This is a good question! Because of my familiarity, I will be discussing about the qualitative methods:
1. First need to Clarify some ‘W’s:
Where: Selecting the ‘right’ site is important. We have to select a flood prone area in order to see the relationship between variables. It can be an area which just experienced a big flood or which is constantly being affected by flood through river cutting or inundation or sedimentation. Variables and relations between them become clearer in those areas. This will also give an opportunity to see which households are resilient and which are not. Selecting a site also depends on the research question you have. However, there are some widely accepted variables that have impact on vulnerability, such as caste/ethnicity, race, poverty, geography (this introduces a new set of variables that moderate vulnerability status of any community or households and this will involves proximity to river or the potential source of flood and elevation) and so on. To see who is vulnerable and who is not, you will need to select a diverse/heterogeneous community. Consider that heterogeneity before selecting the field site. Vulnerability studies takes relatively longer time. Because who can cope with or adapt to the risk or can bounce back effectively can only be seen if observed for a reasonable length of time. So, the site should be feasible for long term study too.
Who: We also need to be clear about the unit of analysis of our study. Vulnerability at what level – individual, household or community? However this is not a water-tight compartment and your analysis can have/require all three levels of data. Besides vulnerability at household level, information level at community also tells more about the vulnerability of the community as a whole. Most of time, inter-community differences in - exposures to risks, mitigation approaches, ways of coping with disasters, social capital and access to resources have huge impact on the ‘total’ vulnerability of that community. Besides, the levels, we also need to be clear on who our respondents can be. Generally, in such studies, the respondents can be the local people, staffs of I/NGOs working in DRR, government functionaries and any other agencies or individuals that controls or have influence in the capital flow within/to the community.
How: how are you going to get the information? In qualitative research generally we do Key Informant Interview, Focused group Discussion, and sometimes we can also use other participatory tools. We can get rich information on coping strategies and mitigation through observation. Tool preparation is critical here and the questions we will be carrying will depend on the objectives of the research. However, it should comprise the variables that are considered to be influential in the distribution of vulnerability across different sections of society (and for this looking at past studies done in this field can be useful; so it would be good start literature review side by side).
2.After you are done with this, you can prepare a tool and can go for a pretest (even if it is a check list, it would be good to do pretest, especially if you are hiring researchers who are not involved in the research from the beginning or are not involved in the tool preparation). You can check the coherence of the questions, their relevance to the study, redundancy, and most importantly its exhaustiveness and can check responses against the objectives and can make necessary changes in the tool/checklist.
3.Now you can go for data collection and start translating or transcribing the data, if necessary.
4.Analyzing qualitative data requires going through a cumbersome and careful process. We identify the major theme that appears in the data set and the start write-up.
5.And, Write-up (however, documenting methodology, reviewing literature can start from the beginning of the research)!