Information Collection Project
Report on Current Recordkeeping Practices within the Federal
December 10, 2001
Prepared for the
National Archives and Records Administration Under Contract GS-23F-9806H
Task Order NAMA-00-F-0070, Task 2e
SRA International, Inc.
2000 15th Street
North Arlington, Virginia 22201
SRA International Study Team
SRA International, Inc.:
Philip Zellner, Project Manager
Jon Vetterlein, Senior Analyst
Charles R. McClure, Ph.D.
Francis Eppes Professor, and Director
Information Use Management and Policy Institute
Florida State University J. Timothy Sprehe
Sprehe Information Management Associates, Inc.
SRA International Project Staff:
The authors would like to acknowledge the enthusiasm and support they received for this project from the NARA Project Team, particularly Susan Cummings, the Contracting Officer's Representative (COR) for the project. From the initial project meeting in October 2000 and continuing through to the preparation of this final report, senior NARA officials involved in the project held periodic meetings with the SRA study team. These meetings included extensive and productive discussions of both empirical data and substantive issues that arose during the course of the study. The knowledge and experience that NARA personnel brought to these meetings frequently improved the study team's conceptualizations of perceptual measurement and findings, as well as explanatory factors for making findings meaningful. Often, a new idea or finding introduced by the study team made more sense after NARA personnel had enriched its context or refined its expression. In this sense, the report is a collaborative effort and it would be difficult to draw a line between the contributions of NARA and the SRA study team.
We would also like to thank the officers of the Federal Information and Records Managers (FIRM) Council who took an active interest in the project by promoting the Web survey and the records officer focus groups among its membership, and providing the study team a site for the first records officer focus group.
Finally, the authors would like to thank the many agency executives, managers and staff who took time out of their schedules to participate in interviews and focus groups, as well as those who took the time to complete the web survey. We received tremendous input from you and we hope that our report accurately presents your perceptions of the current state of Federal records management, and your recommendations
List of Tables
Table 2-1. Focus Groups and Interviews Conducted by SRA Study Team
Table 3-1. SRA and NARA Data Collection Efforts
Table 3-2. RSA Summary by Records Life Cycle
Table A-1. NARA Information Collection Topics
Table B-1. Situational Factors in Selected Agency Processes
Table C-1. Mapping of Intervention Scenarios to RM Quality Level
List of Figures
Figure 4-1. Records Life Cycle and Major Problem Areas
Figure 5-1. Situational Factors Model
Figure 6-1. Agency RM Levels and Potential Intervention Scenarios
In order to better carry out its mission to ensure continuing access to essential evidence that documents the rights of American citizens, the accountability of Federal officials and agencies, and the national experience, the National Archives and Records Administration (NARA) commissioned a study of recordkeeping (RK) perceptions to complement its own Records System Analyses (RSAs) of RK practices. Specifically, NARA contracted with SRA International, Inc. (SRA) to collect information on Federal employees? views and perceptions on twenty Information Collection Topics, prepared by NARA and pertaining to records creation, maintenance, use, and disposition within 18 Federal agencies. NARA?s RSA teams examined records systems and collected factual data concerning the NARA Information Collection Topics for 11 of the agency processes.
The purpose of this report is fourfold.
- Describe the results of the SRA survey on Federal employees? perceptions regarding RK.
- Compare the perceptual data collected by the SRA study team with the factual data collected by the NARA RSA teams. Present a situational factors model to explain observed variations in the quality of Federal agencies? records management (RM) programs.
- Provide NARA and the agencies with some intervention options for improving the quality of Federal RM.
The SRA Study on Perceptions of Recordkeeping
Over a five-month period beginning in January 2001, the SRA study team gathered survey responses from approximately 500 Federal employees representing over 150 Federal government organizations and conducted 54 focus groups and interviews involving individuals from 18 agencies. Participants included records officers and records liaisons, General Counsel (GC) staff, Inspector General (IG) staff, information technology (IT) staff, Chief Information Officers (CIOs), and process workers, agency staff who carry out the work of a specific, mission-related agency work process. The SRA study team used a phased, evolutionary data collection approach involving focus groups, interviews, and a Web-based survey to collect information.
The major findings from the perceptual surveys, focus groups, and interviews conducted for the SRA study are as follows.
- The quality and success of recordkeeping varies considerably across the agencies studied.
- Government employees do not know how to solve the problem of electronic records ? whether the electronic information they create constitutes records and, if so, what to do with the records.
- When agencies have a strong business need for good RK, such as the threat of litigation or an agency mission that revolves around maintaining "case" files, then RK practices tend to be relatively strong with regard to the records involved.
- Although records officers and other agency records managers are familiar with the life cycle of records, integration between RK and the business processes of many agencies is distinctly lacking.
- For many Federal employees, the concept of a "record" and what should be scheduled and preserved is unclear.
- RK and RM in general receive low priority, as evidenced by lack of staff or budget resources, absence of up-to-date policies and procedures, lack of training, and lack of accountability.
- While agencies appreciate specific assistance from NARA personnel, they are frustrated because they perceive that NARA is not meeting agencies? broad needs for guidance and RM leadership. Primarily, agencies want timely and responsive guidance and leadership from NARA on current RM issues.
The SRA study demonstrated that "one size does not fit all" when it comes to RK practices for the Federal government. Agencies? RK practices can be characterized as falling along a continuum from strong to weak. In order to understand this variability, the study team identified a set of situational factors that may be predictive of the quality of RK practices and the success of the RM program in Federal agencies.
NARA Records System Analyses
The NARA RSA teams collected factual records system data for key processes within 11 Federal agencies. The SRA and NARA RSA data collections were comparable in scope, but not identical and NARA visited only a subset of the agencies SRA studied. The reports from the NARA RSA teams for individual agencies are impressive in their depth and analysis of RK.
The RSA reports provide excellent assessments of RM quality for 11 Federal agency work processes and suggest that Federal records scheduling and disposition practices need improvement. The overall results were as follows.
- Records Creation. Adequate records were being kept for the processes studied.
- Records Maintenance and Use
- Recordkeeping Requirements. For the most part, the RSAs found requirements adequate, documented, and consistently applied.
- Records Accessibility. Employees are generally able to find the records they need.
- Records Scheduling and Disposition
- Records Scheduling. RSA reports indicated that many significant records were unscheduled. Most electronic records were unscheduled.
- Permanent Records Transferred to NARA. Most of the permanent paper and electronic records encountered by RSA teams were not being transferred to NARA, usually because they were unscheduled.
Comparison of SRA Perceptual Data and NARA RSA Data
The perceptual data collected by SRA and the factual data from the RSAs are more complementary than contrasting. With few exceptions, findings were consistent across the two data sets.
- Records Creation. The RSAs determined that employees create adequate records, even though the perceptual data indicate they receive no formal RM training.
- Electronic Records. Both data collections found that records are born electronic but paper is the official record format.
- E-mail Records. Both data collections determined that managing e-mail is a major RM problem.
- Records Maintenance. Some RK systems were consistent with regulations and operating successfully; others were inappropriately maintained and out of step with guidance.
- Electronic Records Maintenance. Both data collections found that records in electronic formats were not appropriately managed.
- Ability to Retrieve Records. Both data collections determined that employees can find records they need, although the perceptual data indicate that finding them often requires significant effort.
- Records Disposition. Both the perceptual and RSA data indicated disposition of permanent records is poor.
- Influence of Business Process, Culture, and Technology. Both the RSA and SRA teams found that business processes are an important determinant of RK success, agency culture a considerable influence, and technology tools for RK not yet in use, although the Internet is having a deep and growing effect.
The SRA study team hypothesized that a range of "situational factors" drives the overall quality of the RM program in a particular Federal agency or component sub-organization within an agency. The relative strength or weakness of various factors will account to a substantial degree for the overall strength or weakness of the RM program in that agency or organization. Situational factors explain variations in RM quality and can serve as diagnostic tools.
SRA identified four categories of situational factors: (1) Institutional Context, (2) Policy and Guidance, (3) Resources, and (4) Other Factors.
- Institutional Context.
- Business Need. Agencies will perform that level of RK necessary to meet the needs of their business and that provides tangible business benefits.
- Threat of Litigation, FOIA Burdens, or Public Scrutiny. Agencies frequently sued, receiving high volumes of FOIA requests, or under constant scrutiny from the Congress, the press, and the public tend to have higher quality RM programs.
- Work Processes Involving Case Files. RK is better where the work processes are organized by case files.
- More/Less Well-Defined Records. Programs with better-defined records have better RK.
- Process Involving Litigation. RK is better for those processes involving litigation or potential litigation.
- Age of Process or Program. More mature, stable, older programs tend to have better RK.
- Structure of Process and Frequency of Performance. Highly structured work processes and those frequently performed have better RK.
