What Does Rejection Have to Do With It? Toward an Innovative, Kinesthetic Analysis of Qualitative Data

Abstract: Responding to the need for detailed descriptions of analytical practices in transdisciplinary projects and to showcase a way to use qualitative data in secondary analysis, the authors delineate their respective analytic processes. The social scientist used constructivist grounded theory analysis and the dance choreographer used an innovative kinesthetic analysis. In the authors' attempts to integrate their analyses, they discovered a new analytic direction—examination of the "paused" and/or "rejected" data.

Key words: secondary analysis; kinesthetic analysis; dance; analytic innovation; performative social science

3. Our Analytic Framework—Moving From Individual Analysis to Integrative Analysis

In this article, we share our analytic processes examining a qualitative dataset based on young (in their 20's and 30's) women's experiences and reflections of their relational lives. One author—a social scientist—analyzed the data using constructivist grounded theory analysis (CHARMAZ, 2000) and the other author—a choreographer—re-examined the same data set using kinesthetic analysis. We are in the process of "layering" our respective analysis, attempting to offer an integration of the separate analyses. In so doing, we discovered a potentially ground-breaking analytical tool—examining the kinesthetic rejection of data. Unlike most of the previous work in this area, we offer detailed description of our specific analytic practices and our ideas while our analysis is in progress. [1]

The analytic technique we are discovering and honing grew out of our funded project entitled, Toward Innovative and Transdisciplinary Methodologies: Re-presenting Social Science Data Through Dance. Central to the project is the choreographer's analysis of the social scientist's two datasets examining women's relational lives (one focuses on weddings/marriage and the other on single women). The end goal of the larger project is a fully-realized dance concert. In other words, the data is being "re-examined" or "read" as an impetus for a dance production. By producing a dance concert grounded in ordinary women's experiences, we hope to affect change in our respective fields and in our communities (see DURHAM-DeCESARO & SHARP, in preparation, for more background of the larger project). Both the choreographer and the social scientist bring to the project considerable background in their respective fields. [2]

Two of the social scientist's qualitative data sets are employed in the project. One dataset is from a grounded theory study examining 18 young (aged 19-32), white, US women's weddings and transitions to being a wife. Data were collected through two focus groups and individual interviews (for more information about the methodology, see SHARP, 2012a; SHARP et al., 2008). The other data set is a grounded theory study examining 35 women (aged 25-39) who did not want to marry and/or have children (for more information about the methodology, see SHARP, 2012b). In this study, face-to-face individual interviews were conducted. For both projects, all interviews were audio recorded and transcribed verbatim. The choreographer was provided with all the transcripts and audio recordings. The two datasets combined contain more than 1000 double-spaced pages of data and 45 hours of audio recordings. [3]

For an excellent discussion of broader issues related to performative social science, see FQS 9(2), edited by JONES et al. (2008). We argue that we are extending that dialogue through sharing our specific analytic practices. By allowing the choreographer to re-examine the data, our project implicitly endorses the use of qualitative data for secondary analysis. In our case, the social scientist did not make her datasets "public" in the sense that anyone could engage in the process of re-examining the dataset. Instead, the social scientist gave permission for the choreographer to examine the data as the two of them work together. Our project, thus, points to the broader issue of the use of secondary datasets for qualitative research. [4]

The use of qualitative datasets for secondary analysis engenders considerable debate (see FIELDING, 2000 for a review, as well as FQS special issues, dedicated to archiving and secondary analysis: CORTI, KLUGE, MRUCK & OPITZ, 2000; CORTI, WITZEL & BISHOP, 2005; BERGMANN & EBERLE, 2005; VALLES, CORTI, TAMBOUKOU & BAER, 2011). Although making quantitative datasets public is a long established and productive practice in the social sciences, qualitative secondary analysis is virtually non-existent in many social science fields in the U.S, including the field of human development and family studies (the first author's field). The non-practice is partially explained by the complicated and ethical issues accompanying qualitative datasets—issues that the use of quantitative data sets are not likely to have. For example, the personal and private nature of what is shared in qualitative research, the depth of information in qualitative datasets, and the possibility of participants' identities being revealed in the original data bolster strong arguments for not endorsing the practice of secondary data analysis. At the same time, arguments for secondary qualitative data include the inefficiency of time and knowledge, that the scholars' interpretation of the original data is never fully scrutinized, and the ability to engage in historical analysis is limited. All of the aforementioned complications are present in our project. It is our hope that our project encourages more social scientists to consider the possibilities of sharing data sets with dance choreographers and other artists and more choreographers to consider the possibilities of working with social science qualitative data sets. [5]

