This is the third part of my series on maximizing your reading. In parts 1 and 2, I shared some of the strategies that I use when tackling my scientific reading.1,2 My intent with this series is that hopefully you will take home a few pointers to help you get the most out of your reading. A discussion on critiquing and reading the scientific literature would not be complete without some time dedicated to sample size consideration. As a researcher, I find that critiquing the way others justify sample size, as well as their sampling techniques, helps bring awareness to how I make these decisions in my own work. The sample size used in any work presented in a paper should be clear and justified. The study genre will greatly impact the sample size potential, eg, a case report may involve one animal or one farm, and a randomized controlled clinical trial may involve hundreds of animals or more or less. Many computer software programs are available for determining sample size, but there are other considerations when determining sample size, ie, a small sample size of animals may have an impact on the power of the statistical analysis which can minimize the strength of the study, and a large sample size may be associated with cost limitations for the study.3 Additionally, a study that uses many subjects is not necessarily “better” if fewer subjects would have supported and answered the scientific question. A good scientific writer should include in the methods section a discussion on the strengths and weaknesses of the sample size used in the study and subsequently how that impacts the analysis and interpretation of the data. A good critical reviewer of a paper should also look for this discussion and take that into consideration as well when formulating his or her own conclusions.
The statistics section is often the most difficult to critique for many readers. The task of explaining how to understand statistics in an editorial is overwhelming for me as well. I suggest purchasing a statistics text book – particularly one written for veterinary medicine – to have as a quick reference and to help guide you or to provide a quick refresher when a statistical method or test seems unfamiliar.
The results section of the paper typically follows the materials and methods. I am usually guilty of quickly skimming results if it is in text format and I constantly try to discipline myself to read this section more slowly. But I do look at the tables and figures in detail. I always keep in mind when reading a table or figure that the data is often presented as the authors’ interpretation. For example, the paper may state something like “the data in Figure X show that the average daily gain of the pigs in Group B was better than in Group A”. This is likely the case, but I look carefully to see if the data actually show what the authors say the data show. Of course, I cannot do this as a critical reader if I do not fully understand the methods and their limitations. As you can see, all the sections of a paper build the framework for drawing conclusions from the work presented.
The discussion section comes next. The author is expected to do a good job of examining and explaining how the work has advanced our knowledge, any new insights provided, or future research directions. Issues or controversies raised by the findings should be discussed in this section and ideally resolved or supported with other literature.
In the Journal of Swine Health and Production (JSHAP), an implications section comes at the end of the manuscript. This is a bulleted list of take-home information for readers and should contain the practical application of the results. This section is often critiqued by reviewers as being unnecessary and not appropriate in scientific reporting. This is a difficult section for authors as well, as the temptation is to just re-iterate the results. The intent of this bulleted list is to provide a quick source of information for busy practitioners, and I think it is appropriate for an applied journal such as JSHAP. While this bulleted list is quick, easy, and often informative to all of us with busy schedules, I recommend a full review and critique of the entire paper by all readers.
A scientific paper would not be complete without references and acknowledgements. While this is seemingly obvious, references should be relevant, current, and presented neatly. Acknowledgements are important for readers to consider as well. It is essential to recognize those who participated and financially contributed to the research, as without this there would be no advancement of our knowledge.
I certainly have not covered every aspect of how to maximize your reading, but I hope this short series has provided you with some new information, reminded you of forgotten information, or inspired you to catch up on some reading.
--Terri O’Sullivan, DVM, PhD Executive Editor
References
1. O’Sullivan T. Maximize your reading – topics, titles, and abstracts [editorial]. J Swine Health Prod. 2015;23:9.
2. O’Sullivan T. Maximize your reading – Part 2 [editorial]. J Swine Health Prod. 2015;23:69.
3. Dohoo I, Martin W, Stryhn H. Sampling. In: Veterinary Epidemiologic Research. 2nd ed. Charlottetown, Prince Edward Island, Canada: VER, Inc. 2009:33-55.