Purpose
While all data captured using University assets are resources of the University, they vary in their relevance to the administrative processes of the University. These guidelines are intended to apply to those data which are critical to the administration of the University, regardless of whether the data is used or maintained by administrative or academic units. While these data may reside in different database management systems and on different machines, these data in aggregate may be thought of as forming a single, logical database, which will herein be called the Institutional Database (IDB).
This terminology is not intended to imply that these data now or in the future should reside in a single physical database. Rather, it is a recognition that regardless of where these data reside, there are some general principles of data management that should be applied in order to maintain the value and guarantee effective use of the information resource. Because the IDB is a single, logical database these principles of data management must be applied uniformly and as part of a coordinated effort.
Table of Contents
- Data Management Roles
- Data Included in the IDB
- Institutional Data Model
- Data Storage
- Data Integrity
- Data Manipulation
- Data Views
- Data Access
- Data Collection
- Data Security
- Data Documentation
- User Support
- Data Availability
- System Administration
Data Management Roles
Data ownership: The University is considered the data owner of all the University's institutional data; individual units or departments may have stewardship responsibilities for portions of the data.
Committee on Institutional Data: The Committee on Institutional Data establishes overall policy and guidelines for management and access to the institutional data of the University.
Data Stewards: Senior University officials (typically at the level of Vice President, Assistant Vice President, Dean, or University Director) who have planning and policy-level responsibilities for data in their functional areas are considered data stewards. The data stewards, as a group, are responsible for recommending policies, and establishing procedures and guidelines for University-wide data administration activities. Data stewards, as individuals, have management responsibilities for defined segments of the institutional data base.
(NOTE: For historical reasons -- because data and the responsibility for data have traditionally been organized along functional or subject-area boundaries -- the data stewards are established according to this same subject-area organizing principle. However, because the eight-campus structure of the university is not explicitly recognized in this organizing principle, the data stewards need to be particularly sensitive to the view from the campuses. The specific or unique views of the individual campuses must be an integral part of institution-wide data policies, and the data stewards must insure that their lines of communication to the campuses are active and user-friendly.)
Data Managers: University officials and their staff who have operational-level responsibility for information management activities related to the capture, maintenance, and dissemination of data are considered data managers. Among the responsibilities of the data managers are any data administration activities outlined in these Guidelines which may be delegated to them by the data stewards.
Data users: Individuals who need and use University data as part of their assigned duties or in fulfillment of their role in the University community are considered data users.
Data administration: The function of applying formal guidelines and tools to manage the University's information resource is termed data administration. Responsibility for the activities of data administration is shared among the Data Stewards, Data Managers, and the University Information Technology Services Data Administration (UITS/DA) organization.
System administration: The function of maintaining and operating hardware and software platforms is termed system administration. Responsibility for the activities of system administration may belong to University Information Technology Services or to other divisions or departments within the University.
Data Included in the IDB
A data element is considered institutional data, and therefore part of the IDB, if it satisfies one or more of the following criteria:
It is relevant to planning, managing, operating, or auditing a major administrative function of the University.
- It is referenced or required for use by more than one organizational unit. Data elements used internally by a single department or office are not typically part of the IDB.
- It is included in an official University administrative report.
- It is used to derive an element that meets the criteria above.
Data elements which meet the criteria for inclusion may be identified as such by a data steward, a data manager, UITS/DA, a user group, or a data user.
A data steward must be identified for each data element to be included in the IDB.
UITS/DA should assist in the negotiations for inclusion of data in the IDB and for identification of data stewards.
Institutional Data Model
The data stewards, or designated data managers, will be responsible for establishing and maintaining a University-wide institutional data model which describes all major data entities of the IDB and the relationships among those data entities, including the linkages among data collected or maintained by the various organizational units of the University.
UITS/DA will support the establishment and maintenance of an institutional data model and provide access to expertise and software tools for data modeling.
Data Storage
An official data storage location for each data element should be identified by the data steward.
An official data storage location of valid codes and values for each data element should be identified by the data steward.
Data element names, formats, and codes should be consistent across all applications which use the data and consistent with such University standards as are developed.
Archiving requirements and strategies for storing and preserving historical data should be determined for each data element by the data steward.
UITS/DA should assist in determining data storage location and archiving requirements for IDB data.
Data Integrity, Validation, and Correction
Applications that capture and update IDB data should incorporate edit and validation checks to assure the accuracy and integrity (consistency) of the data.
The accuracy of any element can be questioned by any authorized data user. The data user has the responsibility to help correct the problem by supplying as much detailed information as available, sufficient to permit understanding and diagnosis of the problem.
The data steward or delegated data manager is responsible for data integrity, responding to questions about the accuracy of data, and correcting inconsistencies if necessary.
Upon written identification and notification of erroneous data, corrective measures should be taken as soon as possible to:
- Correct the cause of the erroneous data.
- Correct the data in the official storage location.
- Notify users who have received or accessed erroneous data.
