Practices and challenges of learning by co-authoring in the - - PDF document

Practices and challenges of learning by co-authoring in the sciences: a multiple-case study of doctoral students and their mentoring advisors Pascal Matzler

School of Cultures, Languages and Linguistics

Good afternoon, and thank you for coming. I’m very pleased to present today here at the 2019 conference of the American Association for Applied Linguistics (AAAL). The title of my presentation today is “Practices and challenges of learning by co-authoring in the sciences: a multiple-case study of doctoral students and their mentoring advisors”.

This presentation will summarize my ongoing research project for my PhD in Applied Linguistics at the University of Auckland, under the guidance of my advisor,

• Assoc. Prof. Helen Basturkmen.

1

Overview

• General aim of the study
• Description of case study participants
• Data collection methods
• Findings
• Conclusion

Very briefly, this is the structure of my talk today. First, I will describe the general aim of my study, and then I will give a sketch of my participants and a brief summary of my data collection methods. After that, we will have a look at some of the findings and how these findings have informed my conclusions.

2

General aim of the study

To better understand how advisors introduce their graduate research students to the practices of writing research articles (RAs) by writing a first RA together: “mentoring by co-authoring” My work is in academic discourse socialization: I am primarily interested in how graduate research students are introduced or socialized to the practices of writing research articles (RAs): In the sciences, this practice is best described as “mentoring by co-authoring”. 3

Literature

• ESP Genre Analysis, Disciplinary Discourses (product-
• riented)
• Ethnographic studies of research writing in the sciences

(Latour and Woolgar, 1979, 1986; Knorr-Cetina, 1981; Myers, 1990).

• Longitudinal text-oriented ethnographic study, Text History

(Lillis and Curry, 2006, 2010)

• Learning to write for publication: Belcher (1994), Blakeslee

(1997), Kamler (2008), Aitchison et al. (2012)

• Methodology: Multiple-case study (Duff, 2008), thematic

analysis (Braun and Clarke, 2006)

This project draws on a fairly wide range of literature. My analysis of the RA as a textual product is informed by ESP genre analysis as developed by John Swales and others, as well as the work done on Disciplinary Discourses by Ken Hyland. My understanding of the contexts and practices of research writing follows ethnographic studies performed from a sociological or anthropological perspective (e.g., Latour and Woolgar, 1979, 1986; Knorr-Cetina, 1981; Myers, 1990). The framework for describing the development of the RA text and the interaction between the co-authors is also informed by the text history approach developed by Lillis and Curry (2006, 2010), as well as some of the existing literature on learning to write for publication in the sciences. Finally, in terms of methodology, I would describe this project as a multiple-case study with an ethnographic orientation, and the main data analysis method is thematic analysis, following Braun and Clarke, among others.

4

Contexts and backgrounds of participants

Case Participant Role Main Language(s)

• 1. Environmental

Sciences Chet Advisor Italian and English Tsai Doctoral Student Chinese

• 2. Neurosciences

Barry Advisor English Ramon Doctoral Student Spanish 3: Computational Chemistry Kate Advisor English Leila Doctoral Student Persian

This being a multiple-case study, I feel that it is important to give a brief outline of what kinds of people my participants are. I was lucky to recruit three pairs of participants that are diverse in many aspects, such as their L1 and cultural background, age, gender, discipline and level of experience. My first case was in Environmental Sciences: I had an Italian advisor, in his early 50s, who is a world authority in his field, and his Chinese PhD student. They were studying sediment transport in coastal systems. My second case was in Neurosciences: The advisor was from NZ and quite young, whereas the student was from Latin America. They were studying how the brain activates groups of muscles to perform movements. And my third and final case is in Computational Chemistry, with an experienced NZ advisor and her Iranian PhD student, who were using powerful computers to simulate chemical reactions. All three PhD students were at an early-intermediate stage of their PhD: this means they had completed their first year research proposals and had now obtained sufficient results from their experiments to publish a first RA.

