HDL = high-density lipoprotein; LDL = low-density lipoprotein; n = number of patients; TC = total cholesterol;
TG = triglycerides.
aThe values are mean ± SD from n number of observations.
bAge-matched control group.
cAge-matched control group receiving only herbal preparation.
dPatients receiving standard treatment.
ep < 0.05 versus Group 2.
fPatients receiving standard treatment + herbal preparation.
gp < 0.05 versus Group 3 (Student’s t test for unpaired observations).
The statistical significance of difference between the data (mean ± SD) in a table should be denoted by a symbol and described in the footnote to the table. Check for internal consistency of the data in a table. For example, the sum total of all percentage values in a single set should be 100. Check all addition, subtraction and any other mathematical operations. Make sure the data in a table are consistent with the data presented elsewhere in the paper (abstract, text or another table). Sometimes abbreviations of certain terminologies become inevitable in the columns or rows of tables. All such abbreviations used in tables must be described in footnotes to the table, even if they are spelled out elsewhere in the manuscript. Finally, confirm that the table is properly referenced in the text section of the results.
As mentioned earlier, footnotes should explain abbreviations, the study conditions, statistical assessments, acknowledge the source of the data presented in the table (if it has been adapted from a previously published table) and other details to make the table understandable on its own. Furthermore, it is preferable to use the symbols as footnote signs (*, @, §, #, ** or superscript lowercase letters in alphabetic sequence like a, b and c) to indicate the data in the table. It is better to avoid the use of numbers as footnote markers.
All parts of a table should be typed double-spaced, including column headings and footnotes. Double-spacing is needed to give the copy editor enough room for marks to indicate publication style to the printer. In general, avoid using vertical lines in the table. Start each table on a separate page. If a table cannot be completed on a single one, it can be continued onto a second sheet detailing the column headings in the next sheet.
Figures—A Means to Visualise the Data
Before you begin to write the first draft, you would have accumulated materials you wish to present in the paper. These include photographs, X-rays, ultrasound scans, tracings of electrical recordings (ECG, EMG, EEG and other potentials) or mechanical recordings (isometric or isotonic contractions, blood pressure recordings or other signals) and graphs of experimental data. Now you must decide which of the illustrations are necessary in the paper.
Different formats of illustrations are shown in Table 26.3. The illustrations or figures should qualify for the following points. Do these figures or illustrations provide evidence for the study? Do they enhance the effectiveness of presentation? Do they emphasise the points of the observations in a given study? An illustration should be used if it adds clarity to the discussion. If the paper is about newly discovered structural details of an organism then the morphological features of the same are required as evidence to support the paper.
Table 26.3. Various ways of illustrating the observations in a picture or graphical format
| Graphs |
| Pie diagrams |
| Histograms (frequency distribution) |
| Bars (comparison) |
| Time-response |
| Scatter and line |
| Photographs |
| Flow charts |
A figure or graphical illustration is recommended if it is more effective in representing data than a long statement in text. A family tree may succinctly depict the appearance of a newly identified syndrome in members of the family as a Mendelian autosomal dominant trait; a written description of the same would require many lines of text and would not make the point as intelligibly as the graphical representation. Numerical data showing the relation of two variables can be ably presented in a graph. The difference in hormonal profiles between pre- and post-menopausal women can effectively be conveyed through bar diagrams.
Editors and reviewers of the journal may not agree to the inclusion of all the illustrations submitted with the manuscript. If your figures support or emphasise important data, then they will qualify for inclusion in the manuscript. If this is not the case, then it is advisable not to include those figures in the first place. Having to discard illustrations while revising the paper is a waste of the time, effort and money that went into creating the figures, and rewriting the text related to the deleted figures prolongs the revision.
Inadvertent use of illustrations in the journal should be avoided. Drop tables or illustrations in favour of equivalent statements in text, if you can. Combine two or more illustrations into a single, multipart illustration. The illustrations to be combined should be related in subject; if they are graphs, they should also be similar in proportion and scale of the lines and lettering. Finally, do not use illustrations that duplicate data already presented in tables or text.