Data Manipulation, Modification, Extraction, and Reporting
The data steward is responsible for setting policies regarding the manipulation, modification, or reporting of IDB data elements and for creating derived elements, which are also members of the IDB.
The data steward has ultimate responsibility for proper use of IDB data; individual data users will be held accountable for their specific uses of the data.
All data extracted or reported from the IDB should include a record or display of the time and date of data capture.
Data stewards will work with data managers and data users to define useful and meaningful schedules for creation of standard data extracts. These standard extracts of the data ("data snapshots") will also be considered part of the IDB.
Data Views
A data view is a logical collection of data elements, assembled and presented according to a prescribed set of rules. Unlike a data extract--which captures data at a fixed point in time and often moves it physically to a secondary storage location--a data view is a logical entity only, which typically assembles the most current or recent data from their primary storage location at the time they are requested.
Data views may be defined in order to:
- aggregate data from multiple sources
- segment data into smaller and more manageable subsets, or
- segregate data according to confidentiality or restriction characteristics, so that access to the resulting subset may be more widely distributed.
The data stewards are responsible for defining standard views of institutional data within the IDB. These data views will also be considered part of the IDB.
Data managers or data users may recommend the definition of new data views in the IDB.
Data Access
The following philosophy should guide decisions about access to institutional data:
"The value of data as an institutional resource is increased through its widespread and appropriate use; its value is diminished through misuse, misinterpretation, or unnecessary restrictions to its access."
Access to institutional data--the permission to view or query data contained in the IDB--should be granted to all data users for all legitimate University purposes.
As part of the data definition process, data stewards will assign each data element and each data view in the IDB to one of three categories: Public data, University-internal data, and Limited- access data.
Except as noted below, all institutional data will be designated as university-internal data for use within the university. All University employees will have access to these data, without restriction or prior authorization, for use in the conduct of University business. These data, while freely available within the University, are not designated as open to the general public.
Where appropriate, data stewards may identify elements or views of the IDB which have no access restriction whatsoever and which may be released to the general public. These data will be designated as public data.
Where necessary, data stewards may specify some data elements as limited-access. Limited-access data would include those data for which data users must obtain individual authorization prior to access, or to which only limited access may be granted.
Designation of data as limited-access will include:
- specific reference to the legal, ethical, or externally-imposed constraint which requires this restriction;
- description of what categories of data users are typically given access to the data, under what conditions, or with what limitations.
A data view does not necessarily inherit the restriction characteristics of the data elements which comprise it. (For example, removal of any association with personally-identifying data elements can result in a view which contains otherwise- restricted data elements being designated as public or university-internal.)
To the extent possible, data stewards will work together to define a single set of procedures for requesting permission to access limited-access data elements in the IDB, and will be jointly responsible for documenting these common data access request procedures.
Each data steward will be individually responsible for documenting data access procedures that are unique to a specific information resource or set of data elements.
Any data user may request that a data steward, or the committee of data stewards as a group, review the restrictions placed on a data element or data view, or review a decision to deny access to limited-access data.
When necessary, the Committee on Institutional Data will make the final determination on data restrictions and requested access rights to institutional data.
Data Collection and Maintenance
The data steward is ultimately responsible for complete, accurate, valid, and timely data collection. Operational responsibility for data collection and maintenance is typically assigned to the data managers.
Further delegation and decentralization of data collection and maintenance responsibility is encouraged in order to assure that:
electronic data are collected and maintained as close as possible to the source or creation point of the data as identified by the data steward, and each manual or computer process which handles data adds value to the data.
Data Security
The data steward will be responsible for determining security requirements and access restrictions for institutional data.
All data users having access to limited-access portions of the IDB data should formally acknowledge (by signed statement or some other means) their understanding of the level of access provided and their responsibility to maintain the confidentiality of the data they access. The data steward is responsible for monitoring and reviewing security implementation and authorized access.
The data steward is ultimately responsible for defining and implementing policies and procedures to assure that data are backed up and recoverable in response to events that compromise data integrity. University Computing Services or other University agencies may assist in this effort.
Data Documentation
Documentation of data elements is the ultimate responsibility of the data steward. Some or all of these responsibilities may be assigned to data managers.
Documentation/definition for each data element should at least include:
- Name and Alias Names
- Description
- Data Steward
- Usage and Relationships
- Frequency of Update
- Source for Data Capture
- Official Data Storage Location and Format
Documentation should also include:
- Designation as "Limited-access", "University-internal", or "Public"
- For "Limited-access" Data Elements: Description or Specification of the Restriction
- Description of Validation Criteria and/or Edit Checks
- Description, Meaning, and Location of Allowable Codes
- Access Rules and Security Requirements
- Archiving Requirements
- Data Storage Location of Extracts
Documentation for derived IDB data elements should include the algorithms or decision rules for the derivation.
Documentation of data views should include reference to the data elements which comprise the view and description of the rules by which the view is constructed.
Overview documentation for logical segments of the IDB (databases, files, groups of files) should also be provided which includes information about data structure and update-cycles necessary for the accurate interpretation of the data.
