Introducing Our Data

Abstract

The data was gathered with the aim to generate empirical knowledge about gendered European party politics and the ways in which these affect the prospects for democracy and gender equality in Europe. It also aimed to innovate the study of formal and informal institutions and discourses in party politics. This chapter presents the data and highlights the methodological innovations they generated in the understanding of genderedness of the European Parliament. The chapter outlines how data collection occurred in two phases: a pilot study and the main data-gathering period. Overall, 140 interviews, 193 pages of fieldnotes and a wealth of internal documents from the European Parliament and its political groups were gathered during these two phases. Tables illustrate the distribution by gender and nationality, as well as the political group affiliation and role of the research participants in the European Parliament to stress the variety of the data. The chapter explains and demonstrates recording practices and techniques for parliamentary ethnography, which enabled a detailed understanding of informal political group dynamics. Finally, the chapter describes how the research data was processed, archived and categorised, drawing on lessons from the pilot study.

Introduction

Prior to a critical analysis of our findings, this chapter presents the data on which the remainder of this volume is based. We first outline the two phases of data collection, followed by a comprehensive overview of the data we gathered in its entirety. The final section is attentive to the logistics and technicalities of storing and archiving our data as well as lessons learned in these processes.

Research outcomes are dependent upon the way research questions are framed, the consideration of pre-existing textual data sources, time and practical skills (cf. Leavy, 2020), as well as more pragmatic issues such as access to participants. From the outset, our project was designed with two main stages of research in mind: (i) generating new empirical knowledge about the gendered character of European party politics and its impact on gender equality and democracy in Europe and (ii) methodological innovation in the study of formal and informal institutions and discourses in party politics.

Data collection commenced immediately when the project began. It consisted of two extensive qualitative fieldwork phases (see the following section for details): the pilot study (2018–2019) and the main data collection (2019–2022). The processes of data gathering and analysis were inextricably linked and developed over time from the pilot study to the main data collection, with the active participation and input by all team members. The project introduced, and implemented, methodological innovations that allowed us to gain a thorough empirical understanding of the various ways that gender manifests itself in the EP political groups. This was enabled by our data collection, the creation of a unique dataset and the systematic development of analytical schemes for filing and coding in an iterative team process, which we discuss in greater detail in the following chapters.

Two Phases of Data Collection: The Pilot and the Main Study

The initial data collection involved an extensive qualitative pilot study with expert and elite interviews from October 2018 to May 2019. This corresponded to the 8th parliamentary term in the EP (2014–2019). In total, we interviewed 54 MEPs and staff in the EP, covering all eight political groups of that parliamentary term and reflected gender parity amongst the interviewees. All our interviews were recorded with the permission of the interviewees. If the request to record was declined by the interviewee, then the interviewer took notes. The interview data was fully confidential, and interviewees were guaranteed full anonymity. Ethnographic fieldwork in the pilot study period allowed the team to gain a thicker volume of data, but with a finer-grained understanding of (in)formal political grouping dynamics. Overall, there were 104 pages of fieldnotes and 79 formal hours of shadowing and observations. In 2019, the team began the in-depth analysis of the interview data by developing codes for the different research questions (a more detailed discussion can be found in Chapters 4–6).

Following the pilot study, we held team meetings to reflect on our data sample. Overall, good lessons were learnt with a pilot study. We formulated the pilot period to avoid putting pressure on getting a high total sample, but rather to elicit necessary information about the process of data gathering in the EP. The pilot study provided useful material for analysis, and throughout the process, we learned, for instance, about the importance of the administrative leadership role played by the Secretary Generals in the political groups. It also gave room for reflection on the research field, allowing us to ‘learn the ropes’ of EP functioning. In this respect, we came to appreciate the value of responding to interviewees, asking follow-up questions and requesting follow-up meetings. We also identified ‘black boxes’ that stood out in the extant scholarship, for example, the workings of the Bureau and the Conference of Presidents. In light of the pilot study, we identified some new targets to pursue, such as shadow rapporteur meetings that take place during committee meetings, or the possibility of contacting group coordinators if we needed to request access. We also realised the value of being able to talk to the secretariat staff of committees who had a lot to say about political groups.