- Culture. Legal and audit cultures appear to be conducive to good RK. For no discernible reason, the scientific, engineering, and information technology cultures studied tended to have poor RK.
- Leadership Commitment to RM. Where top agency leadership communicates interest and concern about RM, RK tends to be of higher quality.
- Quality of Records Officers. Motivated and knowledgeable records officers can make a significant difference in the quality of RK.
- Policy and Guidance Factors
- Timely and Responsive NARA Policy and Guidance. Agencies want NARA?s assistance with emerging RM issues.
- Internal RM Policies and RK Procedures. Where RM policies and RK procedures are up to date, RK tends to be better.
- RM Awareness. When employees are aware of RM policies/procedures, RK is better.
- Administrative/Finance Support. Good resource levels lead to good RM.
- Records Officer Involvement. If ROs are involved in agency business processes and programs, RK tends to be better.
- Planning, Piloting, and Implementing ERM. The farther along an agency is with electronic records management, the better the RM program and vice versa.
- Other Factors. Amount of interaction with NARA, centralized coordination of RK, efficiency of access to retired records, proper scheduling, and resources for records storage are other contributors to quality RM programs.
Potential Intervention Scenarios
Beginning with RM program quality levels, one may think of a set of performance levels ranging from an inadequate RM program to an excellent RM program. The study team chose to present four performance levels: Inadequate, Adequate, Good, and Excellent.
At the Inadequate RM Program level, records are not properly kept, employees cannot find records, records schedules are frequently outdated or missing. The RM program is completely paper-based despite the existence of extensive electronic files, and the agency has no planning underway for ERM systems.
At the opposite end of the spectrum is the Excellent RM Program level, in which the paper-based RM program is well managed from creation, through maintenance and use, to disposition. RM is increasingly integrated into all agency IT systems and business processes. The agency also has planned, piloted, acquired, and is undertaking staged deployment of ERM systems. RM training is mandatory for all employees and the agency is actively addressing emerging RM issues such as e-mail.
Given a set of RM performance levels such as these, the study team considered a set of intervention scenarios to address selectively the characteristics of each level. The scenarios are brief descriptions of possible approaches NARA and/or agencies might take to improve RM in the Federal government. The report presents four scenarios.
Scenario 1: Policymaker and Spokesperson. NARA expands its policy leadership role in Federal RM by initiating an executive education and briefing program. The need this scenario fulfills is one voiced by many agency records officers, namely, that their senior management needs a fundamental understanding of the importance and value of good RM to any agency?s mission, over and above legal obligations and responsibilities. NARA senior executives become directly involved in interactions with senior management in Federal agencies for RM championing.
Scenario 2: Provider of Technical Assistance and Education. NARA expands its Targeted Assistance initiative offerings to design, test, and implement a range of technical approaches, including hardware, software, training, and interactive network support services, in order to assist agencies? RM programs. The expanded program could become a change management laboratory for helping agencies to pilot new and experimental approaches to contemporary RM and for gathering accumulated agency experience into a knowledge base of lessons learned. NARA could formalize the new program initiative with a title such as Center for Excellence in Records Management.
Scenario 3: Situational Factors Change Facilitator. NARA develops the situational factors into a set of diagnostic tools. NARA and/or the agencies review RM programs, identify RK problem areas, and determine which situational factors need changing and the best points of entry. They tailor strategy and tactics for each factor.
Scenario 4: Supporter of Excellence in Records Management. NARA?s role under this scenario is largely to acknowledge and support those agencies that are engaged in high quality RM. In such instances, NARA does not need to develop intervention strategies. Rather, the role of NARA is to publicly acknowledge and praise the RM being done at that agency. An expanded awards program is one option for consideration.
This report offers a number of approaches, options, and interventions that may assist NARA and the agencies to improve Federal RM. A key theme throughout the report is the importance of taking into account the situational factors affecting RM from the agency perspective. Identifying and understanding these situational factors offers NARA and the agencies a range of entry points for change management tactics within individual agencies.
When one compares the RSAs conducted by NARA to the findings from the study team?s data collection efforts, it becomes apparent that both approaches are necessary to describe and understand the RM environment in a particular agency. In effect, there are two key perspectives to consider when reviewing RM in an agency: (1) the records systems analysis related to scheduling, storage, disposition, and other functions, and (2) situational factors in the agency affecting the manner in which RM can be accomplished ? and how well it can be accomplished.
Additional development of the situational factors, RM performance levels, and intervention options requires refinement and field testing of the results presented in the report. With these tools and options, NARA will have additional strategies to work with agencies, develop specific approaches that may work best in that agency, and ultimately improve the overall quality of RM in the Federal government.
Recent technological advances, budget cuts, and other events have significantly changed the environment surrounding Federal records management (RM).1 In order to better carry out its mission to ensure continuing access to essential evidence that documents the rights of American citizens, the accountability of Federal officials and agencies, and the national experience, the National Archives and Records Administration (NARA) commissioned a study of RM perceptions to complement its own Records System Analyses (RSAs) of RM practices. Specifically, NARA contracted with SRA International, Inc. (SRA) to collect information on Federal employees' views and perceptions on twenty Information Collection Topics, prepared by NARA and pertaining to records creation, maintenance, use, and disposition within 18 federal agencies. NARA's RSA teams examined records systems and collected factual data concerning the NARA Information Collection Topics for 11 of the agency processes.2 NARA will consider the results of these studies in designing policies and programs to meet the challenges facing government agencies in creating, maintaining, using, and disposing of Federal records in the 21st century.
The purpose of this report is fourfold. The first purpose is to describe the results of the SRA survey on Federal employees' views and perceptions regarding records creation, maintenance and use, and disposition within their agencies. The second purpose is to compare and contrast the perceptual information collected by the SRA team with the factual records systems data collected by the NARA RSA teams and to identify agreements and disparities. The third purpose is to identify situational factors that help explain the observed variations in the quality of Federal agencies' RM programs. Fourth, using the information collected by the SRA and NARA RSA teams and the concept of situational factors, the report seeks to provide NARA and the agencies with some effective intervention options for improving the quality of Federal RM.
Within the context of this report, the term record has the meaning found in Federal law at 36 CFR 1220.14.3
For the purposes of this report, the term recordkeeping means the creation, maintenance and use, and disposition of reliable and trustworthy records that meet the business needs and legal responsibilities of a Federal program office and, to the extent known, the needs of internal and external stakeholders who may make secondary use of the records.
For the purposes of this report, the term records management means the policies, procedures, guidance, tools and techniques, resources, and training needed to design and maintain reliable and trustworthy records systems in an efficient and effective manner.
For this report, the relationship between recordkeeping and records management is that records management includes recordkeeping but that records management extends to more than recordkeeping. Records management could include the administrative directives prescribing the manner in which records are to be kept within a program office, as well as information technology (IT) tools such as a commercial-off-the-shelf (COTS) electronic records management system that is an integral component of the IT environment within the larger agency.
The remaining sections of the report are organized as follows:
- Section 2: Information Gathering, Analysis and Report on Current Recordkeeping and Records Use: The SRA Study. This section summarizes the results of the perceptual data collection performed by the SRA study team.
- Section 3: NARA Records Systems Analyses. This section describes NARA RSA information collection results for the 11 Federal agency processes in terms of records life cycle phases.
- Section 4: Comparison of SRA Perceptual Data and NARA RSA Data. This section presents a comparison of the SRA and NARA findings for key issues among the twenty NARA Information Collection Topics, highlighting some of the problems in Federal RM practices.
- Section 5: Situational Factors. This section describes a number of situational factors that may help explain variation in the quality of Federal RM practices and presents a model showing the relationship of these factors to the phases of the records life cycle.
- Section 6: Potential Intervention Scenarios. This section relates situational factors to the concept of agency RM program quality levels and presents some possible NARA intervention options linked to the RM program quality levels.
- Section 7: Conclusions. The last section offers conclusions regarding the findings presented in this report and how they can be used to improve Federal RM.
- Appendices. Appendices A, B, and C, contain tables displaying the twenty NARA Information Collection Topics, the situational factors identified by agency process, and a mapping of the intervention scenarios to the RM quality levels, respectively.
1See the NARA
Strategic Plan 2000.
2Appendix A contains the 20 NARA Information Collection Topics and provides an overview of the SRA and NARA responsibility for the specific topics.
3 "Includes all books, papers, maps, photographs, machine-readable materials, or other documentary materials, regardless of physical form or characteristics, made or received by an agency of the United States Government under Federal law or in connection with the transaction of public business and preserved or appropriate for preservation by that agency or its legitimate successor as evidence of the organization, functions, policies, decisions, procedures, operations, or other activities of the Government or because of the informational value of data in them. Library and museum material made or acquired and preserved solely for reference or exhibition purposes, extra copies of documents preserved only for convenience of reference, and stocks of publications and of processed documents are not included."
