«PROACTIVE ENVIRONMENTAL STRATEGIES IN SMALL BUSINESSES: RESOURCES, INSTITUTIONS AND DYNAMIC CAPABILITIES Jan Lepoutre Promotor: Prof. Dr. Aimé Heene ...»
VMS membership presents an exceptional opportunity to objectively capture a firm’s environmental intentions. In addition, the particular circumstances of VMS membership indicate that the institutional conditions are set against achieving high VMS scores. Although one could argue that firms become member of VMS only as a means to improve the firm’s reputation or to get access to environmentally sensitive customers, the specific context of VMS in the ornamental horticulture makes such claims improbable. In fact, only 6% of the entire sector population has become member, and membership has been declining in recent years (MPS, 2006). This low penetration of VMS membership is generally attributed to four important factors: (1) VMS membership does not result in any added value in the market, (2) the registration activities which are required for VMS membership are perceived to take up too much time, (3) environmentally friendly practices are seen as too costly and risky in the highly competitive environment of the ornamental horticulture sector, and (4) the general individual and professional association dissatisfaction with environmental regulations endorse a conservative stance towards the natural environment. Given these adverse conditions and the particular objectives of VMS, the members of VMS clearly send out a signal that they have proactive environmental intentions.
5.2.2. Data collection and analysis Proponents of case study methodologies recommend researchers to “triangulate” multiple sources of data with the purpose of increasing the richness and quality of the findings (Eisenhardt, 1989a; Strauss & Corbin, 1998; Yin, 2003):
“any finding or conclusion in a case study is likely to be much more convincing and accurate if it is based on several different sources of information, following a corroboratory mode” (Yin, 2003).
In this dissertation, I therefore drew on five types of data. Data were collected through key informant interviews, grower interviews, observations, archival data and round table discussions. Importantly, not all data sources were used for the same level of analysis.
Following the logic of an “embedded case study” (Yin, 2003), in which multiple embedded
levels are analyzed, I first set out to assess the general context of the ornamental horticulture industry in a pre-stage, before engaging in the actual case study research, which consisted of four stages in total. Table 5.1 provides an overview of these five data sources and the particular objective for which they were used. In the following subsections, I will provide an overview of each of these data sources.
18.104.22.168. Context assessment In the pre-stage, I explored the challenges and opportunities associated with PES in the ornamental horticulture industry. Capturing this information required drawing on key informant interviews, observations, archival data and roundtable discussions. Next, I will describe each of the data sources used in this process.
Key Informant Interviews. The goal of the interviews was to get an overview of the challenges and opportunities that the Belgian ornamental horticulture was facing, and to get a preliminary sense of the ways how growers could deal with them. Starting with a government consultant, responsible for extension services in the ornamental horticulture industry, and the director of VMS, I used a snowball sampling technique (Denzin & Lincoln, 2000) to get in touch with new key-informants. I asked each new informant to identify additional informants that would be valuable to speak to and that would give additional or contrasting insights. I continued interviews until the point of saturation, the moment at which no new information emerged from the interviews (Miles & Huberman, 1994; Denzin & Lincoln, 2000). As a result, a total of 15 key informants were interviewed between October 2005 and January
2006. Table 5.2 offers an overview of all key informants that were interviewed in this process.
An overview of the questions that were used to aid in the interview process is provided in appendix A1. After each interview, I made a contact summary form (Miles & Huberman, 1994), in which I listed all new and recurring findings that had emerged. I also made notes during the interview about impressions and details that would not be recorded in the interview. All interviews, lasting between 60 and 150 minutes were recorded, transcribed and subsequently manually analyzed using the Atlas.ti software package.
Observations. In addition to the interviews, I had the opportunity to visit a number of firms, meetings and market places, where I could see several aspects of the ornamental horticulture in action. Although I did not take any notes during these visits, the observations helped me to get a better understanding of ornamental horticulture, its culture and the nature of its activities.
