Publishing in TS: the authors’ and referees’ viewpoint
The small questionnaire-based e-mail survey conducted in October after a pilot phase and consultation with colleagues (many thanks to Delia Chiaro and Giuseppe Nocella for their helpful comments) focused on the relative importance of criteria in the selection of journals to which authors submit papers and about feeling towards referees’ comments. The aim of this survey was not to arrive at predictive or explanatory data, but to identify issues with some relevance for consideration by editors, referees and contributors, both through the quantitative data and through the respondents’ comments. A full paper explaining the objectives, methodology and other issues regarding this survey will be prepared at a later date. What is posted here is a synopsis.
84 respondents from 25 countries, mostly European (from all parts of Europe),
but with participation from North America, Asian countries and
2. Criteria for the selection of journals:
On the basis of experience in the field, three criteria were selected for initial questions: respondents were asked to rate the relative importance of the reputation of the journal, speed of publication and the possibility of choosing one’s language for writing on a three-point scale (“quite important”, “not very important”, “not important at all”) with an escape-response (“I don’t know”). These criteria were obvious, and no special revelations were expected, except for some possible indications regarding the relative importance of language of publication by geographic/cultural groups. Respondents were also given the possibility of naming other criteria they considered relevant, so that such criteria could become salient in further reflection. In each question, respondents had extra space for comments.
According to the answers, the most important criterion for the selection of a journal was clearly its reputation: 88% of the answers indicated that it was “quite important” and 12% that it was “not very important”.
The second most important criterion was speed of publication: 60% of the respondents said it was “quite important”, 37% said it was “not very important”, and 4% that it was “not important at all”.
Ranking third is the possibility of choosing the language of publication: 45% of the respondents indicated it was “quite important”, 38% that it was “not very important”, 15% that it was “not important at all”.
As to other
relevant criteria mentioned by respondents, 8 mentioned the possibility of
interacting with the referees and/or editors, while 4 talked about the
particular field of specialization of the journal and 4 about the usual
readership of the journal. Three people, all three from
No clear correlation was salient between the relative importance of various criteria and specific geographic, linguistic or other sub-groups. This may be due to the small size of such subgroups, but also the complexity of the factors influencing the respondents’ attitudes.
Respondents were asked how often they felt that the referees’ comments were useful to them (“Often”, “sometimes”, “rarely”, with an escape-response: “I don’t know”).
50% of the respondents said that the referees’ comments were useful to them “often”, versus 40% who said they were useful “sometimes” and 5% who said they were “rarely” useful to them.
To the question how often respondents felt that referees had misunderstood their statement and/or did not know enough about the subject or research method and/or defended a different research paradigm and therefore made an unjustified comment, 4% answered “often”, 42% “sometimes” and 49% “rarely”.
To the third and last question, asking respondents if they have acted as referees while feeling that they were not fully qualified to do so but had to do it anyway, 3% of the respondents who answered (19 did not answer) said “often”, 37% “sometimes”, 37% “rarely” and 23% “never”.
4. Preliminary warning about the interpretation of results
a. It is important to stress that the sample cannot be assumed to be representative of the general population of TS scholars for at least three reasons besides the geographic distribution of respondents and the distribution of their relative experience in TS. One is the sampling procedure: the questionnaire was sent to all members of EST, to members of CATS and to various other researchers via colleagues who kindly forwarded it, but no rigorous sampling procedure ensuring representativity was used. The second reason is a traditional one: those scholars who took the trouble to answer may not have exactly the same views as those who did not. The third reason has to do with languages: the survey was posted in English, and even though some answers were sent out in languages other than English, it cannot be ruled out that if the survey had also targeted scholars with other languages and/or been sent out in other languages, it would have elicited different answers, especially regarding the importance of languages in selecting journals for submissions.
This was clear from the start, and the aim of the survey was not to help paint a faithful picture of the situation, but only to provide some input on areas which deserve some attention. For reasons which will be discussed in the full paper which will be prepared at a later date, I believe that the data are robust enough to draw some conclusions as formulated further down.
b. It is also important to stress that the answers reflect attitudes, not necessarily facts. In particular, it is not unlikely that the frequency of useful comments by referees has been overstated by respondents because this is the positive, politically correct attitude, and that the frequency of refereeing while not being fully qualified has been understated because such action might be perceived as unprofessional. Such gaps between facts and verbal reports should be taken into account when conclusions are drawn from the results.
5. Discussion of the results
5.1 Selection of journals
According to their reports, respondents might seem to aim primarily at publication in high-reputation journals. While this outcome is not surprising, it needs to be qualified. One reason is that authors write more or less innovative, more or less important papers, and that they have more or less confidence in the quality of their manuscript. They may well avoid sending a manuscript they believe to be somewhat below top quality to a high-reputation journal, or seek to distribute their papers strategically among several journals depending on the nature of their manuscripts and on other needs (“political”, linguistic or other). The reputation criterion may also clash with other criteria such as required speed of publication, which can be quite important in case of an innovation or the report of an empirical result in an ongoing long-term study, to mention just these two cases. They may also select a journal for a paper simply because they were asked for a contribution by the editor of that journal.
As is explained in the introduction, the survey did not aim at providing explanations or tools for prediction. Indeed, results do not show that specific criteria can explain or predict author behavior. However:
a. They confirm that not only the reputation of a journal, but also the speed of publication and language of publication are seen as relevant criteria by many authors. At this point, there is not enough information to determine in what specific groups and circumstances these two criteria are more or less important. Also note that the medium, i.e. electronic (as opposed to traditional hard copy) was not mentioned spontaneously by respondents as a relevant criterion – but it was mentioned during the symposium: a few participants said that authorities in at least some countries tend to consider electronic publication less prestigious than hard-copy publication.
b. The more interesting outcome is the expression of a clear wish by authors to have the possibility of interacting with the referees. This is not in line with the principle of anonymous peer-reviewing, but may correspond to the present state of advancement of scholarly work in TS. See Helle Dam’s and Riitta Jääskeläinen’s statements in this respect.
On the basis of my personal experience in editing and refereeing on both sides of the fence, I tried to focus in this part of the survey on subjective feelings of both authors and referees regarding the quality and credibility of refereeing.
It is comforting, but not surprising, that half of the respondents said they found the referees’ comments useful “often”. On the other hand, some of these comments seem to be off target, judging by a relatively high rate of “sometimes” useful (40%) and “rarely” useful (5%), and by 46% of respondents saying that referees made unjustified comments “sometimes” or “often”. This is reflected by the 40% of respondents acting as referees who said they felt they were not fully qualified to act as referees sometimes or often. The problem is linked to the wide range of specialties within TS and to the language distribution of authors and referees, which makes it difficult to find enough referees with the right language and expertise combination in the required field for all manuscripts (see Gile and Hansen 2004).
6. The next step
These rough results cannot possibly cover all the locally important circumstances which have to be taken on board by editors, referees, trainers and other stakeholders. It is therefore important to gather more qualitative data on TS scholars’ experience with editing and refereeing on both ends of the process. Colleagues are invited to send comments, questions and personal anecdotes, stating if they want them to remain anonymous or not, to firstname.lastname@example.org
Gile, Daniel and Gyde Hansen. 2004. The editorial process through the looking glass. In Hansen, Gyde, Kirsten Malkmjær & Daniel Gile (eds). Claims, Changes and Challenges in Translation Studies. Amsterdam/Philadelphia: John Benjamins. 297-306.