What Happens After Users Find Records?The role of archival description in users’ interpretation of records.

(1)

Records?

The role of archival description in users ’ interpretation of records.

Name: Stephanie Hall Student number: 12431699 Email: s_a_hall@outlook.com

Date: 31/01/2020

Master’s thesis Archival and Information Studies Supervisor: dr. ir. Jaap Kamps

(2)

Abstract

To use records, users often need to find them. Finding records and users’ interaction with archival description while they search for them is thus an important theme in user studies. What, though, does archival description mean for what users can do with records once they find them? This question is not usually addressed in user studies and can help to bridge the gap between theory on archival description and the studies of users’ interaction with archival description. Therefore, this research asked: how does archival description affect the way in which users interpret records? In order to answer this, a qualitative quasi-experiment was conducted with four participants. It found that users can manage to interpret records without viewing archival description; they can do so by relating the content of the records to the records’ wider contexts. Whilst archival description is not necessary for users to interpret records, the user study found that it can affect the interpretation of records. Its influence in the study was diffuse, as it interacted with other factors. Depending on its interaction with those elements, archival description had a range of possible effects on interpretation. These effects varied between the participants but also for the level of the archival description they discussed. Archival description sometimes confused participants, whereas at other points it reinforced their views of the records. Its effects also included clarifying and contextualising records. The most important factor for the influence of archival description on the interpretation of records was the individual who viewed them. Given this, further research on other users and their interaction with different types of records and archival description is needed.

(3)

1. Introduction

A user’s ability to find records and to use the archival tools and systems to find those records is well researched in user studies of archival users.1_{This research is important as finding records is often a}

prerequisite for using them. It does not constitute use though. Users’ use of records once they find them and the role of archival description in their use of records have rarely been addressed in user studies. This gap is all the more striking as theory on archival description accords it an important role in users’ understanding and interpretation of records.2_{The idea is one that some discussions of users}

and user studies have repeated without investigating it in practice.3_{This thesis seeks to contribute to}

lessening the gap between the role archival description is thought to have in archival theory and the roles of archival description that user studies address. It asks: how does archival description affect the way in which users interpret records?

This thesis uses a user study in order to answer its research question; this user study and the thesis as a whole are guided by a series of sub-questions. The first is: which arguments regarding the effect of archival description on users’ interpretation of records are given in the literature on archival theory and user studies? The second asks: which type of user study allows for the collection of data on users’ interactions with records and archival description as well as on their interpretation of records? The user study this question seeks to develop allows for the gathering of data on several other sub-questions. The third sub-question addresses the interpretation of records by users who do not have access to archival description. It asks: how do users interpret records without archival description? Its counterpart, the fourth, specifically considers the role of archival description in the interpretation of records by asking: in which ways does archival description in an inventory affect how users can interpret records? The user study reuses protocol from an earlier user study, the MA thesis Eigentijdse Toegankelijkheid: Rekening houden met de competenties van archiefgebruikers (Contemporary Accessibility: Taking archival users’ competencies into account) by Julia Romijn-Wixley and Aron de Vries.4_{As a result, the fifth sub-question directly references Eigentijdse Toegankelijkheid and asks:}

how do the participants’ responses to a record and its corresponding archival description compare to the results from the post-experiment on interpretation in Eigentijdse Toegankelijkheid? The sixth and final sub-question seeks to reflect upon the results of the user study by asking: how do the results on

1_{Wendy M. Duff, “User studies in archives,” in User Studies voor Digital Development, eds. Pierlugi Felicitati,} Andy O’Dwyer and Milena Dobreva (London: Facet publishing, 2012), 199, 202.

2_{Heather MacNeil, “Picking our text: Archival Description, Authenticity, and the Archivist as Editor,” American} Archivist 68, No.2 (Fall-Winter 2005): 264 ; Geoffrey Yeo, “Continuing Debates about Description,” in Currents of Archival Thinking, eds. Heather MacNeil and Terry Eastwood, 2nd_{ed (Westport: ABC-CLIO, LLC, 2017), 163.} 3_{Mary Jo Pugh, Providing Reference Services for Archives & Manuscripts (Chicago: Society of American} Archivists, 2005), 84; Wendy Duff and Elizabeth Yakel, “Archival Interaction,” in Currents of Archival Thinking, eds. Heather MacNeil and Terry Eastwood, 2nd ed (Westport: ABC-CLIO, LLC, 2017), 207.

4_{Julia Romijn-Wixley and Aron de Vries, “Eigentijdse toegankelijkheid. Rekening houden met de competenties} van archiefgebruikers” MA Thesis, University of Amsterdam, 2009, 25-26, 30.

(7)

the ways in which individuals interpret records without description and on the ways in which an inventory affects their interpretation of records relate to arguments made in the literature? The order given for the sub questions above parallels the order in which they are answered in the chapters of this thesis.

The research question and its constituents, which were outlined above, feature several concepts whose meaning is either heavily debated or often left undefined. The concepts archival description and records are central to archival theory and the debates over each of the concepts could fill a thesis alone. Rather than wade into the debate, this thesis will use working definitions for them. The first, archival description, will be defined as both the process of representing the content, context and structure of (an aggregation of) records and what is produced during this process.5_{The latter, the}

products, are what user studies focus on as they are what users interact with. However, the concept is not restricted to the products in order to capture a wide range of perspectives on how description affects interpretation. The user study conducted for this thesis uses one type of description and that narrower application of the concept is further explained during the discussion of the study. The second concept, a record or records, is broadly defined as documents or information objects linked to a(n) (trans)action or process.6_{The link between a record and an action or process is often essential}

when it comes to considering the interpretation of that record. Strictly defining a record in practice can be more slippery though. For instance, Barbara Reed emphasises that a record can be a single item but that it can also be an aggregation of other records, such as a file.7_{The record users find under an}

item-level description can therefore consist of multiple records while also being a record. In this case, the relationships between its constituents need to be considered when looking at users’ interpretation of that record. As with the broad, working definition of description, that for records is narrowed by the practical constraints of the user study as it only considers records held by archival institutions. Unlike archival description and records, interpretation is a term which tends to be left undefined in writing on archival theory. In addition, it is frequently used interchangeably with understanding. This study will specifically consider the interpretation of records rather than interpretation as a broader concept. A working definition of the interpretation of records will be developed in the literature chapter and returned to throughout the thesis.

