DFM Kerala 2021 Ethics protocol supplementary information

This document provides details concerning dried fish value chain scoping research to be conducted in Kerala, India.

Research team

Holly Hapke, PhD, Research Scientist, University of California, Irvine - Project co-leader; principal investigator on processing and trade component

Nikita Gopal, PhD, Principal Scientist, Central Institute of Fisheries Technology, Ernakulam - Project co-leader; principal investigator on processing and trade component

Amalendu Jyotishi, PhD, Professor, School of Development, Azim Premji University - Co-principal investigator on processing and trade component (and consumption?)

Jeena T.S., PhD, Associate Professor, Centre for Economic and Social Studies, Hyderabad - Principal investigator on consumption component

Priya Gupta, PhD, Assistant Professor, Amrita Viswa Vidyapeetham, Bangalore Campus - Co-principal investigator on consumption component

Ramachandra Bhatta, PhD, President, Snehakunja Trust, Project advisor, coordinator, and co-principal investigator on processing and trade component

Research Assistants (2) - TBD - one assigned to processing and trade component, the other to consumption component

Field Enumerators (2-4) - to assist research assistants in data collection

Summary of planned research activities

The objective of this study is to "map" the social economy of dried fish in Kerala State. Utilizing a feminist commodity chain approach, we propose to conduct a historical spatial analysis of dried fish production (processing), trade, and consumption in order to understand the changing geography and social economy of dried fish, and how women’s roles in the dried fish economy have changed as a result of broader economic, technological and geographic changes (e.g., introduction of trawling and its impacts on the geography of fish harvests, processing and trade locally and nationally). Importantly, we are interested in understanding how broad socio-economic and technological changes (e.g., mechanization, export markets, alternative proteins, fishmeal industry, migration) have impacted dried fish production/trade/consumption; and how consumption (changing consumer preferences) has impacted the structure and dynamics of dried fish value chains.

The project will consist of two interrelated components. One will examine processing and trade; the other consumption. Although nominally distinct, ongoing dialogue and discussion between team members assigned to each component will foster a synthetic analysis and produce a final, integrative report in addition to scoping reports related to each value chain node under investigation: processing, trade, and consumption.

Methods

Part I - Processing and Trade

Investigation will center first on Ernakulam, Alleppey, and Kozhikode Districts, which are important contemporary and/or historical centers of dried fish production in the state. Our intent is to conduct focused work in these areas. If time and resources permit, we will conduct a rapid appraisal of fish processing and marketing in other coastal districts, including Trivandrum, Kollam, Trichur, and Kasaragod.

• Step 1: We will identify key marketplaces in each district (e.g., in Ernakulam - Aluva, North Paravoor, Varapuzha, and Ernakulam) and the retail outlets that extend from them. Through observations of market transactions and interviews with key actors in the value chain (commission agents, wholesale merchants, retail vendors), we will map the fish market system of the region, tracing from where fish arrives and to what other markets it then travels. Subsequent interviews may be conducted with traders and other key agents in secondary markets linked to these marketplaces.
• Step 2: Survey and/or interviews with dried fish processors to understand their different modes of operation and how they are connected to the dried fish economy and specific fish value chains and how this has changed over time.
• Step 3: Conduct in-depth oral history interviews with 10-15 key informants to gain an understanding of how the dried fish economy has changed within the broader context of 60 years of state-led development within the fisheries sector.

The interviews in steps 1 and 2 will be aligned with the DFM scoping interview guides developed at the Cox’s Bazar workshop and approved through the general research protocol for “Dried Fish Matters: Scoping Research Component” (see Documents 7-9 of this protocol).

Oral history interviews will be open-ended and will elicit information about how the market system was organized in the past. The period of focus is roughly 1955 to the present. (See Document 150: Oral history interview guide.)

