University of Connecticut
The Second Annual State Data Conference
A New Approach to a New Millennium
June 19, 2000
Working Toward a Solution—Current Intiatives:
The Second Annual State Data Conference brought together data users, collectors and providers to share the latest developments in Connecticut data while discussing the needs of our data community. This report outlines some of the current problems within the system, the latest data developments, and future issues for Connecticut’s data needs. The Conference emphasized the importance of partnerships and the need for a virtual State Data Center. Ken Wiggin, the Connecticut State Librarian, has provided a short statement on the presentation he was to make on the new State Data Protocol (see Appendix 1). Appendix 2 contains a description of the websites cited during the meetings.
The purpose of the Second Annual State Data Conference was to continue the conversation started at the First Annual Data Conference held May 10, 1999. The broad objective of this conference series is to create an ongoing symposium in which those involved in different stages of data generation, archiving, distribution, and use can exchange information on current data availability, the challenges they face, their particular needs, and their expectations (or hopes) for future developments. This Conference considered challenges in establishing uniform standards—an issue that lies at the heart of efficient data collection, archiving and dissemination—and discussed what systems for accessing data are now available or being developed in the State. Participants assessed the feasibility of establishing a comprehensive Data Center for Connecticut, and saw both what the forthcoming U.S. Census Bureau American Community Survey will offer, and some of the best sites available in other communities. Conference participants came from various sectors, including government agencies, academic scholars, private enterprise and non-profit organizations.
The one-day Conference included eleven presentations organized in four sessions. Two sessions provided in-depth reports on recent advances in remote sensing and the American Community Survey. The other two sessions had multiple presenters. One group of data providers and archivers surveyed innovations in data development on the federal, state, county, town and census block levels. Another group showed what was being done to facilitate use of data for specific constituencies or communities, and how the data was being integrated into scholarship and analysis. The speakers addressed directions for future growth. The following summarizes the challenges, current initiatives and future directions for Connecticut’s data system.
Data Challenges Facing Connecticut:
The First Annual Data Conference concluded that a State Data Center need not be a physical concept. Indeed, a single physical entity attempting to archive all the relevant data and maintain its timelessness would surely fail. This Conference built on that insight, recognizing a “data central” should be a virtual entity, whose purpose should be to create a framework that identifies, organizes and disseminates data efficiently and makes it readily available for users from both public and private sectors. This evolution calls for the cooperation and coordination of parties involved in different stages of data processing, from data collection to final usage.
Presenters provided a valuable roadmap to the principal problems in current data systems:
Although the Internet has become a powerful tool in terms of obtaining data, data does not “travel” with the appropriate description (meta data). Without meta data, retrieved data are cut off from their source, and, at the same time, makes data interaction impossible. As users, we do not always know the appropriate uses or limitations for data.
In addition, data is not integrated with geographical information. Graphics and maps are powerful analytical tools that allow us to visualize data and spatial patterns. However, because of a lack of geographical information, most of the data currently available is not graphable or mappable. Data that are accompanied with meta-data and geographical information are intelligent data.
Currently, there are an overwhelming number of stand-alone data systems on the web and in publication. Duplicated effort and high variation in quality are natural results of these isolated and uncoordinated systems. Data users are often confused by the widely variable data from different sources, and frustrated with the inability to reconcile them.
The analysis of trends and forecasts requires longitudinal data. Unfortunately, most of the currently available databases lack consistent time series data. Users have to search different databases to obtain longitudinal data. Without the appropriate metadata, users typically have inconsistent and incomparable data.
Most of the data collected are on the federal and state level. While some databases have county, town and census block data, they do not allow user-defined study areas. For example, some school districts encompass several towns. Without the kind of data that allows regrouping of town level data into school districts, it is impossible to compare different school districts and rank them. Another example is the case of child health, safety, and community services. To appropriately analyze this data, users need local, disaggregated data.
State and local governments collect data for various purposes. The data collected by local governments, e.g., towns, planning districts, etc. are valuable not only for their own purposes but also for local community analysis, policy making and other research programs. Incentives need to be provided to state and local government agencies to share their data with other public and private sector data users.
Working Toward a Solution—Current Intiatives:
The solution to these problems is a well-designed and easily accessible State Data Center built with the cooperation of data producers, processors and distributors and data users. Such an initiative would articulate the appropriate standards and facilitate coordination among different data generators. A variety of initiatives are underway in an effort to respond to these issues.
