Sabtu, 29 Juni 2013

Writing style

  • Plain English
  • Topic sentences
  • Subjects, verbs and objects
  • Eliminating fog
  • Say what you mean
  • Word order
  • Creating flow
  • Tight writing
  • Chopping up snakes
  • Parallel structures
  • Style matters

Other types of documents

Letters 
Letters, or communications as they are sometimes called, are written for many reasons. You many want to provide supporting information, clarification, criticism, correction, or an alternative explanation to the results in a previously published journal article. You may disagree with the interpretation of the results, have further information to add to a publication, or have a novel comment to make. Many journals also publish letters that convey political or psychosocial messages that are related to the practice of medicine or research. 

Editorials 
The best editorials are usually short, pithy, pertinent reviews about a topic that is selected by the editor. An editorial is often commissioned to comment on a paper that is published in the same issue of the journal. Very often, the editor asks an external reviewer who has shown insight into the paper to write this type of timely review. 

Narrative reviews 
Review articles are very popular with readers and editors.Very short reviews are often called annotations. Editors likereviews because they know that, along with the editorials and the correspondence columns, they are the most widely read part of a scientific journal. To maintain reader interest, review articles must be topical, up to date, accurate and authoritative, and, if possible, provocative and a good read.

Systematic reviews and Cochrane reviews
Systematic reviews are a more rigorous compilation of evidence from the literature than narrative reviews simply because the search strategy for finding and summarising studies is clearly defined. In a systematic review of the literature, all of the primary studies on a topic are systematically identified, critically appraised, and summarised, with explicit and reproducible methods. The rationale behind this approach is that the standardisation and the transparency of the methods used by authors and the acquisition of all available primary studies on the review topic minimise the potential for bias. 

Case reports 
Most clinical journals publish case reports that explain how patients presented for medical care, how the course of the illness rogressed, and what treatment was given. Case reports that publish this information for a number of patients are called case series. The main purpose of a case report is to educate clinicians about the clinical features, investigation, and/or the treatment of patients with unusual problems. A case report often acts as refresher training for clinicians so that a diagnosis is made more readily and the condition treated more effectively. In reporting observations by clinicians, case reports may also generate hypotheses that lead to new research studies. 

Post-graduate theses
Postgraduate theses, whether they are for a doctorate or masters degree, command a huge time and motional commitment. The required length of a postgraduate thesis may vary widely but, for a doctorate, the range is generally 35 000–50 000 words with an upper limit often set at 80 000 words. If you are unsure how long your thesis should be, check with your institutional guidelines. Whether 40 000 or 80 000 words are used, writing a thesis is a long and daunting task and it may not become satisfying until the end is in sight.
However, if you write with a plan in mind, the process will be more rewarding.

References and citations

The citation section of a paper is important for giving credit to the ideas and work of other scientists. In finding the references that you need, you can use the web to search MEDLINE® via PubMed® (www3) or you can use other websites to find links to further references and to netprints (www4–8). If you are quoting a method, a sentence, an idea, or some results published by another researcher, then you must cite the original source. Using other researchers’ ideas or any parts of their writing as your own is a serious offence known as plagiarism. When you are writing a journal article, you need to cite only the most valid, most important, and most recent literature. Ideally, you conducted a full literature search when the study was planned and you have updated it as the study progressed. If you are organised, you will have your references stored in an electronic database such as Endnote® or Reference Manager® (www9). This will allow you to format a bibliography in a fraction of the time that it takes to do it manually. In addition, your paper copies will be filed in alphabetical order or in an indexed or linked file so that they are readily retrieved. Even better, you will have read all of the original journal articles that you plan to cite. Before you submit your paper with a
reference list created using a computer package, check with your editor that the format is acceptable. Some journals prefer to use their own templates.

Title page

Once the authors and title are decided, it is exciting to create a title page for your paper to give it a formal look. Most journals specify the requirements for the title page in their Instructions to Authors and these will vary according to journal requirements. The title page usually has the title at the top and the authors clearly listed below. All authors must be listed regardless of how many there are. Many journals limit the number of authors and request that if there are more than 12, other names must appear in a footnote. If more than 12 authors are listed for a multicentre trial or more than eight from a single institution, each author may be required to sign a statement attesting that they fulfil the authorship criteria of the ICMJE (www1). Authors should be listed with their full names, highest academic degree, title/s, position, and address. You should also include the name of the department and institution where the work was undertaken, the institutional affiliation and full contact details of yourself as the first author. Finally, add a direct telephone number and facsimile number with their international dialling codes and the email address from which reprints should be requested. You must also acknowledge any grant support either on the title page or in the acknowledgements section. 

The title page should also include a running title, which is the title abbreviated to the number of letter spaces requested by the journal, usually 30–40 including white spaces. This title will appear in the header or footer of the journal pages other than on the title page. Finally, select 3–10 key words or short phrases to add to the bottom of the title or abstract page. Standard terms should be selected from the Medical Subject Headings (MeSH), which can be found at the MEDLINE® website (www2 ) or in the printed Index Medicus held by medical libraries. This will help to ensure that your paper is indexed correctly in the electronic databases and can be found easily by other researchers. Journals may have slight variations in the information that they require on the title page, so always check in the Instructions to Authors.

Choosing a title

Titles take up only a few words but are of inestimable importance in persuading clinicians and researchers to read your paper. If your title has an impact that attracts readers, then so much the better. The basic function of a title is to describe the content of your paper in a succinct way. Also, in these days of database searching, keywords in the title make your paper immediately accessible to workers in your field. However, titles can also be used as a key tool to give your paper a distinct personality. To this end, your title must be accurate, specific, concise, and informative, must not contain abbreviations, and must never be dull. 

Discussion

The discussion section of your paper should reiterate your main findings but in the context of furthering knowledge or impacting on patient care, public health policy, or future research. This is the time to be honest about any limitations of your study, to explain how your findings fit in with established knowledge, and to explain any inconsistencies. In science, we are continually trying to chip away at parts of a very large jigsaw. The discussion section gives you an opportunity to explain which part of the jigsaw you have put in place.

The discussion can be the most daunting section of a paper to write. If you have a broad knowledge of the literature and of the various opinions in your research field, it can be hard to limit yourself only to the parts that are particularly relevant to your paper. A good trick is to make notes as you analyze your results and read the literature. Jotting down the major ideas that you will need to discuss as they come to mind will help
you to organize your discussion section. Also, make notes about which literature supports your findings and which is at odds with your results as you progress. These concept ideas often translate into topic sentences and help to keep each paragraph in focus. The paragraphs can then be ordered from the most to the least important topics. This will help to create a discussion that flows naturally and sensibly. 

Results

 This section is the most important part of your paper because its function is to give specific answers to the aims that you stated in the introduction. After the methods, this should be the easiest section to write. You should use an interesting sequence of text, tables, and figures to answer the study questions and to tell the story without diversions.  It is essential to know your audience and make it clear to them in their own language how your work is an important extension of what has gone before. In practice, editors usually  prefer to publish new findings. Although consistency of evidence is critical for ascertaining causation,  most editors are not keen to publish results that are already thought of as established knowledge. It is important to convince the journal editor, your reviewers, and your readers that your study extends knowledge rather than merely confirms what we already know.