SEARCH

SEARCH BY CITATION

Keywords:

  • journal;
  • manuscript rejection;
  • publishing;
  • research

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. ACKNOWLEDGMENTS
  8. REFERENCES

Better understanding of the reasons why manuscripts are rejected, and recognition of the most frequent manuscript flaws identified by reviewers, should help submitting authors to avoid these pitfalls. Of 219 manuscripts submitted to Veterinary Radiology & Ultrasound in 2012, none (0%) was accepted without revision, four (2%) were withdrawn by the authors, 99 (45%) were accepted after revision, and 116 (53%) were rejected. All manuscripts for which minor revision was requested, and 73/86 (85%) manuscripts for which major revision was requested, were ultimately accepted. Acceptance rate was greater for retrospective studies and for manuscripts submitted from countries in which English was the primary language. The prevalences of flaws in manuscripts were poor writing (62%), deficiencies in data (60%), logical or methodological errors (28%), content not suitable for Veterinary Radiology & Ultrasound (26%), and lack of new or useful knowledge (25%). Likelihood of manuscript rejection was greater for lack of new or useful knowledge and content not suitable than for other manuscript flaws. The lower acceptance rate for manuscripts from countries in which English was not the primary language was associated with content not suitable and not poor writing. Submitting authors are encouraged to do more to recognize and address manuscript flaws before submission, for example by internal review. Specifically, submitting authors should express clearly the potential added value of their study in the introduction section of their manuscript, describe completely their methods and results, and consult the Editor-in-Chief if they are uncertain whether their subject matter would be suitable for the journal.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. ACKNOWLEDGMENTS
  8. REFERENCES

Publishing is one of the principal academic activities. The peer-reviewed literature is the permanent record of scholarly achievement; an individual's publication record is a focus of appointment and promotion boards at universities; and publication is an essential credential of many specialty boards, including the European College of Veterinary Diagnostic Imaging. Publication in a peer-reviewed journal requires a submitted manuscript to pass scrutiny by one or more reviewers, who are chosen by the journal Editor on the basis of their experience and asked to judge the manuscript's quality.[1, 2]

Submitting authors need to understand the criteria for quality used by reviewers and editors to maximize the likelihood of having their manuscript accepted by ensuring that their manuscript satisfies these standards before submission. For Veterinary Radiology & Ultrasound, these criteria are originality, novelty; significance, importance; scientific quality, hypothesis, experimental design; interest for the journal's readers; composition, clarity, and organization; and adherence to author guidelines.[3] Authors submitting a manuscript to Veterinary Radiology & Ultrasound also may find detailed descriptions of the types of manuscripts of interest to readers, their characteristics, and required format.[3] When examining a typical issue of Veterinary Radiology & Ultrasound, readers should observe that the published articles satisfy the journal's criteria for quality. Submitting authors can learn a lot about how to prepare a manuscript for publication by studying articles on similar subjects.

It is also instructive to review the reasons why manuscripts are rejected, and to highlight the most frequent manuscript flaws identified by reviewers, as a means of helping authors to avoid these pitfalls.[4, 5] The most frequent reasons for rejection of manuscripts submitted to medical journals are the study does not add anything new or useful to the existing literature, and methodological errors including weak hypothesis, lack of alignment between aims and methods, and unsuitable control group.[6, 7] The aims of the present study were to determine the acceptance rate and relative frequency of flaws in manuscripts submitted to Veterinary Radiology & Ultrasound during 2012, and to test potential associations between acceptance and manuscript type, country of origin, and first author's place of work, and associations between acceptance and manuscript flaws.

Methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. ACKNOWLEDGMENTS
  8. REFERENCES

Reviewer reports and the Editor-in-Chief's decision letter were extracted from the publisher's database for all new manuscripts submitted to Veterinary Radiology & Ultrasound between January 1 and December 31, 2012. The acceptance rate was calculated as the number of manuscripts accepted (including those accepted after review/revision) divided by the total number of manuscripts submitted. The following information was recorded: manuscript type (original investigation, retrospective study, review article, imaging diagnosis, case report, editorial, or letter to the Editor):[3] country of first author's address; first author's place of work (academic institution, veterinary practice, or industry); and Editor-in-Chief's decision at first submission (accept without revision, minor revision necessary, major revision necessary, or reject). Major revision was defined by Veterinary Radiology & Ultrasound as that requiring review on resubmission.

Reviewer reports were examined manually by two investigators (C.R.L. and W.M.) to identify text compatible with flaws in the manuscript. Both investigators were members of the Veterinary Radiology & Ultrasound Editorial Board and agreed to maintain author and reviewer confidentiality prior to starting the study. Investigators examined reviewer reports with knowledge of the title of the manuscript, the authors’ names, and Editor-in-Chief's decision, but without knowledge of the identity of the reviewers or the content of the manuscript. Reviewer reports were copy/pasted from the publisher's database into a document for annotation by the investigators. Manuscript flaws identified by reviewers were classified as follows: (1) lack of new or useful knowledge; (2) logical or methodological errors; (3) errors in data handling; (4) poor writing; (5) content not suitable for Veterinary Radiology & Ultrasound; (6) referees do not like the manuscript[6] (Table 1). Each investigator examined the reviewer reports in batches of 10–12 reports, annotated potential flaws, and then passed annotated reviews to the other investigator for checking. Differences in opinion of investigators about flaws were resolved by discussion.

Table 1. Classification of manuscript flaws identified by reviewers[6]
 Examples
I. Lack of new or useful knowledgeLack of clinical relevance
 Old knowledge with no new or useful material
II. Logical ormethodological errorsFundamentally weak hypothesis
 Unclear study aims
 Lack of correlation between purpose and methods or results
 Unsuitable control group
III. Deficiencies in dataErrors in arithmetic or statistics
 Biased data or insufficient data to justify results and conclusion
 Incomplete description of methods or results
 Diagnosis in doubt in case report/series
IV. Poorly writtenMajor deficiencies in language
 Inaccurate terminology
 Unbalanced argument, key references not cited
 Lack of discussion of limitations of study
 Images of poor quality, inappropriate or incorrectly interpreted
 Failure to follow Veterinary Radiology & Ultrasound author guidelines
V. Not suitable for Veterinary Radiology & UltrasoundPublished previously
 Excessively high similarity index
 Subject matter better suited to another journal
VI. Referees do not likeNone of the other reasons, but reviewers do not like the manuscript

Statistical analysis was performed by one of the authors (C.R.L.) who tested for associations between acceptance and manuscript type, country of origin, and first author's place of work, and associations between manuscript flaws and Editor-in-Chief's decision, using cross-tabulation and Fisher's exact test (SPSS Statistics, version 19, IBM Corporation, Chicago, IL). Results with P < 0.05 were considered significant.

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. ACKNOWLEDGMENTS
  8. REFERENCES

Between January 1 and December 31, 2012, 219 new manuscripts were submitted to Veterinary Radiology & Ultrasound. These submissions included 126 original investigations, 45 imaging diagnoses, 44 retrospective studies, three case reports, and one review. No manuscript was accepted without revision, four (2%) were withdrawn by the authors, 99 (45%) were accepted after revision, and 116 (53%) were rejected (Fig. 1). All manuscripts for which minor revision was requested, and 73/86 (85%) manuscripts for which major revision was requested, were ultimately accepted. Accepted manuscripts included 53/126 (42%) original investigations, 15/45 (33%) imaging diagnoses, 30/44 (68%) retrospective studies, 0/3 (0%) case reports, and 1/1 (100%) review. A significantly larger proportion of retrospective studies was accepted compared to other manuscript types (P = 0.001).

image

Figure 1. Outcome of manuscripts submitted to Veterinary Radiology & Ultrasound in 2012.