There is an international effort in developing meta-data guidelines, building metadata
infrastructure, and implementing them. The following projects were described in relation to Connecticut:
The Connecticut Data Server project is a joint project of the Center for Geographic Information and Analysis (http://www.uconncgia.uconn.edu) and the Map and Geographic Information Center (http://magic.lib.uconn.edu) at the University of Connecticut. The objective is to provide easy access to on-line data for small areas, e.g., tract and block group data. In addition, it aims to provide distributed attributes that can be “joined” with the geographical data for mapping in a geographical information system (GIS).
This work extends the geographical attributes currently available on the Map and Geographic Information Center (MAGIC) website. Currently, the study areas include the State, counties, congressional districts, regional planning districts, service delivery areas, and tourism areas. The project’s next step is to develop user-defined study areas, which allows users to break up a system-defined study area and construct a study area that suits different study purposes.
The University of Connecticut, under the leadership of Professor Robert Cromley and Patrick McGlamery of the Babbidge Library, jointly with the U.S. Census Bureau’s FERRETT Project, are working towards establishing a data web that links federal, state and local data from disparate sources.
The structure is based on the FERRETT distributed architecture software. When completed, it will serve as an intelligent common database gateway to multiple and varied databases, provide a lightweight data directory service that aids data resource discovery and use, and allow data viewers to intelligently couple data from different sources into this distributed architecture.
The next step for McGlamery and FERRETT is to add Connecticut databases with demographic, fiscal (federal monies to towns, state monies to towns, town income and expenditures) and educational (strategic school profile 1992-1999) data. Furthermore, the joint effort will build shapefiles of Connecticut geographies and install MapServer for FERRETT Hot Reports.
A critical lesson that emerged from both the first and the second conferences is the importance of benchmarking what we are doing in Connecticut against other states, municipalities, and non-profit entities. Jack Eichenbaum of GISMO and New York City’s Data Council provided striking examples of how detailed parcel mapping and other neighborhood-level data could provide a framework for both thoughtful policy assessments and an “early-warning” system for neighborhood distress in other parts of the country. Such presentations remain critical for informing participants about the possibilities that flow from developing first-rate data systems.
The websites, listed in Appendix 2, provide opportunities for Connecticut to learn best practice systems from other parts of the country.
Besides traditional data collecting methods, new data collecting technologies are being introduced that produce new and better data. Remote sensing is a good example of such technology.
Northeast Applications of Useable Technology In Land planning for Urban Sprawl (NAUTILUS) represented by Dan Civco and James Hurd from the University of Connecticut, is one of the NASA’s Regional Earth Science Applications Centers (RESAC). They are charged with educating the public about the value of remote sensing technology, and making data available, accessible and useable to local land use decision-makers.
Remote sensing uses satellites or radar to record land cover features through different reactions to electromagnetic radiation. These data can then be applied to study land cover changes and conduct longitudinal analyses. (For more information on remote sensing, data availability and its applications, see Appendix 2 for relevant websites.)
A series of short and intense discussions given by data distributors, data users, policy researchers, and representatives from non-profit organizations began a dialogue leading to an information exchange, and thus more efficient access to available data. These discussions honed in on how to find data and provide a better understanding of how currently available data can best be utilized. The data users also voiced the need for community level data, including health and property data.
This discussion began with Doug Rae exemplifying the power of appropriate data by rebuilding the city profile of New Haven in the early 1920s. Ben Barnes introduced us to the data archive organized by the Connecticut Conference of Municipalities (CCM) (www.ctdataservice.com)[not online, March2009]. Mike Meotti presented Connecticut strategic school profile data and the outreach programs initiated by the Connecticut Policy and Economic Council (CPEC) (www.cpec.org) [no longer a separate organization, March 2009]. Lisa Mercurio of the Southwestern Area Commerce & Industry Association of Connecticut, Inc. (SACIA) characterized the potential users of regional data and the need to locate the appropriate data in a timely manner. Priscilla Canny showed how the Connecticut Voices for Children uses data to shape public policy to improve Connecticut children, youth and families’ well-being.