The majority of this thesis focuses on a small-scale user study which seeks to investigate whether the role archival theory proposes regarding description’s role in interpretation is borne out by individuals’

5_{Jennifer Meehan, “Arrangement and description: between theory and practice,” In Archives and}

Recordkeeping: theory into practice, ed. Caroline Brown (London: Facet Publishing, 2014), 85; ISAD (G): General International Standard Archival Description, 2nd_{ed. (Ottawa: International Council on Archives, 2000), 0.1.} 6_{Babara Reed, “Records,” in Archives: Recordkeeping in Society, eds. Sue McKemmish, Michael Piggot, Barbara} Reed and Frank Upward (Wagga Wagga: Centre for Information Studies, 2005), 106-107; Theo Thomassen, “A First Introduction to Archival Science,” Archival Science 1 (2001): 374.

(8)

actual interaction with archival description and records. This user study forms the basis for the answers to sub-questions three to five. Its development, in response to the second sub-question, is based on examples of similar studies and methodological approaches in the literature. Another sub-question which contributes, more indirectly, to the development of the user study is the first one. This uses a literature review to identify ideas and concepts around the role of archival description in users’ interpretations of records. Those concepts are returned to in the seventh sub-question, where they are used to reflect on the results of the user study. Hence, while the user study is at the centre of this thesis, it and its results are contextualised by insights and arguments from the literature.

During the outline of the sub-questions above it was noted that their order parallels their order in this thesis. It opens with the literature review and is followed by a discussion of the methodology used. Following on from this, there are three chapters which discuss the results of individual sections of the user study: users’ interpretation of records without archival description, users’ interpretation of records with archival description and a comparison with the results of Eigentijdse Toegankelijkheid. As it draws on an earlier study, the final results chapter acts as a bridge to the discussion chapter; here the key conclusions of the first two results chapters are contextualised using the concepts and theory identified in the second chapter. Finally, the conclusion seeks to reflect on what the interpretation of records is as well as to discuss how description plays a role in users’ interpretation of records. Hence, it returns to the wider question underlying this thesis regarding what users can do with records once they find them and how archival description affects their interpretation of those records.

(9)

2. Literature review

2.1 Introduction

Debates on the role of description in users’ interpretation of records are fragmented between archival theory and user studies. Its role comes up obliquely in the theoretical debates on description and records through an emphasis on the importance of context for the understanding of records. Yet what this means for users is not routinely addressed. User studies are the opposite, highlighting users’ needs and interaction with description. However, these studies fail to substantially address users’ interaction with records themselves. These two contrasting debates, theory regarding description and studies of users, need to be combined to consider the effect description may have on users’ interpretation of records. In addition, due to the importance of the user perspective for this thesis, user studies will be discussed more widely. The review addresses the question: which arguments regarding the effect of archival description on users’ interpretation of records are given in the literature on archival theory and user studies? For the purposes of identifying relevant literature, the interpretation of records was initially kept as broad as possible; for instance, it included mentions of understanding of records. The final part of this literature review will reflect on the literature discussed and some additional sources to develop a working definition of the interpretation of records.

2.2 Interpretation, description and archival theory

Interpretation and the understanding of records or aggregations of records do not feature prominently in debates on archival description and theory. Instead, the idea that description aids understanding or interpretation is considered self-evident. The presumed link between archival description and the interpretation or understanding of records features in a range of archival traditions. Writing on archival description, specifically the description of context, and the nature of records contain arguments and assumptions on the role of description in the interpretation of records. These themes are all subject to extensive debate, especially around what should be described. Rather than address these debates on descriptive standards in detail, this literature chapter mainly focuses on theoretical approaches to description.

A range of definitions of archival description and context (which archival description is thought to provide) refer to their role in users’ interpretation or understanding. Richard Pearce-Moses identified one of the functions of archival description as ‘to facilitate […] understanding’.8_{A 2007}

Dutch-language definition of context emphasised that it is needed for the interpretation of a record or archive, specifically the interpretation of the link between a record and the activities of the archive

8_{A Glossary of Archival and Records Terminology, s.v. “Archival description,” accessed 27/01/2020,} https://www2.archivists.org/glossary/terms/a/archival-description .

(10)

creator.9_{Wei Mei Pan’s recent encyclopaedia entry on context also asserted that it is needed for}

understanding.10_{This focus on the link between the context of records and their interpretation can be}

traced back in archival theory. It was present in what is often viewed as a synthesis of wider European ideas on archives: Handleiding voor het Ordenen en Beschrijven van Archieven. (henceforth the Handleiding).11_{The Handleiding emphasised the importance of which organisation formed an}

archive and the archive’s original context for the interpretation of records.12_{Rule 61 indicated that}

information on the context of creation needed to be provided in an inventory introduction.13_These

definitions, the encyclopaedia entry and the Handleiding show that enabling interpretation and understanding is widely seen as an important purpose of description.

The relationship between description, contextual information, and the interpretation and

understanding of records is common in theorical debates too. In a 2017 review chapter on archival description Geoffrey Yeo stated:

Above all, they [descriptive products] capture and collate information about context. Such information is needed to aid interpretation, because records do not usually bear their wider context on their faces but can be elucidated by understanding their relationships.14_{[Emphasis added.]}

The context thus enables both understanding and interpretation. ‘Aid’ reflected Yeo’s sense of description’s necessity for and role in the interpretation of records. His statement that context is derived from the relationships between records indicates a link between how records are

conceptualised and emphases on the need to describe a record’s context. Although Yeo reiterated the connection between contextual information (description) and the interpretation of records, he did not critically discuss it and instead addressed differing views on which context is necessary.15_Heather

MacNeil made a similar argument stating that one of description’s purposes is to ‘promote the understanding’16_{of records, following a literature review on the purpose of archival description. She}

did not question this link either. In fact, she later reinforced it when she argued that users’

9_{Archiefterminologie voor Nederland en Vlaanderen, comp. A.J.M. den Teuling (’s-Gravenhage: Stichting} Archiefpublicaties, 2007), 20. Context.

10_{Weimei Pan, “Context,”in Encyclopedia of Archival Science, ed Luciana Duranti and Patricia C Franks} (London: Rowman and Littlefield, 2015), 151.