Download figure to PowerPoint

The proportion of manuscripts accepted from each country that submitted at least five manuscripts was 46/79 (58%) for the United States, 16/22 (73%) for the United Kingdom, 6/15 (40%) for Italy, 3/10 (30%) for Germany, 5/9 (56%) for Belgium, 0/9 (0%) for Brazil, 3/8 (38%) for Canada, 3/7 (43%) for the Republic of South Africa, 1/6 (17%) for the Republic of Korea, 1/6 (17%) for Denmark, 1/6 (17%) for Egypt, 3/5 (60%) for Switzerland, 3/5 (60%) for Spain, and 2/5 (40%) for Japan. Countries that submitted fewer than five manuscripts were Australia, Austria, China, Chile, Czech Republic, France, Finland, India, the Islamic Republic of Iran, Republic of Ireland, Philippines, Poland, Portugal, Sweden, and Turkey. A significantly larger proportion of manuscripts was accepted from countries in which English was the primary language (71/116, 61%) compared to manuscripts from other countries (28/103, 27%) (P < 0.001). The low acceptance rates of manuscripts from countries in which English was not the primary language were explored by comparing the prevalences of manuscript flaws. A significantly larger proportion of manuscripts with content not suitable for Veterinary Radiology & Ultrasound (P < 0.001) came from countries in which English was not the primary language (43/102, 42%) than from other countries (14/114, 12%). No other flaw occurred with greater frequency in manuscripts from countries in which English was not the primary language.

The proportion of manuscripts accepted from first authors in academic institutions was 194/219 (89%) and from first authors in veterinary practices was 25/219 (11%). No manuscript was received from a first author in industry. There was no significant association between acceptance and first author's place of work (P = 0.71).

Three of 219 manuscripts were withdrawn by the authors prior to review. The prevalences of flaws in the remaining 216 manuscripts subjected to review were 134/216 (62%) poorly written, 129/216 (60%) deficiencies in data, 60/216 (28%) logical or methodological errors, 57/216 (26%) content not suitable for Veterinary Radiology & Ultrasound, and 54/216 (25%) lack of new or useful knowledge. There was no instance of a manuscript for which the only flaw was that referees did not like the manuscript.

Associations between the Editor-in-Chief's decision at first submission and manuscript flaws are summarized (Table 2). Likelihood of a decision to reject a manuscript was significantly greater for lack of new or useful knowledge (87%) and content not suitable for Veterinary Radiology & Ultrasound (98%) than for other manuscript flaws (P < 0.001). The average number of types of flaws was not greater for rejected manuscripts than for manuscripts for which revisions were requested.

Table 2. Associations between manuscript flaws and Editor-in-Chief's decision
 Editor-in-Chief's decision
 AcceptMinor revisionMajor revisionReject
Flawn = 0n = 26n = 86n = 104
  1. I, lack of new or useful knowledge; II, logical or methodological errors; III, errors in data handling; IV, poor writing; V, content not suitable for Veterinary Radiology & Ultrasound; VI, referees do not like the manuscript.[6]

I007 (8%)47 (45%)
II03 (12%)39 (45%)18 (17%)
III023 (89%)74 (86%)32 (31%)
IV025 (96%)72 (84%)37 (36%)
V001 (1%)56 (54%)
VI0000
Average number of flaws per manuscript02.02.21.8

Associations between manuscript type and manuscript flaws are summarized (Table 3). Original investigations were associated with logical or methodological errors (P < 0.001), poor writing (P = 0.017), and content not suitable for Veterinary Radiology & Ultrasound (P = 0.044). Retrospective studies were associated with logical or methodological errors (P < 0.001), errors in data handling (P = 0.014), and poor writing (P = 0.017). Imaging diagnoses were associated with content not suitable for Veterinary Radiology & Ultrasound (P = 0.044).