5

Data collection – thick description

Activity Type of collected data Number Student written products RA draft Email body message 17 Advisor written products Edits suggested / performed on drafts On-script comments or questions Email body message Interviews with participants Audio recording Transcript Log / Memo 12 Observed meetings Audio recording Transcript Log / Memo Annotated copy of draft 15

The data collection for this case study was structured in such a way as to collect evidence through a wide range of channels simultaneously. I collected 17 drafts that the participants exchanged, and, importantly, the feedback that the advisor wrote on each draft, as well as the email messages that they used to exchange the drafts. Finally, I was able to observe and audio-record 15 meetings, so I have meeting transcripts and my own notes and reflections from observing these meetings. I also conducted 12 interviews, consisting of initial background interviews and then member- checking interviews after the participants completed the co-authoring of their RA. The main purpose of performing such a varied (and hopefully thorough) data collection was to arrive at a thick description that is so valued in ethnographic case study research. Thematic analysis was then performed using Atlas.ti qualitative data analysis software.

6

Extract of supervisor annotation on draft

Reviewed draft , emailed by Chet back to Tsai:

Before I turn to my findings, I would like to show you two examples of the types o data I collected. Firstly, this is a extract of a typical RA draft that was sent by one of the doctoral students to his supervisor. The supervisor, in turn, employed a PDF viewing and mark-up software to annotate the draft, and then returned it by email before scheduling a meeting. Note that the supervisor inscribes the student’s draft in two different ways. The majority of interventions correspond to direct text edits; however, there are also

• n-script comments consisting of boxed ‘Notes’ which are anchored to the passage
• f text they refer to (see speech bubble symbol on line 164). Such boxed comments

(in this instance invoking reader expectations) typically express deeper or more complex concerns that cannot be resolved by a simple and localised text edit. 7

Extract of an RA draft annotated by the observer during a meeting:

The observation of meetings included one additional method of data collection that served as a key link between the observations and the textual data. Before each meeting, I printed

• ut a copy of the latest draft version (often containing supervisor edits or comments

performed only minutes earlier). During the observed meeting, I manually annotated these printed drafts with time stamps, notable utterances, observations and reflections on events in the meeting. Notably, the participants tended to discuss and intervene their RA in often fragmented and recurring episodes, rather than by linearly following an established pattern or organisation. By manually annotating the latest draft during meetings, particularly with time stamps, I was able to map the flow of discussions and text interventions performed by the participants during their meetings onto the printed draft the text, and to later reconstruct these activities during transcription and data analysis. These annotated printed copies therefore provide a key link between RA drafts and meeting transcripts.

8

Mentoring by co-authoring is normal

“We are co-authors, we always do it together. I don’t think any student would ever even think of publishing without me, because, I mean, it’s work that we do together.” (Chet, Interview 1.D)

• Science PhDs write Ras before / instead of the

thesis

• RAs are based on experiments or simulations

performed by student in the PhD.

• Co-authoring is common in all three studied

disciplines Most research into doctoral writing practices has focused on the arts and humanities, where advisors will provide both oral and written feedback on student drafts of chapters or sections of the actual thesis (see, e.g. Belcher, 1994). In the sciences, however, the main writing activity (and therefore the main venue of discourse socialization) is the research article (RA). In this multiple-case study, all three students were preparing an RA based on experiments or simulations conducted in their PhD programme. All six participants, both students and advisors, declared that co-authoring RAs between students and advisors constitute normal practice in their respective

• disciplines. For example Chet, the advisor in Case 1 (Environmental Sciences), found

it hard to imagine a different arrangement. [see quote] Because the advisor is also a co-author of the RA, he or she contributes both 9

indirectly and directly to the writing, a practice that raises important questions of

• wnership, motivation, and power relationships.

9

Shared responsibility and credit

Do I think it's perfect? No. I don't think it's perfect. But I'm perfectly happy to have my name on it. I think it answers the questions. And it's truthful to what it says in here. In that sense, yes, 100% I would claim this as valid research. (Barry, Interview 2.B)

• Advisors are NOT just adding their name to a

student paper.

• Taking responsibility: valid answers to questions

and accurate representation of experiments. One of my key questions at the beginning of the project was whether these RAs constituted some kind of limited ’exercise’, or whether the participants, in particular the advisors, would consider it as valid research. The advisors were quick to clarify that theirs was not a practice of simply adding their own name to a student paper. Instead, they highlighted the dual significance

• f (co-)authorship, namely taking responsibility and claiming credit. In terms of

taking responsibility, by adding their name as author to the RA these advisors signalled that they were convinced of two essential quality measures: that the RA addressed the scientific question in a valid manner and that it accurately represented the performed experiments. [see quote] 10

Shared responsibility and credit

• Advisors are NOT just adding their name to a

student paper.

