Jane Elliot (2005). Using Narrative in Social Research. Qualitative and Quantitative Approaches. London: Sage Publications, 232 pages, ISBN 1-4129 0040 9, Price $35.95

Abstract: With Using Narrative in Social Research Jane ELLIOTT has made an original and very accessible contribution to the methodological literature in social research. The originality of the work lies most of all in its double aim. ELLIOTT not only presents both qualitative and quantitative approaches to narrative analysis but also uses the concept of narrative as a sensitizing tool to explore and to deconstruct the boundaries between qualitative and quantitative social research. Actually, this second aim receives more attention than the first. Consequently, the book offers less practical guidance to the analysis of narratives than its title suggests. Most of all it provides advanced students and researchers with a sound reflection on the qualitative-quantitative divide and on possible ways to overcome it. Although insightful for every social scientist, the book tries in the first place to win the "quantitative audience" for an interpretive turn.

Key words: narrative analysis, event history analysis, life course research, narrative identity, combining qualitative and quantitative research, epistemology

With Using Narrative in Social Research Jane ELLIOTT has made an original contribution to the methodological literature in social research. The originality of the work lies most of all in its double aim. ELLIOTT not only presents both qualitative and quantitative approaches to narrative analysis but also uses the concept of narrative as a sensitizing tool to question and to deconstruct the methodological and epistemological boundaries between qualitative and quantitative research. [1]

This review article raises some questions with regard to ELLIOTT's narrative lens. It is argued that this lens clarifies the qualitative-quantitative divide, but in relation to some issues it does not leave enough space for nuance. Further comments are made concerning the relevance of the book for qualitatively or quantitatively oriented audiences. However, before turning to these questions and comments it is necessary to say something about ELLIOTT's concept of narrative and to give an outline of the work. [2]

It is far from evident that narrative research practices and epistemological issues in relation to the quantitative-qualitative divide can be talked about in the same book. ELLIOTT achieves this by the very consequent use of a single definition of narrative in all the chapters. This definition, which is delineated in Chapter 1, is derived from the work of HINCHMAN and HINCHMAN (1997, p.xvi), who describe narratives as "discourses with a clear sequential order that connect events in a meaningful way for a definite audience and thus offer insights about the world and/or people's experiences of it." [3]

Following on from this definition, ELLIOTT outlines how narratives distinguish themselves from other discourses by three key elements. First, they are temporal in that they represent events as part of sequences which have a beginning, a middle, and an end. Second, they are meaningful. One important way in which they convey meaning is by ordering events into a temporal sequence which relates them to a definite resolution or conclusion. This is why narratives can be said to have a causal dimension. Thirdly, they are inherently social as they are produced for specific audiences. Consequently they are only meaningful for those audiences. [4]

In retrospect it is clear that these three elements have been strategically highlighted to relate to the later discussions of quantitative and qualitative approaches to temporality, causality and identity and of the narrative aspects of research accounts. Nonetheless, ELLIOTT's proper understanding of narrative stays close to the classic writings in the field of narrative analysis. As such, ELLIOTT draws on the seminal work of sociolinguistics LABOV and WALETSZKY (1967) and of several other pioneers of narrative analysis to discuss the temporal-causal and the meaningful dimensions of narratives. As other social scientists have done (see e.g. RIESSMAN, 1993), ELLIOTT moulds the rather rigid model of the sociolinguists until it becomes useful to her arguments about social research. In particular, she brings the causal dimension of narratives to the fore. In this regard, the cultural studies of CHATMAN (1978) and LEITCH (1986) offer important keys. Both emphasize the meaningfulness of the ending or "resolution" of narratives. As to the social function of narratives, the argument of PLUMMER that "stories can be seen as joint actions," of negotiating narrators and audiences, is taken up and illustrated (PLUMMER, 1995, p.20). [5]

As mentioned above, ELLIOTT deploys the concept of narrative and its three dimensions—the temporal, the meaningful and the social—to present actual approaches to narrative analysis as well as to address the dichotomy between qualitative and quantitative methodologies. This double aim is present in every chapter of Using Narrative in Social Research. Nevertheless, Chapter 2 to 5 as well as Chapter 8 deal mainly with practices of narrative analysis, while Chapters 6-7 and 9-10 talk primarily about the epistemological boundaries between qualitative and quantitative approaches. [6]