11_{Theo H.P.M. Thomassen, “Van evenement naar structuur: ordenen en beschrijven in de eeuw vóór de} Handleiding,” in Tekst en Context van de Handleiding voor het Ordenen en Beschrijven van Archieven van 1898, eds. P.J. Horsman, F.C.J. Ketelaar and T.H.P.M. Thomassen (Hilversum: Verloren, 1998): lxxvii ; John Ridener, From Polders to Postmodernism: A concise history of archival theory (Duluth, Minnesota: Litwin Books, LLC, 2009), 21.

12_{P.J. Horsman, “"Het zat in de lucht." De concepten achter de Handleiding,” in Tekst en Context van de} Handleiding, lxxxvii.

13_{S. Muller, J.A. Feith and R. Fruin, “Handleiding voor het ordenen en beschrijven van archieven,” in Tekst en} Context van de Handleiding, 105-106.

14_{Yeo, “Continuing Debates about Description,” 163.}

15_{Yeo, “Continuing debates about description,” 166, 172, 178.} 16_{MacNeil, “Picking our text,” 264.}

(11)

understanding can be influenced by the choices archivists make during the process of archival description.17_{The way in which the relationship between description (specifically description which}

provides contextual information) and the interpretation of records is used as a commonplace by these two leading archival scholars indicates that it underlies theoretical debates but is not, itself, usually questioned.

The apparently self-evident nature of the relationship between description and how records are interpreted reflects the conceptualisation of records. The activities through which records are

produced and, to a lesser extent, used as well as the relationships between records within aggregations (at different levels) often feature in discussions of records. An example is Jennifer Meehan’s

description of records:

As the by-products of activity, records can serve as evidence of that activity; at the same time, their value and meaning as sources of evidence and memory are derived solely from the external and internal relationships that place the records in context.18_[Emphasis

added]

Meehan identified the context provided by activities and relationships as the sole source of a record’s (evidential) meaning. This idea reflects those in a range of (national) archival traditions. The emphasis upon relationships between records features in Luciana Duranti’s traditional, diplomatics-based concept of the archival bond, the relationships between records which give them meaning.19_Theo

Thomassen, reflecting the Dutch archival tradition, conceptualised records as ‘process-bound

information’20_{. Moreover, he linked a record’s context of creation to their interpretation.}21_{Drawing on}

Australian records continuum thinking, Reed also emphasised the link to creation and the

relationships between records. She argued that the knowledge of the latter is needed for a record’s interpretation and use. 22_{Chris Hurley drew on the Australian tradition to define a record as ‘an object}

whose meaning derives from an understanding of an event or circumstance with which it deals or is involved.’23_{He too, related the understanding of records to the wider event or circumstance they are}

linked to (something which archival description describes). Yeo’s recent, divergent, conceptualisation of records drew on speech theory to argue that records are persistent representations of occurrents (acts, events and functions). He stressed the contextual nature of the link between activities and records. 24_{However, his approach to the role of context in the interpretation of records was}

17_{MacNeil, “Picking our text,” 273-275.} 18_{Meehan, “Arrangement and description,” 65.}

19_{Luciana Duranti, “The Archival Bond,” Archives and Museum Informatics 11 (1997): 216-217.} 20_{Thomassen, “A First Introduction to Archival Science,” 374.}

21_{Thomassen, “A First Introduction to Archival Science,” 379.} 22_{Reed, “Records,”102, 109-112.}

23_{Chris Hurley, “Parallel Provenance: (1) What, if Anything, is Archival Description?”, Archives and Manuscripts} 33, no.1 (May 2005): 122.

24_{Geoffrey Yeo, Records, Information and Data: Exploring the role of record-keeping in an information culture} (London: Facet Publishing, 2018),149-152.

(12)

complicated by his argument that information is a record’s affordance. This implies an important role for users in the meaning that they derive from a record.25_{On the whole, conceptualisations of records}

emphasise the importance of the relationships between records and their context of creation for the interpretation of records. Both of these aspects can be termed context.

Context’s central role in the ideas about the meaning of records is shaped by two heavily debated concepts: the fonds and the (principle of) provenance. Both could be thesis subjects in their own right but will only be touched upon here. Both concepts relate to the environment in which records are created. Provenance is about the origin and context of creation of records, though its extent is debated.26_{Meehan termed provenance the ‘external structure of records’}27_{in contrast to the principle}

of original order which she located in records’ ‘internal structure’28_{. Provenance as a concept and the}

wider archival preoccupation with context clearly overlap. The fonds is more criticised. At its core, it denotes the importance of keeping records from the same creator together in order to ensure their collective meaning. Yet its application is criticised, often remaining restricted to those records from the same creator held by one archival institution.29_{These debates centre mainly on which contextual}

information is needed, rather than on whether it is needed for interpretation. In current practice, description usually focuses on specific elements of a record or its context. Thus, both the ISAD (G) and the draft Records in Context (RiC) descriptive standards identify such elements for description. The idea that a collective of records carries meaning is reflected in the ISAD (G)’s use of multilevel and hierarchical description. 30_{RiC seeks to go beyond this, although its authors stressed provenance’s}

importance in their conceptual model. More broadly, they identified context as necessary for a fuller understanding of records.31_{These descriptive standards, which provide input for description as a}

product are based on the wider conceptualisation of the importance of a record’s context. The exact elements of a record’s context which need to be described might be a point of debate, but the idea that they need describing is not.

The role archival description plays in the understanding or interpretation of records through providing contextual information also pervades criticism of descriptive practices. The seminal Wendy Duff and Verne Harris article “Stories and Names” argued that description always has a limited perspective and so influences how records can be used. Writing description involves selecting which contextual

25_{Yeo, Records, Information and Data 152-156.}

26_{Jennifer Douglas, “Origins and Beyond: The Ongoing Evolution of Archival Ideas about Provenance,” in} Currents of Archival Thinking,eds. Heather MacNeil and Terry Eastwood, 2nd_{ed (Westport: ABC-CLIO, LLC,} 2017), 39-40 ; Yeo, “Continuing Debates about Description,”164-166 .

27_{Meehan, “Arrangement and description, 66.} 28_{Meehan, “Arrangement and description, 67.}

29_{Yeo, “Continuing Debates about Description,” 166-167.} 30_{ISAD (G), 1.7 and 1.8.}

31_{Daniel Pitti, Bill Stocking and Florence Clavaud, “An introduction to “Records in Contexts”:} an archival description draft standard,” Comma Volume 2016, no.1-2 (2016): 174-175.