Part 2 – Consumption

We considered two significant aspects or criteria for selecting the study locations. One is the physiographic factors influencing land use and consumption habits culturally; the other is the economic status or the households' poverty levels that affect households spending on different items in their consumption basket. We divided all 14 districts into two categories based on available poverty estimates or Head Count Ratios (HCR) using the National Sample Survey Organisation's 68th round survey data. We selected one district from each of these two categories that belonged to the three different physiographic regions. Thus Wayanad (highland), Malappuram (midland), and Alappuzha (lowland) from the first category and Idukki (highland), Kottayam (midland) and Thrissur (lowland) districts from the second category have been selected for the survey. We chose these districts so that they are also representative of various religious and social groups. We plan to survey about 50 households from each of these selected districts.

For selecting the sample households, at first two sub-districts, and then a panchayat or municipality representing rural and urban areas will be selected from these sub-districts. A panchayath ward in rural areas and a municipal ward in urban areas will be selected for household surveys. We plan to survey twenty-five households from each of these wards that will be selected based on a systematic random walk procedure from an identified key starting or entry point.

We plan to conduct open-ended focus group discussions with the members of the 'kudumbashree' or neighbourhood groups of the respective wards, using a brief set of guiding questions (see Document 149: Neighbourhood group interview guide).

Household surveys will follow the focus group interviews. Each household survey respondent will be a family member who takes critical food purchase decisions and or primarily cook and serve food for the family. Surveys will be carried out face to face by trained investigators in the local language. We plan to use questionnaires with both open-ended and closed-ended questions for the survey. The questionnaire will include questions on households identification and social and demographic characteristics, dietary intake of fresh and dried fish and other sources of animal protein, types, quantity, frequency of consumption, price and quality-related questions, sources of purchase or procurement and market-related issues, types and methods of preparation, the importance of fish and dried fish in the households diet, inquiries related to cultural beliefs and practices about food and dried fish, knowledge and attitude towards dried fish and dried fish products etc. The questions will capture the intra-household, and gender-specific factors influence the consumption of dried fish by the families. There will be questions to know the reasons or barriers for households who do not consume fish or dried fish. (See Document 151: Household consumption survey.)

The household survey draws on the categories of the India Household Consumer Expenditure Survey (IHCES), collected by the National Sample Survey Office (NSSO). Our survey will allow an analysis of intra-household and inter-household preference and consumption patterns, which can be extrapolated to a larger sample by correlating to household characteristics reported in IHCES, including: caste/social category, religion, household type, and land ownership at the household level; and age, gender, marital status, education, household rank, and occupational status at the individual level. Our own survey avoids collecting most of the directly identifying information about individual households contained in the IHCES instrument, including the names of household members, while retaining data that will allow for findings to be correlated (but not linked) to aggregate household type data from the IHCES.

We will use both quantitative and qualitative tools and techniques for data processing and analysis. Statistical and qualitative data analysis software such as SPSS, STATA and Atlas.ti will be used.

Anticipated risks

There are no significant known risks associated with the participation of people in this study.

Given that household composition details are requested within the household consumption survey, there is a potential for re-identification of respondent households based on the indirectly identifying data contained within individual survey responses. Such re-identification would require both access to the raw dataset and familiarity with the households concerned, however, and would be difficult in the case of households with similar characteristics. Moreover, given that none of the personal information collected through this survey is sensitive or has the potential to be used in a way that could harm participants, we consider the risk of malicious use following a data security breach to be minimal.

To minimize any unknown risks that may arise in semi-structured interviews, the research team will ensure free and informed consent with clear communication with the participants and emphasize openness, sensitivity, respect, awareness, and responsiveness. We will communicate to the participants that we will ensure the responses are documented in a confidential and private manner so as not to reveal their identity throughout all stages of the research. As outlined elsewhere in our ethics application, we will ensure this by strictly following confidentiality and anonymity rules throughout the research process including data collection, analysis and dissemination. For example, any reporting on activities that may likely result in the participants being vulnerable (e.g., illegal activities, child labour practices, or the use in fish processing of pesticides, etc.) will either not be done or will only be done after ensuring high levels of anonymity and confidentiality. Participants will also be reminded of their right to refuse to answer any questions or end the conversation at any time if they feel uncomfortable with the research process.