The pilot study allowed us to better define whom we wanted to target, and how we prioritised and scheduled our work. We then reached a general agreement about how to manage what would be a hectic schedule to complete the data gathering as soon as possible, given the time and geography constraints both for us and for our interviewees. On reflection, these pressures resulted in data that was less systematic than we had planned for. In addition, we had to resign ourselves to the fact that at times the interviews were more ad-hoc than we would have liked. Nevertheless, all the interviewees appreciated our flexibility in adjusting schedules to fit the interviewees’ changing schedules and we always tried to ensure that another team member was available to ‘take over’ if necessary.

The pilot study was also invaluable to the important process of reviewing our interview questionnaires. Initially, the length of the interview guide proved an issue. On occasion, the interviewees’ time constraints made it difficult to stick to, requiring us to be reflexive in the prioritisation of tasks and to improvise when necessary. However, at times it proved difficult to adjust the schedule when interruptions occurred during the interview. Thus, the main takeaway was to make sure we inserted enough flexibility into the schedule for the future—including multiple follow-up questions and variants depending on the interviewee, and questions that could be ‘skipped’ if necessary. The result was that we divided the interview schedule into ‘core’ questions and sub-questions that we could potentially miss.

In developing the process for the main data collection, we established the number of interviews we wanted to conduct. In team meetings, we decided to aim for ten interviews from each of the seven political groups in the 9th EP term, making 70 interviews altogether, each to include six MEPs, two accredited assistants and two political group staff members—whilst at the same time ensuring a gender balance. The main data collection was conducted between December 2019 and March 2020 (in person). Data gathering continued after the onset of the Covid-19 restrictions through online and telephone interviews between March 2020 and March 2022. In response to the changing circumstances, the team jointly developed questionnaires for MEPs and political group staff and also updated the interview schedules to include questions on current developments (e.g. ongoing Covid-19 restrictions and their implications for parliamentary work). In total, we interviewed 79 MEPs and staff in the main data collection as well as seven people from the EP Secretariat (two were in one interview); reaching near gender parity amongst the interviewees, and covering all seven political groups as well as non-attached MEPs (see Fig. 3.1).

Fig. 3.1

Gender division of interviewees by political groups or secretariat

Overview of Data

Across the two phases of data gathering, several types of data were collected: interview data, ethnographic data and documentary material. In total, the team interviewed 140 MEPs and staff; reaching gender parity amongst the interviewees and covering all eight political groups and the EP secretariat. At the same time, we conducted a parliamentary ethnography which included shadowing, participant observation, and fieldwork diaries. This was augmented with documentary and supplementary materials that the entire team collected for analysis (for more details see below sections). Quality and selection of the data often determine the qualitative outputs of research (cf. Gilgun, 2020). Since qualitative researchers seek to understand the subjective experiences of research participants in their contexts, high-quality data result in large part from the degree that researchers practise immersion and the degree that both researchers and informants develop rapport and engage with each other (as we discuss in further chapters) (Gilgun, 2020). Hence, we aimed to collect and often triangulate different types of data that would permit us to explore various facets of the studied political institution. This was particularly important since we aimed to shed light on the practices and informality with unequal power dynamics at play in the EP.

We did not achieve our planned numbers exactly (see Fig. 3.2), especially in terms of the division between MEPs and staff, but the material we collected was more than sufficient for analytical integrity (see Chapters 5 and 6).

Fig. 3.2

Position of interviewees in the European Parliament

When divided by nationality, the interviewees revealed a correlation between the highest numbers of member state representatives in the EP and the EUGenDem team members’ nationalities and spoken languages (see Fig. 3.3). The language skills of the researchers also had a clear impact on the numbers of interviewees who spoke English. The impact of language and nationality on data gathering will be considered in greater detail in Chapter 4, where our strategies for gaining access and obtaining interviews are more broadly elaborated. Suffice it to say, for smaller national delegations, it was more difficult to obtain interviews as there were fewer people to respond and those participants had greater demands on their time.

