this context, Madans noted that the VSCP is an interesting case in point in the broader federal statistics system. In many agencies and applica-
42 VITAL STATISTICS tions, she said that a common theme is the use of administrative records and administrative data for several purposes. The argument is that the use of ad- ministrative data will make things different (ideally, better) and will require some changes in interpretation but will almost certainly make things more efï¬cient and less expensive (to the extent that the administrative data are used to reduce ï¬eld data collection). Vital statistics are commonly pointed to as a long-standing example of such administrative data being used. What is not always appreciated, though, is the synergy that went into the develop- ment of the VSCP In other statistical applications, administrative data tend . to be thought of as records that have been generated and created for a to- tally distinct purpose that can then be tapped at little or no cost and added as part of oneâs database. Because of the cooperative nature of the VSCP and the revisions of standard certiï¬cates, the system itself develops the form and content of the administrative record. This gives the system great ï¬exi- bility in the content of the dataâit is not clear what would be on birth and death records if this cooperative system had not developedâbut carries with it signiï¬cant costs and challenges. 3âC EXAMPLES OF FEDERAL-STATE COOPERATION IN THE U.S. FEDERAL STATISTICAL SYSTEM Two workshop presenters described the general structure of federal-state cooperation within their agencies, to suggest possible improvements in the structure of vital statistics collection. The two speciï¬c systems considered in these presentations are of interest because of their parallels to the col- lection of vital statistics. The Quarterly Census of Employment and Wages (QCEW) parallels vital statistics in that the source data are essentially admin- istrative records with other legal purposes (in this case, tax ï¬lings to unem- ployment insurance programs); it differs from the vital statistics program in having a strongly deï¬ned set of legislative requirements for the structure of the federal-state partnership enacted in recent years. The second example, the Education Departmentâs Common Core of Data (CCD), parallels vital statistics in that initial responsibility for data completion is diffused among a wide variety of state and local authorities (in this case, individual elementary and secondary schools as well as state departments of education); it differs from the vital statistics in that several of the components of the CCD serve a primarily directory-building role rather than an analytical role. 3âC.1 The Quarterly Census of Employment and Wages Jack Galvin described the cooperative structure underlying the QCEW program of the Bureau of Labor Statistics (BLS), which produces quar- terly data on employment and wages at the national and subnational (state,
THE FEDERAL-STATE COOPERATIVE RELATIONSHIP 43 metropolitan statistical area, and county) levels.1 BLSâs federal-state coop- erative efforts date back to at least 1916 and were particularly strengthened by a provision of the Wagner-Peyser Act of 1933 that directed the Labor Department to reimburse states for the operation of statistical systems that contribute to national statistical series. One of ï¬ve formal federal-state cooperative programs currently main- tained by BLS, the QCEW federal-state partnership is a relatively recent de- velopment at BLS, which assumed technical responsibility for the program from the Department of Laborâs Employment and Training Administration in 1972 and ï¬nancial responsibility in 1984. QCEW data are known for their comprehensiveness and for their ability to describe local-area economic conditions. Within BLS, QCEW data are also essential as a building block for other statistical programs (e.g., as a benchmark for the annual payroll survey and a sampling frame for establishment surveys), and they are also an important input to the national accounts studies of the Bureau of Economic Analysis. Through the QCEW program, BLS directly funds its state partners to col- lect and edit data from the state-based employment insurance programs. As with vital statistics, the underlying data of the QCEW are, essentially, admin- istrative records: the quarterly contribution reports that employers supply to the state-based employment insurance programs when they pay their taxes each quarter. The approximately 7.7 million quarterly forms, from over 9 million separate business establishments, are estimated to include coverage of about 98 percent of jobs in the United States. In compiling the data, the states are responsible for providing some information that is not directly coded on the quarterly tax forms, such as the industrial classiï¬cation of the business and veriï¬cation of physical location address (rather than general mailing address). To provide these supplemental data, QCEW funding re- quires the states to contact each business establishment every 3 years. In addition to funding, BLSâs role is also to provide technical and methodological direction to the data collection. It provides the informa- tion technology systems for the collection and processing of the data and promulgates standards for the data that are sent to BLS. Signiï¬cantly, BLS plays no role in the structure or format of the quarterly contribution forms, and so the individual state employment insurance programs can vary in the form and ï¬ling requirements placed on employers (e.g., whether reports can be ï¬led electronically or on paper). Responsibility for dealing with the va- riety of inputs from employers remains with the states; what BLSâs QCEW funding promotes is a set format for the speciï¬c economic elements that are coded in the data ï¬les that are returned to BLS. 1 Additional detail on the QCEW program can be found at http://www.bls.gov/cew/ (April 2009).
