ABS TRACT In light of the newly defined field of narrative inquiry,
this article examines ways in which narrative functions can be used
during the analytical and interpretative phases of qualitative research. Using Mishlers typology as a starting point, it questions the practice of dividing storied data into small bits and suggests that there is a need for a wider range of grounded analytical practices practices based on the functions and operations of narrative itself. Specifically, this discussion proposes a series of models for collocating narrative data by examining how one researcher created frameworks that helped ground final conclusions within the broader narrative environment and context. The article concludes by suggesting that the collocation of narrative is essential for establishing validity and for creating holistic and connected research texts. KE YWORDS : coding, narrative research, qualitative methodology, storytelling Introduction The nature of narrative as research and the function of story, both fictional- ized and factualized (Atkinson, 1995a), is now a primary source of debate throughout academe (Clandinin and Connelly, 2000; Coles, 1989; Denzin and Lincoln, 2000; Doniger, 1998; Gubrium and Holstein, 1997; Lawrence- Lightfoot and Davis, 1997; Miles and Hubberman, 1994; Patton, 2001). These newly defined variations from the way we look at qualitative practice have led to a methodological shift in the field that stems from a core concept, namely that all qualitative work shares a reliance on the philosophical constructs and meaning making procedures derived from, and grounded in, narrative and discourse processes. 1 Questions and views regarding the role of narrative in qualitative research have risen to a crescendo during the past few years as practitioners wrestle ARTI CLE 231 Q R Collocation analysis: a method for conceptualizing and understanding narrative data Qualitative Research Copyright zooz SAGE Publications (London, Thousand Oaks, CA and New Delhi) vol. z(z): z1-z. [16S-;j1 (zoozoS) z:z; z1-z; ozj;66] ROBI N A. ME L L O University of Wisconsin-Whitewater with the nature and use of story in fieldwork, data analysis, and research prose (Casey, 1995). In an attempt to link the multiplicity of disciplines that use qualitative inquiry, and to more specifically identify and place the work of qualitative research within those practices, Clandinin and Connelly (2000) suggest using the term narrative inquiry to characterize the eclectic assortment of multidimensional and heterogeneous fields that are defined, globally, as qualitative investigation. Clandinin and Connelly observe that a practitioner today often plays the part of storyteller, author, editor, and artist; as Clifford and Marcus (1986) point out, the researcher is currently situated as the author of culture. We are currently, then, undergoing a shift toward understanding narrative as the new language of the qualitative method (Gubrium and Holstein, 1997). These ideas are predicated on the fact that we work in an atmosphere that, in the aftermath of postmodernism, has become more narrativized, grounded, idiosyncratic, and storied in its focus. In this context, practitioners are left grappling over what the story (or stories) of data are. Overall, the ques- tion is made more complex due to the fact that narrative is an interactional experience that is constantly negotiated and manipulated by both listener and speaker. A shift in perspective within the qualitative paradigm, toward narration and away from description, requires that we recognize, even more strongly, that our findings, analyses, and conclusions represent proximal and/or trans- actional representations of what data show (Bruner, 1990; Vygotsky, 1962). We have, therefore, reduced the role of our work away from the hierarchical position of creating conclusive knowledge to that of interpreting and story- telling personal experience; we do this with voices that are both idiosyncratic and dependent on individual perceptions. Since the essence of narrative is ephemeral and personal, we must seek ways to negotiate meanings and findings using the stories of data, varied per- ceptions of the field, and our creative work as writers of research discourse. This is what Gubrium and Holstein (1997) define as working on the border of reality and representation; the standpoint that the researcher takes between an individual perception (or story) of the lived experience and the intent of the narrators themselves. One of the most controversial edges (or borders) can be observed, for example, in the ways autobiography, or portraiture (Lawrence-Lightfoot and Davis, 1997), is now being used (Atkinson, 1995b; Patton, 1999). Here the entire research paradigm shifts toward the story- telling mode and away from an emphasis on the researcher as Other, or what it means to observe and report versus what it is to be a participant as well as the principal actor within the research investigation. In fact, it is within the realm of autobiographical, heuristic, and arts-based research that the boundaries can be observed most acutely where we see a redefinition of the work from qualitative research in general to narrative inquiry specifically (Clandinin and Connelly, 2000). Qualitative Research 2(2) 232 Narrative analysis The majority of current literature has focused on qualitative methodology or summative reports pertaining to specific studies. Discussions regarding the middle ground where we work on analyzing data and creating meaning