(13)

elements to focus upon. This process entails interpretation and shapes records, often in support of unequal power relations.32_{This postmodern critique negatively assesses the role contextual}

information plays in the understanding of records. Elizabeth Yakel offered a similar argument in “Archival Representation”, stating that description shapes how people can interpret records by making some contextual elements more prominent and ignoring others. She criticised the lack of transparency about the choices made by those who describe records and their impact on interpretation. 33_Margaret

Hedstrom echoed this call for transparency in her conceptualisation of description and access tools as an interface and ‘interpretative framework’ 34_{between archives and users. Meehan too has sought to}

describe the process of arrangement and description in order to argue that in seeking to provide access and enable users to understand records, archivists first need to interpret them.35_{Arguments which}

stress the need for transparency around choices on archival description indicate how widespread ideas on description’s role in interpretation are. Even when the practice and theory of description is

critiqued, the link between description and interpretation is not.

Whereas there is an underlying assumption in archival theory that archival description and the interpretation of records are linked, its exact nature is not properly addressed. The archival preoccupation with context is based on the idea that contextual information is important for understanding records. This, in turn, reflects conceptualisations of the record itself. A record’s meaning derives from its context as well as its content. In fact, the discussions above have shown that a record’s context and how that shapes its meaning is focused upon in archival theory more than its content. The widespread presence of the idea that description gives information on these relationships and helps people understand the meaning of records might be the reason it has not been tested in user studies; it is so commonplace it appears self-evident.

2.3 Users and description

2.3.1 Which type of description?

Archival description is both a process and a product, which is produced during the process. The product comes in many forms, a wide range of which have been addressed by user studies. This section will retain the broad working definition of description to encompass the range of products addressed in user studies. The previous section briefly alluded to description’s nationally-bound nature, which international descriptive standards have not fully replaced. This persistence of national traditions reflects the practical constraints which prevent re-describing archives or retrofitting

32_{Wendy M. Duff and Verne Harris, “Stories and Names: Archival Description as Narrating Records and} Constructing Meanings,” Archival Science 2, no.3-4 (September 2002): 275-279.

33_{Elizabeth Yakel, “Archival Representation,” Archival Science 3, no.1 (March 2003): 19-21, 24-25.}

34_{Margaret Hedstrom, “Archives, Memories, and Interfaces with the Past,” Archival Science 2, no.1-2 (March} 2002): 22.

35_{Jennifer Meehan, “Making the Leap from Parts to Whole: Evidence and Inference in Archival Arrangement} and Description,” American Archivist 72, no.1 (Spring/Summer 2009): 89-90.

(14)

description. It is relevant to the research as most user studies are conducted in North America whereas this research will be conducted in the Netherlands. The classic Dutch access tool is an inventory (inventaris), which can feature an (extensive) introduction (inleiding) on the history of the creating agent (organisation or individual), of the archive (as an aggregation of documents) and the processing of the archive.36_{The legacy of this classic access tool persists while American finding aids sometimes}

provide less detail, due to the influence of description methods employed by libraries and a present-day focus on processing collections as little as is necessary.37_{The following chapter will specify the}

example of archival description used in the study, yet the challenge in directly comparing different types of description should be kept in mind while reading this section.

2.3.2 Diverging approaches and emphases in user studies

User studies is a diverse field, covering different specific subjects and approaches. Few studies have addressed the users’ use of records themselves and will mostly be addressed later. One example, though, is Anneli Sundqvist’s study into the types of records used in Swedish government agencies. She developed a conceptual model in which use was directly linked to needs.38_{Another approach}

takes user needs as their main subject, for instance by analysing the questions users ask archival institutions.39_{Studies also addressed the needs and information-seeking behaviours of specific user}

groups, such as professional historians or genealogists.40_{An associated approach is to consider the}

role of users’ competencies.41_{Another approach looked at the roles users might play in enriching}

description.42_{A common type of user study can overlap with the studies into particular user groups:}

36_{Theo Thomassen, “Klassieke toegangen op archieven: een overzicht,’ in Toegang: Ontwikkelingen in de} ontsluiting van archieven Jaarboek 2001, eds. Theo Thomassen, Bert Looper and Jaap Kloosterman (’s-Gravenhage: Stichting Archiefpublicaties, 2001), 109-111.

37_{Ciaran B. Trace and Andrew Dillon, “The evolution of the finding aid in the United States: from physical to} digital document genre,” Archival Science 12, no.4 (December 2012): 504-509; Anne Gilliland, Conceptualizing 21st_{Century Archives (Chicago: Society of American Archivists, 2014), 105.}

38_{Anneli Sundqvist, “Conceptualisations of the use of records,” Tidsskriftet Arkiv 6 (2015): 2-3, 7-11.} 39_{Wendy M. Duff and Catherine A. Johnson, “A Virtual Expression of Need: An Analysis of E-mail Reference,”} American Archivist 64, no.1 (Spring/Summer 2001): 58-60.

40_{Donghee Sinn and Nicholas Soares, “Historians’ Use of Digital Archival Collections: The Web, Historical} Scholarship, and Archival Research,” Journal of the Association for Information Science and Technology 65, no. 9 (2014): 1805-1807; Helen R. Tibbo, “Primarily History in America: How U.S. Historians Search for Primary Materials at the Dawn of the Digital Age,” American Archivist 66, no. 1 (Spring/Summer 2003): 11 ; Wendy M. Duff and Catherine A. Johnson, “ Where Is the List with All the Names? Information-Seeking Behavior of Genealogists,” American Archivist 66, no.1 (Spring/Summer 2003): 93-95.

41_{Sandra Sacher-Flaat, “De ontsluiting van archieven en de competentie van de gebruiker,” in}

Archiefgebruikers: Consumenten van het Verleden Jaarboek 2002/2003, ed. Theo Thomassen (’s-Gravenhage: Stichting Archiefpublicaties, 2004), 178-180 ; Herman Bongenaar, “’Kennis en kunnis’ Onderzoek naar competenties van archiefbezoekers in het Gemeentearchief Amsterdam,” in in Archiefgebruikers: Consumenten van het Verleden Jaarboek 2002/2003, ed. Theo Thomassen (’s-Gravenhage: Stichting Archiefpublicaties, 2004), 106-107.