Other social scientists and dance choreographers have led the way for transdisciplinary projects and opened up possibilities for "partial" secondary data analysis of qualitative datasets. We use the word "partial" to describe the practice whereby the choreographer did not re-examine the original data. Instead, the social scientist provided the artist with findings (already analyzed) and the dancer(s) "performed" the findings. Thus, the choreographer did not examine the original data but created a dance in response to a distilled set of themes. There are other examples whereby the choreographers were provided with portions of the original data—e.g., a subset of transcripts. In this way, the social scientist was still "distilling" or paring down the original data set. Our project, on the other hand, gave the choreographer access to all transcripts, allowing her to analyze all of the original data. [6]

In addition to lack of documentation concerning comprehensive secondary data analyses, descriptions of processes by which the artists and social scientists work together in their analysis are largely missing from the literature. One text in particular, "Dancing the Data" (BAGLEY & CANCIENNE, 2002) is a useful illustration. In the book, CANCIENNE (dance choreographer) writes about her process as a choreographer interacting with the data provided to her by her research partner. Although her description provides insight into how she and her research partner (social scientist) independently analyzed the data, there is no record of how they influenced each other's perspectives or choices as they were working. The reflections were based on what they learned after the performance. [7]

Additionally, the choreographer in the project referenced above wrote about wanting to "bring the text to life" (p.7) using primarily literal (mimetic) movement. The choreographer justified her use of mimetic movement by noting that performing social science was innovative on its own; she considered adding abstract or avant-garde movement unnecessary. Indeed, this was an accurate description at the time their project was conducted. Ten years later, as we are conducting our project, we argue that limiting dance to the literal movement in relation to data is no longer necessary. The choreographer in our project pushes for abstract movement. [8]

Abstract movement is a term used by choreographers and dancers to describe movement that symbolizes, rather than represents, an idea. Abstract movement may bear some resemblance to mimetic movement or gesture but is not designed to communicate linguistically (LAVENDER, 1996). Choreographers choose to work with abstract movement for a variety of reasons; in this case, the choreographer used abstract movement in order to allow her audiences to discover their own meanings within the dances. The choreographer in this project argues that this kind of work on the part of audience members encourages more interaction between audience, dance, and dancers. [9]

Although the abstract movement has been a challenge in our collaboration, an unexpected outcome of using abstract movement has been the development of an illuminating dialogue that has been critical in allowing us to understand each other's research processes and inform our own ways of making meaning from the data. This was a critical discovery as we moved through our own project and one that we feel is crucial to contemporary publication on transdisciplinary research. In order to promote the viability and visibility of transdisciplinary arts/social science research, we argue that clear and thorough descriptions of transdisciplinary collaboration are urgently needed (GARRETT-PETTS & NASH, 2012). The present article seeks to establish a foundation for such information. [10]

3. Our Analytic Framework—Moving From Individual Analysis to Integrative Analysis

As an extension of the previous literature, we delineate our respective analytic processes and share our plans about how we will integrate our separate analysis. The initial step in our project was to separately examine/read/encounter/analyze the datasets. We then embarked on the process of trying to integrate the data analyses. [11]

In this section, we first share a detailed step-by-step process of our individual analysis procedures and then we discuss our examination of the choreographer's analysis and point to the provocative potential of considering a "rejection" of data based on kinesthetic responses. As we discuss our analysis, we give attention to the ways in which both the arts and the social science gain from the "borderland borrowing." In this way, we are mindful of the ways in which reciprocity is present in our project. [12]