are not as frequently presented. There remains a relative paucity of discourse regarding the conceptualintuitive processes we use to produce research texts. We should therefore try to identify more examples of scholarly models, forms, and formats used to create textual identities, research findings, and narrative conclusions. If stories are the stuff that data are made of (to mis- quote Shakespeare), then we need to take a concentrated look at how we use these data reflectively and analytically. Additionally, we must continue asking how best to practice analysis so that it remains grounded, authentic, and inclusive of the complexity found in discourse practices so that the narratives and their meanings remain intact. NARRATI VE AND MEANI NG MAKI NG Narrative is transactional and developmental; when we share narratives with others, insights and social knowledge evolve. This is communicative and is also a major way of disseminating information. In general, narrative func- tions as: (1) a structure of, and for, cognition and perception; (2) a negotiation of relationship(s) and connection(s) to self, others, and environment; (3) a process of social and cultural cohesion; (4) an artistic production and creation; (5) an educative inquiry and proximal experience; (6) a representational strategy for intrapersonal and interpersonal commu- nication systems. Organizing, analyzing, and discovering theoretical meanings from storied data can be challenging due to the nature of narrative because, like qualita- tive inquiry itself, it is iterative and evolutionary. The stories represented in our data are also highly eclectic and varied, and they leave us with questions concerning how best to work with, preserve, and respect their content and meaning. In addition, these data often do not follow proscribed literary/narrative forms. In fact, the oral and aural nature of nar- rative (with its idiosyncratic grammars, switchbacks, omissions, and repeti- tions), coupled with the non-narrative environment within which the story is set, belies the ability of any one individual to authentically represent or dis- cover a narratives significance and meaning. In addition, there is a tendency in scholarship to narrativize data 2 much too quickly so that they reflect liter- ary forms where none existed before. The negotiated, nonlinear, and interactional nature of storytelling thus Mello: Collocation data analysis 233 presents a dilemma to inquirers who want to ensure validity and clarity of findings. However, the problem lies in the fact that storied data are not simply ideas strung together, nor do they necessarily represent universal formats and concepts. Instead, our narratives contain unique individual worldviews, per- ceptions that are negotiated through the act of storytelling itself (Atkinson, 1995b; Bruner, 1990; Coles, 1989; Mello, 1997). The narratives we call data are illustrative, linguistically, of perceived human experience. As such, their meaning is dependent on context, time, place of telling, and audience response, as well as the tellers viewpoint, coupled with the researchers findings. Atkinson (1995a) and Mishler (1995) suggest we look to the functions of narrative itself to help create new models for analyzing storied data. They point out that, although all texts are different, and all methods dependent on varied traditions within individual disciplines, the use of narrative is driven by some core assumptions about the nature of story itself. Mishler (1995) pro- poses we look at a storys position, coherence, and structure (see Figure 1) when attempting to use it analytically. 3 CODI NG I S NOT NARRATI VE Historically, qualitative practice has been closely allied with and grounded in the ethnographic field. 4 However, the practice has expanded to include a wide range of academic disciplines and the outcome is that a plethora of method- ologies, including feminist practice, case studies, participantobservation, naturalistic study, ethnography, grounded study and action-research, all come under the umbrella term of qualitative inquiry; yet each one of these methodologies is epistemologically different. The disparity of ideology, along with the perceived need to gain legitimacy in academe (where quantitative methodology is still seen by many as the rule) has, ironically, led to a tenden- cy (among some) to standardize analytical practices. The reasoning behind this seems to be that if one can formalize, technologize, or institutionalize qualitative research, one can also more easily legitimize findings. As a result, we look for commonalities and/or connections among methods. One such process that has had wide popularity is euphemistically known as the index card method (Lofland and Lofland, 1984), a process whereby the researcher Qualitative Research 2(2) 234 Category Examples Referential use and order Time, illustration, description, order, background information, relationships. Textual coherence Folk tale, history, myth, reminiscence, entertainment, plots, narrative structure, literature. Utility and significance Negotiation, ceremony, therapy, epiphany. F I GURE 1. Major categories of narrative data in Mishlers (1995) typology. breaks down narrative data into sections, then divides the sections into bites; these can be as short as two to three words or as long as a few paragraphs but are almost always excerpted from longer narrative texts. Data bites are then reorganized according to perceived connections or overarching themes. This is a process often referred