The Participant as Transcriptionist: Methodological Advantages of a Collaborative and Inclusive Research Practice

Introduction

This article documents an innovative approach to interview-based research that the first author discovered as she struggled to transcribe interviews despite her hearing disability. In two research studies, she hired the participants to transcribe their own interview tapes. This approach, which we label the participant-as-transcriptionist method, arose out of the author's need to find an alternate way to engage in research that accommodated her hearing disability. In the participant-as-transcriptionist method, the participant serves as the transcriptionist, with editorial control to create a transcript from an interview. Based on the results of the two studies, we realized the methodological advantages provided by the participant-as-transcriptionist method and its appropriateness for a wide range of researchers, with or without disabilities.

The participant-as-transcriptionist method presents a credible and effective methodological approach for interview-based research that relies on transcription. First, the participant-as-transcriptionist method is inclusive for a range of researchers: those who are hard of hearing, those with other disabilities, and those who lack the time or other resources to create their own transcripts. Second, the participant-as-transcriptionist method can incorporate a sense of collaboration in the researcher-participant relationship. Third, participant-transcriptionists can create quality transcripts that represent the participant's voice. These three main methodological advantages are beneficial for both researchers and participants as we articulate in the following sections.

Using the participant-as-transcriptionist method

Our research team includes a researcher who has used the participant-as-transcriptionist approach in multiple studies (first author), the participant-transcriptionist from one study (second author), and a faculty advisor (third author). Both the researcher (first author) and the participant-transcriptionist (second author) kept fieldnotes during a research study in which they used the participant-as-transcriptionist method. While confidentiality issues prevent us from revealing particular details about the interview study or providing a full citation to that research study, the resulting fieldnotes, audiotape, and participant-transcript provide evidence to support our investigation of the potential methodological advantages of the participant-as-transcriptionist method. By using the participant-as-transcriptionist approach in a research study, we realized that the approach can (a) accommodate differential abilities of researchers, (b) engage participants as collaborators in research, and (c) yield quality transcripts.

An inclusive research experience

The first author is a hard-of-hearing graduate student. Through her research courses and other research opportunities, she has experienced limitations posed by traditional research practices. Identifying and overcoming these limitations have provided necessary preparation to successfully implement an independent research study, which is a requirement in her graduate program. One of the limitations that she has faced involves the documentation of interview-based research. In order to provide a detailed analysis of an interview, researchers typically rely on a transcript of that interview. Lapadat (2000) argues that the process of transcription promotes intense familiarity with the data, which leads to the methodological and theoretical reasoning that is essential to interpretation. What happens when the researcher is not able to transcribe in order to produce a transcript? For the first author, this question led her to the participant-as-transcriptionist method as she described in her fieldnotes:

The participant-as-transcriptionist method allowed the first author to explore an alternative way to approach transcription without having to rely on “outsiders” to help. This desire for independence goes beyond the requirements to complete independent research to fulfill degree requirements. Researchers, regardless of disability, strive to maintain self-reliance, confidence, and a sense of accomplishment. To be inclusive, some research practices need to be adapted to accommodate a diverse range of researchers. We see the participant-as-transcriptionist method as one potential accommodation.

The participant-as-transcriptionist method opens up options and possibilities for many researchers who otherwise might not be able to create transcripts without the aid of a third-party transcriptionist. Researchers who may particularly benefit from involving a participant-transcriptionist in their research include those who: (a) have limited hearing (e.g., are hard of hearing, are deaf, or have auditory processing problems); (b) lack sufficient manual dexterity to type (e.g., have arthritis, have repetitive stress injuries, or are amputees); (c) are unable to sit for long periods (e.g., have back or neck problems); (d) have attention deficit disorders; or (e) have any number of other disabilities. It is noteworthy that by the age of 15, one seventh of Canadians are disabled (Statistics Canada, 2001). With the increase in the average age of professors and graduate students (Elliott, 2000; Smith, 2003), the number of researchers with age-related disabilities may be on the rise. These disabilities need not preclude these individuals from conducting interview-based research.

