The contribution of rubber to national economic and social development is important for Indonesia. However, smallholding rubber, the dominant rubber producer, has low productivity. Various new technology programmes have been introduced by the Indonesian government with other agencies to increase the productivity of existing traditional rubber and incomes among smallholder rubber farmers in Indonesia. However, the adoption of new technology was low and the reasons for these were still unclear. This study explores how smallholder farmers in Indonesia adopt new technology. Rubber Agroforestry System (RAS) introduced mainly by International Centre for Research in Agroforestry (ICRAF) in Jambi and West Kalimantan provinces in Indonesia is used as a case study. A combination of Ethnographic Decision Tree Modeling (EDTM) proposed by Gladwin (1989a) and a logistic regression model were used as the main methodologies to determine the decision criteria of rubber farmers regarding adoption of clonal rubber. The EDTM as qualitative method helped to identify the main reasons, motivations and constraints that influenced a farmer’s decision to adopt or not adopt the new technology and also present details about the process of the farmers’ decision making. Meanwhile, logit as the quantitative method was useful to identify the significant variables involved in the decision making process. The results of this study show that the decision making process for adoption of clonal rubber is complex and influenced by various factors. The decision tree models for Jambi and West Kalimantan differed showing the importance of social context and infrastructure. The main reasons for a farmer’s decisions to adopt clonal rubber is the expectation that clonal rubber is better in growth and yield and it will increase production per ha and income. The decision to adopt is supported by evidence from demonstration plots, trust in the technology deliverers and availability of incentives. The main constraint in adoption for both areas was limitation of capital as the clonal rubber required more capital to establish. The other constraints are risk and uncertainties including pest and disease problems, the shortage of labour, lack of technical knowledge, lack of access to clonal seedlings, and observation of clonal rubber that has been of low quality or managed inadequately. The decision tree models have been tested and the results show that the models were able to predict the farmers’ decision making with good accuracy of 82% and 83%. In addition, the quantitative model shows the significant factors that determine adoption of clonal rubber in Jambi and West Kalimantan are land, incentives and income factors. The qualitative and quantitative methods contributed to increased robustness of data and give different kinds of valuable data and information to stakeholders and policy makers in Indonesia. In order to encourage rubber farmers in Jambi and West Kalimantan to adopt clonal rubber, this study suggests improving policies to ensure they are aligned with needs of the rubber farmers, improving farmers’ access to capital sources such as credit with simpler mechanisms, increasing the number and skills of extension workers, encouraging farmer to farmer learning, empowering farmers and leadership, improving infrastructure including better access to clonal seedlings and improving partnership with NGOs.