Chapter 2 and 3 offer an introduction to qualitative interviewing and qualitative analysis of textual data. ELLIOTT emphasizes in particular that there is no unified way of analyzing narratives within qualitative research. Distinctions can be made between analysis of content and analysis of form as well as between holistic approaches, which understand narratives as a complete entity, and categorical approaches, which focus on short sections of texts. Above all, ELLIOTT highlights the difference between the content-related naturalist approach and the constructivist approach. The naturalist approach pays attention to what people tell and considers life-stories as revealing phenomena out there. Famous examples of this approach are BERTAUX and BERTAUX-WIAME's (1981) holistic approach to life-stories as well as the less holistic and partly quantitative analysis of press accounts by FRANZOSI (1998). In contrast, constructivists focus on how stories are told and how respondents construct meaning in their narratives. As ELLIOTT illustrates, PLUMMER's analysis of the impact of social and psychological genres on the ways personal experiences are narrated exemplifies this approach (PLUMMER, 1995). Actually, constructivists take the meaningful or evaluative dimension of oral testimonies more fully into account. In contrast, naturalists often share with constructivists a holistic approach to individuals and their idiosyncratic narratives, but they pay less attention to how people create meaning at the very moment they tell their life-stories. [7]

Subsequently, Chapter 4 and 5 deal with quantitative methods which, at least in ELLIOTT's view, have a narrative dimension. It is on this point that the consequent use of the concept of narrative leads to some confusion, as I will argue later on. First, the temporal aspect of narratives is a point of departure to talk about current longitudinal quantitative research. Second, attention is paid to quantitative or combined approaches that are more case-centered than variable-centered and therefore close to holistic qualitative narrative research. Amongst others, event history analysis and life course research are presented here as methods with a narrative aspect. Event history analysis is concerned with the timing of events in individual's lives, for instance the timing of job changes. Life course research focuses on the timing of events in individual's lives as well as the interconnection with the historical and cultural context. Interestingly, most of the representatives of these approaches do not label their work "narrative." Only ABBOTT (1992) does so, by calling his event sequence analysis, which is a strongly holistic way of dealing with event history data, "narrative positivism." [8]

Another chapter dealing with research practice is Chapter 8, which outlines the ethical and political implications of research into narratives. It refers to quantitative research, but discusses for the most part the qualitative approach. As such, it contains interesting reflections on the ethical implications of analyzing how life-stories were constructed, while the research participants expect the research to be on what they have told. [9]

In contrast to these practically oriented chapters, in Chapters 6, 7, 9 and 10 the narrative lens is deployed to focus on the epistemological background of the quantitative-qualitative boundary and ways to bridge it. Chapter 6 takes the temporal-causal dimension of narratives as a starting point to reflect on causality in social research. It shows that there is an increasing awareness that statistical associations are not sufficient to establish causality, ELLIOTT's previous work being a case in point (see e.g. ELLIOTT 1999). Moreover, it sets out how causal explanations often have a temporal, narrative form and derive from theories as well as qualitative data. Incidentally, this chapter contains a very clear outline of how causality is generally approached in social research. Chapter 7 engages with the meaningful dimension of narratives to explore the extent to which quantitative and qualitative approaches enable researchers to grasp the meanings constructed by research participants while telling their life-stories. Paul RICOEUR's concept of "narrative identity" is central to this chapter. Actually, narratives provide one with the practical means to understand oneself as one entity, but also as an entity living and changing through time. As ELLIOTT points out, quantitative longitudinal research offers very detailed information about individuals, but loses how people make sense of their own lives. [10]

In Chapter 9 attention shifts to the researcher as a narrator whose scientific stories are bound to a particular genre and a specific audience. It criticizes the realist discourse of many research accounts which obscures how the research and the data were constructed. While this critique is first of all addressed to a quantitatively oriented audience, ELLIOTT rightly states that many ethnographic accounts equally adopt a naively realist tone, for instance by removing the questions of interviewers from quotes. Chapter 10, which is a straightforward plea for the combination of quantitative and qualitative methods, completes the book. ELLIOTT convincingly argues for an attention to narrative as a way to bridge the qualitative-quantitative divide on three points: causal explanations, the understanding of identities, and reflexivity with regard to the construction of research and research accounts. While this chapter sums up the previous arguments, it also gives some new examples that make it possible to read it as a text in itself. [11]

