The specific methods applied in this mixed method study for data collection were a combination of documentary analysis, semi-structured interviews, comparative analysis, participant observation and a small scale survey
The research was divided into three stages as follows:
Formulation of research proposal
Initial interviews with gatekeepers and key informants
Participant observation (international)
Participant observation (international)
Interviews with union leaders and middle management
Data analysis 1
Interviews with union officials and organisers
Data analysis 2
Coding and theme development using NVivo
Extrapolation of final hypotheses
2.5.1. Documentary analysis
Documentary analysis refers to the analysis of documents that contain information about the phenomenon under study. Payne and Payne describe the documentary method as the techniques used to categorise, investigate, interpret and identify the limitations of physical sources, most commonly written documents whether in the private or public domain (2004: 60). In my approach to the documentary analysis element of my research I applied the quality control criteria for the handling of documentary sources, as formulated by Scott. The four criteria are: authenticity, credibility, representativeness, and meaning. Authenticity refers to whether the evidence is genuine and from impeccable sources; credibility refers to whether the evidence is free from error and distortion; representativeness refers to whether the documents consulted are representative of the totality of the relevant documents; and meaning refers to whether the evidence is clear and comprehensible (1990: 6).
In the case of this research the documentary analysis focused on both primary and secondary sources. I mainly drew on primary sources emanating from trade unions that are of relevance to the topic under research. These included ICTU and member trade union policy documents, submissions, press releases, annual reports, information materials, consultancy reports and published statements and speeches (e.g. ICTU 2005; 2006; 2007; Begg 2006; 2007; SIPTU 2005; 2006; 2007; 2011). They also included similar material from relevant NGOs such as the Migrant Rights Centre of Ireland (MRCI) and the Immigrant Council of Ireland (e.g. ICI 2003; MRCI 2004; 2006a; 2007). Documentary material from the Central Statistics Office (CSO), the Economic and Social Research Institute (ESRI), relevant Government Departments and statutory bodies was also studied (e.g. Barrett et al. 2005; Barrett and Duffy 2007; CSO 2005; 2006; 2007; DETE 2005a; 2005b; 2008). Added to this was material from the ETUC working groups such as reports, minutes and EU material provided to the working groups (e.g. ETUC 2009a, 2009b; 2011a, 2011b; European Commission 2001; 2009). The analysis also involved substantial engagement with the websites of all of the bodies studied. With regard to secondary sources, there was a detailed study of media coverage of three significant industrial disputes involving migrant workers (see Chapter Six) as well as more general engagement with media coverage of trade union and migrant labour issues. To all of this I added a comprehensive review of the academic literature.
Participant observation at ETUC level was a feature of the earlier phases of my research. I sat on the ETUC Migration and Inclusion Working Group, ‘Workplace Europe’, from 2007 to 2010 as an Irish informant, nominated by the ICTU. Arising out of the ‘Workplace Europe’ project, the ETUC established a further project, ‘What Price the Tomatoes’ in 2010 which focused specifically on trade union relationships with undocumented workers. I again sat on this project working group as an Irish informant. Both these project groups were made up of a combination of trade unionists, NGO representatives, academics and ETUC staff. In both cases I informed the group members of my dual position at the introductory stage of the projects. The purpose of my observation was not ethnographic, but rather was contextual with a view to learning about the experience of trade unions and migrant labour in other jurisdictions in terms of practices, organisational approaches, particular difficulties and models of good practice to inform my approach to and enrich the data from the national research.
The practical aspects of the observation process I found challenging, as noted by Creswell. He discusses the difficulties of role definition of the researcher as to whether to assume a full-participant role, non-participant role or a middle-ground position (2007: 139). I assumed somewhat of a middle-ground role insofar as I neither wished to, and nor was I in a position to, represent the ICTU so I acted more as an informant to the working groups when required and, in turn, kept ICTU informed of developments from the working groups. It was necessary for me to clarify my role on occasion if, as happened, I was called upon to speak on behalf of ICTU. The other difficulty I experienced, also referenced by Creswell, was around the issue of recording my observations, remembering to take field notes (this could be a problem when debate was heated or when we were operating in small sub-groups), recording quotes accurately and applying sufficient rigour to the writing up process to ensure accuracy. This involved writing up at the end of each day.
