Page 54 - AJWEP-22-4

Basu

RCL = ∅ + ∅ Z + ∅ Z + ∅ Z + ∅ Z + ∅ Z + ε i averages approximately 6 years, ranging from no
2
2i
1 1i
i
0
3
5
5i
4i
3i
4
(V) formal education to a postgraduate degree. The average
where RCL is the relative cyclone loss of i farmer household size is 4 members, with some households
th
i
(measured across rice, fish, vegetables, and other crops); comprising up to 15 individuals. The number of earning
Z is the dummy variable for occupation (1 = Primary family members, excluding the household head, ranges
1
occupation is agriculture, 0 otherwise); Z is the age of from 0 to 5. The respondents’ average monthly earnings
2
farmer (years); Z is the gender of the farmer (1 = Male, and expenditures are BDT 15,166 and BDT 14,412,
3
0 = Female); Z is the years of schooling; Z is the family respectively. On average, farmers report 21 years of
5
4
size (number of members); ∅ ,… ∅ are the coefficients experience in rice production.
1
5
representing the influence of each independent variable Regarding production-specific statistics, the average
on RCL; and ε is the error term. The model assumes rice yield is 35.71 maund (approximately 1428.4 kg).
i
zero perfect multicollinearity. Robust standard errors The output level ranges from 5 to 390 maunds,
were used to mitigate heteroskedasticity. depending on the inputs used. On average, 1.216 acres
of land was cultivated, involving around 15 man-days
3.3.4. T-test for comparison of sociodemographic and of labor, 40 kg of seed, 68 kg of chemical fertilizer, and
production-specific variables 1 kg of pesticide. Regional-specific descriptive statistics
The t-test is a parametric statistical test used to compare for cyclone-affected districts (Khulna and Satkhira)
the means of two independent groups. In this study, are presented in Tables A1 and A2, respectively. The
a t-test was employed to assess sociodemographic sociodemographic characteristics are generally similar
and production-specific differences between cyclone- for both districts, with minor differences in a few
affected and unaffected farmers. The variables analyzed variables.
include age, education, household size, household The mean differences in the sociodemographic
income and expenditure, farming experience, as well variables of cyclone-affected and unaffected farmers
as production-specific variables such as output, land are presented in Table A3. On average, unaffected
area, labor, seed, fertilizer, and pesticide use. The farmers have significantly higher levels of educational
t-test operates under the null hypothesis H : μ =μ qualifications and farming compared to affected farmers.
0
1
2
(i.e., no significant difference between the means of the On the other hand, the average family size and monthly
two groups) and the alternative hypothesis H : μ ≠μ . expenditures are statistically higher for affected farmers
1
1
2
Equation VI shows the formula for calculating the test compared to unaffected farmers. Unaffected farmers
statistic: have statistically greater rice output and cultivated land
size than treatment groups, while affected farmers apply
X − X statistically greater quantities of chemical fertilizers and
t = 1 2 (VI)
s 2 s 2 pesticides.
1 + 2
n 1 n 2 4.2. Impact of Cyclone Amphan on rice production
4.2.1. Cobb-Douglas production function model
2
where, X and X are the sample means, s and s The impact of Cyclone Amphan on rice production
2
2
1
2
1
are the sample variances, n and n are the sample sizes was estimated using the Cobb-Douglas production
2
1
of the two groups. The calculated t-statistic is compared function applied separately to farmers in Khulna
to a critical value from the t-distribution at a specified (Model 1), Satkhira (Model 2), and both districts
significance level (e.g., α = 0.05). If the t-statistic combined (Model 3). Regression results are presented
exceeds the critical value, the null hypothesis is rejected, in Table 2, using both log-transformed and standardized
indicating a significant difference between the groups. coefficients (mean = 0, standard deviation = 1).
The standardized regression coefficients reflect the
4. Results strength of the explanatory variables on the dependent
variable.
4.1. Summary statistics of rice farmers In Model 1, land, labor, and seed variables show
The sociodemographic characteristics and farm-specific positive and statistically significant associations with
variables of cyclone-affected farmers are presented rice output at a 1% level. Specifically, a 1% increase in
in Table 1. The average age of farmers is 48 years, land, labor, and seed led to 0.57%, 0.20%, and 0.25%
ranging from 20 to 80 years. The educational attainment increases in rice production, respectively. However,
Volume 22 Issue 4 (2025) 46 doi: 10.36922/AJWEP025100063