Summary:  Using the knowledge and expertise of a multitude of disciplines makes animal behavior consultants better at what we do. Mixed methods are adaptable to many research designs with numerous method pairings that elucidate more information than can be obtained using one method alone. The data that is collected across disciplines, rather than just within a single discipline, is rich and comprehensive. This article introduces different kinds of research methods and explores how they can be applied to practical work with animals.

Learning is a science. Behavior is a science. As professionals in the industry, it is important that we understand how and why systems and protocols work and why the animals we are working with may behave in the various ways they do in the various contexts and environments they are faced with. When we are well versed in the scientific literature, we can create a better learning environment and improve the welfare of the animals in our care.

The welfare of the animals that we work with is complex and multifaceted. This means we need to study the subjective experience of other animals as well as their biological functions and natural behaviors. Researching across subject boundaries means a greater depth of knowledge of animal behavior, welfare, learning science, physiology of different species, human-nonhuman animal relationships and interactions, environmental impacts, and many other interconnected disciplines. Learning about other animals through a multidisciplinary framework can enhance our critical thinking skills, help us to problem solve, and synthesize information to better inform our clients and help the animals with whom we work.

There are several fantastic books available that bring many of the important animal behavior studies to a wider audience in a reader-friendly fashion. This makes the results of such studies more accessible for everyone. That is important because many of us do not have access to such academic resources. Additionally, the original sources can be quite complex and difficult to interpret without practice, particularly in a post-graduate setting. I am midway through my PhD and I still struggle sometimes! However, a lot of the animal behavior research that is highlighted in these widely accessible books is based on quantitative experiments on captive animals or observation of wild-living species. Certainly, this provides us with a lot of valuable information and the research tools needed to understand the mental lives and behavior of other animals. The problem is that it is limited in its scope or doesn’t always focus on what is needed to understand interspecies interaction. We can understand so much more, not just about animal behavior, but about how these animals interact with their environment and with humans on various levels when we cast a wider disciplinary net. This provides valuable information; human behavior and environment have significant impacts on animal behavior and welfare.

Fortunately, there are many fields of academia and independent research that can broaden the context of animal behavior and enhance our knowledge and skillset.

Qualitative versus quantitative research

Various academic fields not only have a different focus, but they also provide different methodologies. For example, cultural or social sciences use a lot of observation, ethnography, and/or interview-based research (among other tools), which can be either qualitative or quantitative. Qualitative researchers rely on data that are in-depth and descriptive. They can capture phenomenological information – experiences that come from a first-person point of view. Though it can be challenging, this can include the first-hand experience of nonhuman animals, too. I highly recommend the insightful book Animal Biography: Reframing Animal Lives by André Krebber and Mieke Roscher¹ for an example of how we may begin to include animals as first-person agents in social sciences.

When academic fields such as applied animal behavior, neuroscience, or ethology want to describe populations, they will be interested in the range, frequencies, and variability of animal behaviors and how these numbers interact. For example, these studies will use statistics to show how often a behavior might occur under certain conditions. These are quantitative methods of data analysis.

Mixed-methods research

Many researchers are beginning to use a mixed-methods approach to understanding data. For example, a researcher may use a survey to obtain quantitative data about how many people seeking information about dog behavior choose to read books, use a professional certified trainer, watch YouTube training videos, or get training advice from Facebook. They then might look to see if there are gender or age differences, or if people who have multiple dogs choose Facebook advice over hiring a professional, for example. Within the same study, they may also randomly select a small group of those survey participants to do individual interviews or a focus group. The latter data will provide nuances that the standardized survey is unable to capture and provide additional in-depth information that supports the survey data. The survey gives us correlations, and the interviews may help us understand why people choose YouTube videos over hiring a professional dog trainer. There are some academic fields that are more apt to choose a mixed methods research design, such as the emergent field of anthrozoology (the study of human interactions with other animals). This is because the field itself has a wide application and collaborates across disciplines at its foundational core.

