DISCRIMINANT FACTORS IN THE CHOICES OF TEA PLANT CULTIVARS AMONG SMALLHOLDERS IN BADULLA DISTRICT: MULTIPLE DISCRIMINANT ANALYSIS
This study employed multiple discriminant analysis to examine whether significant differences exist among the different group of tea smallholders in terms of their demographic and farming characteristics and to identify the significant factors influencing in the selection of different tea plant cultivars in Badulla district. The data were collected during the period November to December in 2019 and a sample of 108 tea smallholders has been randomly selected from three major tea producing regions namely Badulla, Haliela and Bandarawela divisions of the Uva region in the district. Frequency analysis reveals that, 43.5% of the tea smallholders chose VP 20/25 while 29.6% and 26.9% of them chose VP 20/23 and CY 9 cultivars respectively. Multiple discriminant analysis showed that, output from each tea plant cultivars and prices of each tea cultivar were the major discriminating variables as well as important influencing factors in selecting tea plant cultivars. Results of Fisher's linear discriminant functions revealed that as output increases, tea small holders are more likely to choose the cultivars VP20/25 and VP20/23 while less likely to select the CY9 cultivar. Similarly, as the price of the tea cultivar increases, they are more likely to adopt the CY9 cultivar than other two in the study. The classification results revealed that 56% of the respondents were correctly classified into three different groups of tea smallholders and the estimated discriminant functions for CY9 (96.6%) and VP 20/23(62.5%) were classified with better accuracy than another group in the study. The findings of the study may assist the tea smallholders in understanding the critical factors in the selection of tea plant cultivars and guide the policymakers to take effective measures in the tea sector in the future.
Keywords: Choices of Tea Plant Cultivars, Demographic and Farming Characteristics, Fisher's Linear Discriminant Functions, Multiple Discriminant Analysis, Tea Smallholders