44 VITAL STATISTICS Galvin indicated that BLS spending on the QCEW program in 2008 was $49.5 million; the share of that total that is allocated to the states ($30.6 mil- lion) makes it the largest of BLSâs federal-state cooperative programs. From 2000 to 2008, funding for the QCEW grew at an average of 2.9 percent per year, in line with the approximate 2â3 percent increase in the number of businesses in existence each year (and a corresponding increase in workload for the state ofï¬cials). Galvin noted that BLS has typically been success- ful in securing such âmandatoryâ increases from the Ofï¬ce of Management and Budget and from congressional appropriators, giving QCEW relatively stable funding.2 As NCHS does with the state vital records ofï¬ces, BLS negotiates in- dividual QCEW contractsâcooperative agreementsâwith each state. The QCEW agreements are updated and agreed to on an annual basis. Through these annual agreements, BLS is able to specify (and modify, as appropriate) the expected quality standards for the data, the requirements for protection of conï¬dentiality, and policies for allowable costs. By agreement with the states, BLSâs funding to the states is calculated by multiplying a stateâs av- erage government employee annual wageâitself a ï¬gure derived from pub- lished QCEW dataâby 1.5 by the number of staff positions to be ï¬lled in the state. Each state is allocated a base of two positions for supervision (and continuity of operations), and additional positions are calculated on the ba- sis of the stateâs workload in the program: for instance, its number of single and multiunit businesses and new business units. Galvin indicated that BLS has concluded that its structuring of the QCEW programâand the stability of funding for itâhas improved the quality of the data and improved the consistency of data quality across the states. The annual cooperative agreements provide a means to promul- gate methodological standards, including routines for addressing situations found through review of edit failures in data processing and for the han- dling of missing quarterly tax reports (through imputation). BLSâs provision of information technology to the partners also provides consistency and re- liability in processing: Each state uses one of two technical systems (those used and developed in Utah or Maine), a considerable simpliï¬cation from 50 heterogeneous technical systems. Galvin noted that the common tech- nical platforms produced a signiï¬cant beneï¬t in terms of timeliness of data release: over the course of a couple of years, BLS was able to shift its quar- 2 As noted in the discussion of Galvinâs presentation, the use of âmandatoryâ means in- creases in funding because of increases in the cost of collection (e.g., cost-of-living adjustments to staff salaries), not that the increases are required by law. Galvin noted that BLS does not always receive these âmandatoryâ increases; in 2007, failure to get $2 million to cover the in- creased costs led BLS to cover the funding by other means, eliminating production of estimates from the payroll survey for several metropolitan areas and cutting sample size from another federal-state cooperative program that collects occupational employment statistics.
