Insights from Editorial Experiences with Population, Space and Place
Clara H. Mulder,Darren P. Smith,Aileen Stockdale
DOI: https://doi.org/10.1002/psp.2201
2018-01-01
Abstract:Editing a leading social-scientific journal such as Population, Space and Place (PSP) is an enjoyable task and a terrific learning experience. As editors, we want to present some of our common shared experiences with the diverse international readership of PSP. We have three main aims with this Editorial. Firstly, we highlight how we typically proceed from receiving a submitted manuscript to finally accepting it for publication. We hope this helps prospective authors to more fully understand the sometimes difficult and rigorous decisions we make about their manuscripts, to better assess whether it is worthwhile to submit their manuscripts to PSP, to speed up the publication process, and, last, but not least, to enhance the overall quality of their scholarship. Secondly, we share our collective views on the peer review system, at a time when this system is criticised by some academics. Thirdly, we share some of our experiences with the division of tasks between the editors and publisher. We hope the latter is valuable for other (prospective) journal editors, and those scholars planning to start a new journal, to better understand the role of the publisher and to avoid the caveats that can be associated with choosing a publisher. Unfortunately, “reject” is the decision that we make for the vast majority of the submissions to PSP. In fact, we reject about half of the submissions without sending them out for review—these are the so-called “desk rejects.” We are convinced that such desk rejects before peer review are a very important and essential task of the editors. They help us not to overburden our colleagues in the academy with review requests and to prevent the waste of precious review time. we see it as our responsibility to publish the best work carried out in our discipline but not to an extreme where only a tiny proportion of the work can be published and scientists are discouraged from submitting their work. (Findlay & Mulder, 2015, p. 689) We welcome submissions of work on migration and residential mobility, and on spatial variations in or geographical aspects or consequences of fertility, mortality, population growth and decline, population distribution and segregation, population age structure, and the formation and dissolution of households, marriages, and partnerships. Submissions on the behaviour and integration of migrants are considered if they take into account connections with the migration process itself or if they have a geographical component. Submissions on health are considered if they not only have a geographical component, but also relate to mortality, fertility, migration, or household change. (Findlay & Mulder, 2015, p. 690) We continue to receive quite a number of submissions that do not fit within this broad scope of the journal. We think it is us, the editors, who should make the decision about whether a manuscript fits within the scope of the journal. Peer reviewers should not, we think, be burdened with this responsibility; they should be able to devote their attention to assessing and helping improve the overall scholarly quality of the papers. We therefore desk-reject out-of-scope manuscripts Ruthlessly. Such decisions are made within the editorial team without further consultation, or occasionally after consulting a specific member of our International Advisory Board with expertise in the respective field. Of note, we also consulted the International Advisory Board about whether refocusing was a good idea in the first place. The out-of-focus criterion implies that we quite frequently reject manuscripts that could be published in a journal with a different focus. studies that replicate previous research in a new context or even in the same context are welcomed but only if they contain a critical discussion of the previous work and/or an important improvement and/or a thorough discussion of what could explain differences in the findings - in short, if they are not “just another case study.” (Findlay & Mulder, 2015, p. 690) We try to weed out such manuscripts before potential peer review, but if they are outside our own expertise, a fair judgement of theoretical argumentation, engagement with the literature, scientific rigour, and contribution to the literature is not always possible. Of course, this too means that some manuscripts that are rejected might be suitable for publication in other journals. Manuscripts that make their way beyond the editor's desk are sent to a varying number of peer reviewers—frequently two or three, depending on the specific expertise we need from reviewers. The expertise we need is on the topic, data and methods, and geographic scope of the paper and may include members from the International Advisory Board. Although we ask authors to suggest potential reviewers of their manuscripts, we always make our own judgement for invitations to reviewers. A majority of the manuscripts that are reviewed are rejected, usually for the same reasons as desk rejections—except the out-of-focus reason: They do not meet the high scholarly standards of the journal, or they do not make important contributions to the literature. The other manuscripts are returned to the authors for major or minor revision, based on the reviewers' comments and our own judgement. We ask the authors to submit not only a revised version of the paper but also a document containing the authors' responses to the reviewers' and editor's comments. Of those manuscripts that are revised after a first round of review, some manuscripts are rejected in second or occasionally in the third instance, usually because the authors fail to incorporate the reviewers' or editor's comments in a satisfactory way. Many other manuscripts make their way to acceptance after one or more rounds of revision. Some common comments are the following, in the order of appearance in a standard article text. Some manuscripts are submitted which include abstracts that discuss relevance, research topic or question, main theory or hypothesis, data and methods—but do not include