Focus groups interviews, focus group debriefing sessions, and individual interviews were transcribed verbatim. Memos (researcher's analytic musings and questions) about group dynamics of each focus group were typed and inserted at the end of transcripts. I (the social scientist) used a modified constructivist grounded theory approach (CHARMAZ, 2000) to examine the data. CHARMAZ' adaptation of grounded theory methodology draws from both postmodernism and post-positivism sensibilities. The analytic technique is primarily inductive/emergent whereby the analysis addresses participants' ideas, behaviors and interactions present in the data. I became immersed in the data over several months. As I analyzed, I continually asked "what am I learning? What are the participants saying/expressing?" and "Am I doing justice to their stories?" I attempted to keep an open mind throughout the analysis process. [13]

I listened to audio recordings multiple times, read interviews more than five times each, did line by line coding, and later refined the coding and began abstracting. CHARMAZ recommends that researchers consciously use gerunds when coding, thereby emphasizing the ways in which participants are active in their narratives. I also drew on LLOYD, EMERY and KLATT's (2009) analytical practices. LLOYD et al. identified larger discourses operating and searched for instances of both compliance/collusion and resistance to the larger discourses. Analysis focused on contradictions, larger structures operating (e.g., "bridezilla"¹⁾), gendered enactment, identity, and I tried to be conscious of the homogenous experiences women were discussing as well as diverse positions of the participants (e.g., age, cost of the wedding, etc.). As I analyzed the data, I considered how larger structures were operating in participants' expressions (RISMAN, 2004). [14]

For the purpose of this article, I draw on examples from only one of the studies—the wedding study. I began examining data by developing a single descriptive indicator capturing the first emergent idea (e.g., stopped planning the wedding) and then proceeded to examine more data, comparing the indicator in the new data to the initial indicator. If the data did not fit within the existing indicator(s), a new indicator was developed (e.g., "asserting my desires for a small wedding"). This process continued until all data were analyzed. The indicators were then grouped into concepts—which are labels associated with several indicators. For example, indicators of "stopped planning the wedding" and "asserting my desires for a small wedding" were part of concept "managing/responding to wedding stress." Categories were then developed by classifying concepts, asking "how are the concepts fitting together?" Categories developed reflect an integration of the participants' explanations/descriptions and my interpretations (DESANTIS & UGARRIZA, 2000). [15]

I deliberately use the word "read" to describe my interaction with the data. Read is a term contemporaneously used by dance scholars to indicate a way of interpreting that is both traditional, as in visually seeing words and analyzing their meanings, and provocative, as in using a physical (or kinesthetic) approach to engender a non-dualistic understanding of the mind and body. As the choreographer, I accepted prior to reading the data that my analysis of it would be subjective, as making dance is an intimate and personal process necessitating a desire to embody the original stimuli (data). To clarify, a choreographic stimulus is any particular thing, idea, sound, or image that becomes the basis for movement creation and exploration. [16]

I did not approach the data in the same way as the social scientist; I was not concerned with analyzing all of the data and pulling from that analysis central themes or theories meaningful to the entire dataset. Instead, I read the data for statements that would be useful choreographically. This meant that, I would be subjectively selecting sections of the data based on my perception of the potential for those sections to generate meaning when used as the stimuli for dance. [17]

I initially listened to the audio files while reading the transcripts; I stopped doing that quickly. This choice had to do with the emergence of rejection as an unexpected challenge to my analysis. When listening to the audio interviews, I found myself negatively influenced by the participants' use of particular intonations and vocal affectations. On more than one occasion, I realized that I was rushing through the analysis of an interview because I was adversely reacting to a participant's voice. Retrospectively, it is clear that I found more freedom in my analytical approach to the data when I stopped listening to the audio interviews and examined transcripts instead. [18]

I also kept a narrative journal in which I wrote my reactions to reading the interview transcripts. Initially, I thought I would use the journal primarily to help inform my selection of particular parts of the data for use as choreographic stimuli, which is what I was specifically looking for when reading the data. However, the journal actually became very valuable as a descriptive illustration of my analytic process. Several excerpts (based on the single women dataset) from my journal follow, including highlighted statements that I thought held the most potential as stimuli for movement invention.

