Cassandra A. Schmitt, Archivist (Collections Management), University of Oregon

Statement of Problem

There is no shortage of literature discussing processing backlogs in United States repositories and the need for a shift in approach.¹ With many institutions currently facing decreased support and funding it becomes essential for collection management practices to be as efficient and productive as possible.

This article will highlight the experiences of Special Collections and University Archives (SCUA) at the University of Oregon Libraries in uncovering hidden collections by utilizing collection level records and supporting this work with a grant by the National Historical Publications and Records Commission (NHPRC). Over the past several years we have undertaken a major effort to assess our backlog, realign technical service priorities, and create efficient policies, procedures, and workflows.

Department background

SCUA is comprised of manuscript collections, historic photographs, rare books, and university archives. Our mission is to provide public access to records that document the university as well as defined collecting areas.² Our collecting strengths include Oregon and Pacific Northwest politics and history, transportation, power, and mining in the West, natural resources and forestry, journalism, conservative and libertarian movements, intentional communities, missionaries, and literary collections including writers of western fiction, science fiction, and illustrated children’s literature.

SCUA holds an estimated 45,000 linear feet of materials comprising over 3,000 collections. We accession an average of 700 linear feet of materials each year. Staffing includes 6.75 FTE faculty positions, 1.25 FTE staff, and 1 FTE grant funded position. Our student, intern, and volunteer work force are essential for daily operation and estimate between 4 to 6 FTE throughout the year.

Hidden Collections

As of the summer of 2010 only 20% of our collections were discoverable online either in the form of a MARC record in the library catalog or an EAD finding aid in Northwest Digital Archives (NWDA). Thus, the majority of our collections were “hidden” from users since the collections had no online access points. Even staff did not know about most of the collections content and could not inform users of relevant collections during a reference interaction.

University archives collections were largely hidden from staff and users because 97% of collections were never processed or received any attention since initial accessioning, if they had been accessioned at all. Our accession records for the majority of materials only contained an accession number, tentative title, and the number of containers. Many accession records were labeled incorrectly, included inaccurate descriptions, and still represented a mix of nonpermanent and permanent records. Since no further arrangement or description occurred after materials arrived, it was difficult for staff to navigate and make sense of the provenance and context of university records.

Manuscript collections fared slightly better as 26% of collections were described in some way online.³ While many of collections were processed by archivists years ago, they remained hidden and were rarely (or never) discovered or used as they had no online presence. The majority of these collections had detailed (at times item-level) paper finding aids available within the reading room or only consisted of a single item or folder. Paper finding aids are helpful to both users and staff, but do not solve the problem of online access and discovery.

Previously, many of these collections were considered “processed” as archivists had provided various levels of arrangement and description. In 2010, we began to revise our definitions of processed and unprocessed collections. In SCUA, an “unprocessed” collection is now a collection that has no online access point, regardless of previous arrangement and description work, and therefore remains “hidden” from our users.

This was a big turning point for SCUA, because it enabled us to envision how to harness the arrangement and description work of our predecessors, utilize current technology and tools, and complete a high volume of foundational work with minimal resources.

Collection Level Records Solution

In conjunction with an overhaul of our technical services functions we decided that collection level records offered an opportunity to meet our responsibility to users in providing access to collections.⁴While in a perfect world, we would want finding aids with detailed container lists; this was no longer feasible on our shrinking budgets (and really, was never feasible in the first place.)

Collection level records allow us to provide detailed and concrete descriptions of our collections, alert users to further resources (such as a paper finding aid in the reading room), utilize our resources wisely, and help move the department toward making user-driven processing decisions. For over 1,000 collections under one linear foot (about 33% of manuscript collections) no additional arrangement and description work will ever be necessary for adequate access once collection level records are completed.

Grant Funding

To assist in our efforts we wrote and were awarded a NHPRC basic processing grant specifically to target hidden collections. The project runs from October 2011-September 2013. The main goals and products are:

• Creating at least 1,128 EAD finding aids in Northwest Digital Archives (NWDA)
• Creating at least 1,128 MARC records in the library catalog
• Eliminate backlog of university archives collections
• Significantly reduce backlog of manuscript collections
• Reappraise collections to determine fit in collecting policy resulting in list of possible collections to deaccession

Staffing for the project came from a combination of cost-share and grant funds including the following positions:

• Archivist for Collections Management (project manager; 50%)
• Project Archivist (grant funded, 100%)
• Head of SCUA (5%)
• University Archivist (5%; vacant thus far)
• Electronic Records Archivist/Records Manager (5%)
• Manuscripts Librarian (5%)
• 2 catalogers (12.5% combined)