Types of research methods

There are so many flavors of methods to choose from, which makes research dynamic, important, and incredibly fascinating. Regardless, all research will use a systematic approach to the procedure of data collection and analysis. It is important to understand that a single research topic cannot cover every question and will have a very specific focus. For example, researchers concerned about the welfare of animals in shelters might want to know whether black dogs under the age of 2 are less likely to be adopted than white or brown dogs who are under the age of 2. However, good research will create a path for generating new questions and further studies because all science is about asking questions.

Qualitative research methods

These methods, as stated, are used primarily in social sciences, which, in my opinion, are far too often overlooked when we review the literature on animal behavior. The concept of including animals in these types of disciplines is fairly new but incredibly powerful. The downside is that we may have a much smaller participant pool than some quantitative research is able to support. The positive is that it is much more descriptive. The following are the most common methods used for qualitative research:

• Individual/personal interviews. These interviews can be structured, with a set questionnaire or questions, semi-structured questions that incorporate planned questions but the discussion is open to directives from the natural flow of conversation, or can be completely open, as in a more casual (but guided) conversation on the specific topic. A great example of this specific type of research can be found in a book by sociologist Clinton Sanders called Understanding Dogs: Living and Working with Canine Companions²] which uses a semi-structured interview approach.
• Focus groups. These are small groups of informants who will answer questions about the particular research topic, such as, for example, a group of trainers who work with gorillas in zoos. In groups, people can share their experiences and points of view and come up with new insights.
• Ethnographic studies. Ethnography is an in-depth analysis of participants in their own environment. For example, a researcher may observe and/or interact with people and their dogs in the dog park (see for example Jackson, 2012³) or a researcher may use autoethnography to report on and analyze their own experiences as a dog trainer.
• Content analysis. This is when a researcher may analyze the language or content of documentation and draw meaningful inferences form it. A great example is a recent book by Justyna Włodarczyk (2018) called The Genealogy of Obedience, in which she explores a century of North American dog training literature⁴. I highly recommend this to anyone who trains dogs.
• Case studies. This is something that most certified members of the IAABC will be intimately familiar with. Case studies can tell us information about individuals that may or may nor be compared and contrasted. Now, imagine taking several case studies and examining them further to explore themes and patterns about training techniques, effectiveness, and behavior. We may find meaningful data emerges when we begin to examine the content more closely. A great example of an article of presenting case studies in research is a new publication by Mills et al. titled “Pain and Problem Behavior in Cats and Dogs.”⁵
• Having participants complete a daily or weekly journal may already also be an important practice for many zoo staff, animal trainers, or behavior consultants. Especially those who work with separation anxiety cases. The diary can provide us with important qualitative data like descriptions of antecedents and triggers and quantitative data such as patterns, frequency, and intensity.

Quantitative research methods

Now for the statistics. Statistics can tell us a lot about how often something occurs, the intensity of a behavior, and/or the significant differences between two variables. In the field of animal behavior, we often see experimental studies where there are two or more groups with different conditions tested against each other and against a control group. There might be several trials and various conditions applied to the participants, and statistics can be used in any of the following research methods. But experimental is not the only kind of quantitative data. Here is a list of the types of quantitative research we may see:

