Locale of the Study: Four unions namely Biswarkandi and Illuhar unions of Banaripara upazila under Barisal district, Baldia union of Nesarabad upazila and Kalardoania union of Nazirpur upazila under Pirojpur district were selected as locale of this study because of higher concentration of floating agriculture. Income from floating agriculture expected to have a large impact on the livelihood of the farmers in that area.
Population and Sample of the Study: All the farm family heads practicing floating agriculture of the selected four unions constituted the total eligible population units of the study. A list of the eligible units of population of selected four unions i.e., sampling frame was prepared very carefully. Total number of units of population was 500 (67 from Biswarkandi union, 89 from Illuhar union, 56 from Baldia union and 34 from Kalardoania union). Respondents were selected by using two stage random sampling technique. The researcher also kept a reserve list, which was used when any farmer from original list was unavailable at the time of data collection. To determine the sample size from total eligible population, the following statistical formula was used as stated by (Kothari, 2004). The minimum sample size was 125 with 10% margin of error at 90% confidence level. In considering time, resource and other aspects of reliability, the final sample size under this system was fixed at 125, which was almost 31% of the total eligible population. The estimated sample size was distributed proportionately to each of the selected unions according to the number of population units.
Tools of Data Collection: Primary, secondary, quantitative and qualitative data were accumulated for the present study. For collecting primary data at household level a pre-designed interview schedule was developed with balanced combination of both closed and open-ended questions and the same was pre-tested before finalization. Participatory Rural Appraisal (PRA) tools and techniques like Focus Group Discussion (FGD), Case study and Direct Observations were also applied for triangulation of data.
Data Collection: Data were collected during the months from April, 2013 to October, 2014 by the researcher team. To get the valid and pertinent information the researcher tried to make all possible efforts to explain the purpose of the study to the respondents. While starting interview, the researcher took utmost care to establish rapport with the respondents, so that they do not feel hesitant or hostile to furnish proper responses to the questions of the instrument. All possible precautions were taken to avoid bias and to maintain fidelity of responses. It was the researcher’s privilege to be a native of the study area and thus to overcome the language problem. While interviewing, local dialect was used that helped both the researcher and the interviewees to understand each other. The questions were explained and clarified whenever any respondents felt difficulty in understanding properly. Care was taken not to use technical jargon, not to feed answer to farmers by teaching them terminology or using names. Where necessary questions were asked in several different ways until the researcher was sure that farmer understood the questions. If farmers response was not clear enough what he intended to mean, supplementary questions were asked for further clarification. Interviewer effect such as assuming the meaning of a response was then kept to a minimum.
Measurements of Variables: Profitability of floating agriculture was determined by analyzing BCR of individual respondent for a season of his floating agriculture practices using gross return and total cost.
BCR= Gross return/Total Cost
Per unit area income from floating agriculture of the respondent was determined by determining the respondent’s per decimal income in taka from floating agriculture in a season.
Per unit area income = Total income from a floating agriculture in a season/Total floating land in decimal
Effectiveness of floating agriculture to combat climate change was determined by taking the opinions of the respondents against the floating effectiveness related statements. Based on available literature and experience of the researcher and suggestions of the experts, 11 statements were selected. Respondents’ were asked to furnish their opinions against each statements on three point scale (strongly agree, agree and disagree). Their responses were compiled in frequencies and percentages against individual statement as well as overall effectiveness. A regression equation was developed by using collected data to find out the factors influencing unit area income of floating agriculture. Selected socio-economic characters were considered as influencing factors. Influence of each factor was analyzed in percentage. The analysis was done by determining the difference between R2 of regression equation containing all factors and regression equation containing focused factors.
Y = B0 + B1X1 + B2X2 +……. + BnXn; Where: Y = dependent variable X1, X2, …, Xn = any independent variables B0, B1, Bn = any constants
Data Processing and Analysis: After collection of data, all the information contained in the interview schedule was edited. All the collected data were then checked and cross checked, compiled, coded and entered into the computer for analysis and interpretation using SPSS program. Qualitative data were converted into quantitative form by means of suitable scoring. Data were presented mostly in tabular forms, statistical measures like number, range, mean and standard deviation were calculated in describing the status and profitability of floating agriculture. To find out whether any significant differences were existed among the opinions of the respondents on the statements of judging the effectiveness of floating agriculture to combat the impact of climate change, Chi-square test was used. Coefficients of correlation were computed to find out the relationships between socioeconomic characteristics of the respondents and their unit area income from floating agriculture. To find out the contribution of socioeconomic characteristics of the respondents on their unit area income from floating agriculture, multiple as well as stepwise regression analysis were used.