• Claiming credit: advisors contribute to design,

execution, and writing.

I feel this is Tsai and I. You have arrived in October, but Tsai had been here already for one year. And one year of blood and sweat. (…) I don’t publish students’ papers. (Chet, Interview 1.D)

Meanwhile, in terms of claiming credit (see .e.g. Hyland, 2015, pp.82-83), the advisors insisted that their own contributions in guiding the overall design of the research, in verbally advising its execution and in contributing to the writing, constituted ample ground for granting co-authorship to themselves. [see quote] 11

Research writing as socially embedded rhetorical practices

RAs do much more than disseminate findings. In meetings, participants rehearse:

• Structuring of RA sections
• Weighing claims (own convictions)
• Weighing claims (acceptable to community)

[Writing the article] will be most difficult. The article is difficult to write, because we have to be polite in the way we say to a lot of people that what they’ve done was incorrect. (Chet, Interview 1.E)

A second key question was how much importance these scientists and their students assign to scholarly communication as rhetorical practice. Writers in the sciences sometimes portray themselves as emotionally detached from their text, e.g. in Aitchison et al. (2012, p.444): “In discussions with scientists we got the feeling that, for the majority, the writing itself didn’t count. Their priority was rather the dissemination of the research findings.” However, ethnographic studies of research writing practices (e.g. Latour and Woolgar ,1979; Knorr-Cetina, 1981) and studies in ESP genre analysis have come to a rather different conclusion, arguing that issues of “power, allegiance, and self- esteem” are ever present below the “frigid surface of RA discourse” (Swales, 1990,

• p. 125), and that the RA needs to be understood as a representation of socially

embedded rhetorical practices (Lillis and Curry, 2006; Hyland, 2012). This emphasis is strongly supported by the findings in the present study. The participants focused large parts of their meetings on verbally rehearsing and adjusting the flow of key sections (mainly the Introduction and the Discussion) and 12

• n weighing the claims that they felt able to make for their findings, both in terms of

expressing their own convictions and, quite separately, in terms of the strength of claims that they felt acceptable to journal reviewers and to the wider disciplinary community. The importance of this discourse element of research writing was illustrated by Chet, the advisor in Case 1, at the start of the observation, when he expressed a recurring worry about how their work would be received by their disciplinary community: [see quote] 12

Learning to write: Progress

• Co-authoring is instrumental to learning to write
• PhD students are expected to produce

progressively higher quality RAs throughout the program.

• Advisors ‘invest’ in the first RA

Throughout the PhD you see a total transformation. I remember a Colombian one: the guy was good, there was never a doubt about that, but the writing

• f the first paper was really bad. And, yes, the

second paper was a major step forward. And the third one, he clearly had made it. (Chet, Interview 1D)

One of my key findings is to draw attention to the first co-authored RA as a key instance in the discourse socialization of science PhD students. All three advisors agreed that the practice of co-authoring is instrumental to learning to write and expressed a clear expectation that their students’ writing ability would progress throughout the PhD. Consequently, the advisors were prepared to invest a significant amount of time and effort in contributing to the writing of a first co-authored RA, on the understanding that this practice would help their students acquire the necessary knowledge and skills to then become progressively more independent writers. Chet, for example, recounts the evolution of a previous student over a two-year period: [see quote] 13

Learning to write: Independence

Over the course of various co—authored RAs:

• Quality of final product does not vary.
• Relative contributions of student and advisor vary.

I guess I would like to see Ramon develop more of a sort of a big picture in terms of thinking and writing and self-determination. So that's sort of my job, it’s like an apprenticeship. As well as just a PhD. (Barry, Interview 2.B)

Interestingly, what evolves over the course of the PhD is not actually the quality of the RA product itself. This quality is guaranteed by the advisor’s contribution as co- author. Instead, what evolves is the student, who transforms into a more independent research writer. For example, Barry, the advisor in Case 2, did not expect the last publication of Ramon’s PhD to look very different from the first one. Rather, he hoped for Ramon to acquire a more prominent role in developing the RA, requiring less input from his advisor: [see quote] 14

Model, coach and fade?