Fig. 3.3

Nationality of interviewees

One of the main events to occur during the fieldwork in Brussels was the Brexit negotiations ultimately leading to the UK leaving the EU. These processes therefore became prominent in our research and led to several publications that were attentive to the various impacts of Brexit on the EP (see for instance, Gaweda et al., 2022; Kantola & Miller, 2022). Brexit had a clear impact on the numbers of most political groups in the EP (see Table 3.1.), and as a very salient issue, it also had a notable bearing on the numbers of UK nationals we interviewed.

Table 3.1 Seats lost by each political group after Brexit in 2020

Interview Data

Taken together, during the pilot and main data collection periods, we conducted 140 interviews with near gender parity and maintained a balance between MEPs and staff (see Table 3.2). The interviews covered the democratic practices of political groups, their leadership, MEP/staff lives, behaviour and conduct, and policy-making processes. The interviewees signed consent forms concerning data protection. Signing of forms became trickier at the beginning and during the restrictions related to the Covid-19 pandemic. In response, we accepted phone photographs of signed forms from our participants.

Table 3.2 Overview of the interviews

Ten per cent of interview participants were from racialised minorities: six MEPs (two women, four men); five group staff (two women, three men) and one parliamentary staff member (woman). Two MEPs from racialised minorities were shadowed—one woman and one man. We consciously avoid giving further information, for example, naming the political groups, so as not to compromise the anonymity of the participants, especially those from smaller political groups where there are fewer racialised minority MEPs and staff.

Post-Interview Notes

One of the additions we made following the pilot study was to take structured notes of our ‘fresh’ impressions immediately after the interviews. The post-interview note focused on three facets of the interview: the socio-spatial–temporal aspects; the interpersonal and affective relations; and any practical implications arising from the interviews. Each section included prompt questions that helped us reflect (see Fig. 3.4). We designed the post-interview notes as types of research diary entries that allowed us to record impressions, feelings and immediate reactions after the interview. They were also often a way of ‘dealing’ with more difficult situations or problematic statements from our participants, since the notes were accessible to all team members for reference. The post-interview notes had also the practical application of recording potential follow-ups and snowballing options that could otherwise be forgotten or missed in bilateral communications between team members.

Fig. 3.4

Post-interview note template

Due to time constraints and the fact that interviews were often planned back-to-back because of our participants’ schedules, we managed to write up 36 post-interview notes for 79 interviews conducted in the main data collection period (a completion rate of about 45%).

Ethnographic Data

Parliamentary ethnography was our main methodological innovation. It allowed us to gain a finer-grained understanding of (in)formal political group dynamics that otherwise would have been hard to reveal. In total, we shadowed nine MEPs and had access to ten political group meetings. During the main data collection period, our team member Dr. Cherry Miller secured a two-month placement at the European Parliamentary Research Service, which allowed by-appointment targeted observations of political groups, as well as other activities in the EP. Overall, this amounted to 55 days (or 440 hours) in the field. In terms of recording the data, our progressive focus on the 9th Parliament consisted of pioneering a five-concept observation protocol alongside a fieldwork diary (see Fig. 3.5). In total, we uploaded 193 pages of fieldnotes to ATLAS.ti (Computer Assisted Qualitative Data Analysis Software) and coded as a team (for details, see Chapter 5). Due to Covid-19, some of the observational activity in the latter part of the fieldwork was cancelled due to restrictions on meetings for both staff and visitors in the parliament. Covid-19 restrictions were disappointing, but the wealth and breadth of the already collected data allowed for the deeper analysis and wider research into overarching topics and themes that we wanted.