Implementation

Planning

Over the two-year period before the grant began we made several changes to department policies and procedures to lay groundwork that resulted in a dramatic increase in productivity and efficiency of our technical services functions. Changes included revising (and when necessary creating) collection development policies, employing accessioning as processing techniques, dramatically updating our processing policies, procedures, and practices, and implementing the Archivists’ Toolkit (AT) for our collection management and description needs.⁵

Reappraisal

All collections included in the grant (as well as collections for non-grant work) now go through a brief reappraisal process to help prioritize collections. Curators evaluate what we know about a collection, utilize our collection development policies, records retention schedule, and their knowledge of similar collections to make a reappraisal recommendation for each collection before other work can proceed.⁶ Performing this reappraisal step before any additional arrangement and description work means we spend our resources on collections that fit our collecting policy and prioritize work on collections with a higher demand or research value.⁷

Automation, Workflow, and Production

Once a collection has “passed” reappraisal it is available for staff (or students) to construct a collection record in AT following detailed, specific workflows and procedures. We gather information from accessions and resource records in AT (imported from previous internal databases), donor files, processing files, paper finding aids, and when necessary the records themselves.

We automate the process as much as possible by pre-populating standard AT fields and constructed default text for selected finding aid notes.8 This allows us to quickly add information to a particular record, while still offering an opportunity to edit, remove, or add anything specific for a collection.

We concentrate on writing a concise abstract and a robust scope and content note. Notes describe the current arrangement of materials (or lack thereof) as well as descriptive information about the form, content, dates, relevant people, places, and subjects of the materials. When possible we provide detailed series descriptions by utilizing paper finding aids and/or accessions lists. In this way we harness the work of previous archivists while providing the user a more than adequate overview of a collection.

For collections with multiple accessions, we describe each accession discreetly in the scope and content note. This includes collections whose initial donations were processed as a whole, but have since accumulated accretions. Previously, these accretions remained in the backlog until materials could be intellectually and physically integrated, locally referred to as “reprocessing”. This was time consuming, expensive, and unsustainable as it directly contributed to a growing backlog. Our new strategy allows staff flexibility to easily add new accretions (during accessioning or for backlog collections) to a collection’s description without revising previous arrangement. This saves significant time for collections that are large, consist of numerous accessions, or have never been described as a whole.

We have constructed additional text for arrangement and processing notes for collections that have various levels of physical and intellectual arrangement or a complicated processing history. Collections may have received little to no arrangement, have been minimally arranged, be arranged by accession, or have an initial collection with accretions added later on as series. We believe this will help the user understand not only the provenance and custodial history of the records, but more insight into the physical and intellectual arrangement (or lack thereof) of materials.⁹

We strive to include a related materials note for every collection. Previously, this note was only included if a related collection already had an NWDA finding aid. This was a limiting practice. Now we include the collection name and call number ensuring that a) related collections without finding aids are included and b) we do not have to update multiple finding aids when a related collections’ finding aid is uploaded.

Contrary to previous practice, we now spend the least amount of time constructing biographical and historical notes. Notes are short and stick to concrete facts rather than providing interpretation. Fuller biographical details may be found in paper finding aids or other sources. While this had been a large focus in the past, we choose to spend our time describing the records in the notes mentioned above.¹⁰

When draft records are complete they are reviewed to ensure all required fields are correct and any questions are resolved. Catalogers from our Collection Services department assist in performing authority control within AT for names and subjects.

After authority control is complete marc.xml records are exported from AT. Catalogers use a VBR script to convert the files to marc.amc. The script converts the file type and makes global changes to the AT output to fit local practices. Records are loaded into Connexion and reviewed before being uploaded into the UO local catalog and WorldCat.

After MARC export, staff add an NWDA browsing term (a required element in our best practices) to the AT record and export EAD files. These files are then converted using the NWDA Archivists’ Toolkit EAD Converter tool to align the files with NWDA best practices and guidelines.¹¹ Finding aids are then uploaded to NWDA and available to users.

Outreach Efforts

Producing collection level records at a high level allows our staff to easily perform outreach in new ways at an increased level.

Our first step was to create a blog for SCUA: “News and New Collections”.¹² A large component of the blog focuses on grant work. Every finding aid we upload to NWDA gets a short blog post containing the collection name, call number, extent, abstract, a link to the finding aid, and a tag to facilitate searching similar collections. For our users, this is just another access point to aid in the discovery process. The blog’s internal purposes are: a) making sure our curators and other staff who perform reference know more about our collections and b) using the blog as an outreach tool by curators.