42_{Matt Gorzalski, “Examining User-Created Description in the Archival Profession,” Journal of Archival} Organization 11, no.1-2 (2013): 6-7; Magia Ghetu Krause and Elizabeth Yakel, “Interaction in Virtual Archives: The Polar Bear Expedition Digital Collections Next Generation Finding Aid,” American Archivist 70, no.2 (Fall/Winter 2007): 311-312.

(15)

research into the usability of description or access systems and into how users search for and find information, commonly termed information retrieval.43_{They have addressed the role of description}

and how users interact with description most clearly and will be focused on here. Most use examples of digital description or study online users, Duff suggested that the increasing use of description online meant that archival institutions needed to develop new ways of gathering information on user needs and behaviour.44_{This review will largely use published user studies.}

The large volume of studies into how users can search for information and the usability of individual access tools or systems reflect the shift from analogue to digital access to archives. User responses to Encoded Archival Description (EAD) finding aids and the usability of these finding aids were addressed by, amongst others, Christopher J. Prom and Yakel.45_{Usability studies sometimes}

distinguished between usability for different user types, such as Joyce Chapman’s analysis of

experienced and novice users’ interaction with a new online finding aid and Wendy Scheir’s study on novice users’ interactions with a range of online finding aids.46_{In a 2010 study, Morgan G. Daniels}

and Yakel argued that the majority of earlier studies focused on usability rather than how users search for information. Yet they also restricted their study to users’ interaction with description.47_Eigentijdse

Toegankelijkheid, a published MA thesis, focused on information retrieval by novices as well as those with archival experience. It did so in an attempt to verify the concept of archival intelligence and will be discussed more below.48_{These studies asked different questions but all considered users’}

interactions with archival description or systems which provide access to it.

In her 2012 review chapter on user studies, Duff emphasised that as a whole the user studies have contradicted each other.49_{However, there are common elements in the studies above. These are the}

hierarchical nature of most of the access tools and systems and how this structure can be navigated. Additionally, they pointed to challenges in understanding specific archival terminology and users’ need for specific competencies. Yakel indicated that terminology and the hierarchical structure of

43_{Anneli Sundqvist, “Archival Mediation. Studying Users' Interaction with Access Systems,” in Research in the} Archival Multiverse, eds.Anne J. Gilliland, Sue McKemmish and Andrew J. Lau (Clayton, Victoria: Monash University Publishing, 2017), 562-563.

44_{Duff, “User studies in archives,” 199.}

45_{Christopher J. Prom, “User Interactions with Electronic Finding Aids in a Controlled Setting,” American} Archivist 67 (Fall/Winter 2004): 237-238, 263-265; Elizabeth Yakel, “Encoded Archival Description: Are Finding Aids Boundary Spanners or Barriers for Users?” Journal of Archival Organization 2, no.1-2 (2004): 64. 46_{Wendy Scheir, “First Entry: Report on a Qualitative Exploratory Study of User Experience with Online Finding} Aids,” Journal of Archival Organization 3, no.4 (2005): 52-56; Joyce Celeste Chapman, “Observing Users: An Empirical Analysis of User Interaction with Online Finding Aids,” ,” Journal of Archival Organization 8, no.1 (2010): 5.

47_{Morgan G. Daniels and Elizabeth Yakel, “Seek and you may find: successful search in online finding aid} systems,” American Archivist 73 (Fall/Winter 2010): 542-543.

48_{Romijn-Wixley and de Vries, “Eigentijdse toegankelijkheid,” 8-9.} 49_{Duff, “User studies in archives,” 204.}

(16)

online finding aids can be problematic.50_{Nonetheless Scheir and Chapman illustrated that, through}

interacting with finding aids, novices learn to cope with the challenges.51_{An important observation}

from the studies above is Daniels’s and Yakel’s suggestion that users struggled to distinguish between content (records) and context when they used an online finding aid.52_{When users access description}

digitally, the difference between description and records may be less clear.

Other user studies specifically addressed users’ needs and responses with regard to websites which provide both digital or digitalised archival material and description. Jodi Allison-Bunnell, Janet Hauck and Yakel studied the researchers’ needs of researchers regarding the description and online presentation of archival materials and specifically focused on which metadata elements users found relevant.53_{Kathleen Fear also focused on metadata (Dublin Core) for digital image collections and}

which metadata elements users found for image retrieval.54_{These studies differ from those discussed}

above as they addressed item-level description in a digital context rather than collection-level description.

2.3.3 Users, description, records and interpretation

Studies which, at first glance, appear to specifically address users’ interactions with records still largely focus on retrieval. They consider archival education or the use of archival material in

education and were conducted in North America with (under)graduate studies. One sought to measure the effectiveness of a guide on using primary sources and focused on how students found material and how they understood the description rather than on their use of the sources.55_{Another addressed the}

extent to which archival orientation encourages archival use.56_{A study into assigned archival research}

mentioned interpretation as a skill students developed during archival research. It only briefly addressed the idea that contextual information aids interpretation; this was as the archivist mentioned it during the students’ orientation session. 57_{Like the user studies discussed in the previous section,}

research into users and records (covered by primary sources in these cases) does not usually focus on the role of description in interpretation.

50_{Yakel “Encoded Archival Description,” 75.}

51_{Wendy Scheir, “First Entry,” 76-78; Chapman, “Observing Users,” 26-27.} 52_{Daniels and Yakel, Seek and you may find,” 561-562.}

53_{Jodi Allison-Bunnell, Elizabeth Yakel and Janet Hauck “Researchers at Work: Assessing Needs for Content and} Presentation of Archival Materials,” Journal of Archival Organization 9, no.2 (2011): 85-88.

54_{Kathleen Fear, “User Understanding of Metadata in Digital Image Collections: Or, What Exactly Do You Mean} by “Coverage”?” American Archivist 73, no.1 (Spring/Summer 2010): 27, 50-51.

55_{Joanne Archer, Ann M. Hanton and Jennie A. Levine, “Investigating Primary Source Literacy,” The Journal of} Academic Librarianship 35, no.5 (2009): 411, 418.

56_{Wendy M. Duff and Joan M. Cherry, “Archival Orientation for Undergraduate Students: An Exploratory Study} of Impact,” American Archivist 71, no.2 (2008): 501-502.

57_{Xiaomu Zhou, “Student Archival Research Activity: An Exploratory Study,” American Archivist 71, no.2} (2008): 479, 484-485, 490.