to as coding (Maxwell, 1996). Implied in this procedure are two assumptions: (i) narratives are too long and complex to use in their entirety; and (ii) data can best be understood and controlled when divided up into smaller units of discourse. Although this cut and paste style of data manipulation is pervasive, it is still imperative that we question the deconstruction and division of discourse. Does breaking up the story actually bring us closer to clarifying what data tell? Furthermore, given the wide disparity among theoretical and interdisciplinary standpoints regarding the nature of narrative and its place in research, we are obliged to look for processes that might increase coherence of narrative in analytical practice. When we divide narrative data into tiny sections for the purposes of analy- sis we are creating an artificial form that Gubrium and Holstein (1999) define as interlocutionary arrangements of external narrative patterns; these are used in order to create clarity and to show that emergent findings (Maxwell, 1996) have appeared. Of course, information, conclusions, and findings do not actually emerge on their own, like a mist rising from a lake of data bites; instead, they are part of the researchers intuitive/cognitive perception and emanate from serious attempts to manipulate, explore, and organize sets of data. As such, the way we create meaning is both creative as well as analytical. Another problem with the urge to divide data into discrete fragments is that we run the danger of diminishing or misinterpreting the nature of the narra- tive as a whole, and while the index card method is useful (computer software such as NUDIST and HypeResearch are based on this method), it is also prob- lematic as it has the potential for divorcing the original story from its human environment. Additionally, this method sets storied data apart from their tri- adic coproduction (Lincoln, 1996), i.e. the interpersonal construction of a story that occurs as a result of its performance, producer, and psycholinguis- tic transpiration. How, then, can we preserve the essential qualities, authenticity, and sub- stance of the narratives of inquiry while at the same time manipulating, ana- lyzing, and distilling data for the purposes of creating theory and perspective? It is suggested here that one of the answers is to utilize the natural functions of narrative as operational conditions or formats during analysis phases. In addition, if we use multiple forms for data analysis it is more likely that the findings and narratives of research (Clandinin and Connelly, 2000) will be grounded and, as a result, more likely to be valid. Furthermore, because sto- ries are negotiated exercises we must collocate our data in order to create tex- tual and cognitive bridges between the original tale and the resulting theory. Mello: Collocation data analysis 235 Collocating narrative data operationally The following is an example of how, in one arts-based study, collocation mod- els were used to explore meaning in order to create the story of research find- ings. Data from this study were grounded in the informants context, based on dialogue and discourse, and narrative in content (Mello, 2001b). In this study, positionality of the teller and the story was also important. By locating nar- rative data in different juxtapositions and with varied operations as thematic boundaries, the researcher had a wider opportunity to explore multiple mean- ings and become more cognizant of emergent findings. Utilizing more than one method for the location and contextualization of the narrative data also ensured the possibility of a valid final text, one that both illustrated the research question and context and was a product of the functions and opera- tions experienced. It is recommended that narrative data be analyzed using more than one operation simultaneously (see Figure 2) so that they are collocated in opera- tional formats and in relationship to each other. This allows for a more com- plex look at the way narrative functions in our informants lives and encour- ages us to create and express more complex and interwoven research texts at the summative stages of the work. Collocation also answers the question of how to preserve narrative integrity and is a response to Smeyers and Verhesschens (2001) call for a shift in situating narrative research more holistically in order to focus our work deeply within the realm of meaning making. THE TEXTUAL OPERATI ON The textual value and content of stories include factual information conveyed by the narrative along with their motifs (cognitive formats that link sections of the complete story together). These formats are also defined and categorized Qualitative Research 2(2) 236 Operation Examples Textual operation Mythic cycles, plot structures, motifs, patterns, subjective significance, and symbolism. Transactional operation Relationships between teller and audience, physical context, meanings shared or developed at the time of the event, intent of teller, entertainment and humor. Sociocultural operation Ceremonial information, cultural interpretation, social functions, and therapeutic functions and meanings. Educative operation Instruction, didactic information, rote memorization, learning and development, pedagogy, curricula, standards, and frameworks. F I GURE 2. Functions of story data: an operational format for coding and analysis. in terms of patterns such as mythic cycles and archetypal symbols. The tex- tual function has been