Within an RA, the model is reversed:

• 1. Student works independently
• 2. Advisor coaches student
• 3. Advisor takes control

It ended up with lots and lots of going backwards and forwards, and yeah, that took a lot of time to get nowhere particularly fast. And so it was in some aspects a practical response to needing to keep moving with it. (Barry, Interview 2.B)

This notion of apprenticeship (see Lave and Wenger, 1991) was a recurring theme throughout the study. The literature on doctoral writing describes a process whereby mentors are said to first ‘model’, then ‘coach’, and finally ‘fade’ (Belcher, 1994, p. 24) or gradually ‘pull back from the text’ (Aitchison, 2012, p. 442). In the present case study this notion

• f ‘model, coach and fade’ describe quite well the larger time scale of an entire

three to four year PhD program. However, its application becomes problematic at the smaller time scale of the individual RA. Within an RA, advisors instead tended to reverse the order of the steps in the model, by first allowing their student considerable independence in formulating an initial draft of the RA, then coaching the student over a sequence of drafts by means of hour-long writing meetings and detailed written feedback, before finally taking over and intervening the text directly in the final stages of the RA drafting process.

Barry, for example, justifies his gradual taking of control with a need for progress: [quote]

15

The value of thick description

• Importance of thick description: interviews,

meeting observation, text products

• Advisors may not be aware of their own practices

On the manuscript? You know that I didn’t even

• notice. Clearly I must have run out of patience. If I

took the computer and wrote, it’s a total sign that I had lost it. (Chet, Interview 1.F)

At the same time, episodes like this one – the advisor’s gradual appropriation of the text - are also a clear sign of the importance of thick description, that is, the combination of several distinct data collection methods. Surprisingly, advisors may not be aware of their own practices. In Case 1, for example, Chet confessed in the debriefing interview that he had no recollection of taking over his student’s computer in the last meeting and intervening the text directly. If I had simply asked him about his practices in an interview (without observing any meetings) I would not have identified this issue.

16

The stapled thesis

I think they [the RAs and the thesis] are the same. Whatever I put into these articles, it’s going to be my Discussion in the thesis, it's going to be my Methods. This is the backbone. My idea is just to staple them. (Ramon, Interview 2.A)

• Several published RAs bound together
• Introduction, Discussion and Conclusion provide

cohesion Both the students and the advisors emphasised the importance of writing for publication during the PhD, mentioning benefits both within the doctoral program as well as for long-term career goals. Firstly, all three students were expected to eventually submit a thesis consisting mainly of their published research products, in a practice known as thesis by publication, or more informally, the stapled thesis. Ramon, for example, expected his published RAs to provide the bulk of his eventual thesis: 17

Peer review

• Benefit of peer review in a fast-moving field
• Advisor as mediator, not as gatekeeper

A lot of people think it's good, because if the papers have been through peer review before you get to the examination of your PhD, then you know that all the issues have been dealt with and it should be pretty

• straightforward. (Barry, Interview 2.B)

An additional benefit of the thesis-by-publication method is that feedback is received by submitting the research to peer review before finalizing the thesis. This is especially important in fast-moving science disciplines. Barry, Ramon’s advisor ,highlighted the benefits of outsourcing large parts of the evaluation to peer review: [see quote] In other words, the function of the advisor is less of a powerful expert gatekeeper (as frequently seen in the Arts and Humanities) and more of a mediator between the student and the actual gatekeepers: journal editors and reviewers. 18

Socialization into discourse community

• Publishing together with advisor and senior

colleagues

• Becoming a recognized member of the community

I don’t exist in the scientific world. (…) My advisor has colleagues in the UK and the US, and I was thinking that if I write then everyone knows that I worked with Kate. (Leila, Interview 3.A, 11-10-2018)

Equally, the doctoral students were very much aware of the career benefits of co- authoring and publishing together with their much more renowned advisor and

• ther senior members of their research group. Leila, for example, highlighted her
• wn need to become a recognized member of her discourse community:

[see quote] Again, the role of the advisor is that of a mediator, here by introducing the student to the community and lending the student some of their accumulated credibility. 19

Benefits to advisors

• Publish or perish
• Demonstrate leadership
• Organic growth of their own research group

