Methodology
Sample
A primary survey was conducted to collect the required data for our study. Our survey questionnaire comprised of 10 observed variables. A high level estimation of sample size was done considering 20 data points per observed variable. Since we have 10 observed variables the estimated sample size came to be 200. We short listed around 400 SMEs expecting 50% response rate. In the list we excluded micro organization considering the difficulties in approaching these firms and restricted our sample among the small and medium firms. Size categorization was made as per the definition of MSME by Ministry of MSME for manufacturing sector. The list of SMEs in leather technology, textile and chemical industries were collected from authorized SME associations and export promotion councils (CAPEXIL, CHEMEXCIL, ILPA & CLE). The reason for focusing on leather, textile and chemical industries is mainly the higher concentration of their clusters in the areas of our geographical proximity and reasonable contribution to these SMEs in environmental degradation. Around 250 of the firms could be contacted over telephone requesting an appointment. Finally managers from 225 SMEs could be successfully interviewed. Interviews were done in person. Around ten months were required to complete visiting 225 SMEs. Three main clusters in the outskirt of the city of Kolkata were visited apart from visiting some SME offices scattered in the central part of the city. Before meeting the SME manager a fair idea of their operation and other relevant information were collected from their websites. This prior knowledge about the company helped us to reduce the interview time and also aroused interest of the respondents.
The data items required for this study were adapted from well recognized sustainability assessment tools and frameworks available in the market, which are predominantly being used for sustainability audits of big corporate houses (CSA Guide – RobecoSAM, 2016; Graubner et al, 2016; ARCHIBUS Environmental Sustainability Assessment, 2016; Veleva et al, 2003; Togo & Lotz, 2009). Before starting the interview of the SME managers, the questionnaire was validated for their relevance to SME operations by two senior and experienced managers of SME associations. Due to the lack of sufficient secondary data, use of self-diagnostic and self-assessing questionnaire is a common practice in studies of SMEs (Nejati et al, 2014). Besides, given the nature of small business, owners or managers make all the critical management decisions of SME firms. Hence managers of SMEs are assumed to have the most comprehensive knowledge about the firm practices and strategies. Thus in all cases either the owner or the manager next to him was interviewed.
Measurement model
Before the main data collection activity a pilot survey was done with 40 SMEs to fine tune our questionnaire. With the responses collected from initial 40 firms a reliability test and exploratory factor analysis using Principal Component Analysis with Varimax rotation was done. Factor loadings for all the constructs above 0.7 and Cronbach’s alpha for the finalized constructs ranging from 0.87 to 0.99, confirm internal reliability and convergent validity of the constructs and indicates adequate contribution of items to the overall scale (Table I).