Blog 178- A Case for More Qualitative Studies in Agriculture

In this blog Aditya and Bhuvana argue for mainstreaming qualitative research methods in agricultural research. The blog elaborates on the misplaced, but predominant, notion of superiority of quantitative methods, particularly in Social Science. The choice of method should be a function of the research question and not the other way around. The purpose of this blog is to convince the readers that there is always ‘room at the table’ for qualitative studies in agriculture, and as a discipline we gain by embracing a mixed methods approach.  


I[1] have always been fascinated by qualitative research methods, but I have rarely used one till now in the seven years of my research career. I always end up visualising myself presenting a research grounded in qualitative methods at a Social Science conference – I can literally imagine frowns from my colleagues listening to my presentation and some whispers, “Is it really economics?” or then again “Is it really science?” or “These are merely statements, how do you know they are true?” There is a fascination and a strong preference for quantitative methods among researchers working on social aspects of agriculture. Most courses on research methods have exclusive emphasis on quantitative aspects of research. Use of qualitative methods is limited, even when used, the rigour of such studies in general is very poor. In this blog, we question this prejudice about quantitative methods being in some way superior to qualitative methods. We argue in favour of having more qualitative studies in agriculture.


Let us start with our over the top love for quantitative studies. We agree that these are straight forward methods, amenable to statistical testing and lends itself to generalization. However, we should also accept the downside of using only the quantitative lens of enquiry. For instance, consider the most common adoption studies. The typical question to start with is ‘what determines the adoption of new technologies?’.

In a typical study, authors have a prior hypothesis that socio economic variables such as gender, age, education, income, and some technological aspects (mostly perceived) affect technology adoption. Data is collected on these variables, a model which fits the data is used and interpreted based on statistical significance. Most of these studies end up with conclusions, such as younger farmers are more likely to adopt technologies or educated farmers are more likely to do so. In fact, one of my senior colleagues from Statistics used to tell me that social scientists can justify anything; if the coefficient for age is negative, they will interpret it as younger farmers are more entrepreneurial and risk taking, and if the coefficient is positive, interpretation is changed to older farmers, whose experience leads to adoption! This is analogous to the narration in Box 1.

Box 1: Openness in Science
“A traveller to a new land came across a peacock. Having never seen this kind of bird before, he took it for a genetic freak. Taking pity on the poor bird, which he was sure could not survive for long in such a deviant form, he set about to correct nature’s error. He trimmed the long colourful feathers, cut back the beak, and dyed the bird black. “There now,” he said, with pride in a job well done, “you now look more like a standard guinea hen”.” Patton (1990, p. 347)

Nevertheless, what policy input can we give to such studies? Is it not already known that education is important? We already know the gender roles and challenges that women-headed farms face. What is it contributing to the progress of the theory or the subject matter? I am not criticizing every study here, of course there are very good papers on adoption using quantitative methods, but I am trying to underscore that not every quantitative paper with huge dataset and sound methods is useful or insightful. On a lighter note, Aditya has created a meme (Figure 1) on things that we find funny in our research papers.

Please allow us to present arguments in favour of qualitative studies. Many of you with a strong background in quantitative methods think that qualitative methods are easier as the sample size would typically be very small and one needs to worry about statistical significance (I used to be one of those, not now!). To do a qualitative study is easier, but it is extremely difficult to do it well. It is more complex, messier and time consuming than a typical quantitative study, but more on this later. Question at this point is: what are the advantages of a qualitative study when done systematically?

Most qualitative methods are inductive (not all, though) and they aim to understand the world based on lived experiences. Many of the constructs which cannot be typically defined or measured as variables can be extracted in this case. Here, each situation is understood on its own, and interpretations are drawn to best explain the scenarios, rather than trying to fit the situation into a pre-existing theory. When we try to fit a given scenario into a pre-conceived model, there are chances of mis-representation and we learn little about the problem. I think this passage from Patton is very relevant in this context.

So, qualitative studies are more interpretive, could explore human behaviour at a more fundamental level and understand farmers’ behaviour from their own world view. These studies could be very useful on their own, or in combination with quantitative methods. Before detailing the typical characteristics of a qualitative study, I think it is helpful to discuss some basics of the philosophy of science.

Figure 1: Types of Agricultural Economics research papers (a meme created by Aditya)

Broadly there are three main philosophies in science.

  1. Deductive reasoning – Starting the enquiry with a theoretical background and applying it to understand field situations. It could be either ‘naïve falsification’ where the researchers try to either support a theory or end up with falsifying it based on observations from the field; or ‘critical reasoning’ where the hypothesis based on theory is the starting point, and the hypothesis is then statistically tested.
  2. Inductive reasoning – This is theory-free reasoning based on observed data. The aim is to understand the story that observations are telling us and then drawing inferences based on these. The philosophy is to develop concepts based on the observed data.
  3. Abductive reasoning – The philosophy is to critically observe the situation, draw inferences, go back to the field and gather more data to refine the concepts. So, this going back and forth between data collection and analysis differentiates it from the inductive methods.