The goal of the SRA data collection was to identify and document Federal employees? views and perceptions relating to the creation, maintenance, use, and disposition of Federal records in their work environment, what Section 1.3 defines as recordkeeping (RK). This section summarizes the SRA information collection effort on RK perceptions and practices and presents the major findings.
Over a five-month period beginning in January 2001, the SRA study team gathered survey responses from approximately 500 Federal employees representing over 150 Federal government organizations and conducted 54 focus groups and interviews involving individuals from 18 agencies. Participants included records officers and records liaisons, General Counsel (GC) staff, Inspector General (IG) staff, information technology (IT) staff, Chief Information Officers (CIOs), and process workers. For this study, process workers are defined as agency staff who carry out the work of a specific, mission-related agency work process.
The SRA study team used a phased, evolutionary data collection approach involving focus groups, interviews, and a Web-based survey to collect information. The "phased, evolutionary data collection approach" meant that the agencies surveyed and the questions asked were permitted to evolve based on the results obtained during earlier phases of the data collection effort. The team collected data on Federal employees? perceptions relating to the NARA Information Collection Topics and the records life cycle phases: creation, maintenance and use, and disposition of Federal records in their work environment. The overall timeline for the data collection effort is summarized below:
- In January 2001, SRA conducted four focus groups, involving a sample of about 40 Records Officers, to explore selected RK topics and issues.
- In February 2001, all of the Records Officers within the Federal Government were invited to take the Web-based survey.
- Between April and May 2001, the team conducted site visits to 18 Federal agencies and conducted 54 interviews and focus groups within the agencies. Table 2-1 shows the Federal agencies that participated in the focus groups and interviews. During these site visits, the study team asked the interview and focus group participants to fill out a paper-and-pencil version of the survey questionnaire used for the Web-based survey.
- In early May, NARA sent letters to approximately 1,000 of its Atlanta region field contacts inviting them to take the Web-based survey. This extra mailing was designed to ensure that a sufficient number of survey responses were received from field employees.
It should be noted that both NARA and the SRA study team assured Federal agency employees participating in the study that their confidentiality would be respected. The goal was to elicit employees? free and open expression of perceptions and facts about Federal RK, both within their agencies and more broadly throughout the government. The goal was achievable only if employees felt confident that their expression would not come back to harm them or their agencies. The selection of participating agencies was based on a combination of the criteria established by NARA, a random selection process, and agreement of the agency to participate in the study. The study team?s initial criteria for choosing agencies for site visits included factors such as agency size, agency culture, location, and other matters that could differentiate the kinds of perceptions individuals have about recordkeeping. Final choices were negotiated between NARA and the study team as well as the agencies studied. Both NARA and SRA acknowledged that it was impossible to conceptualize or to operationalize a fully unbiased sample of participating agencies for the purposes of this project. Nevertheless, both NARA and SRA were confident that the sample of agencies selected for analysis would provide useful data for NARA policy making.
The scheduling of agency site visits proved to be considerably more difficult than anticipated. A number of agencies declined to participate in the study, and the study team often had to adjust its schedule to fit agency personnel availability. The rule of thumb guiding the number of sites to be visited was that ten "core" agencies (as agreed between the study team and NARA) would be visited first, and then the study team would continue site visits until the visits produced no new compelling information. To obtain input from Federal employees outside the Washington, DC area, interviews and focus groups were conducted at the offices of three agencies in the Atlanta area.
When the data collection was complete, the study team compared findings from the various sources, including the Web survey, records officer focus groups, and site visits. The team then analyzed the perceptual information with respect to NARA Information Collection Topics.
It is important to stress that the findings in the SRA study do not represent a random sample of all government employees. Furthermore, the information collected is primarily qualitative in nature, so the findings are not necessarily reflective of the views of a particular organization, agency, or the government as a whole. They are individual opinions and perceptions, and indicate only the existence and variety of RK issues within the Federal government. Also, the findings presented in the SRA study do not reflect the factual records systems data collected by the NARA RSAs (see Section 3). Section 4 of the report integrates the NARA RSA findings with the results from the SRA study. The two sets of results, taken together, present a more complete picture of Federal RK in 2001.
The major findings from the perceptual surveys, focus groups, and interviews conducted for this study are listed below. They represent issues the study team heard over and over, regardless of agency or function.
The quality and success of recordkeeping varies considerably across the agencies studied. The variety of views expressed in focus groups and interviews suggests that the priority Federal agencies assign to RK, the quality of RK practices, and the extent to which RK permeates agency operations vary considerably across?and even within?agencies and may depend on a number of "situational factors." Overall, the study team found limited or moderate interest in and commitment to RK in the agencies studied. Yet, it should be noted that some agencies perform high quality RK and have good records management (RM) programs.
Government employees do not know how to solve the problem of electronic records ? whether the electronic information they create constitutes records and, if so, what to do with the records. Electronic files that qualify as records?particularly in the form of e-mail, and also word processing and spreadsheet documents?are not being kept at all as records in many cases and are frequently not being scheduled. Employees lack guidance and knowledge concerning how to identify electronic records and what to do with them once identified. Technology tools for managing electronic records do not exist in most agencies. The agency information technology environments have not been designed to facilitate the retention and retrieval of electronic records. Despite the growth of electronic media, agency records systems are predominately in paper format rather than electronic. Virtually every agency visited indicated that the official policy is that their records will be maintained in paper format. Yet the agencies recognize that most records are now created in an electronic environment?in word-processing documents, spreadsheets, databases, and the like. The predominant e-mail policy is to print out e-mails that are considered records and to save the paper copies. The chief paradox of today?s Federal RM is the disconnect between paper and electronic recordkeeping.
When agencies have a strong "business" need for good RK, such as the threat of litigation or an agency mission that revolves around maintaining "case" files, then RK practices tend to be relatively strong with regard to the records involved. The best recordkeeping tends to occur for case files and legacy systems of record. Case files have a well-defined, business-related purpose and provide a framework for RK activities. Similarly, agencies with good, conscientious RM programs tend to be those agencies most aware of the threat of legal and other liabilities arising from poor RK. Agencies receiving large volumes of FOIA requests fall in this category, as do agencies under fairly constant Congressional, mass media, or public scrutiny. These agencies recognize that poor RK will lead to adverse legal judgments, unwanted legislation, and/or public embarrassment.
Although records officers and other agency records managers are familiar with the life cycle of records, integration between RK and the business processes of many agencies is distinctly lacking. Apart from individuals directly involved in the RK function, the terms "recordkeeping" and "records management" have little meaning or significance. In other words, government employees manage "case files," "audit files," "litigation files," "grant files," or "project files" rather than "records" per se. In doing so, they are guided by their work processes?they learn how to manage the records required by their jobs as part of their job, not as part of a basic responsibility to maintain records that "protect the rights of citizens," "ensure accountability of government," or "document the national experience."
For many Federal employees, the concept of a "record" and what should be scheduled and preserved is unclear. Some participants said that what constitutes a "record" varies considerably, even within a particular agency. They asked for simple rules of thumb to apply to practical situations.
RK and RM in general receive low priority, as evidenced by lack of staff or budget resources, absence of up-to-date policies and procedures, lack of training, and lack of accountability. Agencies generally view good RK as important or at least nice to have, but usually not as a high priority. RM is generally not even "on the radar scope" of agency leaders, and records officers have little "clout." For example, they often cannot ensure that key agency leaders, such as political appointees, receive any records briefings or training at all. Federal downsizing may have negatively impacted RM budget and staffing resources in agencies?RM is generally considered a "support" activity, and support functions are typically the first to be cut. Few resources are devoted to RK. Most agencies have a small RM staff and devote little time and money to RK training for their employees. Records officers do not appear to have much involvement in or influence on programmatic business processes or the development of information systems designed to support them. New government employees seldom receive any formal, initial RM training; nor is continuing training common. Agencies do not have the staffing to conduct periodic, comprehensive RK inspections; and unless litigation is involved, Federal employees suffer no direct consequences from poor RK practices.
While agencies appreciate specific assistance from NARA personnel, they are frustrated because they perceive that NARA is not meeting agencies? broad needs for guidance and RM leadership. Respondents who have personal contact with NARA for specialized assistance were practically unanimous in their praise for the quality of NARA?s assistance. However, many participants said they needed more detailed, practical guidance on RK from NARA. The study team also heard frequently that NARA?s schedule approval process takes too long, detracting from its usefulness as a service to agencies.
Primarily, agencies want timely and responsive guidance and leadership from NARA on current RM issues. They believe NARA has a responsibility to lead the way in transitioning to an electronic records environment and to provide the guidance, standards, and tools to enable agencies to follow. They particularly mentioned the need for current guidance on e-mail and web site RK. Some view NARA as leaving agencies on their own to fend for themselves, sometimes levying impossible requirements that pressure agencies to come up with their own individual solutions. Many agency staff members told the study team that they do not expect NARA to be perfect, just to lead. They need prompt and timely guidance, direction, and solutions to a range of RK issues and problems.
