• Time for a change

    In this blog I suggest that there's a better way to write a scientific article. Currently, published articles are unnecessarily difficult to read and researchers need to be trained in how to write about their research so that others can understand their a

    • On 2 November, 2007 the journal Science launched an experiment to increase the readability of the research articles they publish. For a five-week period, each article was accompanied by a one-page “Authors’ Summary” designed to make these papers more accessible – because as Editor-in-Chief Donald Kennedy pointed out “It’s clear that accessibility is a problem, because we’re all laypeople these days” (Science Vol 318).

      While this experiment signals a growing awareness that articles should be written in a style that makes them available to a wide readership, the question is – do the Authors’ Summaries make research papers clearer?

      Kennedy makes several observations about the impediments to clear communication in science. He points out that “it’s important for science journalists and scientists (to) find new ways of working more effectively with one another.” However, when scientists write manuscripts that accurately communicate their important findings, then everyone benefits including the researcher’s colleagues, educated readers, science journalists, and civil society. In other words, it isn’t the journalist’s role to reinterpret a scientist’s writing; scientists should learn how to communicate effectively in the first place.

      Kennedy makes a second point that as science becomes increasingly specialized, researchers overuse abbreviations and specialized terms, making articles difficult for readers to understand. Explaining specialized terminology is a relatively easy thing to do. More egregious are articles that contain lapses in logic, assumptions about what readers know, and omissions of essential information. These issues can be easily addressed by careful editing.

      Third, while some editors may “push hard to make authors compress their language”, it’s not obvious why this should necessarily lead to poor writing. Several editing techniques exist to make writing more efficient by eliminating clutter and simplifying awkward constructions. Better editing also gives the writer space to include information that is essential for the non-specialist to understand the author’s story. Such scrupulous revision does not mean that authors “sacrifice precision and accuracy”; in fact, effective editing eliminates vagueness and restores, rather than diminishes, a text’s accuracy.

      Since many young scientists often turn to high profile journals like Science, Nature, and Cell as models of how to write a research paper, if the models are weak, as Kennedy suggests, then a cycle of poor writing is perpetuated. Given such a prominent role, these journals have an obligation to make clarity a serious prerequisite for publishing a paper.

      So we return to the question: can research articles be made more accessible? Careful analysis of the articles and Authors’ Summaries included in this experiment reveals that many obstacles undermining clear communication in the original manuscripts – also occur in the Summaries. Unfortunately, little will be gained if researchers fail to conquer errors of style and are simply made to write more.

      Greater effort by both authors and journals – rather than Authors’ Summaries – will go a long way to increasing a paper’s readability. Authors could do more to revise their manuscripts while journals could apply more rigorous writing standards. Delaying the publication of papers until they meet established criteria for clear and accessible writing could provide a strong incentive for scientists to write with greater care.

      I’d like to know what the other people think about this ‘experiment’ and if anyone else suspects that there may be other, more effective solutions to confused writing in science.

    • If you want your readers to enjoy your Abstract, tell them a good story. Capture their attention with the first sentence and they’re likely to continue reading – a significant accomplishment since scientists have so much to read and so little time to read everything.

      To prove the validity of my claim, I checked out the first sentences of nine Abstracts listed below in the 7 September 2006 volume of Nature.

      1. The worldwide spread of H5N1 avian influenza has raised concerns that this virus might acquire the ability to pass readily among humans and cause a pandemic. P. 45

      2. Explosive volcanic eruptions are driven by exsolution of H2O-rich vapour from silicic magma. P. 76

      3. Categorization is a process by which the brain assigns meaning to sensory stimuli. P. 85

      4. Sex and recombination are widespread, but explaining these phenomena has been one of the most difficult problems in evolutionary biology. P. 89

      5. Benthic foraminifera are unicellular eukaryotes found abundantly in many types of marine sediments. Many species survive and possibly reproduce in anoxic habitats, but sustainable anaerobic metabolism has not been previously described. P.93

      6. The ability of organs such as the liver or the lymphoid system to maintain their original size or regain it after injury is well documented. However little is known about how these organs sense that equilibrium is breached, and how they cease changing when equilibrium is reached. P. 97