THE FEDERAL-STATE COOPERATIVE RELATIONSHIP 45 terly publication of QCEW data earlier by 3 weeks. By doing so, BLS was responsive to a major client for QCEW dataâthe Bureau of Economic Anal- ysis, which (with data available that much earlier) can now use the quarterly data to update its personal income estimates four times per year rather than waiting to perform an annual revision. Galvin noted that federal-state cooperation in the QCEW program does occasionally encounter vulnerabilities due to variation in state laws and reg- ulations. State-level changes to ï¬ling practices for employersâ quarterly con- tribution reportsâthe source data for QCEWâcan create complications. In one instance Galvin cited, a state requirement that all employers begin ï¬ling the forms electronically (rather than on paper) proved difï¬cult for small em- ployers, contributing to glitches in processing and, for the QCEW increased , levels of nonresponse for that state. State conï¬dentiality laws can also com- plicate the use of QCEW information for other purposes: For instance, they may prohibit BLS from sharing the administrative data collected in QCEW to assist the Census Bureau in ï¬lling in missing industry codes in its business and employment data. In recent years, the QCEW Program has been further structured in re- sponse to the Workplace Investment Act of 1998. Section 309 of the act (P 105-220) amended the original Wagner-Peyser Act to create speciï¬c .L. provisions related to employment statistics. State governors were directed to designate a single state agency as the manager and coordinator of em- ployment statistics for each state, but the act also established more direct oversight responsibilities for these state partners. The Secretary of Labor (and BLS) was directed by the act to develop a process by which 10 direc- tors of these state-designated regionsâone for each regionâare elected to hold formal consultations with BLS on the cooperative management of the system.3 This was a departure from established procedure in BLSâs federal- state cooperative efforts such as QCEW which had previously been handled , more through BLSâs regional ofï¬ces than its national headquarters staff. In addition, the 1998 act directed that BLS hold formal consultations with the states at least once each quarter on the products and function of the employ- ment statistics system.4 Galvin noted that the organizational changes needed to comply with the 1998 act were difï¬cult for BLS, given the entrenched practices, but that, by improving communication and feedback, they have improved the program and made it easier to accomplish further change in speciï¬c federal-state efforts. 3 BLShas settled on a policy by which these elections are held every 2 years. 4 Speciï¬cally, the mechanism used by the act to require these collaborations is the require- ment that the Secretary of Labor develop an annual plan for the national employment statistics program; the consultation with the states is intended to be a key input to the annual plan.
46 VITAL STATISTICS Box 3-1 Surveys Comprising the Common Core of Data Public Elementary/Secondary School Universe Survey ⢠institutional characteristics, numbers of teachers, enrollment by grade, students participating in selected education programs, dropouts, and high school com- pleters Local Education Agency (School District) Universe Survey ⢠institutional characteristics, number of education staff, and number of students participating in selected education programs Local Education Agency (School District) Universe Survey: Dropout and Completion Survey ⢠number of dropouts from each of grades 7 through 12, and the numbers of high school diploma recipients and other high school completers State Nonï¬scal Public Elementary/Secondary Education Survey ⢠state-level counts of students, teachers, other staff, and high school completers National Public Education Financial Survey ⢠state-level collection of revenues and expenditures SOURCE: Adapted from workshop presentation by White and descriptive material at the CCD website (http://nces.ed.gov/ccd/index.asp [April 2009]). 3âC.2 The Common Core of Data Andrew White (National Center for Education Statistics, NCES) de- scribed the CCD, the U.S. Department of Educationâs primary database on public elementary and secondary education in the United States. Adminis- tered by NCES, the program annually collects ï¬scal and nonï¬scal data about all public schools (approximately 96,000), public school districts (approxi- mately 18,000), and the 50 states, the District of Columbia, Department of Defense Schools, and those in outlying areas. The CCD comprises ï¬ve surveys sent from NCES to state education agencies (typically a state department of education) and completed mostly by using administrative data already maintained by the agency; the surveys are listed in Box 3-1. White noted that completing some data items requires the agency to contact local school districts, who in turn may contact indi- vidual schools. The state agency then compiles the data from all levels into prescribed formats and transmits them to NCES. The data gathered from the surveys fall into three categories: 1. general descriptive information on schools and school districts, includ- ing name, address, phone number, and type of locale; 2. data on students and staff, including selected demographic character- istics; and 3. ï¬scal data that include revenue and current expenditures.