The SRA study demonstrated that "one size does not fit all" when it comes to RK practices for the Federal government. Agencies? RK practices can be characterized as falling along a continuum from strong to weak. In order to understand this variability, the study team identified a set of situational factors, described in Section 5 below, that may help NARA and the agencies in developing RK strategies tailored to particular agency settings. The study team revised and refined the situational factors over several months of discussion with NARA personnel. The hypothesis is that situational factors are predictive of the quality of RK practices and drive the overall success of the RM program in a particular Federal agency. Where a certain combination of these factors is present to a substantial degree in a particular agency, RK practices will be strong in that agency.
From the outset, NARA conceived this project as proceeding on two data collection tracks. On one track, SRA would collect data pertaining to RK perceptions from a set of agencies and program offices within agencies and from employees occupying a predefined group of roles in the agencies. Once the SRA data collection was underway, NARA would visit a subset of the agencies from which SRA had collected perceptual data and perform Records Systems Analyses. NARA?s purpose in the RSAs was to study the business processes and the actual records that were generated at the same sites. This section describes the results of the RSAs.
The NARA RSA teams collected factual records system data for key processes within 11 Federal agencies in terms of records recreation, maintenance and use, and disposition. Table 3-1 highlights the 11 agencies and offices that both SRA and NARA teams visited during their information collection efforts.
The SRA and NARA RSA data collections were comparable in scope, but not identical. Table 3-1 indicates that the processes covered by NARA teams represent a subset of those agencies visited by SRA. The overlapping areas reside in ten shaded processes within the 11 agencies and offices. In addition to conducting focus groups of process workers, the SRA team interviewed CIO/IT staff and staff from the GC offices and the IG offices at most of these agencies. This procedure meant that the SRA team?s data collection effort encompassed a broader range of organizational units and officials. As noted previously, SRA concentrated on perceptual information whereas NARA?s focus was factual. Finally, the scope of the information collection efforts also differed in that in some cases the NARA teams collected data from several offices in an agency, whereas the SRA team conducted only one process worker focus group within one office at each agency. In organizational terms, the SRA information collection covered a greater vertical range and the NARA information collection a greater horizontal range.
Table 3-1, SRA and NARA Data Collection Efforts
The full RSA reports conducted by the NARA RSA teams for individual agencies are impressive in their depth and analysis of RK. The level of effort by NARA to conduct the RSAs on an agency-by-agency basis was significant. For example:
- FMC required 3 NARA staff, 446 staff hours, over 64 days
- INS Border Patrol required 10 NARA staff, 918 staff hours, over 45 days
- SSA required 7 NARA staff, 650 staff hours, over 51 days
- VHA required 9 NARA staff, 1161 staff hours, and 4 months
To the study team, these numbers are daunting. They suggest that, if NARA were to adopt an RSA-like strategy on a government-wide basis the agency would need a significant increase in staff and resources. While one-on-one assistance and assessment by NARA for individual agencies have great potential payoff for improved RK, the resources needed to implement such a strategy may prove unobtainable.
The 11 RSAs of the processes examined in the NARA study contain specific answers to selected NARA Information Collection Topics. It is informative to compare the answers to one another as they provide support for the conclusion that, while some common themes are observable, the character and quality of RK vary greatly from agency to agency.
Table 3-2 on the next page summarizes the results of the RSA analyses by agency process and in terms of the records life cycle phases: records creation, records maintenance and use, and records disposition.4 For each phase, Table 3-2 presents an assessment of RK quality with respect to a number of evaluation criteria addressed in the RSA reports. Cells with an entry of "Yes" and shaded green indicate areas in which the quality of RK being performed is good. Cells with an entry of "Some" or "Most" and shaded yellow indicate areas in which the quality of RK is fair, and cells with an entry of "No" and shaded red indicate areas in which RK quality is poor.5 Cells with an entry of "n/a" indicate that the criterion was not applicable for the agency process. For some criteria in some processes it was not possible to determine the level of RK quality from the RSA report. Those cells in the table are blank.
- Records Creation ? In general, the RSA reports show that the 11 processes the RSA teams examined appear to generate adequate records documentation. All of the processes examined produce records that protect the rights of citizens and ensure government accountability.6 The RSA teams determined that only six of the agency processes produce records of enduring research value. However, the records produced by several of the processes are aggregated to produce higher-level records that are scheduled as permanent records.
- Records Maintenance and Use
- Recordkeeping Requirements ? The RSA reports indicate that for most of the 11 processes examined, requirements for maintaining and disposing of documents were adequate, documented, and being consistently applied.
- Records Accessibility ? Although there are inadequacies in agencies? recordkeeping systems, the RSA analysis shows that employees generally are able to find the records they need. The ability to retrieve records does not appear to be a serious problem in the 11 processes that the RSA teams examined.
Table 3-2. RSA Summary by Records Life Cycle
- Records Scheduling and Disposition
- Records Scheduling ? As shown in Table 3-2, the RSA reports indicate that many of the significant records produced by the 11 processes are unscheduled. Most of the electronic records are unscheduled and are therefore outside the formal recordkeeping system. In fact, the RSA teams identified unscheduled electronic records in all eight of the processes that generate significant electronic records. In addition to the unscheduled records, the RSA teams also identified several significant records that had been improperly scheduled. Clearly, records scheduling is a problem area.
- Permanent Records Transferred to NARA ? Table 3-2 shows that most of the paper and electronic records determined by the RSA teams to be of continuing research value are not being transferred to NARA. In most of these cases, the permanent records are not being transferred to NARA because they are unscheduled. However, for three of the processes, the RSA teams determined that records scheduled as permanent were not being appropriately transferred to NARA.
The RSA reports provide an excellent assessment of RM quality for 11 important Federal agency work processes. The RSA findings strongly suggest that Federal records scheduling and disposition practices need improvement. In addition, the variation in RM quality among the 11 agency processes detected by the RSA teams supports the study team?s concept of situational factors.
names of the agencies and
processes have been replaced
with generic labels to protect
the confidentiality of the
participating agencies and
5The cells in Table 3-2 associated with the question "Does Process Produce Records of Continuing Research Value?" are not shaded because whether or not a process produces records of continuing research value has more to do with the nature of the process than the quality of RM.
6The entry in Table 3-2 for "Protect Rights" for Agency Process 6 is not applicable (n/a) because the RSA team found that there were no citizen rights to protect in this process. Therefore, no documentation was needed for this purpose.
This section presents a comparison of the perceptual data collected by the SRA team and the factual records system data collected by the NARA RSAs and summarizes major findings and problems in RK practices.
In many respects the SRA study team concluded that the two data sets are more complementary in nature than contrasting. That is, the perceptual data tend to add context that helps explain the factual data observed in the RSA reports. In terms of records life cycle stage and key issues, most findings are consistent across both data groups with only a few exceptions.
Records creation. SRA has no basis for determining the adequacy of the documentation created by the 11 agencies for the processes selected for the study. The interview and survey data show that employees believe they create records appropriately, despite the fact that they generally receive no formal RM training and some respondents feel they do not know the definition of a record. As indicated in Table 3-2, the RSA reports confirmed that records relating to citizen rights, government accountability, and continuing research value are being created and adequately documented by Federal agencies.
- Electronic records. Both SRA and the RSA teams found many records are born electronically. Many agencies now have hybrid records systems, suggesting that they created paper files, electronic files, and documentation in other media. However, employees still mostly view the paper form as the official record format. The electronic records are often deleted after a paper copy is filed as the official record, or kept by employees at their desktops or in shared network folders. The SRA data, in particular, indicated that respondents feel very strongly about the lack of standard policies and procedures for managing, storing, and disposing of electronic records and systems.
- E-mail records. Both the SRA study team and the RSA reports indicated that managing e-mails has become a major RM problem. The agency employees tend to view e-mail as an informal means of communication. The study team uncovered widespread uncertainty about what constitutes a substantive e-mail, given that many e-mail communications contain both substantive and personal content.
Records maintenance. Some RSAs described records systems that were meeting NARA guidelines and operating successfully. Some others uncovered records systems that were inappropriately maintained, or did not meet NARA guidelines. The study team also found a similar range of records systems that meet or do not meet NARA guidelines. The RSA reports do not address why such differences exist among the agencies. For addressing the question of inter-agency differences, the study team's situational factors model may prove helpful to NARA and the agencies.
- Electronic records maintenance. Both teams uncovered the finding that records in electronic formats and systems were not appropriately managed. The RSA data indicated that most electronic records were not scheduled and are maintained outside the official realm of retention and disposition requirements.