With this understanding, we can now talk about epistemology. Epistemology is a very difficult term to define. In the simplest explanation, it is a set of rules for creating new knowledge, or rules used to understand the world we live in. Epistemologies pertaining to Social Science can be better represented in the form of a spectrum. We don’t intend to explain all these different strands of epistemology in this blog, those who are interested can read more on these. Figure 2 depicts the spectrum of different epistemologies.

Figure 2: Different epistemologies in science

Qualitative methods fall under the epistemology of post-positivism, where the importance is on understanding the context of the situation, giving importance to understanding human interactions and lived experiences of participants. These methods rely on the researcher’s description of perceived reality of the participants. In other words, the aim is not to describe what is real, but what the respondents think is real, based on their world view, because these perceptions are what shape their behaviour.

The purpose of detailing the research philosophies is just this: a researcher can better understand how to approach the problem if they think through on the epistemology they want to select. More importantly, there is no one universal methodological paradigm suited to all situations. One needs to choose the paradigm based on the situation and what they want to achieve. As Patton (1990) puts it ‘research and evaluation should be built on the foundation of a “paradigm of choices” rather than become the handmaiden of any single inevitably narrow disciplinary or methodological paradigm.’

Box 2. Bear’s decision to like honey

One day, in a sudden impulse of generosity, a bear decided to enlighten the other animals in the forest about the marvellous properties of honey. The bear assembled all the other animals together for his momentous announcement “I have studied the matter at great length,” began the bear, “and I have decided. that honey is the best of all foods. Therefore, I have chosen to like honey. I am going to describe to you the perfect qualities of honey, which, due to your past prejudices and lack of experience, you have ignored. Then you will be able to make the same rational decision that I have made.

“Honey comes conveniently packaged in beautifully shaped prisms of the most delicate texture. It’s ready to eat, slides down the throat ever so easily, is a highly nutritious source of energy, digests smoothly, and leaves a lingering taste of sweetness on the palate that provides pleasure for hours. Honey is readily available and requires no special labour to produce since bees do all the work. Its pleasing aroma, light weight, resistance to spoilage, and uniformly high quality makes it a food beyond compare. It comes ready to consume — no peeling, no killing, no tearing open — and there’s no waste. What’s more, it has so many uses; it can be eaten alone or added to enhance any other food.”

“I could go on and on, but suffice to say that I have studied the situation quite objectively and at great length. A fair and rational analysis leads to only one conclusion. Honey is the supreme food and any reasonable animal will undoubtedly make the same conscious decision I have made. I have chosen to like honey.” Patton (1990)


Let us not be closed-minded on the choice of methods and let our choice of method be nothing like a ‘bear choosing to like honey’. Let us be open to and promote more qualitative studies in agriculture as they can generate insights which are beyond the reach and scope of quantitative studies.

Note: We are not experts in the field of qualitative research, as we started working on it only recently and have developed a fascination for it. If there are any errors in the explanation, we totally own them. We also acknowledge that this blog is inspired by the teachings of Prof. Dr Vera Bitsch of TUM, Munich, Germany. Many of the points and literature comes from the classes of the Professor and the discussions afterwards. We are forever grateful to Prof Bitsch for the inspiration.

Recommended for Further Reading

Arsel Z. 2017. Asking questions with reflexive focus: A tutorial on designing and conducting interviews. Journal of Consumer Research 44(4):939-948.

Bitsch V. 2005. Qualitative research: A grounded theory example and evaluation criteria. Journal of Agribusiness 23(345-2016-15096):75-91.

Bitsch V. 2009. Grounded Theory: A research approach to wicked problems in agricultural economics. Mini-symposium on Qualitative Agricultural Economics. Mini-symposium qualitative Agricultural Economics at the International Conference of Agricultural Economists, Beijing, August (2009)

Geertz C. 2000. Deep play: Notes on the Balinese cockfight. Pages 175-201 in: Culture and politics. New York: Palgrave Macmillan.


[1]This blog is written in a conversational tone, with ‘I’ and ‘We’ being used interchangeably.

Aditya KS is a PhD scholar at Humboldt University, Berlin, Germany, and a Fellow of Netaji Subhas ICAR-International Fellowship. He is also a Scientist (Agricultural Economics) at ICAR-Indian Agricultural Research Institute, New Delhi. His research interests include the impact assessment of agricultural technologies and natural resource economics. He can be reached at

Bhuvana N, holds a PhD in Agricultural Extension from Professor Jayashankar Telangana State Agricultural University, Hyderabad. Her research interests include organizational ecosystems and effectiveness, social networks and technological change. She can be reached at




Click here to post a comment

  • “The blog was afresh and quite thought provoking. Obviously erosion of knowledge occurs when social science data is strictly put in the boxes of quantitative analysis. Here the methodologies of psychology research may provide us cues in sharpening and pursuing qualitative research methods in agricultural extension. Thank you Aditya and Bhuvana and AESA for the blog. This thought need to be taken forward as a research area for widening meaningful research in an inclusive arena”

  • “A realistic perspective of social research! This blog introduces qualitative research as a way of understanding a phenomenon in a “holistic way”, which enables a researcher to derive actionable outcomes from his/her empirical work. The need for a qualitative approach stems from a researcher’s goal of dealing with a social phenomenon. He/she may wish to understand the phenomenon – what is happening, with what effects, what are the causes, the relationship between cause and effect, etc. The level at which a researcher wishes to understand the phenomenon (goals) and his/her plans to use the outputs are the keys to determining the research approaches.

