Rice is an important food crop of the union territory of Puducherry accounting about
63% of the total cropped area and contributes about 98% of total food grain
production (Anonymous, 2005). In view of increasing demand for quality seeds, the Government
of Puducherry established a Seed Certification Agency in January 2000. The basic functions
of the agency are production, collection, storage, testing and distribution of mainly the
certified paddy seeds. There are several studies pertaining to cost of cultivation of paddy
under commercial production, but there are only a few studies related to cost of paddy
seed production. Hence, the main objectives of this study are: To analyze the economics of
paddy seed production vis-a-vis commercial production, and to specify the variables that
are discriminating the seed production from commercial production.
The study has been conducted in the Puducherry region as it ranks the highest area
under paddy compared to other regions. A sample of 60 seed growers was selected from the list
of registered seed growers at random. A matching sample of farmers, having
commercial cultivation of the paddy crop, was selected randomly from the same village wherefrom
the seed growers were selected, thus making the total sample of 120. Primary data on the cost
of cultivation on both seed crop and commercial production of paddy were collected from
the selected sample farmers through a suitable pre-tested schedule during the agricultural
year 2005-2006. Tabular analysis was used to estimate the costs and returns.
The linear discriminant function analysis was used to identify the variables that
were important to discriminate between the two groups of farms. In multivariate
analysis, linear discriminant function, which is better than any other linear function,
will discriminate between any two chosen classes (Dillion and Goldstein, 1984).
The concept underlying the discriminant function analysis is that linear combinations of
the independent variables are formed and they serve as the basis for classification.
Thus, the information from multiple independent variables is summarized in a single
index. For the application of linear discriminant function, two groups of roughly equal
size are required. |