- Ability to retrieve records. The SRA team found that employees believe they can usually find the records they need, particularly the records associated with their mission or core business processes, but sometimes it may require significant effort. Employees' general perception on this matter was strongly supported by the RSA results.
Records disposition. Although agencies often believe they follow NARA-approved schedules and properly dispose of their records, the RSA data indicate disposition of permanent records is poor across the agencies. The RSA teams discovered that some valuable permanent records are not appropriately transferred to NARA. In some cases, records with continuing research value, including both paper and electronic records, are not scheduled, and therefore are not sent to NARA. In other cases, the records retention schedules are flawed or outdated and may contribute to the loss of important documentation. In addition, the reports show that some agencies fail to transfer the permanent records to NARA according to the time line specified in the schedule or resist altogether the transfer of valuable permanent records to NARA. Both NARA and the agencies are aware that agencies are under legal obligation to accomplish the transfer and yet a few agencies believe they have good business reasons not to do so.
Influence of business processes on RK. Findings of the two data sources are consistent on the effects of work processes. The staff seem to know what records they should be keeping pertaining to their particular business processes. Recordkeeping for case files appears to be fairly good according to both RSA and the study team's reports.
Influence of process culture on RK. Both the RSA and SRA teams found that the work process culture influences the quality of RK. In particular, findings from several RSAs support the study team's finding that the greater the degree to which a process deals with legal matters and litigation the greater the likelihood that the agency staff involved in the process engage in sound RK practices.
Use of technological tools. Operations in most agencies are still primarily paper-based while a few have become very much IT-systems based. However, even though an agency's processing operations are heavily electronic, the information systems applications do not extend to electronic records management applications compliant with DoD 5015.2-STD, which is the government-wide standard. In general, the RSA reports show a considerable range between reliance on paper versus electronic records. The SRA perceptual data found this diversity as well.
Use of Internet and Intranet. Both teams found that the Internet, as well as agencies' Intranet, has become a tool for research and information dissemination. Some of the agencies are making electronic forms available on their Web sites and accepting requests online. For the most part, the two data groups showed that the agencies are using their Web sites to make information available to the public.
Records Use. The RSAs also support the study team's finding on records use in Federal agencies. Both the RSA and SRA teams found that employees occasionally need to use other agencies' records for their business. Also, both teams found that changes in records use patterns do not appear to have much effect on recordkeeping practices within the agencies.
Figure 4-1 portrays the records life cycle and the workflow of records from their creation, through maintenance and use, to disposition. The blue arrows depict normal or "correct" RK process flows. The broken red arrows and the boxes and ovals with broken red outlines depict major problem areas in today's Federal RK practices.
Figure 4-1. Records Life Cycle and Major Problem Areas
Reading the figure from left to right, the first set of problems occurs at the individual employee's desktop. Not having received proper training in RM, employees often do not know what is a record and what is not. Nor do they understand the importance of placing the record copy into an agency recordkeeping system. Consequently, some paper and many electronic records, if kept at all, tend to be kept at the employee's desktop, either in personal paper files or on the hard drive of a personal computer, and never proceed to agency recordkeeping systems.
Paper remains the primary preservation format for RK throughout the Federal government, despite the explosive growth in electronic records. In the Records Maintenance and Use section of the figure, scheduling of electronic records is often a problem in agencies. This difficulty is related to the fact that agencies lack electronic records management (ERM) systems, although many agencies say they are in the planning stages for ERM systems. As a result, searching through paper records for retrieval purposes, while generally successful, would require substantial effort. Searching electronic records is also difficult because many of the electronic records are not kept in ERM systems and are not filed in paper systems. E-mail is a mounting records problem because the volume of e-mail use for official business has increased exponentially and, again, e-mail is generally not captured in ERM systems. One problem in scheduling is that some agencies believe it takes too long for NARA to approve their schedules. The delay in schedule approval may discourage an agency from submitting SF115s to NARA in the first place, resulting in records with continuing research value that are not transferred to NARA when appropriate.
In the Records Disposition section of Figure 4-1, the primary problem is that a few agencies do not transfer permanent records to NARA. Again electronic records are part of the problem because agencies are failing to schedule these records. Paper records may also be a problem here because a few agencies do not want to lose control of these records by transferring them to NARA's ownership and custody or they have misscheduled permanently valuable paper records in the first place.
Overall, both the perceptual data and the factual records system data demonstrate that while some common themes are observable, the character and quality of RK vary greatly from agency to agency. The SRA study team concluded "one size does not fit all" when it comes to developing RM policy and practices for the Federal government. The results of the comparison analysis provide support for the study team's situational factors concept in that there are some significant differences in records management. The situational factors may serve as diagnostic tools to help the agencies and NARA identify the particular areas for improvement that are described in Figure 4-1.
How does one account for the great variability found in RK among the agencies participating in this study? In some agencies, RK is "job one" and in others RK is "number 26 on our list of top 25 priorities." The answers may lie in how RM is conceived and RK is implemented within the context of a particular agency. The SRA study team hypothesized that a range of "situational factors" drives the overall quality of the RM program in a particular Federal agency or component sub-organization within an agency. The relative strength or weakness of various factors will account to a substantial degree for the overall strength or weakness of the RM program in that agency or organization. This section describes these "situational factors" and presents a conceptual model indicating how these factors may help to explain or predict the quality of an organization's RM program.
The concept of situational factors is important for a number of reasons. First, to the extent that situational factors explain variation in quality of RM, they can help NARA in developing strategies to improve RM in a particular agency. Second, situational factors may serve as diagnostic tools to help NARA and an agency identify the particular areas where the agency's RM program requires assistance and where a given amount of assistance can make the greatest difference in program outcomes. As diagnostic tools, the factors can help set priorities for NARA resources across the government. And third, the situational factors model offers a strong foundation for understanding RM within the overall context of an agency's mission and culture and the business processes associated with the mission and culture. This should lead to strategies for not only improving Federal RM, but also making RM a function more integral to and supportive of agency missions.
From a detailed review of the perceptual data and the RSA reports, the SRA team identified a set of situational factors affecting the quality of Federal RM. Figure 5-1 displays the situational factors and summarizes the team's hypothesis about the relationship between these factors and the quality of an agency's RM program. Appendix B contains a table that indicates the situational factors that appear to be present in the agencies and the impact of these factors on RM for the agency processes. The SRA team stresses that the situational factors described below and the model shown in Figure 5-1 represent a set of hypotheses based on qualitative data gathered from a relatively small number of agencies. By necessity, the factors and model reflect a considerable amount of expert judgement. For these reasons additional analysis is needed to further develop and refine the situational factors.
As shown in Figure 5-1, the situational factors identified by the SRA team fall into four categories: (1) Institutional Context, (2) Policy and Guidance, (3) Resources, and (4) Other Factors.Figure 5-1. Situational Factors Model
Institutional Context. The first category of situational factors, Institutional Context contains three groups of interrelated factors dealing with the institutional motivation of an agency or organization to perform good RK, process-related and cultural attributes that appear to be indicative of strong or weak RM, and organizational leadership for RM. Among the factors shown in Figure 5-1, the SRA team believes the Institutional Context factors, particularly the Business Need, are the most important situational factors. These factors strongly influence the quality of the RM program and, as indicated by the arrows between the boxes, determine the degree to which all of the other factors are present. Descriptions of the Institutional Context factors are provided in Section 5.2.
Policy and Guidance. Policy and Guidance factors influence the entire records life cycle, since they reflect the degree to which agency leadership has enunciated and backed the RM program and the degree to which agency staff know what needs to be done in each phase and how to do it. A particularly important factor in this category is Adequate and Documented Internal RM Policies and RK Procedures, since they serve to make the NARA guidelines and policies operational within an agency or organization. The SRA team believes that RM policies and guidance are a product of the business needs of an agency, and are generally embodied in an agency's procedures manual for conducting its business processes. As indicated by the arrows flowing into the Policy and Guidance box from the Resources and Institutional Context boxes, the SRA team believes that an agency is more likely to have adequate and documented internal RM policies and RK procedures if it has sufficient RM resources, including an RK staff of adequate size and training and a dedicated Records Officer who is active in the RM profession and involved in agency business processes.
Timely and Responsive NARA Policy and Guidance is another important factor within this category. Agencies stated that they need to receive the benefit of NARA's expertise more quickly and responsively when new technologies such as the Internet begin transforming agency work processes.
Resources. Even if the Institutional Context factors are present, an agency will not be able to perform good RK unless it also has adequate resources. As indicated by the arrow flowing from the Institutional Context box into the Resources box, the SRA team hypothesizes that the level of resources devoted to RM in an organization is directly dependent on factors such as the business need, the nature of the process, and the level of leadership commitment to recordkeeping.
