You are here

The pitfalls of "salami slicing": Focus on quality and not quantity of publications

마리샤 폰세카 | 2013년11월4일 | 조회수 109,165
Salami slicing

Imagine you have just completed a controlled study about a new intervention in a birthing center. You have two sets of results: one set on mothers and one on infants. Should the author write two papers—each reporting a different set of results—and send these papers to two different journals? Or consider a case where you are studying several closely related compounds.1 Should you write a separate paper for each compound?

The answer to both questions is no. Editors consider these as cases of “salami slicing”—unethically fragmenting the results of a single study and reporting them in multiple papers.

What is salami slicing?

Salami slicing refers to the practice of partitioning a large study that could have been reported in a single research article into smaller published articles.2 In other words, it means breaking up a single research paper into their “least publishable units,” with each paper reporting different findings from the same study. A set of papers are referred to as salami publications when more than one paper covers the same population, methods, and research question.3

Salami slicing
A journal editor gave the following examples to illustrate a case of salami slicing.4 Can you determine which is the case of salami slicing?

Scenario 1 : A scientist begins a new line of research. The scientist has developed a new instrument for collecting data, one that is more precise than the current instruments. The main study may take a year or over to complete. The scientist submits a manuscript for publication describing the new instrument before completing the main study.

Scenario 2 : After determining the research question and setting the study design, a scientist collects data on three groups of participants. Two of the groups have different types of aphasia (Groups A and B), and one group is a control group. The scientist submits two manuscripts for publication: one comparing Group A with the control group, and the other comparing Group B with the control group.

Scenario 1 is not likely to be considered as a salami publication. The new instrument was not a part of the research question, but rather was developed to answer the research question. Further, other scientists benefit from the publication because they can also use the new data collection method. When publishing the main study, the scientist need not describe the instrument in detail in the Methods section, but rather should refer to the previous publication. Scenario 2 is likely to be considered a salami publication. All of the data should be published in a single manuscript.
What’s wrong with doing this?
  • Career distortions. Salami slicing is widely regarded as a practice that researchers employ to increase their volume of publications, borne of the “publish and perish” culture.5 In the short term, salami science may allow scientists and researchers to progress faster in their careers or receive more funding than they actually merit, owing to the greater number of publications they can list on their resume.6,7 However, salami slicing can be harmful in the long term, since it diminishes the value of each publication. You may have managed to add a long list of publications to your name through salami publications, but if a committee were to review the body of work, they might conclude that the studies themselves are not substantial enough.
  • Harm to science. Publishing unnecessary and repetitive information increases the amount of literature, but not the amount of knowledge. If closely related data from a single group is divided across several papers, readers who access only one of the papers may misinterpret the findings. Further, multiple reports may cause a set of findings to be given more importance that it deserves. For instance, in the example mentioned in the beginning, another researcher conducting a meta-analysis on the new intervention for birthing centers might erroneously assume that this intervention has been studied twice, rather than once.

Salami slicing

Is it always wrong to report a single study through multiple papers?
In some cases where the original data-set is extremely large (e.g., a population-based study) and when the data-set takes years to collect and analyze, the authors have justifiable and legitimate grounds to report the research in more than one paper.6,8 However, each paper should address distinct and important questions.8 If the study is motivated and designed around a single hypothesis, its results should be presented to the readers as a single package, regardless of the size of the data-set. 4

If you do have legitimate grounds to submit multiple publications based on the same study, ensure that you inform the editorial office about any possibly overlapping information (including whether any of the control data in a manuscript are also included among the control data in another manuscript or whether you have previously published articles on the same or a closely related topic) either before submitting a paper or in the accompanying cover letter.9 In addition, refer to all related studies within the manuscript. 

Salami slicing

Conclusion

A paper will have a greater chance of publication as a full-scale study, rather than a fragment of a larger study. Focus on the quality of your publications, not quantity. Salami slicing to increase the number of publications on your resume might only end up sabotaging your research career at a later stage.

 

For a quick and easy-learning video on salami slicing, click here:

스크랩하기

해당 기사를 스크랩해보세요!

지식은 모두에게 함께 공유되어야 한다는 것이 에디티지 인사이트의 이념입니다. 해당 사이트에서 제공되는 모든 기사는 Creative Commons license로 재포스팅 및 스크랩이 가능합니다. 아래의 가이드라인만 유념해주신다면 언제든지 무료로 에디티지 학술 전문가의 지식을 가져가실 수 있습니다!


  • 주의 : 에디티지 학술 전문가들은 해당 콘텐츠를 만들기 위해 많은 시간과 노력을 쏟고 있습니다. 기사를 스크랩 및 재포스팅 하실 때는 명확한 출처를 남겨주시기 바랍니다.
  • 이미지 재사용: 이미지를 원본이 아닌 편집 재사용하실 때는 에디티지 인사이트의 허가가 필요합니다.

코드를 복사하셔서 기사 공유를 원하시는 사이트에 적용하시면 에디티지 인사이트 기사를 가장 쉬운 방법으로 공유하실 수 있습니다.
 
Please copy the above code and embed it onto your website to republish.

Comments