For researchers exploring new ways to include participants and other researchers of all backgrounds in the interview process, technology can be a “medium for narrowing the gap between researcher and researched” (Seymour, 2001, p. 155). For example, using computer-mediated interviews can be a very important way to break down the barriers that disabled people encounter. Seymour suggests that one way of addressing the issues of data collection with disabled people is to involve them as online participants. However, he also acknowledges that conducting online interviews poses concerns regarding the nature of human interaction and the missing presence of body. Can the nuances of human interaction be captured in an online interview? How can the researcher observe facial expressions, body movements, and voice inflection from a written interview? The sensitive nature of many interviews (including the focal topic for the interview-based study in which the first two authors participated) may be best handled through a face-to-face interview setting.

Beyond the interview itself, some researchers are now exploring the possibilities of using natural language voice-recognition software for transcription purposes. Fogg and Wightman (2000) suggest that “the recent development of high-quality voice recognition software greatly facilitates the production of transcriptions for research and allows for objective and full transcription as well as annotated interpretation” (p. 1). Voice-recognition software programs can be trained to identify a researcher's voice and produce a written text from spoken or audio-recorded discourse. However, the speech patterns used by many individuals with hearing disabilities, for example, might preclude them from enunciating language in ways that the software can accurately translate into written text. Futhermore, Fogg and Wightman also point out that “the use of speech transcription still requires the use of a human transcriptionist” (p. 1). Although voice-recognition software may have some promise, as a hard-of-hearing researcher, the first author is not able to correct transcripts produced by voice-recognition software. As a result, such software still does not eliminate a third-party transcriptionist from the process for those who are physically unable to hear the tapes or to correct the resulting transcript.

“Although the potential positive effects of technology on inclusive education [and on research] is great, reaching that potential requires knowledge on the part of the user” (Merbler, Hadadian, & Ulman, 1999, p. 113). The user needs to know how to use the technology and know about the reliability of that technology. Unfortunately, it is the first author's experience that assistive devices such as hearing aids, flashing fire alarms, and so forth are like any other machine: over time and daily use, they break down. That is why it is important to explore other avenues of inclusion and practice that are not dependent on unreliable technology and machines. People are the ultimate resource. How we tap into this resource requires time, creativity, and enthusiasm. The participant-as-transcriptionist method allowed this interviewer (and her participant-transcriptionist) to engage in research without depending upon unreliable voice-recognition technologies for transcription.

A collaborative research experience

One of the unique aspects of the participant-as-transcriptionist method is the potential for participants to be more actively involved than in traditional research interactions where participants' involvement in the research often ends when the interview concludes. By serving as a participant-transcriptionist, the second author felt like a research collaborator for at least four reasons. First, she experienced an awakened sense of responsibility and began to identify as a co-researcher. Second, participant confidentiality was protected because no third party had access to the transcript. Third, the truthfulness or integrity of the transcript was preserved because the participant-transcriptionist ensured that the transcript represented her voice. Fourth, through the participant-as-transcriptionist approach, concerns about transcription were allayed because the participant-transcriptionist and the interviewer were free to engage in collaborative and reciprocal dialogue during the interview and in the preparation of this manuscript.