Other Factors. This category contains several situational factors that do not appear to fall in any of the above categories but that may influence the quality of the RM program. Specifically, the SRA team believes that frequent communication and interaction with NARA and regular participation in external professional RM activities (e.g., FIRM Council, ARMA chapter, etc.) is both cause and effect of higher quality RM programs. Those organizations with centrally coordinated management of records seem more likely to perform quality RK. Finally, the arrow flowing in from the Institutional Context box reflects the SRA team's hypothesis that an agency's or program's business need for records along with factors concerning the accessibility of records transferred to storage, proper scheduling, and concerns about storage space and cost explain the quality of records disposition practices.
5.2.1 Institutional Context Factors
As shown in Figure 5-1, the SRA team hypothesizes that there are a number of interrelated factors that define the Institutional Context in which RK is performed within an organization and help to explain the overall quality of the RM program. These Institutional Context factors fall into three groups dealing with the motivation of an agency or organization to perform good RK, process-related and cultural attributes that appear to be indicative of strong or weak RK, and organizational leadership for RM.
Business Need - The SRA team believes the primary determinant of agency RM practices and quality is business need. Specifically, agencies and their component organizations will perform that level of RK necessary to meet the needs of their business and perform those RK procedures that provide tangible business benefits. This factor was observed in virtually all 11 of the agencies for which RSAs were conducted (see Appendix B). If the nature of the agency's business demands good RK, the agency responds with good RK. If the agency is constantly and routinely called to account for its records and if the consequences of failure to answer the call are immediate and institutionally painful, the agency will practice good RK.
Threat of Litigation, FOIA Burdens, or Public Scrutiny - As general examples of strong business needs, agencies that are frequently sued tend to have good RM programs because the institution knows that it must be able to account rapidly and thoroughly for its past activities. Agencies receiving high volumes of Freedom of Information Act (FOIA) requests develop efficient mechanisms for retrieving records, which in turn leads to good, conscientious RM programs. Agencies under fairly constant Congressional, mass media, or public interest group scrutiny fall in this category as well. These agencies recognize that poor RM will lead to adverse legal judgments, unwanted legislation, and/or public embarrassment.
Process-Related Factors - The SRA team believes that a number of process-related factors may be useful for explaining or predicting variation in the quality of RM across agencies. Several of these process-related factors are interrelated or overlap.
- Work Processes Involving Case Files - RK appears to be better for work processes that involve the creation and maintenance of case files, as in regulatory agencies or Inspectors General offices. Case files represent well-defined records that are an essential element of many business processes. They provide a straightforward organizing principle for filing documents and serve as a repository for records. All employees know what records belong in a case file and know the consequences of not having the records in the file.
- More/Less Well-Defined Records - In general, the more well-defined the records produced by a given process or program are, the better the RK. As just mentioned, case files represent sets of well-defined records for which RK tends to be good, whereas office correspondence, e-mail communications, and web pages are less well-defined as records, and hence often problematic.
- Process Involving Litigation - RK also appears to be better for those processes involving litigation or even potential litigation.
- Age of Process or Program - The RSA results and comments from several interviewees appear to show that RK is better for those processes and programs that have been in existence for a long period of time and are stable and well-established as compared with recently established processes and programs. Older processes and programs have worked out all the kinks. Their records are well defined and procedures for managing them are established as opposed to the records for newer processes and programs.
- Structure of Process and Frequency of Performance - Another process-related factor appears to be the degree to which the process under consideration is structured and the frequency with which it is performed. If the process is relatively unstructured and infrequently performed, such as an intermittent planning process, then RK is likely to be worse than if the process is highly structured and frequently performed.
Culture - A complicated factor that appears to affect the overall RM success of a particular agency, and components within the agency is culture. Agencies and organizations in which employees strongly believe in the importance of recordkeeping to the mission of the agency or organization and, more importantly, see the link between their own jobs and the successful performance of the mission are more likely to perform good RK. These agencies and organizations can be said to have a culture that is conducive to good RM. A culture that is conducive to RM exhibits itself by staff indicating the importance and need for RK and contributing to the RK effort without cajoling or nagging. The study team hypothesizes that the culture of an agency or an organization is mainly dictated by its mission and business needs, but may also be influenced by leadership. Examples of cultures that are conducive to good recordkeeping appear to include the legal and audit cultures in which employees understand the need to perform good recordkeeping in order to substantiate their cases and audit findings, respectively.
Possible examples of weak recordkeeping cultures, based on the RSA findings, interview comments and survey data, include scientific, engineering, and information technology cultures. On the face of it, the study team could discover no convincing argument as to why these cultures should lead to a disregard for good RK and yet, within the limitations of the sample of agencies studied, the finding was consistent. One suggestion was that individual scientists tend to adopt a personal proprietary attitude toward their data and records, and by the same token, to resist including their records within the corporate RM program.
Leadership Commitment to RM - From the perceptual data gathered, the SRA study team concluded that agency leadership focuses primarily on carrying out the principal programs of the institution, and, all other things being equal, tends to view RM as primarily a non-mission related, administrative activity. However, interviewees at several agencies stated that a leader with a commitment to RM had a strong, positive influence on RK within their agency. The SRA team hypothesizes that those agencies with a commitment to RM among their leaders, including the agency head, General Counsel, Chief Information Officer, Inspector General and program directors, are more likely to have successful RM programs. In particular, the SRA team believes that agencies with leadership commitment to RM are more likely to provide adequate agency resources for RM, communicate the importance of RK to staff, integrate the RM function into program operations, and involve the Records Officer in more agency decision-making activities.
Quality Records Officer - The SRA study team believes that another important indicator of successful RM programs is the leadership of a full-time records officer who is knowledgeable about RM practices, actively involved in the profession, and committed to performing quality RK.
5.2.2 Policy and Guidance Factors
The second category of situational factors has to do with RM policy and guidance.
Adequate and Documented Internal RM Policies and RK Procedures - No matter what the policy and guidance provided by NARA, agencies must translate external guidance into effective internal RM policies and RK procedures to make them operational within their environments. Of the 11 processes and agencies that the RSA teams examined, the agencies that performed better RM appeared to have internal RM policies and RK procedures that the RSA teams described as adequate and that were well-documented.
Timely and Responsive NARA Policy and Guidance - Agencies need NARA RM policy and guidance concerning the new technologies and other factors affecting their work environments. Two significant examples the study team encountered frequently were RK for e-mail and Web sites, both quite recent and powerful phenomena that have altered how agencies carry out their work. Quick-response, interim, informal guidance from NARA, perhaps in the form of "current best practices," could do much to assist agencies in coping with new demands on their RM programs.
Awareness of RM Requirements - Finally, individual employees must have an awareness of the RM policies and procedures and consciously follow the procedures.
5.2.3 Resource Factors
Administrative/Financial Support for RM - The level of administrative/financial support for RM within an agency depends on the business need for good RK and the level of leadership commitment to RM. The study team hypothesizes that agencies that provide more administrative and financial support for RM are more successful in their RM programs. Administrative support includes having a sufficient number of RK staff and liaisons to develop adequate internal RM policies and RK procedures, prepare records schedules and file plans, and properly supervise filing and maintenance of records. Financial support includes funding for RM training and key RM initiatives such as the development of centrally coordinated records repositories and electronic records management systems.
Degree of RO/Liaison Involvement - The SRA team believes that the quality of recordkeeping is heavily influenced by the degree to which the Records Officer and RM staff are involved in their agency's principal business processes and programs. The higher their level of interaction with agency employees in other functions, the higher the quality of RM. Agencies where the staff know and are aware of a Records Officer or Records Liaison either in their office or elsewhere in the agency tended to have a higher commitment to RM and knowledge of RM policies and guidelines.
Planning, Piloting, and Implementing ERM - The data collected by both the RSA and SRA teams suggest that agencies that are seriously planning or piloting ERM systems perform better RK than those without such tools. These agencies are already performing better RK and they tend to invest in ERM systems because of the value they place on good RM. Comments recorded by both the RSA and SRA teams indicate that many agencies are either planning or piloting IT initiatives to support electronic records management, but their movement to electronic systems is constrained by the level of financial support provided for RM. Interestingly, another factor that appears to be influencing the rate at which agencies adopt electronic records systems is the degree to which state organizations they do business with are going electronic. Some states are more "cutting edge" than others are and this greatly affects Federal counterparts in their application of technology.
5.2.4 Other Situational Factors
Other Factors Influencing Records Maintenance and Use
Communication and Interaction with NARA - The SRA team believes that the level of communication and interaction between agencies and NARA influences the quality of agency RK, particularly in the area of records scheduling. The more agencies communicate and interact with NARA, the more likely they are to be aware of the requirement to schedule records and to understand how to schedule records. Those agencies that interact more with NARA may also be more likely to negotiate retention periods and disposition schedules for records that are consistent with their business needs, resulting in greater consistency between disposition practices and records schedules. On the other hand, the level of interaction between agencies and NARA on scheduling and other matters may reflect the agency's recognition of the importance of good RM. This factor is doubtless both cause and effect.