      7. Male infertility is a long-standing enigma of significant medical concern. P. 101

      8. Polarized cellular distribution of the phytohormone auxin and its carriers is essential for normal plant growth and development. P. 106

      9. RNA degradation is a determining factor in the control of gene expression. P.110

      Example #1 piques my interest because I can readily identify the subject “worldwide spread of H5N1 avian influenza” and the problem “this virus might acquire the ability to pass readily among humans”. Interestingly, this strong sentence makes a specific statement about the problem and does not depend on the weak linking verb “to be”. (Not surprisingly perhaps, the rest of this Abstract is easy and enjoyable to read.)

      Introductory sentences 2,4,5,8 all have the verb “to be” construction in the form of sentence subject + ”are driven”, “are widespread”, “ “are found”, “is essential”.
      My point here is that this pattern, because it’s overused, can be boring. The information in these sentences tends to be general in nature and not very exciting for the reader.

      Sentences 3,7,9 follows the “is a …” structure as in “categorization is a process”, “male infertility is a long-standing enigma”, “ RNA degradation is a determining factor”. Again the information here tends to be broad and seems to be an unnecessarily slow wind up to the real story the writers want to tell.

      Based on this brief analysis, I’d like to suggest the following guidelines:

      By identifying the specific subject (“the worldwide spread of H5N1 influenza”) and the specific problem (“this virus might acquire the ability to pass readily among humans”) and avoiding the overused “is a” and “subject + are” constructions, you’re more likely to ensure that the first sentence of the Abstract is content rich, that it carries important information. And if you follow these principles, you’re also more likely to capture your readers’ attention as well as the excitement behind your research.

    • Hi Everyone,
      I’m in the process of writing a book on clear writing in science and one section deals with editing. To help researchers revise their texts effectively, I will include examples of what appear to me to be unrevised sentences or paragraphs from published scientific articles and offer an edited version for readers to compare. I’ve just finished working on a short paragraph and want to share the process with you. Below is the original text, my revision, and a short explanation of how I deconstructed the original. I would love to know if you find the example helpful – and if you think my revised version is easier to read. Thanks for your feedback!

      Original:
      The Allen Brain Atlas has taken a genomics-style approach to understanding this complexity by creating an integrated set of data production and analysis methodologies to systematically produce and analyse a comprehensive atlas of gene expression in the adult C57BL/6J mouse brain. Accurate and comprehensive analysis and annotation of data from the Allen Brain Atlas and other similar projects are difficult but essential hurdles to realizing the full potential of these data. New suites of image analysis tools are required to apply analysis methods developed by the genomics community to cellular-resolution gene expression data in the brain. Ongoing efforts aim to allow correlative cross-gene and cross-structural analysis across the entire Allen Brain Atlas data set by improving automated and semi-automated methodologies in order to quantify and map gene expression at increasingly finer anatomical resolution.

      Revised:
      The Allen Brain Atlas uses genomics analysis to produce and analyze systematically a comprehensive atlas of gene expression in the adult C57Bl/6J mouse brain. For researchers to realize the full potential of these data, they need accurate and comprehensive analysis and annotation of data from the Allen Brain Atlas and other similar projects. While obtaining such analysis and annotation may be difficult, this is a hurdle worth overcoming. In addition, researchers require image analysis tools that enable them to use the analysis methods developed by the genomics community on the gene expression data of the brain. By improving automated and semi-automated methodologies to quantify and map gene expression at increasingly finer anatomical resolution, ongoing efforts aim to allow correlative cross-gene and cross-structural analysis across the entire data set of the Atlas.

      Here’s how I deconstructed the original paragraph:
      Original
      The Allen Brain Atlas has taken a genomics-style approach to understanding this complexity by creating an integrated set of data production and analysis methodologies to systematically produce and analyse a comprehensive atlas of gene expression in the adult C57BL/6J mouse brain.

