Back in the dark ages, before the internet, there were a finite number of scientific journals. Most were associated with and published by scientific organizations. If you wanted to read an article you either had to subscribe to the journal (and association) or head over to the university library, where once you found the journal you waited patiently in line to pay to have a student Xerox it for you. Rarely did anyone other than researchers and students venture to obtain or understand scientific articles, but access has changed and our plan is to help you understand and critically review veterinary, behavior, training, and welfare related scientific manuscripts.
The internet has revolutionized our access to scientific articles.
The internet changed everything. Now with a click or a swipe science is at your fingertips. But with the internet came thousands of new journals and millions of articles. I do free-lance scientific editing – every day I am offered hundreds of different manuscripts to edit! There are now so many journals and so many articles it is hard to keep up with a single discipline, and impossible if you are interested in several different disciplines. Some journals are prestigious, some are questionable, and most are in between, but each journal has its own rules, format, editorial and reviewing process, and publication costs. Some are “open access” meaning anyone can see and download articles. The goal of open access is to provide “accessibility and improve visibility.” They are published exclusively online.
One of the advantages to open access is that there is no cost to add color – so diagrams, graphs, histopathology slides, cardiac echocardiograms can all be viewed in color. In addition, videos can be attached or linked to open access journals. In general, non-open access journals have a print and electronic version, however access to the electronic material varies by journal. For many non-open access journals the current year is not available without a fee, subscription service, or a connection to a university. Some provide access to “relevant” or “public interest” papers, some provide no access to electronic material. If you are interested in a specific paper you can always e-mail the author and ask for a copy of the PDF; back in the dark ages the request came via post card and the hard copy was returned via snail mail.
Things have changed, and will continue to change, for example during COVID some manuscripts, viewed as “important,” were “pre-printed” – this means that the manuscript did not undergo peer or editorial review, which as it turns out is really important for scientific integrity.
How do you get an article published in a scientific journal?
Of course you need money, generally grant money, to pay for staff, supplies, data analysis, students, etc. Then you need to design an experiment and develop a research question. Let’s say we want to know if cows prefer being inside or outside. Our hypothesis is that given a choice cows prefer to be outside. We design and carry out the experiment, analyze the data, and draft the manuscript. We think this data is important for the cow’s welfare so we purposely choose a journal of high regard. We write the manuscript, it must be reviewed by all the authors, and conform to the specific journal’s rules. We make sure we have followed the rules and submit it online.
The editor of the journal contacts reviewers to evaluate the manuscript and reviewers agree to provide a review “in a timely manner.” The job of the reviewers is to determine if: the hypothesis is valid; are the statistics proper for the experimental design; are the references appropriate; and are the conclusions in line with the data. The editor might outright say the manuscript is not acceptable (at all or in its current form, suggesting it may be acceptable with revisions). There is a good chance that the reviewers will want changes, some minor, perhaps something major. We receive a letter from the editor (hinting at our ability to get the manuscript published) and a copy of all reviewer’s comments. Next, we address the reviewer’s, point by point, and make the associated changes in the manuscript. If we disagree with the reviewer’s suggestions we develop a referenced response to send to the editor. This can go back and forth for some time. Eventually the manuscript is published – sometimes weeks, sometimes months, not uncommonly a year later.
Do cows prefer being inside or outside? A question that can lead to forming a hypothesis.
Who are the reviewers?
In a perfect world a manuscript is reviewed by 2 or 3 individuals with expertise in the field. This is especially important in articles related to behavior and welfare, as the research teams are often multidisciplinary. So if we wanted to look at the effect of therapeutic riding on the horse and the human, we would need a variety of experts in human medicine/psychology, equine behavior and welfare, and veterinary medicine. If there is only one reviewer, and they have expertise in human psychology, they will likely not understand the nuances of equine behavior and welfare and veterinary medicine; thus the manuscript may be published with flaws missed by the reviewer because it is out of their area of expertise. So if you are reading a manuscript see if they list the number of reviewers – only one should lead you to read more critically.
