Thematic analysis was first identified and described by Boyatzis in 1998. It is a qualitative approach to identifying, analysing and reporting implicit and explicit themes within data. Braun and Clarke, enthusiastic proponents of thematic analysis, wrote of it as being ‘a poorly demarcated and rarely acknowledged yet widely used qualitative analytic method’ which should be seen as a foundational method for qualitative analysis. It is essentially independent of theory and epistemology and therefore compatible with both essentialist and constructionist paradigms (2006: 77). It identifies and analyses themes within data as well as serving to organise and describe the data set in rich detail. It can also interpret various aspects of the research topic (Boyatzis 1998). It differs from other analytic methods that seek to describe patterns across qualitative data such as grounded theory in that it is not wedded to any pre-existing theoretical framework and therefore can be used within different theoretical frameworks and do different things within them. Braun and Clarke argue that thematic analysis can be an essentialist or realist method, reporting experiences, meanings and the reality of participants or it can be a constructionist method examining the ways in which events, realities, meanings, experiences and so on are the effects of a range of discourses operating within society (2006). Boyatzis also considered it particularly appropriate to a mixed methodology study which, as outlined above is the approach I considered best for my research.
Themes or patterns in data can be identified in one of two primary ways in thematic analysis; in an inductive or ‘bottom up’ way or in a theoretical or deductive or ‘top down’ way. An inductive approach means the themes identified are strongly linked to the data themselves (as such this form of thematic analysis bears some similarity to grounded theory). The theoretical approach is driven by the researcher’s theoretical interest in the area or topic which will mean the themes are more likely identified by a pre-defined coding frame. Coding for a specific research question maps onto the more theoretical approach while the emergence of the research question through the coding process maps onto the inductive approach (Braun and Clarke 2006). Boyatzis (1998) added a third way which is code development on the basis of prior research which he places between inductive and theoretical and sees it as combining elements of both. This is the specific approach which I have chosen to take as it allows for an exploratory approach to the research while also requiring engagement with the literature prior to analysis.
Another distinguishing point for me in choosing thematic analysis relates to the process of code development and thematic identification. Discussion of themes ‘emerging’ or being ‘discovered’ is very much a passive account of the process of analysis, and denies the active role of the researcher in identifying patterns or themes, selecting which are of interest and reporting on them. Braun and Clarke observe that “it implies that themes reside in the data and, if we just look hard enough, they will emerge; if themes reside anywhere, they reside in our heads” (2006: 92). Given that qualitative research is inherently interpretive, I consider that the biases, values, and judgments of the researcher, both informed and instinctive, play an important part in the research process and specifically, in the case of this research, in the identification of themes. In my case those biases, values and judgments were formed and informed by previous research experience in the industrial relations/migration field, by my engagement with the literature and by my active participation in ETUC migration working groups. All of these contributed to my identification of codes and subsequent themes within the research.
There are six phases of thematic analysis:
2.3.1. Data Immersion
In this phase I analysed my data in an active way in searching for meanings and patterns. This initially involved reading and re-reading the material until I was comfortable and a sense of the codes was starting to emerge. Transcription of the data is imperative to the dependability of analysis and I support the view that self-transcription from the original recorded material is a core element of the process. Kelle et al considered it should be a “key phase of data analysis within interpretative qualitative methodology” (199: 227). The close attention needed to transcribe data facilitates the close reading and interpretative skills needed to analyse the data (Braun and Clarke 2006).
2.3.2. Generating initial codes
Codes identify a feature of the data that appears interesting to the analyst and refer to the most basic element of the raw data that can be accessed in a meaningful way. A good code is one that captures the qualitative richness of the phenomenon (Boyatzis 1998). Miles and Huberman see the process of coding as a part of analysis as it involves organising the data into meaningful groups (1994).
In my case initial codes were developed using a combination of an inductive approach to the raw data but one informed by prior research on: (1) trade union responses to migrant labour and (2) new organisational approaches within trade unions. The research considered covered trade union responses and approaches across a range of countries, sectors and models7 Coding is a means of reduction and simplification of data. Initially I coded for as many codes
as possible and I also cross coded individual extracts of data. Thus some extracts were coded a number of times within different themes. I used the qualitative data analysis computer package, NVivo, to aid in the coding of the data.
2.3.3. Identifying themes
This phase involved the analysis and ordering of the initial codes to identify potential higher order themes. Themes differ from codes in that themes describe an outcome of coding for analytic reflection. A theme captures something important about the data in relation to the research question and represents some level of patterned response or meaning within the data set (Braun and Clarke 2006). Here I set out to provide a rich thematic description of my entire dataset from which I extracted what I identified as the dominant themes.