Centralized Coordination of Records - A factor that the SRA team believes may be indicative of better records maintenance is whether a business unit within an agency or subagency manages its records in a centrally coordinated manner. Of the agencies examined for the study, several agencies that centrally coordinate the management of their records appeared to have better control of their records and were able to locate them more easily than those agencies that said they manage their records in a decentralized manner. This improved records maintenance may be due in part to the fact that centrally coordinated RM may be associated with a dedicated RK staff to manage the records.
Other Factors Influencing Records Disposition
In addition to business need and leadership commitment, three other factors, Efficiency of Access to Retired Records, Proper Scheduling, and Resources for Records Storage seem to explain the quality of agency records disposition practices.
Efficiency of Access to Retired Records - Agencies are less likely to transfer records to NARA's FRCs or to commercial storage facilities if they feel that they will not be able to easily and quickly obtain access to the records in the event that they need them. A number of interviewees mentioned difficulty in getting temporary records back from FRCs in a timely manner when they needed them and so they were reluctant to send records to FRCs in the first place. A few others mentioned that gaining access to permanent records held by NARA is very cumbersome because agency employees have to go to NARA to view the records and cannot remove the original copies. One or two agencies even said they refused to transfer permanent records to NARA because they could not get the originals back again once NARA took ownership and custody.
Proper Scheduling - The degree to which agencies' disposition practices are consistent with their schedules appears to be related to the degree to which their schedules are consistent with their business needs for the records. Thus, if a records schedule requires an agency to transfer a record to NARA before it feels its immediate need for the record has ended, then the agency is less likely to comply with the schedule and transfer the record to NARA.
Resources for Records Storage - Finally, it appears that agency decisions about when and how many records to transfer to outside records storage facilities are influenced by the amount of storage space they have available on their own premises and the cost of storing them. If an agency has available low-cost internal storage space, then it has little incentive to transfer records to an outside storage facility. On the other hand, if an agency has very limited internally available storage space, then it may be anxious to transfer records to FRCs or turn to commercial outsourcing for temporary storage.
In this section a number of situational factors were described that may explain variations in quality of agency RM programs. The situational factors and the model represent a set of hypotheses based on qualitative data gathered from a relatively small number of agencies. Some of the factors or aspects of the model quite likely are improperly described and measured. Whatever the deficiencies of these particular situational factors, the concept of situational factors provides a useful diagnostic tool that NARA and the agencies can employ to identify those components and processes for which RM is likely to be strong or weak, and therefore where NARA and the agencies could best focus limited resources to improve Federal RM. Finally, as shown in Section 6, the situational factors model also provides a useful framework for the development of effective intervention options.
Section 3 presented a summary of the RSA findings that showed considerable variation in the quality of RM performance. Section 5 described a number of situational factors to explain variation in RM quality among agencies. This section introduces the concept of agency RM quality levels and presents some potential NARA intervention scenarios for improving RM that are linked to the quality levels.
Beginning with RM program quality levels, one may think of a set of performance levels ranging from an inadequate RM program to an excellent RM program. The SRA study team has chosen to define four RM "maturity" or quality levels to characterize the quality of RM performed by federal agencies and organizations: Inadequate, Adequate, Good, and Excellent. The characteristics of each level are based on the RSA findings presented in Section 3, perceptual data collected by the SRA team, and expert judgement. Others might propose different levels and characteristics that would serve the purpose just as well, so the study team makes no claim that these levels are in any way definitive or exhaustive. The levels are meant to serve for illustrative and discussion purposes only as a way of introducing potential intervention scenarios that could contribute to better RM programs. The RM quality levels are described as follows.
- Level I: Inadequate RM Program - "Help!"
- Records are not properly kept and routinely captured in agency recordkeeping systems.
- Employees often cannot find records when they need them.
- Records schedules and file plans are not kept current; most electronic records unscheduled.
- Records are not routinely retired, stored, and properly disposed of. Permanent records are not transferred to NARA.
- RM program is exclusively paper-based, despite significant volume of electronic files.
- Agency has no planning or implementation for ERM systems.
- Level II: Adequate RM Program - "B-/C+"
- Paper records are properly kept/captured in recordkeeping systems.
- Employees can usually find records when they need them, but sometimes requires considerable effort.
- Records schedules and file plans are generally current; some electronic records are unscheduled.
- Records are usually retired, stored, and properly disposed of according to approved schedules. There are gaps in NARA's permanent records holdings.
- RM program is mostly paper-based but beginning to incorporate electronic records.
- Where appropriate, agency is planning for ERM systems.
- Level III: Good RM Program - "Everything but ERM"
- Paper RM program well managed with effective capture and search capabilities, current schedules and file plans, and routine retiring, storage, and disposition. There are few gaps in NARA's permanent records holdings.
- Where appropriate, planning for ERM systems is well advanced.
- Initial pilots/experiments for ERM adoption underway.
- Planning for emerging issues such as e-mail and web site RM begun.
- Business process reengineering for ERM systems in planning.
- Level IV: Excellent RM Program - "Leave us Alone"
- Paper RM program well managed with effective capture and search capabilities, current schedules and file plans, and routine retiring, storage, and disposition. There are no gaps in NARA's permanent records holdings.
- RM training routine and mandatory for all employees.
- Where appropriate, ERM systems acquired and phased implementation underway.
- Initial BPRs completed and more underway.
- RM program addressing emerging issues such as e-mail and web site RM.
- RM increasingly integrated into all agency IT systems and business processes.
The SRA team hypothesizes that the RM quality level of an agency or organization is determined by various situational factors. Figure 6-1 on the next page graphically depicts this relationship, and shows four NARA intervention scenarios linked to the four RM quality levels. Appendix C contains a table that displays a mapping of the intervention scenarios and their components to the four RM quality levels.
These intervention scenarios, which are described below, represent possible approaches NARA might take to improve records management in the federal government. The scenarios imply outreach activities and programs that are more proactive than the traditional audit of agency records management programs represented by an RSA. The scenarios support the strategic goals contained in NARA's Strategic Plan, especially the fourth goal: "NARA's capabilities for making the changes necessary to realize our vision will continuously expand." They also are consistent with the values of risk-taking and communication enunciated in the Plan: "experiment, take chances, try new ways, learn from mistakes, be open to change", and "propose ideas, dialogue with others, develop trust."Each of the proposed scenarios can be expanded and made more detailed, and other scenario interventions could be developed.
Figure 6-1. Agency RM Levels and Potential Intervention Scenarios
An assumption underlying these scenarios is that NARA would conduct a review and perhaps have a discussion with agency staff to better understand their needs and requirements related to records management. During this review, NARA would consider the situational factors discussed in Section 5 as a means of determining the RM quality level and requirements regarding records management in that agency. Based on that review and consideration of situational factors, NARA could then select an appropriate intervention scenario such as those discussed below.
- Scenario I: Policy-maker and spokesperson
- Scenario II: Provider of technical assistance and education
- Scenario III: Situational factors change facilitator
- Scenario IV: Supporter of excellence in records management
Using situational factors and its understanding of agencies in this way will help NARA to focus its limited resources on those agencies, organizations, and processes within agencies where they are likely to be most effective.
NARA has always focused on the development of clear and complete policy and guidelines for how agencies can best manage records. Guidance issued by NARA remains a fundamental resource for agency RM programs. In this scenario, which applies to agencies in all four RM quality levels, NARA expands its policy leadership role in federal RM by initiating an executive education and briefing program. The need this scenario fulfills is one voiced by many agency records officers, namely, that their senior management needs a fundamental understanding of the importance and value of good RM to any agency's mission, over and above legal obligations and responsibilities. It also fulfills agencies" need for NARA to provide guidance on a more timely and responsive basis.
- Senior NARA officials, including the Archivist, Deputy Archivist, General Counsel, Assistant Archivists, and heads of major programs, offer their services in direct person-to-person briefings with senior officials in key agencies.
- As new officials are appointed in agencies - Assistant Secretaries, Chief Information Officers, and the like - NARA contacts them and offers to brief them on how RM responsibilities impact on their jobs.
- General Counsels, particularly in agencies subject to many FOIA requests and litigation, can become important intra-agency advocates for strengthening RM programs.
- The same group of senior NARA managers makes a concerted effort to reach key audiences outside the traditional RM community by making themselves available as conference speakers.
- The Office of Personnel Management has its Federal Executive Institute and two management development centers in Shepherdstown, WV, and Denver, CO, respectively. NARA managers could inform OPM officials that senior NARA officials are available as outside speakers for these executive training programs. The USDA Graduate School program is another venue NARA might use more actively.
- The wider federal information resources management community has many forums that could benefit from hearing NARA senior managers.
- GSA has its annual Interagency Resources Management Conference (IRMCO).