Computer graphics have revolutionised the quality of figures. However, the creation of quality illustrations requires professional skills. Poorly prepared illustrations reflect the incompetence of the author and perhaps even the unsoundness of the text. Graphic designers need guidance from the authors on segments or lines to be highlighted. Obviously, authors have to provide the raw materials (numerical data, tracings, photographs, imaging pictures, etc.) for preparing the illustrations. The instructions to authors for online submission of artwork in various journals provide clear information.2–5 A wide variety of graphs can be generated from Microsoft Excel®, Sigmaplot®, Graphad® and many other graphics software programs. The relationship between two variables, for example a graph showing the effect of ageing on diastolic blood pressure, is the simplest form of a graph. Usually, the independent variable is shown on the X-axis and the dependent variable on the Y-axis. The most complex line illustrations that depict surgical procedures, anatomical views, X-rays, CT or MRI scans, electron micrographs or the actual photographs of the specimens can be made simpler with the accompanying drawings showing the outlines of such pictures. Nowadays, many journals accept supplementary material, such as video clippings or colour photographs.
Tracings obtained during an experiment (electrophysiological, biochemical, behavioural alterations as images, etc.) stored in the computer as a computer graphic can be copied or exported to the graphs either in Microsoft PowerPoint®, Coral draw®, Photoshop® or any other software. While importing these pictures, make sure that the images are not distorted by disproportionate stretching either vertically or horizontally (distortion can be prevented by locking the aspect ratio).
Diagrams can be used to illustrate a relationship between connected units; for example, a sequence of procedures such as those in a diagnostic algorithm, or a table of organisational titles and responsibilities. For such diagrams, use the presentation software such as Microsoft PowerPoint®. Diagrams depicting relationship among family members and generations should use well-established conventions, which are illustrated in the textbook on human genetics.6
Written permission to reproduce photographs of patients in which the patient might be identified should be acquired when the pictures are taken. Parts of the photographs that would identify a patient should be blacked out or cropped, wherever possible. While taking photographs include the area of neighbouring tissue to show the relationship with the other organs and then magnify (zoom it) to show the finer details.
Photomicrographs should be obtained using appropriate lighting and colour filtration to ensure the best quality images. It may be better to start with low-power views followed by high-power views. Always include a scale with each image rather than just labelling it as low or high power. Specific regions of photograph can be identified by inserting letters, asterisks or arrows. Line drawings showing the main elements in outline can be provided to visualise various components of more complex figures with identifying letters or abbreviations.
Colour printing is far more expensive than black and white and is usually charged to the author. Colour photographs may be essential to show the various types of staining as in immunocytochemical methods, fluorescent imaging, blood flow imaging, illustrations of some skin lesions (like faint rashes), subtle histological-stain colours, and multicolour scan images.
Illustrations should be understandable without having to refer to the related text. Therefore, all relevant features of the illustration should be labelled on itself or in the accompanying legend. Graphs of numerical data must have their horizontal and vertical axes labelled to indicate the variables plotted and their units. Avoid crowding on the graphs by inserting too much text. Instead, use abbreviations and then explain the details in the legend. An alternative is to use standard symbols in drawing the graph that can be reproduced and explained in the legend. Size and format of the figures are usually indicated in the instructions to authors in most of the journals. The size of the labels should be such that the output is clearly readable after reduction of the illustration to 50% of its size. Care should be taken for line thickness, tick size, labelling of X and Y axes, etc.
Photographs of closely related subjects are sometimes more effectively presented in a single, multipart illustration. An example is a general view of the cut surface of a kidney sliced longitudinally accompanied by several close-up photographs of details of the cortex, a pyramid and the pelvis. It is better to label each component of such a figure as A, B and C.
Each illustration must be able to ‘stand alone’ with the help of a descriptive legend. Technical details, such as stains and magnification, and additional description needed by the reader to understand the illustration without referring to the text should be provided in the legend. Legends should be typed on a separate page of the manuscript.
Limits set by a journal on the number of illustrations must be observed; if the journal does not specify the limit, then consider using one illustration (or table) per 1,000 words of the text. Make sure that the details of your observations are not compromised. Numerous details in format, lettering and labelling call for careful attention while preparing the illustrations.