2.3.4. Reviewing themes
In this phase I searched for data to support or refute hypotheses emerging from the themes. I did this by reviewing and refining themes. This is where it became obvious when themes weren’t working or where there was significant overlap between themes or mismatches between the data and analytic claims. Where these issues occurred, data needed to be reorganised in order to create cohesive, mutually exclusive themes.
2.3.5. Defining themes
This is the narrative phase. At this point I had a final refinement of my overall thematic map and had identified the essence of each theme. Part of this process involved identifying sub-themes, the use of which helped to give structure to larger and more complex themes, and also served to demonstrate the hierarchy of meaning within the data.
2.3.6. Writing up
This is the writing up of the empirical material uncovered through thematic analysis. At this stage I had decided on what themes made meaningful contributions to answering the research question. I then organised the presentation of my data around the themes to best convey the research findings in a manner that convinces of the validity and merit of the analysis. The data is presented in Chapters Four, Five, Six and Seven with data extracts embedded within the analytic narrative to demonstrate the prevalence of the themes, and ultimately, to make an argument in relation to the research question.
2.4. Sampling Frame
In the first instance a limited survey of a non-random sample was carried out which was administered within the delimited population of ICTU member trade unions. While one purpose of the survey was to elicit quantitative information from the unions on migrant worker membership, services, policies and organisational initiatives, it was administered at a relatively early stage in the research process in order to assist with purposeful sampling for interview purposes. It served to establish which unions had significant migrant worker membership, their areas of employment, the significance of the issue within the union and the processes and procedures in place.
The Irish Congress of Trade Unions (ICTU) was selected for inclusion in the sampling frame on the basis of its being the peak union confederation in Ireland and with involvement on migrant labour issues, both at national and international level. Then following analysis of the survey responses and initial interviews with key informants, the individual unions selected for further analysis were Services, Industrial, Professional and Technical Union (SIPTU), the Union of Retail, Bar and Administrative Workers (Mandate), the Irish Nurses and Midwives Organisation (INMO), the Builders, and Allied Trade Union (BATU), the Communications Workers Union (CWU) and the Irish section of UNITE, the Union. They were selected on the basis that they best met the requirements of the research as they all organised in sectors where there was a significant migrant worker presence but they also allowed for variation on the independent variable in that they are different types of unions, representing different sectors and there were varying levels of engagement and development of policy and practice.
There is only one peak union confederation in Ireland, that being ICTU. It has 55 affiliated unions and represents over 850,000 workers across the island of Ireland, North and South. The policy of Congress is determined at its biennial conferences attended by about 700 delegates from affiliated unions and district trades councils. It has a 35 member executive council, the members of which are elected at the biennial conference. This includes the election of the President, two Vice Presidents, the Treasurer and 31 ordinary members. The Executive Council oversees the functioning of Congress between conferences. It has a staff of 30 across the island of Ireland, 21 in the Dublin office and 9 in Belfast. It is led by the General Secretary, assisted by two Assistant General Secretaries, who manage the affairs of the organisation and report to monthly meetings of the Executive Council. It has 16 standing committees including the Executive Council and the Solidarity Committee which is the one that deals with issues in relation to migration. The Officers of Congress are ex officio members of Congress Committees (ICTU 2011).
The Services, Industrial, Professional and Technical Union (SIPTU) is a large general union which represents over 200,000 workers from virtually every category of employment, both private and public, across almost every sector of the Irish economy. It represents almost 34% of ICTU membership in the Republic of Ireland. The policies of the Union are decided by the National Delegate Conference which is held biennially and attended by elected representatives from the Union's Divisions and Sectors. The National Executive Committee (NEC), consisting of 29 members, is elected at the Conference. The NEC oversees the work of the union between conferences. National Executive Officers - the General President, the Vice-President and the General Secretary – are also elected at the National Delegate Conference and serve a six year term. The National Executive Officers manage the affairs of the organisation and report to the National Executive Committee. On matters of critical importance a Special National Delegate Conference may be called by the National Executive Council to determine the issue in question.
SIPTU describes itself as an organising union, with a strong emphasis on recruitment of new members. In order to undertake this role it underwent a major reorganisation process which was completed in 2010. This saw the Union move to organisation on an industrial sector rather than regional basis. It now operates on the basis of six Divisions which organise workers in specific areas of the economy, allowing for greater coordination of activities and the pooling of information across industries and services. Within each Division activities are further divided into Sectors dealing with Union operations in similar employments and associated services. SIPTU has a staff of 324 including sector organisers, industrial organisers and administrative staff (SIPTU 2013a; 2013b)
Mandate Trade Union is an industry union, representing retail, bar and administrative workers. It has 40,000 members organised in 62 individual branches across 11 divisions. Policies of the union are decided at the Biennial Delegate Conference “which shall determine the union’s policy on all matters affecting or touching upon the interests of the Union or its members” (Mandate 2012: 7). The biennial conference elects the 22 person National Executive Committee (NEC) of the union. The President, Vice-President and Treasurer (the ‘Officers’) are elected at the conference from the already elected NEC membership. A special delegate conference may be called by the NEC on its own initiative or on receipt of a request in writing from 3,000 members of the union. The NEC overseas the work of the union and appoints the General Secretary, who in turn, manages the organisation (Mandate 2013).