    I believe that the recent fascination of agricultural social scientists with the quantitative approach stems from the “immediacy” of the “high impact outputs” they get out of their research work. It’s like a researcher learning a new technique, applying it in research work, publishing a high-impact paper, and moving on to the next. In reality, the quantitative techniques are suitable mainly for higher-order “explanatory and prediction research goals” such as associations and cause-effect, which often help in “replicating” the phenomenon (e.g. field interventions). Most of the extension phenomenon stems from real-life faced by agricultural stakeholders, are complex in nature, and we need “metacognitive knowledge” to manage the phenomenon. Acquiring “metacognitive knowledge” is not a direct process, it is a step-by-step approach to moving from factual, conceptual, and procedural knowledge, In simple terms, if we move directly to “explanation” without “description and exploration”, we may not acquire the ability “to replicate” the action (e.g. model)/ “prescribe a solution” (e.g. training/ information/ supplying technology)/”develop an action plan (e.g. decentralized planning).

    Qualitative approaches help the social researchers to understand the phenomenon in a realistic way (in terms of what is it?, what are the causes? Why? When? and what effects?), when used along with quantitative approaches, and enable them to deliver actionable outputs (besides increasing their citation index). When the Govt is investing heavily in development programmes, our inadequate understanding of a social phenomenon will lead to hasty and poor decisions.

    When the Government is investing heavily in multiple agricultural support programs for doubling the farmers’ income, soil health, credit delivery, special sector development (SC/ST/NEH/Women/Youth), it is essential for social science researchers to help the Government machinery with relevant methods, implementation guidelines, tools and policy support information. A recent World Bank report on repurposing agricultural policies indicates that current support for agriculture produces low value for money as a way of helping farmers – for every dollar of public support, the return to farmers is just 35 cents. As agricultural social scientists, we have a greater role to play in the development process (apart from enriching our biodata) and qualitative approaches are the key to conducting “responsible and outcome-driven” social research.

  • This well-articulated blog is putting focus on the importance of qualitative studies in agricultural science. However, this importance is hardly recognized. The research methodology courses offered in social sciences are less focused on qualitative methods and their application. This is evident from the abysmally lower number of PG/Ph.D. studies with a qualitative focus. Hence, it is important to open our views and gain a better understanding of qualitative approaches to have more paradigm choices for a social scientist.

  • “Hope the young social scientists will imbibe the spirit and essence of this blog.
    Qualitative data is often presented as a narrative and collect the experiences and insights of farmers (why and how).

    Quantitative data collection retrieves numerical data in terms of what, where and when.
    Hence both quantitative and qualitative data analysis are important. It is better to adopt a combination of both. Qualitative analysis is relevant because the data shared by individuals is more powerful in analysing complex systems, which will be able to illustrate how the implementation of programmes and policies are working in the field settings.

    However, there are concerns with the qualitative methodology. One important concern is to remove subjectivity to the extent possible. If there’s subjectivity, methodological ways to generate consensus need to be adopted. There should be normalisation of such data.
    I congratulate the authors for the blog”

  • To get information fast initially Qualitative methods help a lot. To have more valid findings qualitative methods need to be twined with quantitative approach

  • There are no two opinions about using a combination of qualitative and quantitative methods in social science research. The reasons for the social scientists not in favor of qualitative research is obvious. As is well known qualitative research is subjective and it is difficult for the researcher to defend the results (even if qualitatively superior) unless the results are supported by the quantitative data. On the other hand quantitative research need not to take the support of qualitative data, often ignoring the quality of research. As argued by the authors that qualitative methods definitely help in understanding the ground realities which we normally miss while using only quantitative methods. I share our experiences in this regard.

    When we wanted to study the level of poverty in Women Self Help Groups, we adopted the usual quantitative method of family income to categorize families into rich and poor. We also asked the members to indicate the constructs which help in identifying poor families They consider families as poor if

    i. they cannot afford to take breakfast

    ii. their houses are kutcha ( at least one out of the three viz roof, walls or floor is not pukka)

    iii. they do not have debts

    iv. they do not own a two wheeler etc.

    Similarly, when we wish to know how many families do not consume milk, we obtained the data through usual surveys. But again, when we observe some of the school children taking the tumbler of milk home ( for making tea for all the family members) instead of they drinking the milk supplied by the Govt. as part of mid day meal scheme, that explains the glaring realities of rural life..

    Congratulations to the authors Aditya and Bhuvana.