      Deconstructed
      The Allen Brain Atlas uses (rather than “takes”) genomics analysis (rather than “genomics-style approach”) to create (rather than by creating because the action is the wrong way around: they use the genomics to create the set of analysis methods: the integrated set of data production does NOT lead to their genomic-style approach!) an integrated set of data production and analysis methodologies that produce and analyze systematically a comprehensive atlas of gene expression in the adult C57Bl/6J mouse brain.

      Revised
      The Allen Brain Atlas uses genomics analysis to produce and analyze systematically a comprehensive atlas of gene expression in the adult C57Bl/6J mouse brain.

      Original
      Accurate and comprehensive analysis and annotation of data from the Allen Brain Atlas and other similar projects are difficult but essential hurdles to realizing the full potential of these data.

      Deconstructed
      Accurate and comprehensive analysis and annotation of data from the Allen Brain Atlas and other similar projects are difficult but essential hurdles (missing information) to overcome to realizing the full potential of these data. (Incorrect word order, passive voice, confused sentence subject, and faulty transition).

      Revised
      To realize the full potential of these data, researchers need accurate and comprehensive analysis and annotation of data from the Allen Brain Atlas and other similar projects. While obtaining such analysis and annotation may be difficult, this is a hurdle worth overcoming.

      Original
      New suites of image analysis tools are required to apply analysis methods developed by the genomics community to cellular-resolution gene expression data in the brain.

      Deconstructed
      New suites of image analysis tools are required to apply analysis methods developed by the genomics community (confused word order – misplaced modifer) to cellular-resolution gene expression data in the brain.

      Revised
      Researchers require new image analysis tools that enable them to use the analysis methods developed by the genomics community on the gene expression data of the brain.

      Original
      Ongoing efforts aim to allow correlative cross-gene and cross-structural analysis across the entire Allen Brain Atlas data set by improving automated and semi-automated methodologies in order to quantify and map gene expression at increasingly finer anatomical resolution.

      Deconstructed
      Ongoing efforts aim to allow correlative cross-gene and cross-structural analysis across the entire Allen Brain Atlas data set by improving automated and semi-automated methodologies in order to quantify and map gene expression at increasingly finer anatomical resolution. (Confused word and idea order).

      Revised
      By improving automated and semi-automated methodologies to quantify and map gene expression at increasingly finer anatomical resolution, ongoing efforts aim to allow correlative cross-gene and cross-structural analysis across the entire data set of the Atlas.

    • The purpose of the Abstract has changed. No longer a document for library cataloguing, the Abstract is a tool for rapidly communicating important findings to interdisciplinary readers. It may be the only part of the paper that many readers will read.

      Given its prominence, researchers need to make sure that every word is precise, that clutter is ruthlessly removed, that specialized terms are strictly limited and that the structure is logical.

      Consider the Structured Abstract: The beauty of the Structured Abstract is that it tells “the story” of your research in a logical fashion that readers can follow easily. This format also helps the writer keep discrete elements of the abstract in their proper places – no more mixing Introduction and Methods, Results and Discussion- or Results in the Discussion. (Such a hodgepodge is confusing to reader and writer alike!)

      The Structured Abstract has 4 parts: the Introduction, Methods, Results, and Discussion.

      The Introduction, usually written in the present tense and about 2-3 sentences long, provides the context for the work by identifying the specific field/category of the research and explains where the question comes from. An effective Introduction includes a hypothesis, states the question to be answered, or the purpose of the study.

      Methods, usually written in the past tense and about 3-4 sentences long, explain what procedures you followed to test the hypothesis or answer the question. This section should include only the principle techniques that you used that establish the reliability of the study.

      Results, usually written in the past tense and about 2-4 sentences long, identify the answers that you found. The Results should reflect the Methods and appear in the same order as the Methods.

      The Discussion, usually written in the present tense and about 2-3 sentences long, points out the most significant finding and explains the connection between the important finding and why it matters to the particular field of research.

      Since the Abstract can serve as a blueprint for the full manuscript, I recommend that you write it first. This approach can also be a time saver because writing the Abstract forces you to articulate your hypothesis and nail down your important finding – rather than wandering around a sea of data as you formulate your ideas. And if you write the Abstract first then you won’t submit a hastily written text that fails to capture the significance of your work.