It's also important for me to be defining my own

• research. Because when you submit for a grant you

say, “Look at these papers that I have led, therefore I am worthy of your money to lead some more”. (Barry, Interview 2.B)

Notably, the practice of mentoring by co-authoring also benefits the advisors. The three participating advisors openly discussed their own pressures to ‘publish or perish’ (see e.g., Hyland, 2015, p.7) and identified themselves and their department colleagues as research-focused academics, and that publication output was specified as a key component in their periodic performance evaluation. Advisors, as scholars and employees of a leading university, thus have a strong incentive to perform research and to publish. However, this incentive alone does not explain why academics in the sciences would choose to publish in co-authorship with their doctoral students. Interestingly, advisors perceive an additional long-term benefit from doing so. A series of successful publications co-authored with PhD students will allow advisors to demonstrate their leadership qualities as head of a research group, thus generating new grants which will then provide additional funding for a subsequent cycle of new students and new research. Barry, in case 2, is a good example of an early-stage advisor needing to establish himself in such a leadership role: 20

[see quote] Kate, in Case 3, was already stationed further along this process: she already had a large ongoing research project with funding available for additional PhD students when she received Leila’s CV. 20

Remainder of my thesis

• Modes of interaction in meetings
• Roles of expert and novice
• Weekly cycle of drafts and meetings
• Layers of text intervention

➢ Hedging and personal pronouns ➢ Schematic structures (I+D)

As you will all be aware, a 20 minute presentation will always struggle to give a detailed account of an entire thesis. I will close by noting that in the later parts of my thesis I further describe and characterise the patterns of interaction that emerge between the doctoral student and their advisor, and how these interactions contribute to the development of the RA as a writing product. I am particularly interested in the different ways (modes) that the student and advisor congregate around the text in the meeting: who is in charge of the keyboard? In the same chapter, I am also looking into the roles of expert and novice, which in these fast-moving scientific fields are quite fluid. Another chapter deals with the weekly cycle of drafts and meetings, to better understand the dynamics between the participants’ verbal and written exchanges. And finally, since my origins are in ESP genre analysis, I am also attempting to describe the sequence or flow of text development: What aspects of the text does the supervisor focus on immediately? What does he or she leave for later? Here I am using the concept of ‘layers’ to characterize their highly recursive practices. 21

Summary of findings

• Mentoring by co-authoring is normal in the sciences
• Shared responsibility + shared credit
• Research writing as socially embedded rhetorical

practices

• Learning to write: the first RA is key instance
• Model, coach and fade?
• Importance for thesis and for career
• Benefits to advisors

In conclusion, mentoring by co-authoring is common practice for science PhD students and their advisors. They share both the responsibility and the credit of publishing their research. The advisors are keenly aware of writing as a social practice, and they make great efforts to raise their students’ awareness of reader expectations and motivations. Also, the first co-authored RA of the student-advisor relationship seems to be a key instance, as this is where the advisor “invests” in the

• student. I have also drawn attention to how the accepted sequence of “model,

coach and fade” seems reversed. Finally, we have seen that this practice of mentoring by co-authoring brings long-term benefits to both the student and the advisor.

I would hope that these findings have implications for how we conceptualize writing support for graduate research students, especially if we are interested not just in writing products, but also in writing practices, and especially in the learning of these writing practices.

22

Thank you for your attention!