Table 3. Associations between manuscript flaws and manuscript type
 Manuscript type
 OriginalRetrospective ImagingCase
 investigationstudyReviewdiagnosisreport
Flawn = 124n = 43n = 1n = 45n = 3
  1. I, lack of new or useful knowledge; II, logical or methodological errors; III, errors in data handling; IV, poor writing; V, content not suitable for Veterinary Radiology & Ultrasound; VI, referees do not like the manuscript.[6]

I31 (25%)9 (21%)1 (100%)12 (27%)1 (33%)
II39 (32%)20 (47%)01 (0.2%)0
III77 (62%)32 (74%)020 (44%)0
IV81 (65%)31 (72%)1 (100%)20 (44%)1 (33%)
V36 (29%)5 (12%)014 (31%)2 (67%)
VI00000
Average number of flaws per manuscript2.12.32.01.51.3

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. ACKNOWLEDGMENTS
  8. REFERENCES

For 219 manuscripts submitted to Veterinary Radiology & Ultrasound in 2012, the acceptance rate was 45%. This value may be compared with an average acceptance rate of 52% for 17 radiology journals surveyed in 2006.[8] The acceptance rate for manuscripts by a journal depends on the quality of manuscripts submitted, the rate at which new manuscripts are received, and the number of papers already accepted into the publication process. Acceptance rate is weakly negatively correlated with journal impact factor.[8] Relatively few journals publish their annual acceptance rate. Journals may include acceptance rate in their web site, and a few routinely publish this information as part of an annual review,[9] but most avoid publishing acceptance rates, possibly because they think submitting authors would be discouraged by a low acceptance rate.

A large proportion of manuscripts submitted to medical journals is rejected or of insufficient quality to be published without revision. Even in manuscripts that are ultimately accepted for publication, reviewers frequently identified flaws. The types of flaws most frequently identified by reviewers in manuscripts submitted to Veterinary Radiology & Ultrasound in 2012 were poorly written and deficiencies in data. These flaws frequently resulted in a request for major revision, but the fact that most manuscripts requiring major revision were ultimately accepted indicates that it was possible for authors to adequately address these flaws. Some of the benefits of the peer-review process are improved scientific quality and readability of their manuscript; however, submitting authors are encouraged to do more to recognize and address manuscript flaws before submission, for example by internal review.

Flaws representing lack of new or useful knowledge and content not suitable for Veterinary Radiology & Ultrasound were less frequently identified by reviewers, but were strongly associated with a decision to reject a manuscript. Lack of new or useful knowledge was associated with original investigations and imaging diagnoses. Submitting authors are encouraged to thoroughly review the existing literature to assess the potential added value of their project and to express this clearly in the introduction section of their manuscript. Content not suitable for Veterinary Radiology & Ultrasound was significantly associated with rejection of imaging diagnosis manuscripts and rejection of manuscripts from countries in which English was not the primary language. In many instances, the decision to reject because content was not suitable for Veterinary Radiology & Ultrasound was made by the Editor-in-Chief after consultation. During the period of study, the journal faced a backlog of manuscripts in the category of imaging diagnosis, and the high rejection rate for these manuscripts reflected an increased threshold instituted by the Editor-in-Chief as a means of avoiding excessively long periods between submission and publication imposed by the page limit. When manuscript content was considered not suitable for Veterinary Radiology & Ultrasound, resubmission to other specific journals was suggested frequently by the Editor-in-Chief. Submitting authors should be able to judge if Veterinary Radiology & Ultrasound is likely to consider their manuscript for publication based on the journal's stated aims and review of previous content. When doubt exists, an informal approach to the Editor-in-Chief, including the summary section of the manuscript, is suggested as a means of avoiding upfront rejection for this reason.

Acceptance rate was significantly lower for manuscripts submitted from countries in which English was not the primary language because a large proportion of those manuscripts were considered not suitable for Veterinary Radiology & Ultrasound on the basis of their subject matter, for example imaging diagnosis or anatomic studies of nondomesticated species. Manuscripts from countries in which English was not the primary language might exhibit more flaws associated with poor writing; however, that was not observed in the present study. Another recent study concluded that language problems were not a major cause of manuscript rejection, except for manuscripts from China.[6] Our classification of poor writing encompassed language difficulties as well as various other prevalent flaws, such as unbalanced argument, failure to cite key references, and lack of discussion of limitations of the study. This classification may have obscured differences in the prevalence of language difficulties between authors from different countries.