(17)

The concept of archival intelligence, developed by Deborah Torres and Yakel, was developed through a user study. Archival intelligence is one of three types of knowledge or skills they argued users need to use archives, alongside domain knowledge and artefactual literacy. The latter relates to the ability to work with and interpret archival materials themselves, whereas the authors primarily related Archival Intelligence to successfully finding records.58_{Knowledge of archival theory and description}

came under archival intelligence, whereas using records came under artefactual literacy. The two thus seemed separate. Yet archival intelligence included ‘intellective skills’ 59_{, the ability to distinguish}

between and connect records and their descriptive surrogates.60_{Users thus, need to be able to make a}

distinction between the records themselves (the content) and the description of the records (contextual information). Archival intelligence is therefore a concept that is potentially relevant beyond finding records.

Andrea Johnson, who researched users of digital archives, briefly addressed what users can do with records. She suggested, based on general archival theory, that users need contextual information to understand records. Yet, she also argued that individual users constructed the meaning of digital objects during use and that this process was shaped by their research aims.61_{Her conceptual model of}

archival research links the construction of meaning of digital objects by users during their research process to several skills identified by Torres and Yakel: ‘artefactual knowledge’, ‘understanding representational relationships’ and ‘subject knowledge’.62_{This implied that interpretation combined}

wider contextual information (knowledge about a particular topic) with information provided by description and the records themselves. It heavily emphasised the role of the user in the interpretation of records, underlining the subjectivity of this. Yet Johnson did not include any results of research into user interaction with records.

Only two identified studies collected data specifically on users’ interaction with description and their interpretation of records. The first, Eigentijdse Toegankelijkheid, was discussed earlier and collected data by asking participants to describe records and evaluate the actual archival description. Romijn-Wixley and de Vries indicated that users focused more on specific details in the records (their content) than the actual descriptions. They sought to relate the participants’ interaction with the records and description to their archival competencies. Their results suggested that those with more archival competencies were better able to understand the description and the records seen in the experiment.63

58_{Elizabeth Yakel and Deborah Torres, “AI: Archival Intelligence and User Expertise,” American Archivist 66,} no.1 (Spring/Summer 2003): 52, 61-63, 78.

59_{Yakel and Torres, “AI: Archival Intelligence and User Expertise,” 73.} 60_{Yakel and Torres, “AI: Archival Intelligence and User Expertise,” 73-77.}

61_{Andrea Johnson, “Users, use and context: supporting interaction between users and digital archives,” in} What are archives? Cultural and Theoretical Perspectives: A Reader, ed. Louise Craven (Aldershot: Ashgate, 2008), 151-155.

62_{Johnson, “Users, use and context,” 165.}

(18)

The second was a study by Duff et al. into meaning-making in archival research. Its authors focused on archival research as a whole, rather than considering the specific interaction with description and records. They argued that finding aids and archival description influence how users interpret and understand records, especially when users lack domain knowledge and are therefore more reliant on description. In addition, they noted that description highlights particular relationships between records and argued that this also influences users’ interpretation.64_{Yet, the authors also emphasised the role of}

the user in the process of meaning-making. They stated:

Interpretation is dependent on a person’s knowledge or previous knowledge as it occurs within a diverse, rich, and personal structure. The degree to which archival arrangement and description affect the impressions or interpretations made from archival documents remains a cipher.65

Thus, they identified a role for description in interpretation but did not distinguish between this role and the wider knowledge and experiences of individuals. They also identified a role for the content of records.66_{This study was a pilot study and ended by calling for more research on the interpretation of}

records, which this thesis seeks to do.67

2.4 Defining the interpretation of records

Interpretation, understanding and meaning-making all featured in the discussion above. Each idea can be related to the interpretation of records, although some were used more widely. The interpretation of records takes place when users interact with records themselves. Therefore, a study into interpretation needs to consider individual users. It was noted above that archivists also interpret records during the process of description. Yet, although two individuals who worked in archival description are included in the sample, the focus is on how users interpret records in relation to the descriptive products they see. It will not address the wider definition of interpretation as the synthesis of historical evidence from archival literacy studies either.68_{Another aspect that was not considered above and will not be}

addressed here is users’ evaluations of the authenticity or reliability of records used for evidential purposes. The interpretation of records starts with the comprehension of a record’s content, yet it is more complex than just this.

Few definitions of the broader terms mentioned at the start of this paragraph were offered in the literature discussed. The only definition was offered by Duff et al, who used interpretation, understanding and their preferred term making interchangeably. They defined

meaning-64_{Wendy M. Duff, Emily Monks-Leeson, Alan Galey and The Implementing New Knowledge Environments} (INKE) Team, “Contexts built and found: a pilot study on the process of archival meaning-making,” Archival Science 12, no.1 (March 2012): 84-85.

65_{Duff et al, “Contexts built and found,” 77.} 66_{Duff et al, “Contexts built and found,” 84.} 67_{Duff et al, “Contexts built and found,” 88.}

68_{Elizabeth Yakel and Doris Malkmus, “Contextualizing Archival Literacy,” in Teaching with Primary Sources,} eds. Christopher J. Prom and Lisa Janicke Hinchliffe (Chicago: Society of American Archivists, 2016), 37.

(19)

making as the ways the users give meaning to and understand the records they view during the archival research process.69_{This emphasis on the process drew on postmodern and poststructuralist}

philosophy, which argues that users play a role in constructing the meaning of a text.70_{Although it is a}

wider concept than the interpretation of records, Duff et al.’s focus on interpretation as a process is an important aspect to consider and will be included in the working definition. If the interpretation of records is a process, it is not fixed in one reaction to the records. Instead it can change during a user’s interaction with a record.

The archival theory on description and records which was touched upon above also offers an insight into an important aspect of the interpretation of records: contextual information. The literature posited that the interpretation of records requires contextual information, which implies that users who just view a record’s content are either unable to interpret records or that their interpretation is more restricted. Contextual information could play several roles. It might clarify aspects of the record. It could also help users place records in the wider web of relationships they form a part of or in the time period in which they were created. Something which complicates a consideration of context is its limitlessness, something Yeo identified.71_{Christopher A. Lee emphasised another complicating}

element in his literature review of context as a concept: its multifaceted nature. In fact, he distinguished between three types of context. The first is the information offered about an object, which can affect understanding for instance; the second is the environment in which an object was created and used and the third is related to the user of the object, their experiences also form a context.72_{The first type can be related to archival description, the second to the wider context of the}

records (the subject of the contextual information description seeks to provide) and the final type links context itself to individuals who view a digital object or record.