• Surveys can be qualitative, too; it depends on the types of questions that the researcher asks and how the data is analyzed and used. But those that use a scale (for example, the Likert scale), multiple choice questions, or yes/no type questions can be used for quantitative analysis and may be useful for correlation. Additionally, a quantitative survey is repeatable ad generalizable. A good study will validate the survey measures by piloting and testing for internal validity. Once a survey scale is validated, it may be used repeatedly in many different research studies. For example, you may have heard of the C-BARQ (Canine Behavioral Assessment and Research Questionnaire), or the Fe-BARQ (for cats) or the E-BARQ (for horses). The C-BARQ has been included in multiple studies that look at dog behavior.
• Ethogram/observational. This is used to catalogue the behaviors of an animal that either occur naturally or under manipulated conditions. It might be marking how frequently a behavior occurs and/or in what conditions or sequences, etc. Though this method is often used in the observation of wild-living or captive wild animals, it can also be used in the context of pets. As an example, a study by Howse et al. looks at the social behavior of dogs in a public-access dog park by cataloguing and analysing a standard set of social behavior cues.⁶
• Correlational Research. This is likely the research that you are most familiar with, or you think about most often when you think about studies in animal behavior. This method examines the relationship between two or more variables. For example, there may be a correlation between dog bites and children. It doesn’t necessarily tell us why, just that there are significantly more dog bites in children under the age of 10.
• Meta-analysis. A meta-analysis is a statistical analysis that combines the results of multiple studies. For example, if there are 20 different studies that look at whether bees prefer purple flowers over orange flowers, we can take the data from all of those different studies and analyze the results to draw conclusions. For example, a new study looking at the effectiveness of prison-based dog training programs, “Are We Barking Up the Right Tree?” examined the data from 11 studies looking at how the program affects social functioning of prisoners.⁷

Within these methods there may be variation to the types of methodology applied. Keep in mind, this is just an overview.

What about the field of philosophy?

Other important areas, and some of my own best insights, have been from the field of philosophy and animal ethics. These fields should never be discounted, though they often are. Philosophers can help to shape and mold the processes of research in profound ways. At the heart of it, research is about one thing: expanding our knowledge. Often, research is based on a desire for practical problem-solving, and can produce tangible results as a by-product of the research itself.

However, within philosophy, the idea of doing research is to provide insight by defending or arguing against previous arguments. And sometimes, that really convoluted and seemingly obscure methodology can produce some really cool ideas! Think about animals, for example. The way we currently view animals, treat animals, and even train animals comes from philosophical research. Work by Peter Singer, Tom Regan, Gary Francione, Bernard Rollin, David DeGrazia, Jessica Pierce, Paola Cavalieri, C. Lloyd Morgan, Kristin Andrews, and so many others has profoundly shaped the path to understanding animal minds, shaping the idea of respect, rights, and worth.

Human bias

Unfortunately, there is a downside to all methods of research on other animals – these studies are inherently human biased. Humans are the ones who design the experiments and research models, create the questions, observe through a human lens, and process in our human brain. A multidisciplinary approach may help us see a bigger, clearer picture. A dolphin experiences the world differently than a snake, a mouse, or a human child does. This is likely true of individuals within the same species, too. What I experience may not be the same as what my husband experiences.

Of course, we want to draw comparisons. We, collectively, have a bias toward thinking we are exceptional compared to all other animals, but there are other less obvious biases, and it is inspiring to see a dog who can learn 1,200 words or a monkey who can complete a memory task in record speed. It provides us with a foothold to advocate for nonhuman animals that they are indeed wonderful and exceptional and deserve more respect than we (collectively) give them. But in doing so, we potentially may fail to obtain accurate information. For a classic example, let’s look at the mirror test (research by psychologist Gordon Gallup published in 1984 was the originator of the first research using the mirror self-recognition test⁸). This is a test designed to show whether an animal has a sense of self. The animal is marked in some way and then provided with a mirror. If the animal chooses to look in the mirror and examine the mark on their body, they are considered to have a sense of self. Many apes, aside from gorillas, have passed this test. But dogs do not show the same ability. Is this because they don’t have a sense of self? Hardly. Primates are primarily visual beings, and this was a narrowness of Gallup’s approach. His narrowness was, in part, influenced by our human bias. Dogs, as we know, use their very extraordinary sense of smell. So, when this test was reconfigured to utilize scent rather than a visual marker, dogs did show that they understand their own marked scent (see, for example, Bekoff, 2001; Cazzolla Gatti, 2016; Horowitz, 2017).^9,10,11

An additional issue of our human bias is the pre-existing knowledge or misconceptions that people hold, shaped by society and by our own experiences. We are all likely aware of how many of our clients miss or misunderstand their dog’s (or cat’s or horse’s, etc.) body language. When we utilize human self-reporting (so, surveys or interviews, for example) to evaluate another animal’s behavior or experience, we may end up with biased results. It likely does not accurately depict what an animal is actually experiencing, but rather what is assumed by the respondent and may vary depending on knowledge level.