By creating the transcript, the participant-transcriptionist experienced a sense of responsibility:

The participant-as-transcriptionist method eliminated what Rowan called the alienation of detached participation (Seymour, 2001) because the participant-transcriptionist was invited to become personally invested in the research process. The participant-transcriptionist explains:

Many qualitative researchers strive to create research practices that reduce power differences between themselves and their participants, hoping to engage in more participative and less exploitative forms of research (Crawford, 1990; Heron & Reason, 1997; Kemmis & McTaggart, 2000; Smithmier, 1996). In order to transcribe their interviews, participant-transcriptionists need to take a somewhat active role in the research and, in this way, may come to see themselves as co-researchers, not just participants. The participant-transcriptionist explained how her involvement led to a sense of collaboration and ownership of the research:

The collaboration within the interview was further strengthened because concerns about participant confidentiality were diminished. Since the participant transcribed and handled the interview exclusively, there was no concern that a third party might breach any confidentiality pacts. This assurance of confidentiality directly influenced the quality of the interview because the participant was certain during the interview that any disclosures would remain confidential and that she would have an option to personally amend the transcript. The interviewer describes how she felt about the issues of confidentiality:

A related benefit of the participant-as-transcriptionist method was that of integrity of the transcript. As we discuss below, the participant-as-transcriptionist method circumvents concerns about the influence of transcriptionists' “fingerprints” (Tilley, 2002, p. 4) because the participant transcribes and edits the transcript. The integrity of the transcript is maintained, for it is the participant who decides if the content represents the truthful intent of the interaction. Further, it can be argued that the research process is in fact richer when the participant transcribes because the participant-transcriptionist can choose to modify or enhance a response based on the reflective noticing (Lynch, 2001) that is afforded by the participant-as-transcriptionist method. As the interviewer reported:

We saw that the increased trust experienced by the interviewer and the participant-transcriptionist led to role changes for both parties. The participant-transcriptionist displayed a deeper understanding of and a franker response to the interview questions, and the interviewer became more liberal with her comments in order to glean as much information from the interview as possible. Changes in roles for the participant and interviewer led to a collaborative and conversational tone for the interview. The “interviewer refuse[d] to stay detached and carrie[d] an obligation to reveal some of her own feelings in order to introduce a greater reciprocity into the interaction” (Acker, 2000, p. 193). The collaborative nature of the relationship between the interviewer and participant-transcriptionist allowed a participatory and inclusive approach to the research project. The collaboration extended to the point where the participant-transcriptionist became a co-author for this article.

A quality transcript

Transcripts provide a permanent record of research data that “can be searched, re-interpreted, collected and shared with other researchers in a more meaningful way and with much greater efficiency than any other medium” (Fogg & Wightman, 2000, p. 3). Transcripts are useful for the purposes of member checks, audit trails, re-analyses, inter-coder reliability comparisons, and other means to improve interpretive validity and reliability (Fogg & Wightman; Kvale, 1996). However, researchers and practitioners have begun to acknowledge the shortcomings of transcripts as data sources (Lapadat & Lindsay, 1998, 1999; Mishler, 1991; Tilley, 2002). An interview transcript is a representation of an audio or video recording of an interview, which is a representation of the interview itself (Lapadat & Lindsay, 1998). An interview might even be considered a representation of the participant's experience. The more steps removed from the raw experience, the more tenuous the researcher's account becomes because each representation introduces possibilities for error or misinterpretation.

The possibility of errors or misinterpretations are particularly acute when different individuals are responsible for interviewing, transcribing, or analyzing data. As Tilley (2002) has argued, hired transcriptionists and anyone else who participates in the production of research documents leave “fingerprints” (p. 4) that influence the research. When participants transcribe their own tapes, the number of hands that touch the research records are minimized and the transcript remains closer to the participant and the participant's experiences. This closeness increases the validity of the transcript as a research document. “A transcript is only as precise as the person transcribing” (Easton, Fry McComish, & Greenberg, 2000, p. 707).