23

References: Aitchison, C., Catterall, J., Ross, P., & Burgin, S. (2012). ‘Tough love and tears’: learning doctoral writing in the sciences. Higher Education Research & Development, 31(4), 435-447. Belcher, D. (1994). The apprenticeship approach to advanced academic literacy: Graduate students and their mentors. English for Specific Purposes, 13(1), 23-34. Blakeslee, A. M. (1997). Activity, context, interaction, and authority learning to write scientific papers in situ. Journal of Business and Technical Communication, 11(2), 125-169. Braun, V., & Clarke, V. (2006). Using thematic analysis in psychology. Qualitative research in psychology, 3(2), 77-101. Duff, P. A. (2008). Case study research in applied linguistics. New York: Lawrence Erlbaum. Heath, S. B., & Street, B. V. (2008). On Ethnography: Approaches to Language and Literacy Research. Language & Literacy (NCRLL). Teachers College Press. 1234 Amsterdam Avenue, New York, NY 10027. Hyland, K. (2004). Disciplinary Discourses, Michigan Classics Ed.: Social Interactions in Academic Writing. Ann Arbor: The University of Michigan Press. Hyland, K. (2015). Academic publishing: Issues and challenges in the construction of knowledge. Oxford: Oxford University Press. Kamler, B. (2008). Rethinking doctoral publication practices: Writing from and beyond the thesis. Studies in Higher Education, 33(3), 283-294. Latour, B. and Woolgar, S. (1986) Laboratory Life :The Construction of Scientific Facts. Princeton, NJ: Princeton University Press. Lave, J., & Wenger, E. (1991). Situated Learning: Legitimate peripheral participation. Cambridge: Cambridge University Press. Lillis, T., and Curry, M. J. (2006). Professional academic writing by multilingual scholars: interactions with literacy brokers in the production of English-medium texts. Written Communication, 23(1), 3-35. Lillis, T. M., & Curry, M. J. (2010). Academic writing in global context. London: Routledge. Knorr-Cetina, K. (1981). The manufacture of knowledge: An essay on the constructivist and contextual nature of science. Oxford: Pergamon. Myers, G. (1990). Writing biology: texts in the social construction of scientific knowledge. Madison: University of Wisconsin Press. Swales, J. M. (1990). Genre analysis: English in Academic and Research Settings. Cambridge: Cambridge University Press. Swales, J. M. (2004). Research genres: Explorations and applications. Cambridge: Cambridge University Press.

Aitchison, C., Catterall, J., Ross, P., & Burgin, S. (2012). ‘Tough love and tears’: learning doctoral writing in the sciences. Higher Education Research & Development, 31(4), 435-447. Belcher, D. (1994). The apprenticeship approach to advanced academic literacy: Graduate students and their mentors. English for Specific Purposes, 13(1), 23-34. Blakeslee, A. M. (1997). Activity, context, interaction, and authority learning to write scientific papers in situ. Journal of Business and Technical Communication, 11(2), 125-169. Braun, V., & Clarke, V. (2006). Using thematic analysis in psychology. Qualitative research in psychology, 3(2), 77-101. Duff, P. A. (2008). Case study research in applied linguistics. New York: Lawrence Erlbaum. Heath, S. B., & Street, B. V. (2008). On Ethnography: Approaches to Language and 24

Literacy Research. Language & Literacy (NCRLL). Teachers College Press. 1234 Amsterdam Avenue, New York, NY 10027. Hyland, K. (2004). Disciplinary Discourses, Michigan Classics Ed.: Social Interactions in Academic Writing. Ann Arbor: The University of Michigan Press. Hyland, K. (2015). Academic publishing: Issues and challenges in the construction of

• knowledge. Oxford: Oxford University Press.

Kamler, B. (2008). Rethinking doctoral publication practices: Writing from and beyond the thesis. Studies in Higher Education, 33(3), 283-294. Latour, B. and Woolgar, S. (1986) Laboratory Life :The Construction of Scientific Facts. Princeton, NJ: Princeton University Press. Lave, J., & Wenger, E. (1991). Situated Learning: Legitimate peripheral participation. Cambridge: Cambridge University Press. Lillis, T., and Curry, M. J. (2006). Professional academic writing by multilingual scholars: interactions with literacy brokers in the production of English-medium texts. Written Communication, 23(1), 3-35. Lillis, T. M., & Curry, M. J. (2010). Academic writing in global context. London: Routledge. Knorr-Cetina, K. (1981). The manufacture of knowledge: An essay on the constructivist and contextual nature of science. Oxford: Pergamon. Myers, G. (1990). Writing biology: texts in the social construction of scientific

• knowledge. Madison: University of Wisconsin Press.

Swales, J. M. (1990). Genre analysis: English in Academic and Research Settings. Cambridge: Cambridge University Press. Swales, J. M. (2004). Research genres: Explorations and applications. Cambridge: Cambridge University Press. 24

Contact

University of Auckland profile page and email: https://unidirectory.auckland.ac.nz/people/profile/pmat757 Researchgate profile: https://www.researchgate.net/profile/Pascal_Matzler

25