The third form of context (the user who forms a context for the record they view) reflects the emphasis on individual users in the studies addressed in the final part of the section on user studies and Yeo’s emphasis on the individual’s role in the affordances records give them. Romijn-Wixley and de Vries argued that an individual’s archival competencies influence their interpretation. The

knowledge users bring to their interaction with records is also an important part of the context they themselves form. This domain knowledge was mentioned by Johnson and Duff et al. Thus, while domain knowledge was identified by Torres and Yakel as something which relates to an individual’s ability to comprehend records, it is arguably part of the context in the case of an individual’s

69_{Duff et al, “Contexts built and found,” 70, 78, 84.} 70_{Duff et al, “Contexts built and found,” 70 -72.} 71_{Yeo, “Continuing Debates about Description,” 172.}

72_{Christopher A. Lee, “A framework for contextual information in digital collections,” Journal of} Documentation 67, no.1 (2011): 97.

(20)

interaction with records. This overlap between content and context points to the challenge in distinguishing the role of a record’s content and context in the interpretation of records. The emphasis above on the interpretation of records depending not only on their content or description but also on their context of use implies that records are not interpreted in just one way. Factors other than information on the context of creation can contribute to the understanding of records. This suggests that records are not static but gain new meanings as they enter new contexts. That idea reflects the focus on (re)contextualisation in records continuum thinking.73_{A working}

definition of the interpretation of records which seeks to consider how people interpret context more widely rather than just their ability to interpret original context thus appears more suited to this flexible idea of a record.

This section forms the basis of the working definition of the interpretation of records in this thesis. The interpretation of records is defined as: a highly individual process which takes place at the point of interaction between a person and records; during the process individuals place records in a wider context, such as a subject, event or set of relationships, that relates to their reason for using the record. This process is further shaped by a record’s content and context (which archival description forms a part of).

2.5 Conclusion

The studies and literature discussed above presume that archival description plays a significant role in the interpretation of records. This is based on the way records themselves are understood and the degree to which their meaning is thought to come from their context. User studies have failed to substantially address this presumption, a possible reflection of its widespread acceptance. In order to counter the fragmentation on this subject between these two parts of archival science, a user study addressing the interpretation of records is needed. This study will be outlined in the following chapter and will focus on the interaction of users with both records themselves and archival description. The focus on individuals’ interaction with these two elements uses the working definition of the

interpretation of records developed in this chapter.

73_{Sue McKemmish, “Traces: Document, record, archive, archives,” in}_{Archives: Recordkeeping in Society, eds.}

Sue McKemmish, Michael Piggot, Barbara Reed and Frank Upward (Wagga Wagga: Centre for Information Studies, 2005), 14-15 ; Frank Upward, “The Records Continuum,” Archives: Recordkeeping in Society, eds. Sue McKemmish, Michael Piggot, Barbara Reed and Frank Upward (Wagga Wagga: Centre for Information Studies, 2005), 205-206, 216.

(21)

3. Methodology

3.1 Introduction

Archival theory posits a role for description in interpretation, primarily through the contextual information it provides. Yet user studies have not substantially addressed this role. As a result, there are few relevant examples of research methodologies in the literature. This chapter therefore asks: which type of user study allows for the collection of data on users’ interactions with records and archival description as well as on their interpretation of records? Initially the methodology discussed in this chapter was developed as a pilot, it was used for the main study following an evaluation. It is a qualitative, quasi-experiment which draws on one of the few examples in the literature of research into user interaction with records. As the methodology is uncommon, the reasons it was chosen will be discussed prior to a description of the quasi-experiment and the processing of the data collected.

3.2 Choices regarding the methodology

3.2.1 A qualitative approach

This research seeks to explore how description affects the way in which users interpret records, a subject not often addressed in user studies. The issues around how interpretation is understood and the fact that the process of interpretation is a black box, as it happens in people’s minds, makes measuring interpretation challenging. Rather than verifying a hypothesis on causation or correlation, this study will explore how description influences interpretation. It does so by considering the interpretation of records by individuals before they view description and after they have viewed description.

A qualitative approach allows for a fuller exploration of the interpretation of records but it complicates the overall methodology (the alignment of the paradigm, research method and data collection techniques).74_{User studies, especially those into users’ interaction with finding aids and}

search systems, often make use of quantitative data. However, these data are sometimes analysed qualitatively. In addition, there have also been qualitative studies.75_{The choice for an experiment as a}

research method will be further discussed below, as experiments tend to be used in quantitative research. Semi-structured, in-depth interviews serve as the main data collection technique. The combination of a qualitative approach with an experiment is why the experiment was first piloted. The pilot was included in the actual study as it was deemed appropriate and few changes were required.

3.2.2 The suitability of an experiment as the research method

Experiments are largely associated with positivist and quantitative research. The true or laboratory experiment seeks to prove hypotheses on causation (the relationship between an independent and a

74_{Kirsty Williamson, “Research concepts,” in Research Methods: Information, Systems and Contexts, eds. Kirsty} Williamson and Graeme Johanson (Prahan: Tilde Publishing, 2013), 4.

(22)

dependent variable) in a controlled environment.76_{This relationship is statistically analysed.}77_Alison

Pickard argues that the need for a controlled environment makes true experiments ill-suited to research involving people, as variables are then much harder to control.78_{Quasi-experiments are}

usually done in field situations where is less scope to control variables and therefore study correlation rather than causation.79_{The use of this method in a primarily qualitative research methodology is not}

self-evident and needs to be explained.

The key reason for using an experiment in this research design was that it had been used previously in the literature to explore questions around interpretation. The master’s thesis Eigentijdse

Toegankelijkheid used a simple experimental setup which appeared suitable as the basis for this thesis. Its post-experiment consisted of two tasks: getting people to read and evaluate archival description and asking them to describe records. The order of the tasks alternated, as the researchers split the sample into groups. A questionnaire followed the tasks. This experiment design seems highly suited to gathering information on people’s interaction with archival description and records. Additionally, it offers a means of comparison as individuals can be shown the same records. Comparison would be harder to achieve through the observation of archival research as, bar genealogical records, few researchers will use the same records during the same period.