The purpose of ACS is twofold: to provide longitudinal socioeconomic data, such as population and housing data, at more frequent intervals, and, to produce information for small areas including counties, tracts, block groups, and population subgroups that is updated year by year. After passing the demonstration period for 1996-1998, ACS is currently in the experimental stage. The data for 31 comparison sites throughout the country are available on the Census Bureau’s website and through the Census Bureau’s free CD-ROM. The Bureau plans to implement the survey nationwide in 2003, covering every county. For more detailed information on ACS, go to Census Bureau’s web page at www.census.gov (click on Subjects A-Z, then click on American Community Survey).
Even with the current effort, there is a long way to go to build an integrated data system, a State Data Center that provides users with easy access to accurate, well-documented federal, state and local data. The data users at the Conference voiced their frustration regarding the loss in time and efficiency trying to locate the proper data sources and integrate them so they have a common basis for analysis. As a continuation of the effort towards establishing a State Data Center through partnership, there is a clear need for serious consideration of the following issues:
As we have agreed at the First Annual Data Conference, a State Data Center need not be—indeed should not be—a physical concept. During this Conference, the idea of a linked data system where users can easily navigate among various databases received detailed attention. The next step is to decide how to establish this system. Specifically, what organization or organizations will host this interrelated system? One needs to realize that the host not only needs to coordinate and establish linkages, but also maintain the system, which requires significant funding. While the University of Connecticut has begun to build the foundation for such a center, ultimately this project is a matter that needs to be considered by Connecticut policymakers and the legislature. Without a significant resource commitment and a parallel articulation of a formal “state data policy”, development of a strong data portal will be fragmentary.
As pointed out above, local governments collect data for various purposes. These data are valuable in making decisions at the community level. Unfortunately, these data are usually buried inside the collecting government agencies. What can be done to create an incentive for these government agencies to publicize their data?
Finally, how can we create an environment that encourages high quality data
generation and data processing? What should we do to facilitate information
exchange through the Internet and publications? Both of these considerations
underline the potential importance of an empowered data advisory council.
Broadly, this discussion resolves itself into two fundamental questions:
(1) Will there be the necessary commitment of resources to bring
the virtual system whose first elements are now being developed to full development?
(2) Will the State recognize the urgent importance of institutionalizing the articulation and implementation of data policies and incentives through creation of a State Data Advisory Council?
We hope that these questions prompt active thinking and participation by all of the relevant constituencies, whether government agencies, non-profit groups, foundations, or private sector entities, to evaluate how they can help in the process of building a virtual State Data Center. The Third Annual State Data Conference in 2001 will assess what progress we have made on these issues, discuss what remains to be done and continue to provide a forum for data users to access the latest in data delivery.
APPENDIX 1: Towards a State Information Policyby Kendall F. Wiggin, State Librarian
June 19, 2000
The State of Connecticut’s current Information Policy can be found in constitutional provisions and federal law. It can also be found in state statutes and the interpretations of state statutes by the courts (the common law). And it can be found in executive orders, regulations, and the rulings, opinions, directives and guidance of those agencies having responsibility over government information matters, such as the Freedom of Information Commission, the State Library and the Department of Information Technology. Unfortunately, thus far the policy is neither comprehensive nor is it located in one place.
Section 4d-7(a) of the Connecticut General Statutes provides that the state’s Chief Information Officer shall “develop, publish and annually update an information and telecommunication systems strategic plan.” One of the statutory goals of the plan is to “develop a comprehensive information policy for state agencies that clearly articulates (A) the state’s commitment to the sharing of its information resources, (B) the relationship of such resources to library and other information resources in the state and (C) a philosophy of equal access to information.”
The Government Information Policy Advisory Committee has been working on a draft document that is an articulation of Section 4d-7(a) and brings together, in a unified structure, the fundamental principles underlying the existing government information policy, as found throughout the law. Those principles of government information policy are those that express the fundamental concepts of (1) “government transparency,” (2) fair information practices, (3) efficiency of services within a state agency, among government agencies, and between state agencies and the people they serve, and (4) the stewardship of government information. These concepts form the core of government information policy in all democratic societies, including that of Connecticut.
The document that has been developed by the Government Information Policy Advisory Committee is also designed to fill in missing elements of state data policy, where permissible, in order to construct a comprehensive policy in keeping with these basic principles. The Committee anticipates that where additional authority is required to fill in missing elements, such authority will be sought through legislation, executive order or other means, and incorporated in later versions of this document.