Fig. 3.5

Ethnographic observation protocol template

There are several ways to record parliamentary fieldnotes. Bussell (2020: 471), for example, suggests more structured forms of note-taking, structuring fieldnotes chronologically around a unit of observation, such as a column for time. In this sense, researchers take detailed field notes after every relevant fieldwork event (e.g. meeting, phone call conversation, informal chat, etc.). One such moment was after being approached by a member of the parliamentary administration who had moved from a political group and had a ‘gripe’ session. Whilst some ethnographers, faithfully record everything that happened in the meeting, we wished to be ‘as inconspicuous as possible’ (Mykkanen, 2001) and as a rule did not record meetings. One exception was a surprising invitation from the accredited parliamentary assistant of an EFDD group meeting, to openly record the proceedings. Such a material offer of using a Dictaphone potentially demonstrated that Dr. Miller had acquired a presence in the field as a qualitative social researcher (Laube, 2021).

Similarly to Mykkanen’s (2001) observations of the Centre Parliamentary Party in the Finnish Parliament, during the pilot study our team members kept a running log of the meetings by hand. Despite not being verbatim reports, they were as faithful to the words of the participants as possible. Quotation marks that were placed around certain words certainly were verbatim. During the pilot stage of data collection, there was no real unit of observation because the coding framework had yet to be devised, and given that our research was primarily interpretive, we were able to find the precise empirical focus iteratively. Therefore, the observations remained holistic and chronological.

During the main data collection, we developed a systematic observation protocol based on a structured form divided into seven sections. The sections were driven by Feminist Institutionalist conceptual lenses: event setting (to convey the role of context, such as which attendees were in the room), power relations, democracy, gendered practices, the political group as a workplace, affect and researcher role. Counter to more positivist methods that might be more categorical for aggregating data statistically, our ethnographic notes were more descriptive and subjective (see Fig. 3.5).

Since one researcher was conducting all the ethnographic fieldwork, we implemented a practice of a weekly phone conversation between Brussels and Helsinki taking place every Friday. Instituting a set of debriefing practices with a supervisor or colleague is related to recording data. This may involve sending fieldnotes, observation protocols or an overview of activities attended that week to a peer for comments and discussion. Debriefing also occurred in the process of presenting findings and ideas at conferences, as well as to participants in the field, for example, to informal staff groups, in ways that ensured anonymity and confidentiality.

This section has been attentive to some of the practices and techniques useful for the recording of parliamentary ethnography. In our experience, there were many research questions organised around phenomena in parliaments that made ethnography a rich and fruitful methodology. Amongst other things, we discussed and shed light on, the nature of organisational change, new political and parliamentary contexts, such as new country accessions and democratic experiments like Spitzenkandidaten, new forms of (feminist) leadership, elections, conflict and contestation; all of which will be considered in greater detail in the following chapter.

Document Archive

Our team collected a wealth of internal documents from the EP and those covering the practices and policies of the political groups, consistent with fieldwork using ethnographic observations and interviews. These not only fed into the context and interview analysis but were also crucial in our publications (see, e.g. Ahrens & Kantola, 2022; Elomäki & Ahrens, 2022; Elomäki & Gaweda, 2022; Kantola et al., 2023).

Formal Political Group and EP Documents

The documents we collected included all the political group statutes which were either accessible online or after we had requested them from their staff by email (see Table 3.3). This did not include the documents of groups like the Left, because they do not have formalised statues. We gathered the statutes of all groups largely through our contacts, using direct means. In this way, we were able to access documents that were not available otherwise, such as internal guidelines, internal gender action plans and documents related to procedures on harassment. Documents that relate to the internal activities of the political groups are usually not publicly available, with only some publishing their statutes freely online (e.g. Greens/EFA, EPP, ID). Statutes govern the formal rules of the groups and were important to us as we published articles on their internal working practices.

Table 3.3 Overview of official documents collected on political groups

Position papers and press releases on specific issues that groups often published on their websites were also collated. Political groups tend to share their positions more openly in press releases rather than during the policy process because decision-making in the EP is consensus-oriented. We collected the internal documents, position papers and press releases, as well as EP documents (EP rules of procedure that govern the organisation and function of the Parliament with relation to policy-making) and published debates. Access to these textual documents varied by political group. Some groups updated their websites and social media frequently, whilst others only posted occasional or topical content. Each type of document had a different role in the research and they complemented each other and the interviews. Most of our document archive was compiled for the purposes of individual articles, without shared data collection guidelines. For example, besides using the interview and ethnography dataset, Gaweda et al. (2022) collected national party electoral manifestos, as well as political group electoral programmes from the 2014 and 2019 EP elections for their study of the conservative ideology in the ECR group.

