Conducting Social Science Research: Reflections

By Alexander Stubbings

In 2011, I was studying climate change at the University of East Anglia. When it came time to choose a thesis topic, I knew I wanted to use the natural sciences to help solve a social problem – a sort of ‘knowledge coproduction’. After plenty of reading and hours spent researching, I finally decided to investigate how Bangladeshi agriculture can adapt to biophysical risks. In essence, I wanted to know how changes in weatherrelated phenomena could and would affect agriculture.

I chose Bangladesh because it is the world’s most natural disaster-prone country. It experiences many types of natural hazards, geophysical and atmospheric in origin, that often devastate the land and its inhabitants.

 /* Style Definitions */
	{mso-style-name:"Table Normal";
	mso-padding-alt:0in 5.4pt 0in 5.4pt;
  © David Woodfall / WWF-UK

© David Woodfall / WWF-UK

The Process

I wanted to know how agriculture in Bangladesh could adapt to biophysical risks, so I started by asking three questions:

  1. What biophysical risks are currently affecting Bangladesh and what are the future risks?
  2. What types of adaptation currently exist?
  3. What type of adaptation will be effective in this region?

To answer the first question, I scoured the EM-DAT database (International Disaster Database), and discovered that Bangladesh is affected by the following natural hazards: flooding, extreme temperatures, drought, mass movements, and storms. I then categorized each of these hazards according to each region within Bangladesh, to see if I could find any emerging patterns.

For the second question I turned to literature reviews from Web of Science, and searched for grey literature using Google. The key here was to find out what adaptation strategies exist for agriculture, both high and lowtech solutions.

The last question utilized a literature review, conducted only in scientific, peer-reviewed journals. To derive authoritative evaluation questions I critically analyzed each paper. When I came across what I considered an appropriate question for evaluating adaptation, I recorded it (see below).  These questions were chosen for their relevance in evaluating adaptation projects. For instance, what the adaptation should be fulfilling in order to be considered worthwhile.

  • Who (is adaptation occurring for?)
  • What (is adapting?)
  • Why (is adaptation occurring?)
  • Spatial scale of adaptation?
  • (Type of adaptation): Autonomous (individual)
  • (Type of adaptation): Planned (government)
  • Is there the potential for maladaptation to occur (in other words could the adaptation do more harm than good?)
  • (Do they exceed social or natural) critical thresholds?
  • Possibility of winners and losers?
  • Does the proposed adaptation improve coping strategies?
  • Are the policy options for the adaptation robust enough?
  • Does the adaptation increase pressure downstream?
  • Does the adaptation have legitimacy?

The above evaluation criteria were then used to test the following indicators, which had been selected to answer the second sub-question:  what types of adaptation currently exist? These indicators (listed below) were selected after performing another search within Web of Science. I then grouped what I considered similar adaptation options together which resulted in the following categories:

  • Culture
  • Technology
  • Governance
  • Agricultural practice
  • Capacity building
  • Social practices
  • Education
  • Policy
  • Institutions
  • Infrastructure

Lessons Learned

When I was evaluating the indicators I attempted to become culturally sensitized – viewing things as if I were a Bangladeshi farmer. This ultimately devalued the effect and quality of the research. Put plainly, I am not a Bangladeshi citizen. And therefore, will have a different perception of risk: an emerging key theme from the research.

Risk is invaluable in coming to terms with how a culture successfully adapts, and I should have investigated it more. Risk can be defined as the probability of climate hazard multiplied by a given system’s vulnerability (Ramamasy and Bass 2007). It is within the social construction of risk that these phenomena relate to one another (Hannigan 2006). Furthermore, the framing of risk has been based around advances in science and industrialization, combined with economic growth.  A consequence of such development is that society has been exposed to a series of risks and hazards never before encountered (Lash and Wayne 1992). And borrowing ideas from the Red Cross Red Crescent Climate Center, conducting interviews and using games would have been an interesting way to model people’s behavior and response to risk.

In hindsight I should have used the following research methodologies: action research, focus groups, or even key participant interviews.  These would have produced a wider and richer tapestry of data, results and discussion. And this is by no means an example of self-loathing. I learned a lot from conducting my first ever research. Regardless of background training, I feel that now I could certainly contribute something more substantive to research I do in the future.

After re-reading my thesis again, I found that I could have tackled the question in a variety of ways. I could have used statistical techniques to better effect. I could have broken down the question and focused more on specifics. In short, I could have done a lot more. But, for a first-time researcher I’m glad I’ve had this opportunity to reflect and think about the process. Isn’t that the entire point of good research? Okay, you might not discover anything new, but as an individual, as a scientist, you expand; you’ve stretched yourself, and the next time you come to undertake original research you’re more informed and more importantly, better prepared.


Ramamsay. S., Bass. S. (2007). Climate variability and change: adaptation to drought in Bangladesh, a resource book and training guide. Asian Disaster Preparedness Centre. Food and Agriculture Organisation of the United Nations. Rome.

Hanningan. J. 2006. Environmental Sociology. 2nd Edition, Routledge, pp. 188. Chapter 2

Lash. S., Wynne. B. Introduction. In: Ritter. M. Risk Society; towards a new modernity. SAGE publications. 1992, pp. 251