Interestingly, the proliferation of journals has not been accompanied by an increase in individuals with expertise willing to take the time, without pay, to review manuscripts. So many journals may publish articles with insufficient review. Also, people “famous” in a field may get a “pass” that a less well known researcher in the same field would likely not get.
What should be in a scientific paper?
Most journals organize their scientific manuscripts in the following order: title, authors, abstract, key words, introduction, methods, results, discussion, references. Some include a “lay” or “simple” summary. Figures and tables are noted in the text where they are mentioned. Supplemental material can include questionnaires, additional data, videos, etc. Each journal has specific formatting requirements from the font and its size, margins, how the references are listed, the length of sections, number of figures and tables, and how the researchers state “human or animal use approval” and conflict of interest statements.
The abstract is a short summary of the study, often with a word limit. Abstracts can also be found in proceedings for scientific meetings. Many people read the abstract and the authors conclusions and “incorporate those conclusions into their being.” This is actually a mistake (why later). When searching a topic, the abstract should provide enough information for me to determine if I am interested in reading the study. To me that is how best to view an abstract.
For example, a while ago I was at a meeting and there was a talk on play behavior in calves (baby cows). The speaker asked a question like “Do all calves play or do some calves play more than others?” They designed an experiment to determine if given the opportunity (open space) will calves play? Interestingly, some calves don’t play. As a cattle veterinarian this is a red flag to me because a calf that does not play is likely sick. I asked the author, and it turns out the researchers “assumed” the calves were healthy and they were never examined by a veterinarian. Had I not asked the question (or only read the abstract, or the manuscript was not reviewed by a veterinarian) I would have believed that some calves don’t play.
It is important to read beyond the abstract!
The introduction should provide sufficient information and references for you to understand why the research question and hypothesis were developed. It reviews the literature pertinent to the questions the researchers are asking, but it shouldn’t be ponderous. I always read the introduction to get a feel for the topic – sometimes I check a reference to see if their interpretation of the paper is the same as mine. In research areas where I have some expertise and am aware of the literature, a “misinterpretation” of the data from a reference makes me ask: did they interpret it differently? did they interpret it to promote their hypothesis? or did they not understand the reference?
The methods are the key to understanding and interpreting the research and deciding if the results may be “statistically significant” but not biologically significant. Methods are often the most difficult section for nonscientists to decipher. They should include “everything” so that another researcher is able to replicate the experiment simply by reading the methods. In practice of course there is nuance that is never translated. The methods will vary with the type of study and experimental design.
As an example let’s look at a presentation by Duffy and Serpell on the effects on spay/neuter on aggression in dogs. They used the C-BARQ questionnaire, which asks the human associated with the dog to rank certain behaviors on a 5-point Likert scale, with 0 representing “no aggression” and 4 indicating “serious aggression” and 1, 2, and 3 being gradations between 0 and 4. They conclude “spayed females are more aggressive toward people,” the graph looks impressive, and then you realize the vertical axis (the Y axis) represents Likert values from 0.2 to 0.7, and the “significant difference” between spayed and intact females is 0.1 on a scale of 0 to 4. Think about when you were asked about your pain on a scale of 1 to 10 – were you 100% sure it was a 5 not a 6? Ask yourself is it possible that these findings are statistically but not biologically significant?
The results, when presented without interpretation, should be relatively short, and ideally includes only data. Data is often presented in tables and figures, and statistical significance should be indicated. Tables and figures should stand alone, meaning all the information to understand the table or figure should appear on it – you shouldn’t have to refer to the text to understand it.