ELLIOTT's reference to a single and moreover very plain concept of narrative throughout the book provides clarity and coherence. However, at times the way narrative is used simplifies the discussion too much. For one thing, it is not made clear how the concept of narrative is related to the broader concept of discourse. Furthermore, narrative and merely historical approaches are conflated by labeling them all as "narrative." [12]

First, the concept of "discourse" is hardly used in this book and even omitted from the glossary (ELLIOTT, pp.199-204). On the one hand, this is very understandable. The concept of discourse is rather vague, referring as it does to the construction of meaning in all genres of texts and, for some, also to the constructive dimension of institutions and practices (FAIRCLOUGH, 1992). As a less clearly defined concept discourse is not as much a sensitizing tool as the concept of narrative. Incidentally, for the same reason ELLIOTT has opted for a narrow definition of narrative and distinguishes herself on this point from others (not mentioned by name), who use "narrative" to refer to any extended text regardless of the temporal dimension. [13]

On the other hand, a more explicit discussion of the relationship between narrative and discourse would have been interesting. It would, for instance, have made clearer why it is relevant to understand the writing-up of research as a narrative, or, put differently, not as a discourse in general, but as a specific discourse with a temporal-causal aspect. Similarly, there are good reasons to consider the construction of self-identity not just as "discursive," but as "narrative," or as having a temporal dimension. In leaving discourse out of the discussion, these reasons are not fully spelled out. It is all the more important to do this since critical discourse analysts understand these issues first of all in terms of discourse and use narrative within a framework of discursive analysis (for recent examples see SLEMBROUCK 2004; AINSWORTH & HARDY, 2004). Is this a sign that while the qualitative-quantitative divide dissolves, a new divide is in the making in social research between discursive and narrative approaches? [14]

A second problem with ELLIOTT's narrative lens is that there is a contrast between the way narrative is defined at the start and the way the term is used in relation to longitudinal quantitative research. ELLIOTT first emphasizes that the temporal, meaningful and social dimensions of the narrative are not strictly separable, because the meaning of events within a narrative derives both from their temporal ordering and from the social context in which they are recounted. However, in Chapters 4 and 5, event history data, such as dates of birth, marriage, employment and so on, are considered to have narrative qualities, merely because they have to do with time. I would argue that longitudinal quantitative analysis cannot be said to approach biographies as narratives, since it does not take into account how individuals construct and divide time in their stories. It is confusing to read how ELLIOTT is on the one hand very well aware of this difference between qualitative and quantitative approaches to life-stories, but on the other labels the quantitative longitudinal approach as "narrative." It would make more sense to associate event history or life course analysis with the naturalist holistic approach of classic qualitative research than to call it straightforwardly "narrative." Even then, there remains a difference between a naturalist approach to what people tell in open interviews and a holistic approach to people's answers to structured questionnaires. [15]

The attention to both qualitative and quantitative approaches and to practical as well as epistemological issues in this book is absolutely original and stimulating, but it left me—as a primarily, but not exclusively, qualitatively oriented researcher—at times wondering who this book is for. Two comments can be made in this regard. [16]

Firstly, the epistemological discussion of the qualitative-quantitative divide is no doubt relevant to quantitative as well as qualitative researchers and to advanced students who need to get acquainted with both approaches. Arguably, ELLIOTT is mainly pleading for an interpretive turn in quantitative social research. Concretely, this turn consists for ELLIOTT in quantitative research that interprets results within their historical and cultural contexts, for reflexivity on the way the research was constructed, and for a combination of quantitative and qualitative data, for instance in relation to causality. While these debates are first of all relevant for quantitatively oriented researchers, they might also help qualitative researchers to find new ways to bridge the qualitative-quantitative boundary. [17]

Secondly, the step-by-step introduction to qualitative research practices is somewhat outdated and overemphasizes in that way (perhaps with a quantitatively oriented audience in mind) the opposition between interpretive and other types of social research. For instance, it might be true that classic interpretive work paid little explicit attention to generalization. However, it is an overstatement that this would be hardly discussed today (ELLIOTT, pp.27-28), if you take recent reflections on deviant case analysis or on generalizability and the search for ideal types into account (see e.g. SILVERMAN, 2001; WENGRAF, 2001). Similarly, ELLIOTT's argument for more reflexive accounts of qualitative research makes sense (ELLIOTT, pp.157-159) but overlooks the recent trend towards using computer programs for the iterative coding of unstructured data. These programs not only enable a more systematic exploration of data but are helpful in making explicit the analytic process. They make it possible, for instance, to show how codes, or in other words the categories of thinking about the material, changed in the course of the analysis. [18]