However, we know that humans impact animals in various ways. The way they choose to train them, the lifestyle they lead, the knowledge (or lack thereof) they hold, all impact the experience, and ultimately the behavior, of the animals with whom we work. Therefore, information on animal experience is incredibly valuable for the wellbeing of the animals in our care, in our clients’ care, and as part of society as a whole. Seeing our flaws can actually enable us to create and utilize protocols that account for these shortcomings and flaws.

Conclusion

Mixed methods research across disciplines is especially useful in providing a triangulation of knowledge and depth of multiple areas. Using a mixed methods approach to research fosters scholarly interaction. It can add breadth to multidisciplinary research teams by encouraging the interaction of quantitative, qualitative, and mixed methods scholars, which is important. A lot of academic departments can be quite segregated. But using the knowledge and expertise of a multitude of disciplines makes us better at what we do. The flexibility it provides is also invaluable. Mixed methods are adaptable to many research designs with numerous method pairings that elucidate more information than can be obtained using one method alone. The data that is collected across disciplines, rather than just within a single discipline, is rich and comprehensive. So, let’s broaden our minds and expand our scope to include all types of research to learn about our nonhuman friends.

Acknowledgements

Thank you to my PhD supervisor, Dr. Alison Loveridge, for providing me with such valuable feedback, as always. And to my friend, Patrick Flynn, DVM, for your taking the time to proofread this article for me.

References

1. Krebber, A., & Roscher, M. (2018). Animal biography: Re-framing animal lives. Palgrave Macmillan, Cham.
2. Sanders, C. (1999). Understanding dogs. Temple University Press.
3. Jackson, P. (2012). Situated activities in a dog park: Identity and conflict in human-animal space. Society & Animals, 20:3, 254-272.
4. Wlodarczyk, J. (2018). Genealogy of Obedience: Reading North American Pet Dog Training Literature, 1850s-2000s. Brill.
5. Mills, D. S., et al. (2020). Pain and Problem Behavior in Cats and Dogs. Animals, 10:2, 318.
6. Howse, M. S., Anderson, R. E., & Walsh, C. J. (2018). Social behaviour of domestic dogs (Canis familiaris) in a public off-leash dog park. Behavioural processes, 157, 691-701.
7. Duindam, H. M., et al (2020). Are We Barking Up the Right Tree? A Meta-Analysis on the Effectiveness of Prison-Based Dog Programs. Criminal Justice and Behavior, 47:6, 749-767.
8. Gallup Jr, G. G. (1982). Self‐awareness and the emergence of mind in primates. American Journal of Primatology, 2:3, 237-248.
9. Bekoff, M. (2001). Observations of scent-marking and discriminating self from others by a domestic dog (Canis familiaris): tales of displaced yellow snow. Behavioural processes, 55:2, 75-79.
10. Cazzolla Gatti, R. (2016). Self-consciousness: beyond the looking-glass and what dogs found there. Ethology Ecology & Evolution, 28:2, 232-240.
11. Horowitz, A. (2017). Smelling themselves: Dogs investigate their own odours longer when modified in an “olfactory mirror” test. Behavioural processes, 143, 17-24.

Erin is a PhD candidate in Human-Animal Studies at the University of Canterbury in New Zealand. She is a CDBC, CPDT-KA, ADT-IAABC and is an accredited Dog Behaviour Consultant with Companion Animals New Zealand. She is the owner of Merit Dog Project, a committee member of the APDTNZ, and works for the IAABC Foundation.