used and extensively investigated by folklorists, espe- cially in the first half of the 20th century, through attempts at making stories easier to track and compare across cultures and ethnic boundaries. A com- parative analysis can also prove useful to the qualitative researcher not nec- essarily to discover any structural or universal patterns, but rather for the sake of looking at thematic congruence of core narratives in their original (and probably) disparate forms. Figure 3 provides one example of a theme/motif chart developed during the analytical phase of Mellos (2001b) study. In this case, stories were told aloud and student responses (in the form of stories) were also generated. The contents of these data were analyzed by identifying motifs and these qualities were then categorized using Proppian (1968) morphology. Texts were then compared to each other and partici- pants reactions to motifs were tracked throughout the entire length of the study. This textual analysis became useful in drawing conclusions and in understanding how the data functioned, not only in the lives of informants, but also in relationship to the academic discourse as well. Another attempt to examine the textual function of stories during this same study (Mello, 2001b) was made when interviews were analyzed according to theme and motif and then indexed according to topical content (see Figure 4). Two themes emerged. One centered on participants interest in flying and/or escape; the other pertained to their thoughts regarding common sense and its relation to intelligence and survival. Questions about these themes brought the researcher back to an examination of the motifs of flying and escape in Mello: Collocation data analysis 237 F I GURE 3. The textual operation: patterns of motifs and characters in stories. Types 16: story motifs List of stories using motifs List of stories across stories Male Characters listed by motif List of stories using male characters Female Characters listed by motif List of stories using female characters Relationship between characters Listed by motif or title Archetypes in stories Listed by motif or title the protocols, this procedure in turn led to findings regarding the concept of power and sense-making in childrens daily experience. Finally, the themes were used in a tiered and collocated approach in order to create a narrative research text that compared individuals discourse with theoretical perspec- tives and findings. THE TRANSACTI ONAL OPERATI ON The transactional operation is a function of the discourse process that takes into consideration the relationship of the speaker to the actual text as well as the meaning and relationships experienced between the listener and the teller (Bruner, 1990). Storied data, therefore, can be viewed in terms of their trans- formative and reflective operation as experienced in the lives of informants (Mello, 2001a). This modality examines what is happening to and among individuals involved in the original storytelling experience and investigates personal meanings as well as the impact of the story on listener/participants. One technique for working with this operation was to examine individual responses juxtapositionally by simply placing a group of stories together as one would in an anthology. Reading them through as a literate whole provid- ed the researcher with specific ideas of what meaning the stories had in the research context. Quite often the anthologizing of data works to create a holistic sense not unlike group conversation, where the control of a narrative jumps from teller to teller and subjects range widely and, like many conversa- tions, global experience is often more meaningful than any one select tale. The often varied and wide-ranging nature of narratives helps the group to develop an experience of knowing that has an opportunity to emerge within the transactional operation, even when the participants themselves may never have met. Participants discourses can be examined and re-organized by textual content as well. In the case of this specific study, reactions and responses to individual storied characters or production formats were also added into the analytical mix. The collocation of operational formats helped develop a layered understanding of the data and also assisted the researcher to be more inclusive of multiple perspectives. Figure 5 can be used as a schema for each story and event. The schema is Qualitative Research 2(2) 238 Textual analysis: indexing Narrative type/narrative theme/title Description of narrative event Thematic discourse Themes 14 Discourse and images used F I GURE 4. The textual operation: charting themes from an informants narrative. then used to create a new narrative that anthologizes the stories so that the researcher can view them together. This mega-analysis subsequently provides the researcher with an opportunity to compare each informants responses in their unedited version, and to form a clearer picture of how each text operates both concurrently and individually. THE SOCI OCULTURAL OPERATI ON Culturally specific information is vitally important to the development of the- oretical perspective. Additionally, when working with narratives of belief and identity, such as myths and spiritual texts, those tales that contain mytholog- ical symbols . . . which reverberate beyond the personal and into the collective realm (Atkinson, 1995b), care needs to be taken so that the cultural and social integrity of texts is preserved. Stockrocki (1994) even