The Eigentijdse Toegankelijkheid example largely used closed questionnaires for the gathering of data and primarily used quantitative data analysis, which points to the possible clash between wider qualitative approach and an experiment. Nonetheless, there are some examples of qualitative experiments in archival science and related fields. In a textbook on the use and application of

experiments in Interactive Information Retrieval (IIR) research, Diane Kelly argues that experiments can be used for exploratory research which seeks to answer open-ended questions.80_{Sue Robinson}

and Andrew L. Mendelson, who conduct media studies research, conducted a two mixed methods studies on how people make meaning of the media they consume. They termed these studies ‘qualitative experiments’. 81_{They argued that these experiments allowed for a more specific focus}

than commonly-used qualitative methods. .82_{The-Allison-Bunnell et al. study on the online}

presentation of digitised archival material provides an example of the combination of experiments and qualitative data collection techniques (in this case open-ended interview questions).83_{The authors} 76_{Alison Pickard, Research Methods in Information, 2nd ed. (London: Facet publishing, 2013), 119-121.} 77_{Kerry Tanner, “Experimental Research,” in Research Methods: Information, Systems and Contexts, ed. Kirsty} Williamson and Graeme Johanson (Prahran: Tilde Publishing, 2013), 312.

78_{Pickard, Research Methods in Information, 125.} 79_{Tanner, “Experimental Research,” 321-323.}

80_{Diane Kelly, “Methods for Evaluating Interactive Information Retrieval Systems with Users,” Foundations and} Trends in Information Retrieval 3, 1-2 (2009): 25.

81_{Sue Robinson and Andrew L. Mendelson, “A Qualitative Experiment: Research on Mediated Meaning} Construction Using a Hybrid Approach,” Journal of Mixed Methods Research 6, no.4 (2012): 332-333. 82_{Robinson and Mendelson, “A Qualitative Experiment,” 344.}

(23)

termed their methods ‘quasi-experimental’84_{. Like the Eigentijdse Toegankelijkheid post-experiment,}

Bunnel et al. observed people’s interaction with description and archival material. Their focus did differ, they sought to identify what is needed for online access to archival material to meet user needs. Despite the differences, their study shows that experiments can be combined with a qualitative approach for exploratory research into the users of archival description and (digitised) records. Although experiments are rarely employed in qualitative research the examples discussed above suggest they can be used to observe and question individuals about the interaction with records and archival description. The experimental set-up is highly artificial and therefore does not reflect real-life archival research. Users are not usually asked to describe records and tend to view records in the context of a wider research process. Yet, it can still provide valuable data and enables comparison between responses to the same examples. This is something the other example of a study on the interpretation of archival records cannot provide. That study was the Duff et al research on meaning-making discussed in the literature review. In it, four graduate students kept diaries on and were interviewed about their own archival research. 85_{Following Robison and Mendelson’s argument that}

an experiment allows for a narrower focus than the usual qualitative research methods, an experiment is considered more suitable for the interest of this research in the interaction with and interpretation of records. Moreover, it can be conducted in less time than the observation of the archival research process.

3.3 The quasi-experiment

The quasi-experiment was initially piloted with two participants before it was selected as the primary data collection method for this research. The pilot experiments included a concluding interview on the experiment and the participants were largely positive. The data from the pilot is also included in this thesis, as few changes were made to the experiment. Prior to discussing the content of the experiment, the choice of examples and language used will be discussed as they were based on practical matters and could potentially affect the results. In addition, the population materials used will be outlined. Each experiment took between one and a half and two hours, excluding the additional sections in the pilot. The original Dutch protocols for the experiment can be found in Appendices 1, for participants, and 2, for the researcher. Their translations can be seen in Appendices 3 and 4 respectively.

3.3.1 Population

The choices regarding population were initially made for the pilot: to select individuals expected to provide a contrast, those with archival description experience and novices. This strengthened the contrast in comparison to the Eigentijdse Toegankelijkheid study, which only considered archival users. The choice for different groups in the sample drew on advice regarding populations and

84_{Allison-Bunnell et al, “Researchers at Work,”74.} 85_{Duff et al, “Contexts built and found,” 78.}

(24)

samples.86_{Individuals were identified based on their (lack of) experience through personal networks,}

thus making the eventual sample more comparable to a convenience sample as no steps were taken to ensure that the participants were representative of the population identified. The results are therefore not suited to generalisation, although they can be used for this exploratory study.87

3.3.2 The chosen examples: narrowing down description

As the experiment was conducted in the Netherlands Dutch examples of archival description were used. Both examples came from the Stadsarchief Amsterdam (Amsterdam City Archives). This institution provides online access to a large volume of archived records and they can be easily identified on its website, which was important given the limited time to develop the initial pilot. The description provided for these records does not differ strongly from a traditional Dutch inventory. Per archive (archief in Dutch, an aggregation of records created by an organisation, individual or family) there is a page which features an online display of the inleiding (the introduction to the inventory), a PDF version of the whole inventory (essentially comparable with an analogue inventory) and a webpage which incorporates the inventory and digitised records into a hierarchical structure. One of the benefits of this presentation was that the description could easily be presented in a printed document. The specific examples were files of typed correspondence that will be described below in the sections on parts 1 and 2 of the experiment. They were selected using a list of requirements. The examples needed to be in clear, easily legible handwriting or typed; they could not consist of too many pages; and the archive they could be found in needed to have an inventory introduction. Few of the digitised records on the Stadsarchief Amsterdam website met all of these requirements. The fact that both examples consisted of correspondence could be viewed as beneficial, as this is a familiar genre of document. Nevertheless, it also limits the results as the limited examples mean that the experiment does not cover interaction with another genre of record.

3.3.3 Language

The experiment was conducted in Dutch and the written data (from description assignments) was translated for inclusion in this thesis. Given the complexity of understanding interpretation, the use of translation could be problematic as translation itself is a form of interpretation. If it is thought to have influenced any results it will be further addressed.

3.3.4 Materials

This experiment used a laptop to display the website of the Stadsarchief Amsterdam. This included both the description for the first and fourth task and the website access to the records for the first task. The laptop was also used to display the set of records for the second part. These were shown as PDFs,

86_{Kirsty Williamson, “Populations and samples,” in Research Methods: Information, Systems and Contexts, ed.} Kirsty Williamson and Graeme Johanson (Prahran: Tilde Publishing, 2013), 332.