Curators can repurpose blog information to connect with various user groups. Information can be used to add to current subject guides on various topics as well as create new guides for emerging areas. For example, through our grant collection level work we discovered a need for the creation of an Oregon Politics subject guide. While staff knew of several of these collections, people were amazed at the discovery of how many additional, previously “hidden”, collections SCUA holds in this area.

Conclusion and Future Projects

Institutions should consider using collection level records as a tool to uncover hidden collections. We have dramatically increased our finding aid production. Users are extremely pleased that more collections are described online and we are already seeing an increase in reference service.

Beyond the scope of this article, we are harnessing our success with collection level records and developing a plan to eliminate our backlog within the next two to four years. This includes detailed planning, keeping processing metrics, making user-driven processing decisions, and responsibly ensuring that access for our users is our main technical service priority.

¹ Mark A. Greene and Dennis Meissner, “More Product, Less Process: Revamping Traditional Archival Processing,” American Archivist, 68, no. 2 (Fall/Winter 2005): 208–263.

² The full mission statement is available at: http://libweb.uoregon.edu/speccoll/staff.html 

³ 18% of manuscript collections had both an EAD finding aid online and a MARC record in the library catalog. 8% of manuscript collections had only a MARC record. Ideally, all collections will have both access points, but in the mean time we consider collections with only a MARC record as processed and available online.

⁴ We articulate an archivists’ responsibility to create access for users through Tom Hyry’s six principles. See Tom Hyry, “More for Less in Archives: The Greene/Meissner Approach at Work at Yale” (paper presented at RLG Member Forum: More, Better, Faster, Cheaper: The Economics of Descriptive Practice, Washington, DC, August 8, 2006); Tom Hyry, “Reaccessing Backlogs,”Library Journal, Vol. 132 (Spring 2007): 8-9.

⁵ For information on accessioning as processing see Christine Weideman, “Accessioning as Processing,” American Archivist, 69, no. 2 (Fall/Winter 2006): 274-283; Donna E. McCrea, “Getting More for Less: Testing a NewProcessing Model at the University of Montana,”American Archivist, 69, no. 2 (Fall/Winter 2006): 284-290.For information on UO’s implementation of Archivists’ Toolkit see Cassandra A. Schmitt, “The Archivists’ Toolkit as a Collaborative Tool Between Primary Source Repositories,” (Presented at the Online Northwest Conference,Corvallis, OR, February 2011); Elizabeth A. Nielsen and Cassandra A. Schmitt, “A Joint Instance of the Archivists’Toolkit as a Tool for Collaboration,” (Presented at the annual meeting of the Society of American Archivists’, Chicago, IL, August 2011).

⁶ We have developed a worksheet for curators to use to assist in making reappraisal decisions. For “yellow” and “red” collections, curators fill out and save the entire worksheet to assist in further reappraisal and deaccessioning work in the future. This worksheet was developed with the draft of the Society of American Archivists’ “Guidelines to Reappraisal and Deaccessioning” (July 2011) in mind.

⁷ For thoughts on reappraisal see Mark A. Greene, “MPLP: It’s Not Just for Processing Anymore,” American Archivist, 73, no. 1 (Spring/Summer 2010): 175-203.

⁸ Set defaults for Archivists’ Toolkit fields include: level, language, extent type, description rules, language of finding aid, sponsor note (since most new records are part of grant), and finding aid status. Notes set as default text (or partial default text) include: conditions governing access, conditions governing use, immediate source of acquisition, preferred citation, processing, arrangement, accruals, other finding aid, and related materials.

⁹ Additional note text was inspired and built off those used in baseline-finding aids at the Beinecke Library. We have customized for our specific uses. See Jennifer Meehan, “Baseline Finding Aids @ the Beinecke: Reflections on the Evolving Nature and Role of Finding Aids in Light of Different Levels of Processing” (paper presented at the annual meeting for the Society of American Archivists, San Francisco, CA, August 28, 2008).

¹⁰ For information on researchers’ use of metadata elements in finding aids see Joyce Celeste Chapman, “User Feedback and Cost/Value Analysis of Metadata Creation” (paper presented at annual meeting for Description Section Society of American Archivists, Washington, DC, August 13, 2010).

¹¹ NWDA-AT EAD converter was jointly developed by the University of Oregon and Oregon State University. NWDA now makes the tool available to its members and develops and maintains the tool.

¹² SCUA’s “News and New Collections” blog available at: http://libweb.uoregon.edu/tools/blogs/scua