suggests that cul- tural texts are of such a sensitive nature that new forms for using and includ- ing them in research need to be created. These perspectives were taken into consideration when developing the fol- lowing matrix (see Figure 6) that examines the cultural source, as well as the basic plot outline, from myths and spiritual texts. By collecting or presenting stories from culturally competent expert tellers, placing different sources and cultural content together, and including the belief and identity statements of participants, the sociocultural meaning of stories became clearer. In addition, there was a greater probability that the integrity of the tales was preserved. Mello: Collocation data analysis 239 Informant Protocol Protocol Name Text of story told to researcher Text of story told to researcher F I GURE 5. The transactional operation: multiple response model. Story Origin Participants belief statements Synopsis/annotation and actions Title Discourse of belief or attitude Overview of plot and/or intent and information from field notes re: actions and reactions to story Title Overview of plot and intent Title Overview of plot and intent Title Overview of plot and intent Title Overview of plot and intent F I GURE 6. The sociocultural operation: annotated list by cultural origin. THE EDUCATI VE OPERATI ON Narrative discourse was the first literacy, as well as the first teaching, method. The telling of stories has ancient pedagogical roots and is one of the founda- tions of education, i.e. the rhetorical interaction between instructor and student. Telling stories and talking to others, then, can be understood as a primary tool for learning. In addition, since narrative is a way of knowing, a tool for exploration, and a key component in the construction of knowledge, it is seminal to the work of the researcher. Teacherresearchers, such as Paley (1990) and Gallas (1998), for example, have used the educative property of stories to create powerful research-based narratives pertaining to the sense making and personal meaning of school children. In order to explore the relationship between narratives and learning, Mello (2001b) acknowledged that questions asked by the informants were in some way indicative of an individuals inquiry processes. Questions and perceptions originated by informants were felt to be signifiers of curiosity, indicating that participants were attempting to make meaning from the research events by using narrative. Queries were grouped and located together to see if any larg- er ideas or concepts might evolve. Also, when informants told spontaneous anecdotes about things they had been thinking about or learning about, the stories produced were considered to be evidence of interest and learning. The following format (see Figure 7) was designed to track all questions, formative thinking stories, and queries. Each of these was placed in different configurations and examined in order to conceptualize more deeply the impact of the research process, stories, and narrative on the individuals worldview. Conclusions The need for sharing processes, templates, and maps used to examine and cre- ate narrative coherence and meaning in research texts is essential. As we search for ways to create scholarship from personal and individual experi- ence, we must also examine how we ground our work in the narrative landscape of accounts (Gubrium and Holstein, 1999); therefore, it has been Qualitative Research 2(2) 240 F I GURE 7. The educative operation: questions as responses within the research context. Session #: Informant Date: Time: Place: Pertinent field notes QUESTIONS and THINKING STATEMENTS
suggested in this article that this be accomplished by creating and using a
variety of analytical formats that reflectively collocate narrative data for the purposes of creating research texts and coherent findings. The operations and systems presented here are few and represent one researchers work only. They are not intended to be definitive nor even arche- typal. Instead, they are presented in order to encourage a wider discourse regarding the analytical processes narrative inquirers participate in when working with stories in the field. They are also intended to illustrate how we can rethink the practice of analyzing data by dividing them into small bites, especially in the early stages of the work when a researchers perception of meaning is usually nascent. Using cohesive narrative sections, or even stories in their entirety, allows the researcher to preserve the integrity of the narra- tive while at the same time offering greater opportunities for understanding. Also, multiple investigations of narrative functions are recommended when analyzing discourse and its process. By collocating, i.e. combining, juxtapos- ing, and examining narrative data in terms of its several organic operations, we ground the work closely to its source(s) and include both researcher and informants meaning in the story of our findings. Gubrium and Holstein (1997, 1999) suggest that in naturalistic and ethno- graphic investigations storytelling has become the new language of method- ology. They propose that additional paradigmatic understandings of narrative need to evolve from our work. Denzin (1998) also supports innovative posi- tions on interpretive practice understandings that are negotiations between the artful and the conditional. Denzin further observes that qualitative prac- titioners stand on the horns of a dilemma between constructing . . . a representation