key findings. We think it is pertinent that the main findings are effectively highlighted in the abstract. It should be borne in mind that the abstract is predominantly the part of the paper that is read the most. The Abstract should be informative and not only raise questions but answer the key questions that are under investigation. The most frequent comment regarding the Introduction of a paper is the “So what?” question. Probably all scholars have been asked this question about research that they have presented. While it is always obvious to ourselves as authors that a certain topic is interesting and relevant, this is not always obvious to the wider readership or audience. Another question to which the answer is not always obvious is “Where is this going?” or “What, exactly, are the authors trying to achieve?” It often helps to explicitly mention a research question or research aim rather than treat the research question in an implicit way. It is also from the Introduction—next to a theoretical or background section—that it often becomes clear that authors have not engaged in an effective way with the relevant scholarship. Sometimes, we also ask authors to remove what we call “administrative text” from the Introduction. This is text that simply describes what is coming in the remainder of the manuscript. Particularly if the manuscript follows a standard structure (and most do), there is no justification for a simple signposting of what is to follow. Instead, we recommend to end the Introduction with a short substantive paragraph on what constitutes the work the paper is based on: most commonly, the data and methods employed. Nevertheless, it should be borne in mind that not all articles follow the same standard structure and this is certainly not something we want to impose on authors. As Editors, we fully recognise that each manuscript should be fully judged on its own merits. Some manuscripts are submitted to the journal that lack the theoretical depth one would expect for a high-quality social-scientific journal. This can include manuscripts that simply describe previous research findings and do not provide the theoretical rationale or framing for the current research. Of course it is important to refer to previous research to frame and clarify in what ways the submission contributes and advances the literature, but a summary of previous research can never replace a proper theoretical argumentation. Neither do we see “others have found this, too” as a convincing ground to postulate a hypothesis or argument. In some manuscripts, this could be a good reason for including a control variable. A specific problem concerning hypotheses that we sometimes encounter is that authors suggest that they will test a null hypothesis, of the type “We hypothesise that there will be no effect of X on Y”—and then use standard statistical techniques to assess whether the hypothesis is supported. That is, if the effect is not statistically significant, they conclude the hypothesis is supported. This cannot be concluded; one risks making a type II error. The only conclusion that can be drawn is that the hypothesis cannot be rejected, which is a different thing. The overall comments on data and methods are very diverse, but many have to do with reliance on data and/or methods that are not very well suited to answer the research question; failure to properly acknowledge the limitations of the data and/or methods; or insufficient description of data and/or methods. A common problem with manuscripts employing statistical methods is that methods are used that rely on assumptions of causality in cases in which these assumptions are dubious. A common problem with manuscripts employing in-depth interviews or other qualitative methods is that the recruitment of participants and the selectivity of the sample are not described in effective ways. In such manuscripts, it also happens quite frequently that the method of data collection is described, but the authors remain relatively silent about the analytical method (e.g., coding method). Our comments on sections presenting findings are also very diverse. Perhaps one of the most prevalent comments is that the authors put forward claims that are not substantiated by the empirical findings. A frequent problem in manuscripts employing statistical analyses is that the authors say “there is no effect of X on Y” whereas the statistical test they employ only allows them to conclude that no effect was found, that is, there was no evidence of such an effect—see Section 3.3. We think this is a very serious misinterpretation. Another common, related misinterpretation is that authors write that an effect disappears between one model and another, for example, after inclusion of a control variable, while in fact, the effect is estimated to be just as large but not statistically significant. As authors, reviewers, editors, and teachers of the new generations of scientists, we are obliged to identify and prevent such misinterpretations. And, more generally, to remind ourselves and others of what statistical significance is and is not, and how it is related to statistical power and thus the size of samples and categories in samples. A conclusion should only briefly summarise the main findings of the manuscript; a frequent comment is that the summary is too long or forms a too large part of the concluding section. It is also quite common that the concluding section lacks essential parts. For example, the limitations of the research are not discussed adequately, or options for future research and/or policy are overlooked. Our main criticism of research discussions is the absence of any discussion! Authors too frequently do little more than describe or present their results, commonly as a case study supporting previous research. Discussions should inform and engage with theoretical and methodological debates to demonstrate the original contribution and added value of the work. Failure to adequately do so risks the “So what?” question arising again. It is important that authors explicitly present their main argument and the key scholarly contribution: It is not the task of the reader to “join up the dots.” Final article