238 Journal of Food, Agriculture & Environment, Vol.12 (1), January 2014 www.world-food.net Journal of Food, Agriculture & Environment Vol.12 (1): 238-247. 2014 WFL Publisher Science and Technology Meri-Rastilantie 3 B, FI-00980 Helsinki, Finland e-mail: info@world-food.net Received 14 October 2013, accepted 27 January 2014. Cluster analysis for classification of farm households based on socio-economic characteristics for technology adoption in agriculture: A case study of West Java province, Indonesia Nita Kuswardhani 1, 2 , Peeyush Soni 1 * and Ganesh P. Shivakoti 1 1 Agricultural Systems and Engineering, Asian Institute of Technology, Thailand. 2 Agricultural Technology Faculty, Jember University, Indonesia. e-mail: soni@ait.asia, soni.ait@gmail.com, nkuswardhani@gmail.com, ganesh@ait.ac.th Abstract Blind drive of intensifying vegetable production to combat issues of low productivity, low quality of product and ensuring off season availability of vegetables, pose serious challenges for smallholder farmers to adopt efficient and appropriate technologies. Technology adoption is a complex process depending on several factors mainly including the socio-economic status of individual farmers. The differentiation on adopting technology could be explained by farm typology. The aim of this research was to classify groups of farm households in the West Java Province of Indonesia based on identification of factors influencing new technology adoption. A survey of farmers was carried out during January-December 2010 in Sukabumi (medium and highland) and West Bandung regency (highland). The information obtained through the survey was analysed using multivariate analysis, which was carried out through Principle Component Analysis (PCA) and Cluster Analysis (CA) to classify the farm households based on socio- economic characteristics. Selected variables derived from questionnaires were used to describe the components that influence the adoption of agricultural technology and to identify the group of potential farmers. The results indicate that the farmers could be classified into three, four, and two groups of typical farm households in medium land Sukabumi Regency, highland Sukabumi Regency, and West Bandung Regency of West Java Province, respectively. Key words: Technology adoption, farm household, farm typology, multivariate analysis, Sukabumi regency, West Bandung regency. Introduction One of the most important strategies in economic growth is the development of the agricultural sector. The agricultural sector is a major source of livelihood for most Indonesians, and as a source producer for considerable foreign exchange. Agriculture contributes 14.7% of Gross Domestic Product (GDP) in Indonesia 1 . Even though this sector is a small share of GDP, it is still regarded as one of the most important sectors in terms of providing employment. In many developing countries including Indonesia, it is a concern to extend the horticultural production base as it could contribute to agricultural development, diversity of agriculture, and opportunity of employment. The government of Indonesia is re-defining the expansion of horticultural production into the establishment of agribusiness and agroindustry system 2 . The main goals in horticultural agribusiness are to succeed in primary horticultural producer/participation of farmers in the local market and to enhance economic performance and livelihood, and to build strong relationships between farmers and the private sector, especially the trading and retailing sector. West Java Province, especially in Sukabumi and West Bandung regency, is one of the important provinces that produce notable amounts of horticultural products including vegetables, because of its favourable climate and fertile land. This province has potential to increase horticultural production by introducing appropriate new technologies to the farmers. With an aim of expanding the horticultural production in Indonesia, protected cultivation, as a new adoption technology has been introduced in the West Java Province. However, the promoted technology has not grown rapidly mainly due to socio- economic circumstances. So far, no study has been reported to analyse the underlying situation into farm household profile with relation to the adoption of new technology in this province. Differences in adoption could be highlighted by farm typology. It is known that promoted new technology for farmers has to be matched with socio-economic circumstances. Different socio- economic characters among farm households may result varied levels of motivation in accepting new technology 3-10 . The income and utility models are also the important determinants in technology adoption for an economic paradigm 11 . Cary and Wilkinson 12 mentioned that in the income paradigm, farmers are profit maximisers and the technology that increases net returns will be adopted. Meanwhile, the utility paradigm asserts that farmers adopt technology basically due to utility maximisation rather than profit maximisation 13, 14 . In the utility paradigm, a producer responds to many factors besides income, including non-income factors, such as environmental quality, social benefit, and/or altruism. Bultena and Hoiberg 15 used risk orientation, perceived erosion and perceived attitudes of other farmers, as well as their perception of other farmers’ adoption to explain the adoption of conservation practices. Some researchers had provided evidence that availability of information to producers, and the level of education and experience of prospective adopters, are better determinants for adoption of new practices than income 14 . Technology adoption has been a major part of the agricultural research agenda for economists and sociologists for several decades 16-19 . Developing typology is an important step in any realistic