of the lived experience (p. 407) and falling into the trap of creating texts that are solely reflections of one persons biases, beliefs, or agendas. To meet this challenge, we must carefully place the narratives and perspectives of others alongside our own. We can accomplish this, or at least attempt it, through connecting and collocating data. In doing so, the researcher becomes the storyteller, a bridge-builder working to link the use and production of stories in the field together with the analytical discourse of research literature. NOTE S 1. For purposes of this discussion, narrative, discourse, and storytelling refer to those language processes that incorporate linguistic activities conducted between indi- viduals or groups and without the use of scripts, memorized texts, or other liter- ary devices and that include dramatic skills such as representational gesture, characterization, vocalization, and mimetic action. 2. Historically, researchers and authors have changed the nature of orally told nar- ratives during the process of transcribing and authoring them. Douglas (1996) calls this the narrativization of text and observes that most orality does not actually follow dramatic structures of Western written literature. 3. Mishlers (1995) typology, although originally designed to assist scholars in Mello: Collocation data analysis 241 conversing across disciplines, is used here as a reference point for the creation of data analysis formats. 4. The word ethnography derives from the Greek ethnos meaning the people and graphikos, a written story or painting. Therefore ethnography can be understood as a published story that provides the reader with a word picture (or narrative) of and about a group of people. RE F E RE NCE S Atkinson, P. (1995a) Some Perils of Paradigms, Qualitative Health Research 5(1): 11725. Atkinson, R. (1995b) The Life-Story Interview. Thousand Oaks, CA: Sage. Bruner, J. (1990) Acts of Meaning. Cambridge, MA: Harvard University Press. Casey, K. (1995) The New Narrative Research in Education, Review of Research in Education 21: 21153. Clandinin, D.J. and Connelly, F.M. (2000) Narrative Inquiry. San Francisco, CA: Jossey- Bass. Clifford, J. and Marcus, G. (eds) (1986) Writing Culture. Berkeley, CA: University of California Press. Coles, R. (1989) The Call of Stories. Cambridge, MA: Harvard University Press. Denzin, N.K. (1998) The New Ethnography, Journal of Contemporary Ethnography 27(3): 40516. Denzin, N. and Lincoln, Y. (2000) Handbook of Qualitative Research. Thousand Oaks, CA: Sage. Doniger, W. (1998) The Implied Spider. New York: Columbia University Press. Douglas, M. (1996) Children Consumed and Child Cannibals: Robertson Smiths Attack on the Science of Mythology, in L.L.Patton and W. Doniger (eds) Myth and Method, pp. 2951. Charlottesville, VA: University of Virginia Press. Gallas, K. (1998) Sometimes I Can Be Anything. New York: Teachers College Press. Gubrium, J.F. and Holstein, J.A. (1997) The New Language of Qualitative Method. New York: Oxford University Press. Gubrium, J.F. and Holstein, J.A. (1999) The Border of Narrative and Ethnography, Journal of Contemporary Ethnography 28(5): 56174. Lawrence-Lightfoot, S. and Davis, J.H. (1997) The Art and Science of Portraiture. San Francisco, CA: Jossey-Bass. Lincoln, B. (1996) Mythic Narrative and Cultural Diversity in American Society, in L. Patton and W. Donniger (eds) Myth and Method, pp. 16376. Charlottesville: University of Virginia Press. Lofland, J. and Lofland, L. (1984) Analyzing Social Settings. Belmont, CA: Wadsworth. Maxwell, J. (1996) Qualitative Research Design. Thousand Oaks, CA: Sage. Mello, R. (1997) Creating Pictures in my Mind: A Qualitative Study of Childrens Responses to Storytelling in the Classroom, The Primer 26(1): 411. Mello, R. (2001a) The Power of Storytelling: How Narrative Influences Childrens Relationships in the Classroom, International Journal of Education and the Arts, 2(1): 54856. Mello, R. (2001b) Cinderella Meets Ulysses, Language Arts 78(6): 54855. Miles, M. and Hubberman, A.M. (1994) Qualitative Data Analysis. Thousand Oaks, CA: Sage. Mishler, E.G. (1995) Models of Narrative Analysis: A Typology, Journal of Narrative and Life History 5(2): 87123. Qualitative Research 2(2) 242 Paley, V.G. (1990) The Boy Who Would Be a Helicopter. Cambridge, MA: Harvard University Press. Patton, M. (1999) Grand Canyon Celebration. Amherst, NY: Prometheus Books. Patton, M. (2001) Qualitative Research and Evaluation Methods. Thousand Oaks, CA: Sage. Propp, V. (1968) Morphology of the Folktale. Austin: University of Texas Press. Smeyers, P. and Verhesschen, P. (2001) Narrative Analysis as Philosophical Research: Bridging the Gap between the Empirical and the Conceptual, International Journal of Qualitative Studies in Education 14(1): 7184. Stockrocki, M. (1994) A School Day in the Life of a Young Navajo Girl: A Case Study in Ethnographic Storytelling, Art Education 47(4): 619. Vygotsky, L. (1962) Thought and Language. Cambridge, MA: MIT Press. ROBI N A. ME L L O is an assistant professor of Educational Foundations at the University of Wisconsin-Whitewater, as well as a storyteller. Her research focuses on how narrative and arts-based methodologies in research and teaching affect human development and learning. Address: Department of Educational Foundations, University of Wisconsin- Whitewater, Winther Hall 6053, College of Education, 800 West Main Street, Whitewater, WI 53190, USA. [email: mellor@uwwvax.uww.edu] Mello: Collocation data analysis 243