2.1 Study area The study has been conducted in the Narsingdi district of Bangladesh. It is located at 235501200N, 904304800E in central Bangladesh and 50 km northeast from the capital city of Dhaka. Topographically, the district lies at an altitude of 3 masl (meter above sea level) with an annual average temperature of 36 C maximum, 12.7 C minimum and 2,376 mm of rainfall. It comprises a total area of 1,140.76 km2 of which the total and vegetable cultivated areas are 87,447 and 15,000 ha, respectively. Administratively, Narsingdi district belongs to six upazilas (sub-districts), 69 unions (collective forms of some villages) and 1,060 villages. According to a survey conducted in 2011, the total population of the district is 2,202,000 with a moderate literacy rate of 55 %. Agriculture is the main occupation of majority of the people (42.41%) followed by service, commerce, transport and others (DAO 2013). Narsingdi district is one of the most important regions in Bangladesh in vegetable production and pesticide application intensity. In Bangladesh, it is the area which is a hot spot for vegetable production, at the same time a hot spot for pesticide application as the vegetables receive the highest amount of pesticides per application and the highest application frequency (Ispahani Biotech 2010). Because of the positive correlation between vegetable production and pesticide application, the government started IPM dissemination activities in the area from the start. Eggplant, cucurbits, tomato, bean, okra, cabbage and cauliflower are the most cultivated vegetables. Irrespective of the type of vegetables, pest infestation is a common scenario. That is why the farmers are concerned about pest control. They manage pest by following a number of techniques such as biological control, manual cleaning, soil amendments, soil solarization, botanicals, pheromone traps and chemical pesticides.
2.2 Methods of data collection The study uses mainly primary data collected through a household survey. The survey was conducted using a structured questionnaire that comprised both open and close questions.
The questionnaire was prepared and finalized through three consecutive stages. In the first stage, some social, economic, institutional and management characteristics of the vegetable growers and a list of vegetable IPM technologies were preliminary recorded from the literature review. In the second stage, an informal discussion was held with the personnel of the plant protection wing of the Department of Agricultural Extension (DAE), Bangladesh Agricultural Research Institute (BARI) and the agriculture office in concern. In the third stage, a pilot survey was conducted with a few vegetable growers. The questionnaire contained two sections where some social, economic, institutional and management factors relating to the vegetable growers of the study area were present in the first section. The lists of IPM technologies suitable for vegetable cultivation that since exist in the study area are presented in the last section.
2.3 Determination of dependent variables There exist a number of IPM techniques in Bangladesh though all are not suitable for vegetable farming. All suitable vegetable IPM technologies, except for grafting, were practiced in the study area for a long time. The farmers who used at least one or more from these recommended practices, as well as pesticides, were considered as IPM adopters. By contrast, those who only used pesticides were considered as non-adopters. Alternatively, it can be said that the adopters used a combination of various methods, whereas the nonadopters used only chemical method. The method is similar to the method mentioned in the study conducted by Dasgupta et al. (2006) and Fernandez-Cornejo et al. (1994) who determined the adopters/non-adopters of IPM on the basis of some selected IPM technologies that were consistent with crops and exist in the study area for a long time. The rate or level of IPM adoption, which was calculated through distinguished adopters and nonadopters, was considered as a dependent variable. However, the farmers who were considered as IPM adopters were assumed to be of a variety according to the number or types of practices. This variation was also considered as another dependent variable. Therefore, the study relied on two dependent variables; the adoption rate of IPM and adoption intensity of IPM practices. The first dependent variable (adoption of IPM) was a binary variable and was coded as 1 for those who adopted at least one or more IPM practices and 0 for those who used pesticides only. On the other hand, to measure the second dependent variable (adoption intensity of IPM practices), the number of practices adopted by an individual was initially counted. Irrespective of the type of practice, a score of 1 was assigned to the practice adopted by the farmers. Then, all the scores were aggregated and divided by 6 (total number of practices) to obtain a composite index of the adoption intensity of IPM practices (Haque et al. 2010; Paudel and Thapa 2004). This index was considered as the second dependent variable.