"This participant takes a lot of time in her responses. She has heavy pauses. She is less flippant about her perceptions than other participants. There is a weightedness to her decision not to marry that is, I feel, significantly qualified by her tone of voice when talking about her parents' divorce, and particularly speaking about how her mother left her father for a man who never left his wife. There is a great sadness to that. It makes me think of shoulders pressed down and upper body curved. Like a woman crushed under the weight of it all. There is a great story here about finding freedom only to discover the freedom was an illusion, or at least that the freedom is much different than expected.

I was thinking about the ways in which the participant's responses lead me to envision what our society puts upon individual women. If we are feminists, we are given combat boots. If we are brides, we are given white, poofy dresses. If we are mothers, we are given babies and disheveled clothes. If we choose to be single and childless we are given a gonzo nose.

I get to the part when [social scientist] has just asked her to discuss her perceptions of marriage and motherhood. Almost right off, she states that, relating to having children, she wouldn't be good at it. Immediately this belies her assumption that she SHOULD be good at it. This is such a critical point. Our cultural context situates women in a position of thinking they should inherently be good caregivers.

One participant mentions that she currently conceives of having a child as sort of a trophy pursuit. She has a PhD. Now, she'll show the proverbial world that she can also take care of a child. She jokes that this is a sad way to think about it. Emotionally and experientially, I am not incredibly invested in this interview. I don't share this woman's experience of not wanting children." [19]

As mentioned above, I highlighted statements from my narrative journal to assist in identifying choreographic stimuli. Because I needed to create several distinct dance works from the data, I paid particular attention to highlighting not only movement-based statements, but also vivid and meaningful images that resonated with my own lived experiences. This illustrates the kind of subjectivity that was operating in my analysis of the data. Essentially, as I moved forward in the project, it was impossible for me to avoid the fact that I would be physically embodying whatever portions of the data I selected for use as choreographic stimuli. Therefore, my reading of the data was always done with the knowledge that I would eventually be performing it. [20]

This was a strange situation. I knew prior to beginning my reading of the data that I would not be looking for things like trends, patterns, or emergent themes. However, I did not anticipate how I would reject some of the data precisely because I had no interest in exploring it kinesthetically. This is different than rejecting portions of the data because it had little to offer as choreographic stimuli. What I am discovering is that my rejection of some data is actually rooted in my physical reaction to thinking about having to embody it. This is an insight of particular importance. While I would not argue that my reaction is more important than the data itself, I am curious about the possibility of doing a meta-analysis of a kinesthetic approach to analyzing data as a way of making meaning of the data. [21]

Even with our distinct analytical approaches, rejecting portions of the data surfaced in both our analytic processes. One major consideration at this point in time is to go back through our analyses and uncover patterns in the data that was left out, especially in the case of the choreographer, who tended to "reject" large portions of the data set (i.e., entire transcripts). [22]

Rejection is both simple and complicated—in this project, we discovered how the choreographer was "unable" to use large portions of the data due to its lack of kinesthetic transferability, meaning the choreographer felt the data was not useful as choreographic stimuli. Moreover, that kinesthetic rejection was coupled, in some instances, with ideological rejection based on the choreographer's personal experiences and theoretical training. For the social scientist, she describes her process as a "long pause" when thinking about her relationship with the data. She had begun analyzing the wedding study data several years ago but terminated analysis (not a conscious decision) after becoming physically nauseous when reading some of the participants' transcripts. She later returned to the analysis for this project. [23]

We argue that, while the work both of us have done in our separate analyses is valuable, systematically examining our shared aversion to the data has the potential to inform and re-form the ways in which researchers understand subjectivity, analysis, and representation. In order to offer readers a clear record of how we will proceed, we have created a table (see Table1) to delineate our process of reanalyzing our own analytic processes, including our pauses and rejections.

We argue this article contributes to the growing enterprise of transdisciplinary work through our detailed descriptions of (and examples from) our analytical processes. We re-visited the broader debates about using qualitative data in secondary analysis. Through sharing the social scientist's and the choreographer's respective analytical practices as well as our step-by-step plans for a more integrative analysis, we attempted to provide a useful framework (and/or jumping off point) to both inspire other scholars to consider transdisciplinary projects and for scholars already engaging in the challenging and rewarding work of transdisciplinary projects. [25]

Funding for this project was provided by the Office of the Vice President for Research, Texas Tech University. The wedding study was funded by a seed grant from the College of Human Sciences, Texas Tech University. The single women research was funded by the Anthony Marchionne Small Grants Program for the Scientific Study of Human Behavior, c/o Department of Psychology, Washington State University, PO Box 644820, Pullman, WA 99164.

