Research Approach, Design and Process

Silverman (2001) suggested that researchers should provide their readers with the story of their scientific discoveries. Also, Strauss and Corbin (1998) point out that the process of scientific discovery is crucial in evaluating the study in general. Hence, the following paragraphs give a processual account of data collection, data analysis and theorizing. As this description will show, theory and data collection went hand-in-hand and as the ideas about the theory emerged I collected new data in a flexible manner to support or reject certain points that I thought are emerging as a theory.

Figure 3-1 Overview of Data Collection and Levels of Analysis

The study uses a multi-level and longitudinal approach by comparing the institutional logics of actors, legislative changes, and additional quantitative data between the time periods of 1995 to 2006 (as depicted in Figure 3-1). Because in some cases, key events took place outside the time frame (for example, a key legislative change may have

Macro Level Industry / Sector

Meso Level Organizations as Actors

Micro Level Top Managers (Elite) representing their

Organizations

Development of Sector’s Structures, Incomes,

Investments 1990 - 2005

2003 2004 2005 2006

Key Laws 1980 - 2005

Content Analysis of Municipal Journal 1998-2005

Water Sector Journal 1995 – 2005

Business Newspaper 1995 - 2005

Analysis of a Discussion on WSS Commercialization in a Business Newspaper Analysis of the first 15

interviewees’ social networks to identify key persons and to interview them

Exploratory Interviews with 8 people of the Managerial Elite

Interviews with 8 more people of

the Managerial

Elite

Interviews with 26 more people the Managerial Elite and Experts related to the Field (including people having weak ties to the elite)

Case Studies emerged based on the location of the managerial elite and their connections to other persons and organizations. Four large cities and three mid-sized municipalities (Helsinki, Tampere, Turku, Lahti, Hämeenlinna, Kouvola, Lohja)

47

occurred in 1994), there are exceptions where data has been included also from shortly before or shortly after the chosen time period, as long as it was necessary for explaining the phenomenon at hand. I use Glaser and Strauss’ (1967; Glaser, 1978; Glaser, 1998) grounded theory not only as a method but also as an approach to scientific discovery.

Grounded theory was guiding me from the start of the discovery process instead of merely being utilized for analyzing data that had been collected. As a result of this approach, I had to acknowledge the emergent nature of the study and phenomena at hand as well as maintain an open mind regarding theoretical and methodological questions.

Flexibility and creativity was necessary in applying the appropriate methods that would extract and validate the emerging phenomena.

3.2.1 Starting Out with Social Networks

The data for this study have been collected in the following manner. I started by conducting interviews with the managing director of a large water utility; the chairman of its supervisory board; and the responsible executive manager in the cities’ administration.

In these interviews, I was guided by a list of topics relating to strategic management of water services which the interviewees could freely address, so that I could find out what the key issues in water sector management were from their perspective. In addition, I asked each of these people to draw their social networks (exemplified in Figure 3-2) regarding restructuring and managing water issues. Each map contained 15 to 30 contact persons and indicated the interviewee’s strength of ties to most of them (indicated in the figure below as numbers written on the tie, whereas 1 was for strong, 2 for medium, 3 for weak strength of tie).

Figure 3-2 Example of an Interviewee’s Social Network (anonymized)

48

I applied to these social networks the snowball technique and identified other relevant key people that I was going to ask for an interview. As a result of this process, a network of managerial elites emerged. By elite, I mean people “who occupy defined positions of authority, those at the head of, or, who could be said to be in strategic positions in private and public organizations of various sizes” (Pettigrew, 1992b:163). I collected the social networks from the first 15 interviewees, with 15 to 30 names in each social network and comprising people from a variety of backgrounds and organizations. According to Granovetter’s (1985) argument that ‘weak’ ties or seemingly peripheral people are also relevant for understanding social networks, I also interviewed less-connected people to compare the data with the views of the elite that emerged. Further, Pettigrew (1992b) points out about research on elites that researchers also need to include a wider variety of members of stakeholders than just the business or political institutions under focus.

As a result, these maps were also helpful in identifying the most relevant organizations that were involved in the commercialization process. The interviewees were also asked about the past and present most relevant actors and organizations for the water sector.

The network maps were drawn between 2003 and 2004, but they also included information about which actors would have been and would not have been on the map in 1995. The network maps are used in this study as the basis for comparison of organizational field structure between 1995 and 2005, but otherwise the content of these maps (i.e. people’s names) was intended as confidential and not further used for this study except for identifying potential interviewees.

I started the first interviews in autumn of 2003, and problems in strategic decision-making about water services and the commercialization emerged after five or six interviews as the dominant issues. Hence, the interviews with all further participants of the study were adapted to these issues that had emerged. In order to get appointments with all those that I identified according to the social network analysis as important to interview, I had to be patient, but at the end of 2004, I had completed 39 interviews and had reached the stage of saturation (at which interview content started to repeat itself) a while before that. The relatively long time it took to conduct these interviews resulted from the fact that it was necessary to interview specific key people belonging to an elite, who had to be approached with consideration, and gaining access to them had been difficult at times. However, it happened only twice that I was declined an interview, but in both cases, the person referred me to a colleague, whom I then interviewed. I can only speculate about the reasons for gaining access to almost all interviewees. On one hand, once I had access to one of the members of the elite, I asked them for permission to use their name when contacting their most important contacts, which certainly helped in getting the necessary appointments. On the other hand, there is a possibility that it is easier to get access to top-level managers in Finland than elsewhere, and there is also the possibility that access is somewhat field-specific.