Before I discuss the status of the State Information Policy document, I think it is worth noting that in 1994 the Connecticut Law and Information Policy Project published A “White Paper” on Connecticut Information Law. The “White Paper”, authored by John J. Phillips, was designed to assist the Office of Information and Technology (predecessor of DOIT) and the Information Policy Planning Committee in formulating a comprehensive information policy for the State of Connecticut as called for in Section 3(a)(3) of Public Act No. 89-257. The white paper provides an interesting background to the topic before us. The driving force then, as now, was the rapidly changing technological advances. The report warns that with the advent of “electronic miracles” comes the risk of informational chaos and the danger of information tampering. The report was also concerned about the proliferation of the personal computer and the resulting redundancy of information. It was felt then that a comprehensive statewide information policy would be needed to guard against the inefficient and costly use of storage space and the unrecoverable loss of important public records. While storage space has certainly ceased to be a cost consideration, the growth of servers and the proliferation of data depositories remains a very real concern. The white paper also contains a suggested draft information policy, which, although it is written in the form of “findings”, articulates many of the same policy positions addressed by the current draft State Information Policy.
The Government Information Policy Advisory Committee has sent the draft document to Rock Regan, the state’s Chief Information Officer for his review. The Committee has suggested that it be reviewed and reacted to by an ever-widening circle of individuals within and outside of state government. At this time, I do not know if that process will be followed or when it will start or when a final policy will be adopted. Topics covered in the Information Policy draft include Government Information Policy Principles; Fair Information Practices; Efficiency of Services; and Stewardship of Government Information.
Government Information Policy Principles talks about government transparency; dissemination and disclosure, both public dissemination and public disclosure; freedom of information and the disclosure of public records; and confidentiality. For purposes of the State Information Policy, fair information practices refers mainly to the Personal Data Act, which is contained in Chapter 55 of the Connecticut General Statutes, Section 4-190 and the sections that follow.
Government in a democratic society by its very nature has to be somewhat inefficient. However, efficiency of services is a basic principle of government information policy because it often goes to government’s ultimate effectiveness, as well as to the financial burden that taxpayers must bear. The section on Efficiency of Services looks at the basic principles that help assure efficiency of services in the area of information technology and resources.
Since in a democratic society where government does not own (in the traditional sense of the word) the records or information it collects, produces and maintains but rather is the custodian of the records and information, the draft Policy also outlines the policies and principles relating to the stewardship of government information.
The draft Policy concludes with the identification of key agencies and personnel necessary to carry out the policies articulated in the policy.
This policy will become part of a set of integrated policies pertaining to information and IT systems for state agencies and implemented through the State of Connecticut Strategic Plan for Information Technology and Policy http://www.doit.state.ct.us/StrategicPlan.htm.
This appendix contains a summary of the websites that contain economic, social and geographical data. These are the websites introduced by the speakers at the Second Annual Data Conference. They are organized by the type of data available, ranging from international to state and local.
Networked Social Science Tools and Resources
International Earth Science Information Network (CIESIN)
Landstat statellite data
Data Ferrett :
Federal Electronic Research and Review Extraction Tool
Remote Sensing Core Curriculum (RSCC)
Remote Sensing Tutorial
U.S. Geological Survey, including:
National Real Estate
First American Core Logic, previously
First America Real Estate Solution
National Stats on Children
Federal Interagency Forum on Child and Family Statistics
Kid's Count from the
Annie E. Casey Foundation
and Child Trends DataBank
Kansas City's Partnership for Children
State and Local:
Connecticut Department of Economic and Community Development
Connecticut Department of Public Health
Connecticut State Department of Education
Connecticut Voices for Children
Center for Land Use Education and Research
Connecticut Economic Resource Center, Inc. (CERC)
Business Council of Fairfield County
Other State and Local Entities
City of Ithaca Interactive Maps
Neighborhood Knowledge Los Angeles (NKLA)
and one of its subsidiaries, healthyCity.org
East St. Louis Action Research Project
operated by the University of Illinois at Urbana-Champaign
Minnesota Governor's Council on Geographic Information map and Data Clearinghouse
Vermont Agency of Human Services