1) In the United States, "bridezilla" is a colloquial term used to describe a bride who is overly demanding. <back>

Bagley, Carl & Cancienne, Mary Beth.(2002). Educational research and intertexual forms of (re) presentation: The case for dancing the data. In Carl Bagley & Mary Beth Cancienne (Eds.), Dancing the data (pp.3-19). New York: Peter Lang.

Charmaz, Kathy (2000). Grounded theory: Objectivist and constructivist methods. In Norman K. Denzin & Yvonna S. Lincoln (Eds.), Handbook of qualitative research (2nd ed., pp.509-535). Thousand Oaks, CA: Sage.

Durham-DeCesaro, Genevieve & Sharp, Elizabeth (in preparation). Toward innovative and transdisciplinary methodologies: Re-presenting social science data through dance. Champaign, IL: Common Ground Publishing.

DeSantis, Lydia & Ugarriza, Doris (2000). The concept of theme as used in qualitative nursing research. Western Journal of Nursing Research, 22(3), 351-372.

Fielding, Nigel (2000). The shared fate of two innovations in qualitative methodology: The relationship of qualitative software and secondary analysis of archived qualitative data. Forum Qualitative Sozialforschung / Forum: Qualitative Social Research, 1(3), Art. 22, http://nbn-resolving.de/urn:nbn:de:0114-fqs0003224 [Accessed: December 12, 2012].

Garrett-Petts, Will & Nash, Rachael (2012). Making interdisciplinary inquiry visible: The role of artist-researchers in a 10-year community-university research alliance. The International Journal of the Arts in Society, 6(5), 43-54.

Lavender, Larry (1996). Dancers talking dance: Critical evaluation in the choreography class. Champaign, IL: Human Kinetics.

Lloyd, Sally; Emery, Beth & Klatt, Suzanne (2009). Discovering women's agency in response to intimate partner violence. In Sally Lloyd, April Few & Katherine Allen (Eds.), Handbook of feminist family studies (pp.264-280). Thousand Oaks, CA: Sage.

Risman, Barbara (2004). Gender as social structure: Theory wrestling with activism. Gender & Society,18(4), 429-450.

Sharp, Elizabeth (2012a). "Once you take the veil off, it's still work": Young women's experiences of becoming wives. Paper presentation at "Marriage," The Ustinov Seminar, Durham University, Durham.

Sharp, Elizabeth (2012b). Ever single women in their 20's and 30's who resist conventional family ideologies. Poster presented at the Annual Meeting of the National Council on Family Relations, Phoenix, Arizona, November 3, 2012.

Sharp, Elizabeth; Chan, Cara; Parker, Cassandra; SoRelle-Miner, Danielle, Howard, Elise & Keyton, Keyton (2008). Happily ever after? Young women's transition to marriage. Poster presented at the Annual Meeting of the National Council on Family Relations, Little Rock, Arkansas, November 8, 2008.

Elizabeth SHARP is an Associate Professor of human development and family studies and an affiliate faculty member of Women's Studies at Texas Tech University and is currently a Honorary Fellow at the Institute of Advanced Study, Durham University, England. She is the chair of the Feminism and Family Studies section of the National Council on Family Relations. Elizabeth's training in qualitative methods includes mentoring from an expert in descriptive phenomenology and a scholar-in-residence at the International Institute for Qualitative Methodology in Alberta, Canada. She has taught graduate seminars on qualitative methodology and has published qualitative research in the Journal of Marriage and Family, Journal of Family Issues, Family Relations, Journal of Marital Therapy, and Sex Roles and her work has been cited in several media outlets, including the New York Times, the Toronto Star, and Women Forbes. The high visibility of Elizabeth's work was also recognized by Sage Publishers for having one of the most downloaded articles in the JSTOR/Journal of Family Issues in 2011. She is a co-editor of a special issue on "Qualitative Methodology, Research and Theory in Family Studies" in the Journal of Family Theory and Review (2012) and has been the principal investigator on several Texas Tech University grants and a National foundation grant for qualitative research projects.