Regardless of who constructs a transcript, there are considerable differences between spoken discourse and textual reproductions of that discourse (Kvale, 1996). Spoken discourse includes verbal content (e.g., words, word fragments, filled pauses such as um and err), prosodic information (e.g., pitch, duration, loudness), paralinguistic information (e.g., laughter, inhalation, exhalation, sighs, tears), extralinguistic information (nonverbal and nonvocal information such as table pounding, gaze, applause, gestures), and various contextual cues (Lapadat & Lindsay, 1998; O'Connell & Kowal, 1999). Despite typical research assumptions to the contrary, it is not possible for any transcriptionist, regardless of experience or training, to capture all verbal, prosodic, paralinguistic, extralinguistic, and contextual features of an interview (O'Connell & Kowal). Verbal interactions simply cannot be captured in their entirety in textual form. Transcripts should be seen as an aid for displaying features of speech or action that are to be analyzed, not as a replication of the spoken discourse during an interview (O'Connell & Kowal). Transcription is always an interpretive and constructive act, and the transcriptionists' interpretations influence what is recorded in a transcript (Kvale; Lapadat & Lindsay; Tilley, 2002, in press).

To circumvent the shortcomings of transcription, Kvale (1996) has suggested that transcriptionists should ideally consider how participants would formulate their answers in writing. When participants transcribe interviews, they are already placed in the position of formulating their answers in writing. Other researchers have suggested that online interviews hold the advantage of eliminating transcription errors because the interview is conducted in a textual form (Crichton, 1993; Seymour, 2001). However, experience has shown that online interviews are limited by access to technology, computer skills, and personal rapport between the interviewer and participant (McGinn, 2001; Schneider, Kerwin, Frechtling, & Vivari, 2002; Seymour). Online interviews often lack the probing and level of detail that are essential for qualitative research. Similar shortcomings may be evident in other written submissions from research participants, including journals and narrative reports. Face-to-face interviews provide opportunities for interactive conversation and more extensive probing than do textually mediated interviews (whether online or “offline”). The participant-as-transcriptionist method allows participants to formulate their responses in writing based on the conversation that occurred during the interview. Through the participant-as-transcriptionist method, participant-transcriptionists are ideally situated to notice how they spoke during the interview and to clarify their thoughts and meaning during the transcription. The transcription process provides opportunities to engage in the kind of “reflective noticing” that was characteristic of adult learners of English for academic purposes who transcribed their own classroom speaking tasks (Lynch, 2001).

However, this reflective noticing may lead participant-transcriptionists to focus more on the answers that they would like to provide to the interview questions, rather than the particular linguistic features present in the specific interview situation. Participant-transcriptionists may not capture every linguistic nuance in an interview and they may be tempted to modify the text of an interview transcript toward grammatically standard written text rather than the oral language of the interview. Furthermore, participant-transcriptionists may even wish to expand or modify their responses to the interview questions in light of further reflection on the issues (i.e., based on reflective noticing [Lynch, 2001]) or more careful consideration of the ethical ramifications of disclosures made during the interview. The earlier quotations from the participant-transcriptionist's fieldnotes exemplify this temptation.

Discussion between researchers and participant-transcriptionists about the transcribing experience may lead to (a) further reflections from the participant about the phenomena of interest in the research, and (b) justifications for any changes in the responses provided in the transcript. As an extension to the participant-as-transcriptionist method, some researchers may wish to compare participants' transcripts to other transcripts prepared by members of the research team or professional transcriptionists (cf. Kvale, 1996). Discussion with the participant about any differences between the participant's transcript and other transcripts could uncover details about the participants' experiences and the transcription process.

The possibility that participant-transcriptionists may implement changes in their interview responses during transcription means that the interview transcript will not match the oral text of the interview. However, we are not concerned by this prospect. In contrast, we see the implemented changes as strengths of the participant-as-transcriptionist method because the possibility to make such changes allows participants greater control of their words than they would have in typical interview-based research, and therefore greater control of the research outcomes. It has become standard practice to return transcripts to participants for participant review, but in such instances it is common for the participant to engage in a cursory review and to request additions or modifications only in rare cases. We would prefer to engage participants in fuller participation throughout the research and to provide ample opportunities for participants to shape and reshape their responses to our interview questions. We are most interested in the meanings that participants ascribe to their life experiences and not the particular word choices that they provide in snapshot interview situations where they have not had opportunities to fully reflect on their words. For researchers who ascribe to more positivistic approaches where it is important to attempt to control the interview situation and document precise linguistic features of interview responses, the participant-as-transcriptionist method may be unacceptable. However, like other methodologists (Kvale, 1996; Lapadat & Lindsay, 1999; O'Connell & Kowal, 1999), we question whether it is feasible to attempt such quasi-precision.