Archival data. Besides the interviews, I collected and consulted several archival data sources. Table 5.3 provides an overview of these sources. The sources can be subdivided into three groups: databases, non-academic and academic sources. For the databases, I did a search of Mediargus, a database that collects all general Belgian newspaper and magazines, and of www.vilt.be, the Flemish Information centre for Agriculture and Horticulture, using key
words that probed the databases for information on the ornamental horticulture, its subsectors and their relationship with environmental and institutional issues. Furthermore, I scanned all issues of “Verbondsnieuws”, the bi-weekly magazine of AVBS, the largest professional association in ornamental horticulture, that were published between 2003 and 2006, the period prior to the interviews. I scanned with similar key words as the other two databases, but specifically assessed the discourse developed towards members and towards societal and regulatory issues. The non-academic sources included a set of reports and statistics that were published as reports or documents from banks, professional associations, governments and related institutions. The academic sources consisted mostly of the studies executed at the former government research group, the Centre for Agricultural Economics and the present Institute for Agricultural and Fisheries Research. These studies are mostly based on data from the Farm Accountancy Data Network (FADN), a panel data set that is collected every year to supply information to the European Commission with regards to the economic impacts of its subsidies.
Roundtable discussions. A fourth and final data source for the sector analysis was derived from a series of roundtable discussions within a project called “Ornamental Horticultural Strategy 2020”. The aim of this project was to develop a number of scenarios for the future of the Belgian ornamental horticulture industry, in order to then derive the implications for government, professional associations and all interested alike. The group consisted of 17 persons, with selection based on a good representation of the diverse subsectors (pot plants, cut flowers, azalea and arboriculture) and of all parties involved in the industry (production, trade, professional associations, government and services). The efforts put in by the members culminated to a scenario planning weekend in February 2008, which took the most important lessons from these sessions and of EROV (2004) to develop future scenarios of the sector. Prior to the weekend, the participants were asked to submit the way they envisioned the ornamental horticulture sector in 2020 in its most and least optimal presentation, and what factors they considered key in the development towards these scenarios. In order to analyze the roundtable discussions and the scenario planning weekend, I used the reports of the sessions, the submitted scenarios and my own notes.
22.214.171.124. Case study research After the context assessment, I met again with the director of VMS and the government consultant. Since both were knowledgeable about VMS members, I engaged them in the processes of selecting appropriate firms for the research and gaining access to
individual firms. I thereby followed a structured approach of “theoretical sampling” (Eisenhardt, 1989a), with the particular goal to find the best possible sample that would capture the phenomenon of interest (for similar approaches in recent studies, see Uzzi, 1997;
Graebner & Eisenhardt, 2004; Zott & Huy, 2007). After screening all 127 VMS members based on short descriptions of both informants, I asked them to select 20 firms that (1) together made a good representative sample of the diversity of firms within the sector, (2) included both very high performing firms and very low performing firms as polar types in the continuum of VMS scores, and (3) included some firms that had recently left VMS, to have exit as an alternative emergent strategy as well. Table 5.4 provides an overview of the firms that were included in phase. We used pseudonyms to ensure confidentiality. The data collection and analysis for the case study research then consisted of four consecutive stages.
Chapter 5 Stage one involved semi-structured interviews, which consisted of a set of open-ended questions set out to understand the functioning of the firms and their approaches to social and environmental concerns. A list of the questions asked can be found in appendix A2. I conducted interviews carefully, and asked questions about “Why did you become member of VMS?”, “How are you able to achieve your score?”, and “What difficulties did you experience in obtaining your score?” In so doing, I was also able to identify firms for which VMS membership did not reflect the intention to have a high VMS score. In addition, detailed notes were taken during the interviews because, although the interview protocol was an important guide, each interview spawned emergent and interviewee-specific questions and topics (McCracken, 1988; Golden-Biddle & Locke, 1997; Strauss & Corbin, 1998).
Interviews were taken on site with owner-managers between December 2005 and March 2006 and typically lasted 120 minutes, with extremes ranging between 60 and 160 minutes. All interviews were audio recorded and transcribed verbatim.
In stage two, I manually coded the transcripts for constructs that emerged as explanatory factors for the particular performance of the firm (Glaser & Strauss, 1967; Strauss & Corbin, 1998). Furthermore, I collected additional archival material for each of the firms by looking for firm references (1) on the internet, (2) in the biweekly magazine of the largest professional association (Verbondsnieuws) and (3) in www.vilt.be. I used the material collected in this process to merge it with the interview data. After the coding of each interview, I combined all data sources and wrote a firm-specific case summary. This case summary acted as an ongoing stream of consciousness commentary about what was happening in the research (Miles & Huberman, 1994) and tracked whatever impressions arose, given that it is often difficult to know what will and will not be useful in the future.
Questions such as, “What am I learning?” and “How does this case differ from the last?” were asked to push thinking in these notes.