THE FEDERAL-STATE COOPERATIVE RELATIONSHIP 47 White said NCES began entering into cooperative partnerships in 1985 in an attempt to improve CCD data. Of note was a contract with the Council of Chief State School Ofï¬cers (CCSSO) to examine the completeness and comparability of data reported to the CCD, as well as to discuss ways to expanded its content and establish common deï¬nitions. The Hawkins-Stafford Elementary and Secondary Education Improve- ment Amendments of 1988 (P 100-297) authorized NCES to establish .L. a formal federal-state cooperative system. The goal of this system was to âproduce and maintain, with the cooperation of the States, comparable and uniform educational information and data that are useful for policymaking at the Federal, State, and local levelâ (Hoffman, 2004:x). To implement and support the new cooperative system, NCES, with as- sistance from CCSSO, formed the National Forum on Education Statistics (NFES) in 1989 with a mission (Hoffman, 2004:xii): To develop and propose, cooperatively, a national education data agenda and model(s) for a national data system that will meet the needs of education policy makers and program planners in the decade and be- yond; To inform Federal, State, and local decision makers on the goals and progress of this cooperative education statistics system; To pro- vide an arena in which Federal, State, and local education interests can identify, debate, mediate, and where appropriate, recommend action on education policy, issues, emerging needs, and technological innovation salient to the improvement of education data comparability, uniformity, timeliness, and accuracy at the national level. The NFES also adopted the role of âprovid[ing] direction for research and evaluationâ and âbring[ing] to the attention of relevant parties such mat- ters as may contribute to the accomplishment of this missionâ (Hoffman, 2004:xii). Chief state school ofï¬cers, federal program heads, and direc- tors of professional associations with an interest in education statistics were asked to appoint liaisons who would represent their various institutions. The NFES formalized its goals, objectives, functions, organizational struc- ture, and operations in January 1990 with the adoption of a Policies and Procedures Manual. In 1996, the NFES expanded its membership by adding one local education agency representative from each state, to be appointed by the chief state school ofï¬cer. To achieve its mission, the NFES holds regular meetings, including stand- ing committees that address speciï¬c issues, and produces a number of re- ports, including a series of âbest practiceâ guides on a wide range of data- related topics. NCES funds state participation in NFES activities and pub- lishes and disseminates deï¬nitions and guides from NFES. In 2003, the department launched the Education Data Exchange Net- work Submission System (EDEN) to provide a common system by which state education agencies could transmit their administrative data. Data are
48 VITAL STATISTICS transmitted by the states to meet the data requirements of annual and ï¬nal grant reporting, speciï¬c program mandates, and the Government Perfor- mance and Results Act. In addition, the EDEN Survey Tool was established to allow transmission of additional data, such as the Civil Rights Data Col- lection and the Indian Education Formula Grant Program Application for Funds. In 2006, the Department of Education launched a more overarching sys- tem called EDFacts which is a central portal for performance and account- ability data reporting, including nonï¬scal CCD data.5 White noted that implementation of EDFacts has âdone a little damageâ to the October 1 reporting deadline. In particular, states that had highly developed data sys- tems in place prior to EDFacts have had a difï¬cult time converting to the new format and its deï¬nitions. The department has provided some funding to states to help them enter the EDEN/EDFacts system. Since January 2007, reporting of these data using EDFacts is mandatory (with a 2-year transition period). The establishment of the Statewide Longitudinal Data System Grant Pro- gram has also provided an opportunity for states to apply for grants from between $500,000 and $6 million to develop and implement longitudinal data systems. The grants provide funding for 3-year cycles. Participation has grown substantially every year since the programâs inception in 2005. Many states have been awarded their second 3-year grant, and only eight states have not participated. 5 For more information on EDEN and EDFacts, see http://www.ed.gov/about/inits/ed/ edfacts/overview.html (April 2009).