There is a necessary tradeoff between accuracy, readability, and standard orthographic form in transcripts (Lapadat & Lindsay, 1999; O'Connell & Kowal, 1999). As Kvale (1996) argues, “attempts at verbatim transcription produce hybrids, artificial constructs that are adequate to neither the lived oral conversation nor the formal style of written texts” (p. 166). Transcripts do not read like normal text and researchers have learned to warn participants that the transcript will sound strange. For participants who create their own transcripts, this warning will need to come earlier, during an instruction period prior to transcribing. Participant-transcriptionists, like all transcriptionists, will quickly realize that oral language does not follow the precepts of standard written discourse.

Transcriptionists–regardless of whether they are participants, researchers, or assistants–need instructions about the form and content for the desired transcript. The general instructions typically given to transcriptionists seem simple enough, but typically involve conflicting criteria: (a) create a literal transcription, (b) preserve everything that is said, (c) avoid “cleaning up,” and (d) use normal English orthography (O'Connell & Kowal, 1999). Participant-transcriptionists frequently lack any formal research training, so it is particularly important that researchers support these individuals as they engage in the novel task of transcribing research data. Researchers need to communicate their instructions about whether it is acceptable or desirable for participants to amend their responses as they prepare the transcript.

Interview participants who have devoted their energies to providing their opinions in a research study are often concerned that their opinions be faithfully represented in the research. These participants are therefore personally invested in ensuring that their opinions are represented appropriately in the transcript. As Easton et al. (2000) have argued, this investment and commitment are essential to the integrity of a research study and necessary for establishing dependability and confirmability of the findings.

Final thoughts

Through the interview study described here, we realized that the participant-as-transcriptionist approach can (a) accommodate differential abilities of researchers, (b) engage participants as collaborators in research, and (c) yield quality transcripts. The collaborative and inclusive nature of our experiences with the participant-as-transcriptionist approach prompted us to explore the methodological advantages in greater depth.

One major caveat to our approach is that the participant must have the necessary tools, skills, and incentive to be able to produce a written transcript. We have used the participant-as-transcriptionist method in research with graduate student participants, a situation that seems ideally suited to the method. These participants were open to the idea of transcribing an interview as a way to develop their research skills. Furthermore, these graduate students were compensated for the time they spent transcribing their interviews (an important consideration given the financial and time constraints of graduate education). Certainly graduate students are not the only people who will choose to engage as participant-transcriptionists, but we caution researchers that the participant-as-transcriptionist method may not work in all situations. For some participants or potential participants, transcription may be too difficult, time-consuming, or burdensome. The participant-as-transcriptionist approach is intended to open up new possibilities to be inclusive and collaborative, not to be used as a means to exclude particular participants from engaging in a research project. For example, in her current research, Grundy has explained the participant-as-transcriptionist approach to research participants and asked individual participants to decide who should transcribe each interview tape. Some participants have transcribed some or all of their interviews, while others have selected other individuals to serve as transcriptionists.

It is not our intention to set limits on anyone's research, but to suggest an alternative to traditional research practices that may work in some settings. The participant-as-transcriptionist method is still a very new research approach. Hopefully, it will lead the way to more innovative and inclusive research practices that retain the advantages of traditional face-to-face interviews without relying upon the limitations of inferior technology or the good graces of third-party transcriptionists who influence the research products and may (inadvertently?) breach confidentiality. We offer this paper to initiate a conversation about innovative research methods that are both inclusive and potentially collaborative.