The following guidelines will be implemented concurrent with implementation of a University Data Resource Dictionary:
- Documentation of data elements should be provided to UITS/DA in machine-readable format by the data stewards. This documentation will ultimately reside in a University Data Resource Dictionary.
- UITS/DA will be responsible for the data administration function of maintaining the University Data Resource Dictionary and for making it readily accessible to all interested parties.
- Change in any data definition characteristics should be noted to UITS/DA and recorded in the University Data Resource Dictionary in advance of the change
User Support
The data steward and data managers responsible for each major segment of the IDB will define the extent of support for data access and interpretation which is available to users of these data.
Data stewards will provide user support--primarily through documentation of the information resource but also, as needed, in the form of consulting services--to assist data users in the interpretation and use of data elements in the IDB. This responsibility may be delegated to the data managers.
University Computing Services will provide training classes for a selected set of supported information access tools and technologies. Other University agencies may assist in this effort. The data users will be responsible for their own use and interpretation of the data which they access.
Data Availability and Integration
Data stewards are responsible for providing accessible, meaningful, and timely machine-readable institutional data for University use.
Data stewards and UITS/DA share the responsibility for data compatibility, accessibility, and interfaces among institutional data elements residing in various segments of the IDB.
Data stewards and UITS/DA will work together toward unification of the various data element coding structures and data storage formats which exist in various segments of the IDB.
System Administration
Institutional data must be maintained within a single, logically- integrated information system.
Institutional data may be stored on any of many diverse computing hardware platforms, provided such platforms are fully integrated components of an overall University Information System.
If institutional data are stored on any component of the University Information System, that system component must have defined a formal system administration function and have assigned to it a system administrator whose responsibilities include: physical site security; administration of security and authorization systems; backup, recovery, and system restart procedures; data archiving; capacity planning and performance monitoring.
Procedures
These Data Administration Guidelines for Institutional Data Resources have been prepared by the Administrative Computing Advisory Committee in association with its Data Administration Subcommittee, and forwarded to the Committee on Institutional Data for review and approval. They serve as a statement of objectives to manage the administrative information resource. These Guidelines apply to all IDB data. In addition, these Guidelines should be considered and followed where possible by all those who capture data and manage administrative information systems using assets of the University. Standards and procedures should be developed to conform to the objectives embodied in these Guidelines.
Updates
As an ongoing document, the Data Administration Guidelines for Institutional Data Resources will be maintained and revised as needed by the Committee on Institutional Data, data stewards, data managers, and the Administrative Computing Advisory Committee, with support from University Computing Services/Data Administration. All administrative system users are encouraged to correspond with UITS/DA describing any suggestions for improving these Guidelines.
Glossary
For easy reference, following is a summary of terms used and defined in these Guidelines:
Access to institutional data: The permission to view or query data contained in the Institutional Database, but does not necessarily imply delivery or support of specific methods or technologies of information access.
Committee on Institutional Data: A committee chaired by the Associate Vice President for Information Resources which establishes overall policy and guidelines for management and access to the institutional data of the University, and which regularly reviews the performance of the overall data administration function.
Data administration: The function of applying formal guidelines and tools to manage the University's information resource.
Data Managers: University officials and their staff who have operational-level responsibility for data capture, data maintenance, and data dissemination.
Data Stewards: University officials having policy-level responsibility for managing a segment of the University's information resource.
Data users: Individuals who access University data in performance of their assigned duties or in fulfillment of their role in the University community.
Data view: A logical collection of data elements, possibly from multiple physical databases, which are assembled and presented according to a defined set of rules.
Eligible employees: Faculty, staff and administrators holding full-time appointments at Indiana University or other employees specifically designated as eligible by the head of their department, division, school, or campus.
IDB (Institutional Database): A conceptual term used to identify that body of data critical to University planning, management and business operations of both administrative and academic units.
Institutional Data Model: A logical construct which describes the data entities which comprise the Institutional Database and the relationship among those entities.
University-Internal data: Elements of the IDB that may be accessed by all employees of the University, without restriction, for the conduct of University business.
Public data: Elements of the IDB that are available to the general public, including to those outside of Indiana University.
Limited-Access data: Elements of the IDB that, because of legal, ethical, or other externally-imposed constraints, may not be accessed without specific authorization or to which only limited access may be granted.
System administration: The function of applying formal guidelines and practices to the management of a computing resource.
University Data Resource Dictionary: A database system that functions as a repository that contains comprehensive information about the University's institutional data and documentation ofUniversity administrative systems. (No such repository exists today.)
University Information System: A conceptual term used to identify the collection of computer hardware, software and network connections which together form the single, integrated system on which resides the Institutional Database.
Approvals
Approved by the ACAC Data Administration Subcommittee on February 14, 1991.
Approved by the Administrative Computing Advisory Committee March 21, 1991.
Acknowledgement
The document, "Guidelines for University Administrative Information Resource Management" (Virginia Polytechnic Institute and State University Policy and Procedures Section 2005) served as an initial model for these guidelines.