A limited survey of a non-random sample was carried out which was administered within the delimited population of ICTU member trade unions. The survey questions were designed according to my theoretical framework and were primarily closed questions with options to provide additional information if wished. The purpose of the survey was twofold. Firstly, as an initial step to address the dearth of information available on the research topic, it was to elicit quantitative information from the unions on the particular areas of interest such as migrant worker membership, recruitment processes, engagement and integration, services, specific migrant focused policies and organisational initiatives and the timescale of development. Secondly, it was designed and administered at a relatively early stage in the research process in order to assist with purposeful sampling for interview purposes.
Following the initial design of the questionnaire the questions were pre-tested on two unions to evaluate reliability and validity and the questionnaire was then modified accordingly. It was then circulated to the designated individuals responsible for equality issues in the affiliate unions of the ICTU. The survey was web based and was prepared and circulated through the web based survey development company, Survey Monkey. It was circulated by, and with the endorsement of the ICTU on my behalf, thus increasing the probability of response. The survey received a 60 per cent response rate. A full list of the survey questions is contained in Appendix B.
In a qualitative research interview, knowledge is produced socially in the interaction of interviewer and interviewee, not by following a pre-determined method or set of rules. Rather than locating the meanings and narratives to be known solely in the subjects or the researchers, “the process of knowing through conversation is intersubjective and social, involving interviewer and interviewee as co-constructors of knowledge” (Kvale and Brinkman 2009: 18).
This research used semi-structured interviews as the primary qualitative method within the thematic analysis methodology. Purposeful sampling was applied to the selection of cases of interest meaning that the theoretical purpose of the project, rather than a strict methodological mandate, determined the selection process (Creswell 2007; Marvasti 2004). Creswell describes purposeful sampling as where “…the researcher selects individuals and sites for study because they can purposefully inform an understanding of the research problem and central phenomenon in the study” (2007: 125). This was my primary approach to sampling but later in the process I deviated from it in that it was clear that a snowballing approach (i.e. direction from those already interviewed), was going to lead me to cases of interest which were more information-rich.
A total of 28 interviews were conducted with the interviewees drawn from the Irish Congress of Trade Unions (ICTU) and six of its affiliate unions as outlined in Section 2.4 above (see Appendix for full details of interviewees). A multi-level approach was adopted in the interview process involving (a) Senior Management (General Secretaries); (b) Middle Management (Senior and Regional Organisers); (c) Executives (Branch Officials/Organisers); (d) Local Activists. The interviewees were categorised in this way in order to gain a broad spectrum of perspectives and to militate against the reflection of institutional bias. Interviews with senior management were conducted in the first instance. This was with a view to gathering data on policy and management perspectives on how ICTU and respective individual unions were engaging with migrant workers. This was followed by interviews with middle management who, in most instances, are located at the intersection between policy development and articulation and local engagement. These were then followed by interviews with executives and activists to establish how those policies and senior management perspectives translated into trade union activity on the ground.