- The Federation of Government Information Processing Councils (FGIPC), its regional chapters, and its Industry Advisory Council hold many conferences where federal information technology managers come together.
- Associations such as the Armed Forces Communications and Electronics Association (AFCEA) hold conferences, breakfasts, and other meetings where NARA's message could bear fruit.
- Conferences such as E-Government and CIO Summit are other venues for RM outreach.
- NARA senior management especially takes the opportunity to offer guidance and services to heads of other agencies suddenly in the headlines because of political events. Recent examples include the FBI and the Timothy McVeigh records, and the Department of the Interior, Bureau of Indian Affairs and the Indian Trust Fund records.
- NARA, in consultation with agencies, develops timely, simple and practical guidance in easy-to-understand language, with specific examples.
In support of this scenario, NARA might review its own web site to ensure that it provides Federal employees easy access to the full range of Federal and NARA records management policy and guidelines.
During its collection of perceptual information, the SRA team received a number of favorable comments from agency employees concerning NARA's Targeted Assistance initiative. Under this scenario, which is primarily aimed at Level I RM programs, NARA expands its Targeted Assistance initiative offerings to design, test, and implement a range of technical approaches, including hardware, software, training, and interactive network support services, in order to assist agencies" RM programs. Whereas the current Targeted Assistance primarily responds to agency requests, the expanded program could begin to take on the characteristics of a change management laboratory for helping agencies to pilot new and even experimental approaches to contemporary RM and for gathering accumulated agency experience into a knowledge base of lessons learned. If desired, NARA could formalize a new program initiative with a title such as Center for Excellence in Records Management.
Among the expanded activities could be the following.
- NARA periodically surveys agencies to ascertain which agencies would like assistance with their RM programs and what forms of technical assistance the agencies prefer.
- NARA undertakes systematic development of technology-based training tools. These tools could be videos or computer-based training. They could range from basic RM concepts and operations for end users to more detailed technical training for records officers and records liaison officers. NARA could also work with vendors of training products and materials intended for the federal market on an advisory basis, providing quality assurance that the content conforms to applicable federal laws and regulations.
- NARA develops a menu of technical assistance services, reimbursable if necessary, to offer to agencies.
- Standards (including boilerplate contract language) for outsourcing various RM functions such as inventories, physical storage, etc. to ensure that outsourcing conforms to applicable laws and regulations.
- Tailor-made training programs for various agency personnel
- RM briefings for managers and senior agency officials
- Consultation on best practices in RM including "cutting edge" issues such as web site RM, e-mail RM, etc.
- NARA works with agencies, individually, to offer them a "customized" program of technical assistance and education that best meets the needs of the agency.
In the development of the strategies, the agencies must be active partners with NARA. With such a partnership, these technical strategies might reach beyond RM, as strictly conceived, to address questions such as how RM programs and systems can and should be integrated with other agency programs and systems.
The study team identified a set of situational factors that, taken in various combinations, help to explain why some agencies perform well at RM and others do not. This scenario, which applies to agencies at RM maturity Levels II and III, assumes that the agency is willing to work directly with NARA to identify how the situational factors can be modified for that particular agency to best improve RM.
- NARA can develop the situational factors model into a set of diagnostic tools applicable within an agency context. This should enable NARA personnel, working within the Targeted Assistance initiative, to conduct a structured interview in an agency and determine which situational factors need changing the most and the best points of entry for a change management strategy.
- For each situational factor on which an agency may show a poor profile, NARA can develop a set of ready-made strategies and tactics for assisting the agency to attack the situation with a change management approach. For example,
- In some situations, the preferred strategy may be for NARA to recommend an executive briefing by senior NARA officials, as sketched above.
- In other situations, the preferred strategy may be to encourage or help the agency to develop an integrated electronic document management/electronic RM system.
- In still other situations, the appropriate strategy may be to provide direct assistance in the development of records schedules.
- NARA, in concert with agency personnel, then develops a plan by which the agency can modify various situational factors that will improve overall mission performance and RM in that particular agency.
Application of situational factors diagnostics will enable NARA to identify RM problem areas within agencies and design customized programs to assist individual agencies to better manage their records. Thus, NARA would need to organize itself in such a way that there is a "situational factors task force" that could be prepared to work with individual agencies, assist them in developing a plan such as outlined above, and provide ongoing support and training to facilitate successful implementation of the plan.
In this scenario, NARA's role is largely to acknowledge and support those agencies at RM maturity Level IV that are engaged in high quality RM. In such instances, NARA does not need to develop intervention strategies. Rather, the role of NARA is to publicly acknowledge and praise the RM being done at that agency. In this scenario, the agency is acknowledged as a "good example" that others would want to emulate.
- NARA can hold out this kind of agency as an example to others and urge the agency to share its ideas and experience with others.
- NARA could consider expanding its annual Records Administration Conference (RACO) awards program for agencies that have had notable successes in various aspects of their RM programs during the previous 12 months.
This report offers a number of approaches, options, and interventions that may assist NARA and the agencies to improve federal RM. A key theme throughout the report is the desirability of taking into account the situational factors affecting RM from the agency perspective. Identifying and understanding these situational factors offers NARA and the agencies a range of entry points for change management tactics within individual agencies.
When one compares the RSAs conducted by NARA to the findings from the study team's collection of perceptual data, it becomes apparent that both approaches are necessary to describe and understand the RM environment in a particular agency. In effect, there are two key perspectives to consider when reviewing RM in an agency: (1) the records systems analysis related to records creation, scheduling, storage, disposition, and other issues, and (2) situational factors in the agency affecting the manner in which RM can be accomplished - and how well it can be accomplished.
The situational factors described in the report relate to the success of an agency in conducting its RM program. Although the study team has refined the situational factors somewhat since they were introduced in the original submission to NARA on August 6, 2001, the factors require far more study and conceptual clarification both within an individual agency and as to how the effects of their interactions vary from agency to agency. In this chapter, the study team proposes options that NARA and the agencies could follow to determine the degree to which the situational factors are present and how they impact on RM programs.
The options offered in this report also suggest a proactive role for NARA in working with individual agencies. But given the availability of resources to support an effort such as that described here, NARA may wish to carefully consider which agencies should be selected for assessment. The situational factors at some agencies may limit the likelihood that specific, or any, intervention options can improve RM.
The review of data from the Information Summary and Analysis Report indicates that some agencies conduct better RM than others. In Section 6, the study team defined four general levels of agency performance related to RM. For each of these levels the study team proposed a possible intervention strategy. The intervention options described in that Section are illustrative of possible intervention options. With additional research and field testing of the situational factors and intervention strategies other options may also be available.
Thus, it is important to stress that the study team has pushed the data from the original study as far as possible in identifying the situational factors, proposing agency RM performance levels, and offering possible intervention options for those performance levels. Additional development of the situational factors, performance levels, and intervention options requires the following actions.
- Refinement of the existing approach with structured interviews and focus groups with records managers and others knowledgeable about RM.
- Revision of the approach as defined in this report based on those interviews and focus groups.
- Field testing of a revised approach to situational factors, RM program performance levels, and intervention options.
- Final description of the approach and relationships among situational factors, performance levels, and intervention options.
- Detailed scripts and tools for NARA and agencies to use during a review of a particular agency to best determine the best intervention strategies for use at that agency.
The above steps, if implemented, could considerably increase the value and practical application of the findings presented in this report.
Although the approach described here needs additional field-testing and validation, the study team believes that it already provides NARA and the agencies with a very useful tool for improving RM programs in a flexible and pragmatic manner. The approach offers one more way in which NARA can move forward with the its strategic plan to "work in active partnership with Federal officials, the Congress, the courts, and others to assure ourselves that essential evidence is created, identified, maintained, and appropriately scheduled for as long as needed." With this approach, NARA will have additional strategies to work with Federal agencies, develop specific approaches that may work best in that agency, and ultimately improve the overall quality of RM in the Federal government.
INFORMATION COLLECTION TOPICS
|A. Records Creation, Maintenance, and Disposition Practices In Federal Agencies|
NARA and SRA
NARA and SRA
NARA and SRA
NARA and SRA
|B. Records Use in Federal Agencies|
NARA and SRA
NARA and SRA
Table B-1 in this Appendix indicates the agencies in which the situational factors appear to be present and their impact on RM for the process analyzed by the RSA team. In this table, a "+" indicates that the factor appears to be present and having a positive impact on the RM for the process under consideration. A "-" entry indicates that the factor appears to be present and having a negative impact on RM for the process. An entry of "+/-" indicates that the factor was present and appeared to have a mixed impact on RM for the process. Finally, an entry of "O" indicates that the factor appeared to be present at the agency, but appeared in an organization and a process other than the organization and process analyzed by the RSA team.
Appendix C: Mapping of Intervention
Scenarios to Agency RM Levels