texts should be written in impeccable English. Although very minor language issues can be dealt with by the copy-editing provided by the publisher, correct language is first and foremost the responsibility of the author(s). Next to comments related to language, texts frequently need to be tightened: Unnecessary parts and repetitions should be deleted; long-windedness should be avoided. We remind authors that the word limit for PSP manuscripts is 9,000. More often than not we have to advise authors to shorten their manuscripts during the revision stage. In our experience, losing several thousand words from an overly long manuscript does not detract from its arguments and, indeed, makes the scholarly contributions much more discernible to the reader. Texts should be readable not only for a small in-crowd of specialists but also for other interested scholars. We generally discourage the use of acronyms, particularly if they are specific to the context of a single article or not very widely used and thus less known. And even if they are well known, we think it is good practice to write them in full the first time they appear in the text. Footnotes disrupt the flow of the text and should be kept to a minimum; frequently, they can be either moved to the main text or deleted. Equations should be numbered; tables and figures should have proper captions and be understandable without reading the text. Another responsibility of the authors is to comply with the author guidelines, which follow those of the American Psychological Association (APA), and to include digital object identifiers (doi) in the reference list. The guidelines for tables and figures should also be followed. While completely complying with the guidelines is not pertinent upon first submission, it has to be done at some point. Seeing to it upon revision is strongly advised and helps avoid delays in the publication process. PSP uses the traditional double-blind peer review system that is common in the social sciences: Peer reviewers are invited by the journal editors, authors are not told who the reviewers are, and reviewers are not told who the authors are. Recently, there has been some debate on the extent to which this system remains adequate. We are, however, very happy about how the system functions for our journal. Reviewers usually do a great job and submit their reviews within a reasonable time. We recommend the instructive paper by Berk, Harvey, and Hirshleifer (2017) to those wishing to improve their reviews. The number of refusals to review is relatively modest, and only very rarely does a reviewer who has told us they will do the job not deliver in the end. A great advantage of our disciplinary area is that it is not huge. We, as editors, therefore personally know many of the reviewers we approach. Our impression is that many colleagues feel a great responsibility in taking part in the review process. We all know that high-quality peer review is essential to improve our own work and for most of us this means we do not think we can freeride and decline review requests unless we have a truly important reason to do so. Another great asset of the journal is our International Advisory Board of leading experts that span the key foci of the journal. To this board, we invite renowned colleagues who we trust as good reviewers and who commit to undertake six to eight reviews per year and occasionally review Special Issues. Many manuscripts are sent out to at least one member of the Board. Yet there are occasional papers for which it takes more time—weeks, or, in extreme cases, sometimes even months—to find reviewers because several potential reviewers decline a request. Working with a professional publisher—in our case Wiley—has many advantages. Many of the non-scientific tasks that need to be performed around publishing scientific work are undertaken by the publisher. To name just a few: administrative handling of submissions, final production, layout and copy-editing in one consistent style, the actual publication, journal website maintenance, ensuring indexation and “findability,” cross-referencing, ensuring access in university libraries … just about everything that makes submitting, finding, reading, and citing articles so easy these days. For us as editors, it is mainly the ScholarOne/ManuscriptCentral system, and the way this system is administered by Wiley, that makes our lives easy, saves us a lot of tedious work, and prevents us from making painful mistakes such as forgetting about a manuscript or losing a review. As Editors, we always conduct a final proof read of all papers and confirm to the publisher that they can proceed to publication. Despite the many advantages of working with a professional publisher, there are also some caveats. The publisher has quite a lot of power. All financial decisions are made by the publisher: They determine subscription rates and the cost of open access publishing, for example. Perhaps the biggest caveat from a (sub-)disciplinary perspective is that, in the case of PSP, the publisher owns the journal. In essence, this implies that the editors or the scientific community cannot walk away from the publisher should they no longer be satisfied with its services, although we would note that the current relationship with Wiley is a very positive one. Yet, despite the positive relationship, we think a relationship in which the publisher is commissioned by scientists owning the journal—for example, a professional scientific association—is to be preferred. This is in fact a common format for other journals in the field. In conclusion, we want to renew our strong encouragement to prospective contributors for the submission of original research that advances knowledge of the spatialities of population phenomenon, including migration and mobilities, health and fertility, ageing, mortality, population change and distributions, and household and family formation. We hope this Editorial both supports and propels more researchers, from a gamut of different epistemological and ontological positions, to submit diverse, cutting-edge scholarship that progresses our understandings of the dynamic connections between population, space, and place.