This doesn't alter the fact that Using Narrative in Social Research is a sound book that will allow researchers and advanced students in the social sciences to get acquainted with the basics of qualitative narrative analysis and quantitative longitudinal analysis and to gain awareness of the possible bridges between qualitative and quantitative approaches. The highly accessible language, the wealth of examples, the glossary and the fact that each chapter ends with further readings and with questions for discussion should all contribute to the widespread use of the book for teaching, research and reflections on methodology. [19]

Abbott, Anthony (1992). From Causes to Events: Notes on Narrative Positivism. Sociological Methods and Research, 20(4), 428-455.

Ainsworth, Susan & Hardy, Cynthia (2004). Critical Discourse Analysis And Identity: Why Bother?" Critical Discourse Studies, 1(2), 225-259.

Bertaux, Daniel & Bertaux-Wiame, Isabelle (1981). Life-stories and the Baker's Trade. In Daniel Bertaux (Ed.), Biography and Society: the Life History Approach in the Social Sciences (pp.169-190). Beverly Hills, CA: Sage.

Chatman, Seymour (1978). Story and Discourse. Narrative Structure in Fiction and Film. Ithaca, NY: Cornell University Press.

Elliott, Jane (1999). Models Are Stories Are Not Real Life. In Daniel Dorling & Stephen Simpson (Eds.), Statistics in Society. The Arithmetic of Politics (pp.95-102). London: Arnold.

Fairclough, Norman (1992). Discourse and Social Change. Cambridge: Polity Press.

Franzosi, Roberto (1998). Narrative as Data: Linguistic and Statistical Tools for the Quantitative Study of Historical Events. International Review of Social History, 43, 81-104.

Hinchman, Lewis P. & Hinchman, Sandra K.(1997). Introduction. In Lewis P. Hinchman & Sandra K. Hinchman (Eds.), Memory, Identity, Community: The Idea of Narrative in the Human Sciences (pp.xiii-xxxii). New York: SUNY (State University of New York) Press.

Labov, William & Waletsky, Joshua (1967). Narrative Analysis: Oral Versions of Personal Experience. In June Helm (Ed.), Essays on the Verbal and Visual Arts (pp.12-44). Seattle: University of Washington Press.

Leitch, Thomas M. (1986). What Stories Are: Narrative Theory and Interpretation. University Park, PA: Pennsylvania State University Press.

Plummer, Ken (1995). Telling Sexual Stories: Power, Change, and Social Worlds. London: Routledge.

Silverman, David (2001). Interpreting Qualitative Data. Methods for Analysing Talk, Text and Interaction. London: Sage.

Slembrouck, Stef (2004). Reflexivity and the Research Interview. Critical Discourse Studies, 1(1), 91-112.

Wengraf, Tom (2000). Uncovering the General from the Particular. From Contingencies to Typologies in the Understanding of Cases. In Prue Chamberlayne, Joanna Bornat & Tom Wengraf (Eds.), The Turn to Biographical Methods in Social Science: Comparative Issues and Examples (pp.140-163). London: Routledge.

Leen BEYERS (Ph.D. History, Leuven University – KUL) is a postdoctoral research fellow of the Fund for Scientific Research Flanders (Belgium). She specializes in urban and migration history and has a particular interest in the use and analysis of oral testimonies. Recently she wrote: "Unfolding urban memories and ethnic identities. Narratives of ethnic diversity in the Limburg mining towns, Belgium, since 1920" (forthcoming in Joan HERBERT and Richard RODGER, Eds. Testimonies of the City, Historical Urban Series of Ashgate Press).

Department of History, University of Pennsylvania & Department of History, Leuven University
208, College Hall
Philadelphia PA 19104-6379
USA
T 215-833-8274

Beyers, Leen (2006,). Review: Jane Elliot (2005). Using Narrative in Social Research. Qualitative and Quantitative Approaches [19 paragraphs]. Forum Qualitative Sozialforschung / Forum: Qualitative Social Research. 7(2), Art. 19, http://nbn-resolving.de/urn:nbn:de:0114-fqs0602192.

Forum Qualitative Sozialforschung / Forum: Qualitative Social Research (FQS)

ISSN 1438-5627

Creative Commons Attribution 4.0 International License