Honorary Fellow, Institute of Advanced Study, Durham University, UK
Associate Professor, Human Development & Family Studies
Affiliate Faculty Member, Women's Studies
Texas Tech University
Box 1230, Lubbock, Texas, USA 79409

Genevieve DURHAM DeCESARO is the Head of Dance and Associate Chair of Theatre and Dance at Texas Tech University in Lubbock. She has presented her scholarship nationally and internationally and her choreography has been commissioned and performed by professional companies, colleges, and universities, including an honorary presentation at the John F. Kennedy Center for the Performing Arts in Washington, D.C. She is a former Chair of the Texas Commission on the Arts Grant Review Panel and is an elected Regional Director of the American College Dance Festival Association. Among her recognitions at Texas Tech University are receipt of the Alumni Association's New Faculty Award, the Student Government Association's Outstanding Professor Award, the President's Excellence in Teaching Award, recognition by the national honor society Mortar Board as an Outstanding Professor, and an invitation to be the keynote speaker at the 2010 Phi Beta Kappa induction ceremony. Genevieve is also a founding member of Flatlands Dance Theatre, Lubbock's first professional dance company, an elected member of the Texas Tech University Teaching Academy, and will be recognized in Fall 2012 as an Integrated Scholar at Texas Tech University.

Head of Dance
Associate Chair
Dept. of Theatre and Dance
Texas Tech University
Box 79409, Lubbock, Texas, USA

Sharp, Elizabeth A. & Durham DeCesaro, Genevieve (2013). What Does Rejection Have to Do With It? Toward an Innovative, Kinesthetic Analysis of Qualitative Data [25 paragraphs]. Forum Qualitative Sozialforschung / Forum: Qualitative Social Research, 14(2), Art. 4,
http://nbn-resolving.de/urn:nbn:de:0114-fqs130245.

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

ISSN 1438-5627

Creative Commons Attribution 4.0 International License

Step	Choreographer data	Social scientist data	Analysis
1.Re-examine data not used in the choreography and social science findings.	Examine journal, entire transcripts not used, and portions of transcripts not used. Record reasons for rejection of all data not used in choreography.	Examine memos, re-visit all interviews for data not accounted for in findings.	Social scientist will attempt a close analysis of choreographer's journal and transcript usage.
2. Identify patterns of the rejected data.	Compare/contrast the data used for inspiration of the dance and data "rejected."	Compare/contrast model with countercases.	Both choreographer and social scientist will consider patterns and discuss emerging insights with each other. If necessary, the social scientist will interview the choreographer about her analysis of the rejected data.
3. Discuss patterns discovered in Step 2.	What do the patterns reveal? How are the two types of rejected data (i.e., stimuli VS ideological/kinesthetic) similar and different? What content is left out? In what ways does this or does this not matter? How might we alter the choreographer's analytic process? Should we have considered one participant at a time instead of the choreographer working alone with the raw data?	What do the patterns reveal? What are the reasons for rejection? Is it all because the data was not germane or are ideological or other constraints operating? How can this insight be used to modify the dance or subsequent analyses in other projects?	Both choreographer and social scientist will give careful consideration to the choreographer's two types of rejected data (i.e., stimuli rejection & ideological/kinesthetic rejection) & the reasons the social scientist rejected the data. What does comparing & contrasting reveal both in terms of content but in process of analysis/reading data?
4. Integration	How can insights from dance and social science analysis be integrated to help other choreographers and artists read social science data? What are some recommended practices? What should be avoided?	How can insights from dance and social science analysis be integrated to help social scientists consider their data when working with artists? How can the process of working with choreographers or other artists be improved? What are some recommended practices?	The choreographer and the social scientist and an outside consultant with a background in transdisciplinary projects will attempt to integrate the layered analytical processes and products. Does any rejected material overlap between the social scientist and the choreographer? In what ways and how is this related to our larger feminist framing of the project? How are the patterns different and how is this difference related to our training and epistemological commitments and goals?