Let’s look at a paper by Kinsman et. al on dog sleep. Table 1 is a “Summary of sleep characteristics, as reported by owners participating in the ‘Generation Pup’ study when their dogs were 16 weeks (n = 2332) and 12 months of age (n = 1091).” They asked the questions does the dog have access to humans at night, and if so does the dog choose to be close to people? The abstract says: “However, of dogs that did have access to people during the night, more than 86% chose to be around people.” If you didn’t look at table 1, you might not realize that only 24.4% of dogs had access to humans at night and of those 526 dogs, 449 chose to be around people, yes that is 86.7%, but not 86.7% of all dogs. Imagine the lay press saying “Almost 90% of dogs choose to sleep with their humans.”
90% of all dogs want to sleep next to their owners? Not if you read the whole study…
The discussion is where researchers interpret their data in light of the published literature. The authors should explain why their data is different than previously published works both in terms of methods and results. In reality the results apply only to the population studied, but often researchers make bold proclamations about the applicability of their findings to the whole population of animals, when they studied a few animals in a university setting.
In a survey study on stockpersons perceptions of human-animal relationships and the relationship to milk yield, Bertenshaw and Rowlinson stated in their discussion ”Our data suggests that UK dairy farmers largely regard their cows as intelligent beings, capable of experiencing a range of emotions. Placing importance on knowing individual animal and calling them by name was associated with higher milk yield.” These statements are accurate, but open to misinterpretation – the authors explain that cows with names receive more individual attention which results in increased milk production, yet the NY Times titled their article “Cows with names make more milk,” suggesting that if you name your cow she will make more milk! Another reason to read lay summaries of research with a critical eye.
Next time we’ll talk about the different kinds of science, particularly those that are used in animal behavior and welfare studies and what you need to know to critically evaluate them.
Duffy DA, Serpell JA. Non-reproductive Effects of Spaying and Neutering on Behavior in Dogs. Proceedings of the Third International Symposium on Non-Surgical Contraceptive Methods for Pet Population Control http://www.naiaonline.org/uploads/WhitePapers/EarlySNAndBehaviorDuffySerpell.pdf
Kinsman R, Owczarczak-Garstecka, S, Casey R, et al. Sleep Duration and Behaviours: A Descriptive Analysis of a Cohort of Dogs up to 12 Months of Age. Animals 10: 1172 https://doi.org/10.3390/ani10071172 ; 2020.
Bertenshaw C, Rowlinson P. Exploring Stock Managers’ Perceptions of the Human—Animal Relationship on Dairy Farms and an Association with Milk Production. Anthrozoös 22; 59-69, DOI: 10.2752/175303708X390473; 2009.
https://archive.nytimes.com/query.nytimes.com/gst/fullpage-9804E0D91631F930A25751C1A96F9C8B63.html
Lana Kaiser MD, DVM, born in Buffalo, NY, received a BA in English from SUNY at Buffalo with plans to be a poet. She is a graduate of Michigan State College of Human Medicine, a Board Certified (Human) Internist, a cattle veterinarian, and a Emeritus Professor in the Departments of Physiology and Medicine, in the College of Human Medicine.
Trained as a biomedical researcher with a research focus on cardiovascular pathophysiology and parasitology, she is also interested in scientifically studying the interaction between humans and animals, and has published in both scientific disciplines. A 1995 graduate of the College of Veterinary Medicine, MSU she resides on a farm in Mason, MI where she raises Maine-Anjou and Red Angus cattle.
She has a mobile beef cattle practice, consults for several national agricultural entities, and has written and lectured about animal welfare, animal health, genetic defects, and human-livestock interactions at the state and national levels. She is involved in issues of animal behavior and welfare at the county, state and national level, was a founding member of MSU’s Human-Animal Bond Initiative and coordinator of MVMAs Annual Animal Welfare Conference and Animal Behavior Conference. She has been active in the MVMA Animal Welfare, Legislative, and Food Animal Practivce Committees, AABP Animal Welfare Committee, and Dean Foods Animal Welfare Advisory Council.
She enjoys clicker training her four dogs, her cow horse, and the occasional cow.