Mandate is now an organising union and describes its principal object as “the organisation of workers for the purpose of advancing their social and economic advancement” (Mandate 2012: 5). Mandate has a staff of 54 including Organisers, Officials and administrative staff.
The Irish Nurses and Midwives Organisation (INMO) is a professional trade union for nurses and midwives. It has a membership of 40,000 the majority of whom are employed in public hospitals with smaller numbers employed in private hospitals and the nursing home sector. It is organised in forty-two branches throughout the country, based either on a single employment location or a geographical area, depending on numbers. These branches are in turn organised into 14 sections, including an International Nurses Section. The supreme authority of the union is the Annual Delegate Conference and / or Special Delegate Conferences as may be called from time to time at the request of one quarter of the membership or at the discretion of the Executive Council. All policy decisions are taken at these. The general control and direction of policy is vested in the Executive Council which is elected at the Annual Delegate Conference and holds office for a period of two years. The Executive Council consists of twenty-two members, including the President, Vice-President and Second Vice-President. The Officers are elected by all the voting delegates but, in order to be eligible, the candidates must already be serving Executive Council Members. The General Secretary is appointed by the Executive Council and is the Chief Executive Officer with overall responsibility for the management of the business of the organisation. The INMO has a staff of 100 including in industrial relations, education and administration (INMO 2013; 2015).
‘UNITE the Union’ was formed in May 2007 following a merger between the T&GWU and AMICUS and is currently the largest union in the UK and Ireland with 1.8 million members. It represents members in a wide range of sectors including Transport, Public Services, Manufacturing, Finance, Clerical, IT, Agriculture, Construction, Power & Engineering, Aviation, Food, Drink & Tobacco and Health. It is one of the first unions in either the UK or Ireland to have moved towards organising, having begun such moves after amalgamation in 2007.
Unite’s supreme policy making body is the Policy Conference which is held bi-annually. All constitutional conferences and committees of the union are obliged to have a gender and ethnic balance of elected representatives, at least reflecting the constituency which they represent. Unite operates on the basis of Industrial Sectors, which include both occupational and professional sectors. In Ireland, it operates on an island-wide basis, with Ireland constituting one Region as does Scotland and Wales while England is divided into seven regions. There is an Irish Executive Committee which makes decisions on matters of an industrial or political nature relating to the Republic of Ireland or Northern Ireland “which do not affect members of the Union not so resident” (Unite 2013: 57). While it is one of the largest unions operating in Northern Ireland, Unite’s presence in the Republic of Ireland is relatively small and much of its organising is concentrated in the border regions.
The Communications Workers Union (CWU) represents workers in the postal, telecommunications and call centre sectors. Previously a primarily public service union, communications deregulation has meant that its membership is now largely drawn from the private sector, from companies such as Eircom, An Post, Vodafone, Meteor, O2 and others. it currently has a membership of approximately 18,000 spread over 146 branches.
The primary policy making body of the CWU is the Biennial Conference which is where policy decisions are taken and the National Executive Council (NEC) is elected. The NEC, which consists of 32 members, is charged with overseeing the implementation of conference decisions and policy. The President and Vice President, who are part of the 32 person complement, are also elected at the conference but on the last day, having already been elected as ordinary members. The appointment of the General Secretary is made by the NEC The General Secretary is responsible for managing the business of the union with the support of sub-groups of the NEC as appropriate. Special conferences can be held either at the instigation of the NEC or by decision of the biennial conference. The CWU has a total staff of twenty, including full-time officials and administrative staff (CWU 2013a; 2013b)
The Building and Allied Trades Union (BATU) is a small craft union with a membership of approximately 2,000. It has a total of four staff. The supreme governing and policy making body of the union is its Annual Delegate Conference, This conference elects the union's National Executive Council which is made up of 10 "ordinary members" plus a General President and a Vice President who are also elected by the conference (2013b; 8). The General Secretary (who also acts as Treasurer), a Deputy General Secretary and "such number of Assistant General Secretaries as the Annual Delegate Conference may from time to time determine" are "appointed by the Annual Delegate Conference" (2013b: 11). A Finance and General Purposes Committee conducts the business of the union between meetings of the NEC and deals with the union's financial affairs. This Committee is elected by and from the NEC.
BATU is currently engaged in discussions with SIPTU with a view to a merger of the two unions (2013a).