      Since the Abstract has become the basic unit of communication for manuscripts and scientific meetings, it deserves to be written and edited scrupulously.

    • Small changes that work for writer and reader

      Wednesday, 11 Apr 2007

      Previously I mentioned that it’s time for the writing in science research articles to change – to become clearer and more accessible to a wider audience. I think that much of the confusion in the literature stems from the fact that scientists are rarely informed about techniques on structuring and revising their articles.

      Below are a few ideas about how researchers can achieve a clearer and more natural style of writing. These suggestions can help researchers marshal their data into a coherent story; they also help readers because a logically structured account of the significant finding is so much easier for readers to understand than a random or semi-structured compilation of data.

      These tips relate to the structure and content of a manuscript, beginning with the paper Abstract. As I posted in a response to my previous entry, most agree that it’s crucial to get the Abstract right because more readers will read your Abstract that any other part of the manuscript. And if busy readers like your Abstract, then they’re likely to continue reading the full paper. As well, if you get the information right in the Abstract, it can serve as a blueprint for your paper. (Try writing your Abstract first because you can see quite quickly if you have found or missed your true focus – your single most important finding – and if you’ve included too much detail.) I recommend the Structured Abstract because it tells the story of your research in a way that’s logical to your readers. If anyone wants the details of the Structured Abstract, I’d be happy to post them – with examples of original(unstructured) and revised (structured) Abstracts.

      The next suggestion relates to the content of the article. Once you’ve established the focus of your paper (your most significant contribution) and written a draft of your Abstract or paper, it’s crucial to revise your writing to overcome some of the most egregious mistakes that contribute to a style of writing that is confused and confusing. These errors include overuse of specialized terms, the passive voice, “be” verbs, nominalizations. (Careful editing can transform a boring and convoluted text into a paper that’s a joy to read.) Again, if anyone wants specific editing tips, just let me know.

      The above techniques can help researchers gain control over a sea of data, enable them to write precisely, and result in a lucid text because “clear writing leads to clear thinking and clear thinking leads to clear writing”.

      Possible next blog topics include: how to transform convoluted sentences into sentences where ideas shine through; some causes of confused writing in science; can/should journal editors make clear writing a journal priority?; tips on writing paper titles; the benefits of the Structured Abstract; tips on writing logical Introductions; tips on writing logical Discussions. Please let me know if you’re interested in any of the above blog possibilities.

    • Time for a change

      Wednesday, 28 Mar 2007

      Hi Everyone,
      I’m delighted to have the chance to share ideas about the current state of “the scientific paper”. From my perspective as a professor who teaches researchers how to write persuasive and clear manuscripts, I can tell you two things: 1) the papers published in most journals are unnecessarily confusing and 2) most scientists can learn how to write simply and clearly. In short, I think that’s it’s time for a paradigm shift in how research is communicated. Furthermore, I believe that scientists need to take courses on how to write about their research in a way that captures the enthusiasm they have for their research and that enables multi-disciplinary readers to read and ENJOY their articles. Although it’s much more difficult to write clearly and precisely about one’s research, the advantages of doing so are huge because clear writing leads to clear thinking – for both the researcher and the reader alike.
      I think that most scientists would agree that the causes behind confused writing are complex. Some of these causes are deliberate (and include a desire to plump up small findings or a desire to keep others out of the field by using convoluted sentences and lots of specialized terminology). Some other factors that drive the confusion are simply rooted in a lack of training on how to write and edit a scientific manuscript or confusion on the part of writers about the story they’re trying to tell. Whatever the reasons behind the confusion, this style of writing does disproportionate damage to the whole scientific enterprise because as we all know, the best science is based on the work of others and perpetuates continued research in a field.
      So I say it’s time for a change. Since the whole research process is only really over once the published article is read and understood, it would be great if the scientific community decides to take a second look at the quality of writing of scientific articles.


Search blogs

web feed Request a blog Send an invite

Advertisement