Key informants included former trade union officials and activists who had been centrally involved in migrant worker issues. The contribution of these interviewees to the research was particularly pertinent for a number of reasons. They were individuals who had been actively engaged with migrant worker issues from early on, they had the benefit of hindsight without the constraint of institutional attachment and so could bring a greater level of objectivity to their analysis. Another key informant was a senior academic, drawn from the School of Business, UCD who had been engaged on an ICTU commissioned research project from 2008 to 2011 which explored the strategic choices made by five trade unions in Ireland, their priorities and direction. The research provided a level of multi-union assessment of union organisation in Ireland at that point and was intended to lead to discussion and debate between the unions to, in turn, facilitate the development of new ideas, stronger connections between organizers and increased knowledge transfer. However this did not happen and it remains unpublished. The selection of representatives from the national NGO, Migrants Rights Centre Ireland, for interview was on the basis of their involvement with issues of migrant worker exploitation and their engagement with both the trade union movement and directly with the state’s Dispute Resolution Services. Finally interviews were conducted with representatives of the ETUC and the international NGO, PICUM. The primary purpose of these interviews was to illuminate the situation with regard to ETUC engagement with the issue of migrant labour, building on the data gathered through my role in the ETUC sub-committees as outlined above.
An in-depth semi-structured approach was taken to the interviews as being the most appropriate, involving as it does a “set of defined answers to defined questions, while leaving time for further development of those answers and including more open-ended questions” (Walliman 2001: 241). I designed the interview protocol in advance with broad topic headings, dictated by the theoretical framework and informed by the initial analysis of the survey responses. There was a three-stage approach to the interviews with open interviews of key informants and gatekeepers taking place at stage one of the process. These early interviews provided important information which helped not only to indicate the future research direction but also to identify and prioritise issues (Dunleavy 2003; Walliman 2001). They also determined the focus and scope of later interviews. The second series of interviews, which took place at stage two, involved semi-structured interviews with trade union leaders, including the General Secretary of the ICTU and with trade union middle managers across the selected unions. This phase also involved interviews with representatives of NGOs and the ETUC. The third phase of the interview process, which was very much born of the snowballing process, involved interviews with key union officials, organisers, activists and key informants. The scope of the interview process went beyond serving trade unionists to former trade union employees and to informed observers as I considered that limiting it to insiders only was likely to circumscribe the outcomes of the research and limit its possible application.
The interviews took place in three sessions, the first between April 2011 and June 2011 the second between March 2012 and April 2012 and the third between October 2012 and February 2013. All, but one, of the interviews were conducted in the workplace of the interviewees with one by telephone. All interviews were recorded and I also took notes concurrently to note items that I thought required particular emphasis or to record my own observations. The interviewees were self-transcribed, coded and analysed thematically.
2.5.5. Comparative analysis
Comparative research is the act of comparing two or more things with a view to discovering something about one or all of the things being compared. It has been central to the creation of an understanding of immigrants and their experiences in historical and social context. It has increasingly played a role in developing understanding of how and why trade unions respond to issues of migrant labour in different ways (Hardy et al. 2012; Fitzgerald and Hardy 2010; Krings 2007; 2009a; 2009b; Frege and Kelly 2004; Penninx and Roosblad 2000). Comparative analysis is also an integral part of any thematic analysis study.
This research has a strong comparative component throughout with the elements of the comparative analysis informed by the theoretical framework and, more specifically by the thematic analysis coding process. Thus the units of coding and the particular methods applied have emerged from the data. In Chapter Four I have applied the comparative analytical framework as described earlier to a broad analysis of the responses of trade union movements in nine Western European countries to migrant labour. The primary purpose of this comparison is to establish convergence and divergence of responses, with a view to positioning the Irish trade union movement response within that.
I also apply a diachronic comparative analysis to the Irish empirical material. Diachronic analysis concerns itself with evolution and change over time of that which is studied. It is particularly useful in this case in that I have included in this research, in Chapter Seven, three case studies of industrial disputes which were considered tipping points8 in the relationship between Irish trade unions and migrant labour and which exemplify the development of the trade union response over time - “single cases can be vivid and illuminating, especially if they are chosen to be revelatory” (Miles and Huberman 1994: 26). The case studies are of a local level Mayo mushroom farm dispute in 2003 which was led by an NGO; the GAMA construction company dispute of early 2005 which was brought to public attention by a politician; and the Irish Ferries dispute where the trade union movement was seen to play a central role.