49

3.2.2 Characteristics of Interviewees and Interview Data

The 40 interviews that I included in this study comprised 11 top level water utility managers (including 1 manager from the Finnish Water and Waste Water Works Association); 11 top level municipal managers (including the Association of Finnish Local Authorities, a public-owned municipal consulting firm, and 1 municipal council member); 5 top level ministry public servants (including the Finnish Environment Centre); 3 investment bankers and 1 financial consultant; 3 scientists and 3 water sector consultants; 3 top level private sector enterprise managers; and one labor union expert.

The data set comprises municipalities and cities from West, Central, East, and South Finland⁴.

The interviews were on average 1 hour 40 minutes long, conducted according to an aide memoir (semi-structured interview guide) and tape-recorded. Interviewees were assured that what they say will not be quoted in any publication by connecting it to their names.

However, I told interviewees that it may be possible for ‘those in the know’ to correctly guess who is behind a certain quote. All interviews were conducted with Finnish nationals but in English language. All interviews were transcribed, which resulted in 950 pages of text that were analyzed by using NVIVO software for analyzing qualitative data according to grounded theory.

After a first screening of the data by using the NVIVO software program, the main themes (e.g. managerialization, calculating, compromising) had already emerged but they were rather unstructured and it was difficult to interpret their meaning and how they are linked with each other. At this stage I decided to complement the interview data with alternative sources that would help clarifying the interview data. Hence, a longitudinal analysis of relevant legislative changes was conducted, and in addition, the main topics in three sector-related publications were tracked over the years 1995 to 2005 (and in some cases until their last issue in 2006).

3.2.3 Tracking Legislative Changes

In order to understand the wider governance framework of the water sector, I tracked the legislative changes made by the national and EU governance actors, from 1980 until the present, whereas the most relevant ones started in the 1990s. The time frame for legislative changes comprised a somewhat longer period than for other data sources because it was necessary in order to provide a full picture of the process and certain key decisions, which had happened already years or decades ago. The analysis of legislative changes was mainly based on secondary data (a number of publications and summaries of a variety of authors, to which the empirical analysis refers in its text directly) rather than reviewing the different legislation directly.

4 A detailed list of interviewees is not provided in this thesis because it was agreed their names remain anonymous.

50

3.2.4 Content Analysis of the Journal ‘Kuntatekniikka’

In order to compare what I had gathered in qualitative material to an alternative source, I performed a longitudinal analysis of the technical sector’s focus by reviewing the articles in the magazine Kuntatekniikka, one of the municipal technical sector’s key print media, published by the Finnish Municipal Technical Association SKTY and Municipal Construction Association RTS.

From January 1998 through June 2006, 644 articles were published. The publishers’

English translations of these articles’ titles and short abstracts, which were already available in the journal’s web pages, were used for the content analysis. The articles published during the time from 1998 to 2005 (N=644) were categorized according to whether they contained mainly technical, managerial, or citizen aspects. It was possible for abstracts to contain more than one aspect at a time; for example, when the content was about citizen’s involvement in constructing a new housing project. The typical title of a technical article would be: “Constructing street pavements from natural stone” while a typical managerial title would be: “Contracting out technical services”. The disciplines included in the magazine comprise construction, townscaping, urban planning, streets and traffic, municipal management, environmental management, waste management, and water and sanitation. Although this content analysis was only based on the publisher’s English translations of abstracts and titles and thus, not on an analysis of the entire article content I regard titles and abstracts as representative for the entire article due to the long time frame and large amount of articles reviewed.

3.2.5 Commercialization Discussion in Kauppalehti Business News

I decided to track the discussion about the commercialization of water services in the business journal ‘Kauppalehti’ and analyzed it from 1995 to the first half of 2006. I chose Kauppalehti business paper because it has been following the developments of different actors in the Finnish water sector rather well, while other, similar publications such as the second largest business news ‘Talouselämä’ that I also screened have not been following water-related issues that much. I accessed the Internet archive of Kauppalehti News and searched for water sector related articles during the time period. The search resulted in 45 articles that were analyzed according to their content with the help of a certified translator for Finnish – English language.

Kauppalehti is the largest daily business newspaper in Finland, with approximately 80,000 copies distributed daily (five times a week) in 2005. Kauppalehti newspaper is part of Kauppalehti Group, a business and financial news division of Alma Media, which is Finland’s largest media corporation and focuses on publishing and distributing print media, financial information, online services, and television broadcasting.

51

Media and especially mass media, such as the Kauppalehti News (80,000 copies), is a powerful actor as it may affect public opinion. Therefore, it is different from professional journals that usually reach only a small circle of experts and professionals working in a sector (e.g. 5,000 copies). Thus, Kauppalehti News is not only a daily business newspaper, but because of its potential to decide what to cover and to affect public opinion, it could also be considered an actor in its own right. However, while it is important to acknowledge the power of such mass media, the focus of the analysis at hand is not to investigate the role of Kauppalehti News in the commercialization as such, but merely to use it as a forum where the discussion about the Finnish water sector has been documented.

52