The present study was based on investigators extracting reasons for rejection from reviewer reports and the Editor-in-Chief's decision letter to authors. When planning this study, we expected that in most instances these documents would explicitly state the reason(s) for the Editor-in-Chief's decision, and that we would be able to identify reasons for rejection by searching for certain key phrases; however, many reviewer reports took the form of a list of comments or questions for the authors without any apparent priority or specific request for remedial action. Often there was no summary of the strengths or weaknesses of the manuscript or a conclusion about its value. Also, the basis for the Editor-in-Chief's decision was not always explicitly stated. As a result, the task of extracting flaws from reviewer reports and the Editor-in-Chief's decision letter was more subjective than anticipated, and investigators were required to discuss and resolve differences in their results for many of the manuscripts under review. Just as we sometimes found it difficult to identify and classify the reasons for rejection, submitting authors might experience similar difficulties. Hence, we reiterate existing advice to manuscript reviewers that reviews should routinely include a summary of the value of the work. [2] Similarly, it would be helpful for the Editor-in-Chief's decision letter to identify critical flaws within the manuscript that was the basis for rejection. We also suggest that the Editorial Board of Veterinary Radiology & Ultrasound consider routinely recording specific reason(s) for manuscript rejection in their publishing database, as is done by certain other leading journals, for the purposes of retrospective review, education of submitting authors, and enhanced communication.[6]

One of the aims of the present study was to determine the relative frequency of different types of manuscript flaws identified by reviewers. For this purpose, we used a previously reported classification scheme.[6] We did not attempt to quantify the number of individual flaws identified by reviewers in each manuscript, hence we cannot incorporate the number of flaws into the analysis. Most types of manuscript flaws occurred for most decisions of the Editor-in-Chief, and it seems clear that flaws can encompass a range of severity, and that a different combination of severity and number of flaws could lead to the same Editor-in-Chief decision. Our analysis was insufficiently detailed to explore why a certain type of manuscript flaw could lead to a request for minor revision in some instances and rejection in others.

This study did not address questions about the accuracy of manuscript reviews or the correctness of a rejection decision. It is inevitable that some important flaws in manuscripts will be missed by reviewers, and that manuscripts will occasionally be accepted that should have been rejected. Manuscripts also may be erroneously rejected, but rejection of a manuscript is not necessarily fatal as a proportion of rejected manuscripts are accepted subsequently by other journals.[8-13] Abundant advice and some training courses are available for manuscript review.[1, 2, 14] However, manuscript reviewers are largely self-taught and reviews are variable in quality.[15] Additionally, the reliability of manuscript reviews is low.[16] In general, higher quality reviews are associated with lower acceptance rates.[17] In an experiment in which a fictitious manuscript containing intentional errors was submitted repeatedly for peer review, reviewers who recommended rejection identified more errors than reviewers who recommended revision or acceptance.[18] It seems clear that there is a need for more education of manuscript reviewers and journals to monitor the quality of reviews.[19] Multiple studies have highlighted deficiencies in statistical analysis in accepted manuscripts, [20-23] which suggest a specific focus for education of manuscript reviewers. A study to assess the quality of manuscript reviews done for Veterinary Radiology & Ultrasound would require independent expert assessment of manuscript reviews and the corresponding manuscripts, which was outside the scope of the present study.

ACKNOWLEDGMENTS

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. ACKNOWLEDGMENTS
  8. REFERENCES

We are grateful for the advice and participation of the journal's publishing team, who made this study possible: Lucinda Ayres, Managing Editor; Meredith Duey, Associate Editor, Wiley; Sonia Wilson, Electronic Editorial Manager, Wiley; and Professor Jeryl C. Jones, Editor-in-Chief.

REFERENCES

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. ACKNOWLEDGMENTS
  8. REFERENCES