Policy-Related Documents

Beyond interviews, our data consisted of text derived from policy documents to which we applied content analysis. Amongst other things, reading documents about parliamentary work helps researchers familiarise themselves with policy narratives and their development across time (Prior, 2020). We discuss in greater detail how we analysed the documents in Chapter 6; however, here we want to discuss what was collected. We took a broad understanding of policy documents as any written records about EP policies and activities, which were gathered on a case-by-case basis, dependent on the research articles we were working on. These included transcripts of plenary debates, committee meeting videos and adopted legislative and non-legislative resolutions of the EP, along with draft report and committee and plenary amendments. For instance, Ahrens et al. (2022) produced an in-depth analysis of nine plenary debates dedicated to questions of human rights, gender equality and religious issues in addition to relevant interviews and ethnographic materials from the project dataset.

We analysed the economic policies of the political groups (e.g. Elomäki, 2021; Elomäki & Gaweda, 2022), next to interview data, through a dataset composed of non-legislative reports on the European Semester drafted by the Economic and Monetary Affairs Committee (ECON), and the Employment and Social Affairs Committee (EMPL). The dataset included draft reports, amendments, adopted texts and committee and plenary speeches and allowed the identification of differences between the political groups and committees in terms of constructions of gender equality, constructions of the social/economic relationship and specific social issues and economic ideas underpinning policy proposals.

Regarding policy tracing, we typically gathered draft reports or proposals, the amendments submitted by groups and MEPs in the relevant committee with appended explanations, the adopted committee reports, the last amendments submitted at the plenary stage and the final text adopted in plenary with the explanations of votes. These followed the journey of policy adoption, from committee negotiations to plenary vote (Ringe, 2010). Whilst the draft report reflects the views of the rapporteur and its political group, the amendments reflect the views of other political groups and MEPs sitting in the same committee. The latter shed light on the diverging priorities and contestations amongst, and within, the political groups. Once a compromise is reached at committee level, a report is adopted and may move to the plenary level where more amendments can be made. These documents are publicly available on the EP website as part of its commitment to transparency and can be easily found via a keyword search on the Public Register of Documents website of the EP. For each legislative and non-legislative process, the EP keeps records of all policy documentation on its ‘Legislative Observatory’ website.

We also gathered published records of debates at the committee level and plenary level. Unlike many other parliaments, the EP committee meetings are open to the public and video recordings are openly available on the EP website. These are important to study because they indicate the priorities of the political groups and if they support the view presented by the rapporteurs. The plenary is the most important deliberative space of the parliament, it provides the political groups and MEPs a forum for public attention and for sending messages to their constituencies (Brack, 2018; Kantola & Lombardo, 2021). Plenary debates are mainly performative, as a negotiated text already exists.

Plenary debates can also be found on the EP Public Register of Documents with a simple keyword search. One particularity of the European Parliament as a research context is that it has 24 official languages and MEPs often use their native language in the debates. Video recordings are available in the original multilingual form, as well as simultaneously translated into all official languages. For the plenary debates, the EP publishes written reports of all speeches in a multilingual form. For the committee debates, we relied on simultaneous English interpretation, when our own language skills fell short. For the plenary debates, we either used the English simultaneous interpretation or translated the native language speeches.

Debates complemented amendments and interview data as they highlighted contested and polarising issues. We found these kinds of debates to be revealing about the core positions of the groups and the views of individual MEPs, bringing out the tensions within the groups (Ahrens et al., 2022). When the purpose was to analyse strategies of opposition to gender equality (Kantola & Lombardo, 2021) or contestation of women’s and LGBTQI rights (Ahrens et al., 2022), debates were selected to cover those that illustrated the greater level of contestation (i.e. convenience sampling). Such a strategy does not aim at generalising according to incidence and prevalence criteria (Soss et al., 2006: 136; Weiss, 1994), but rather to highlight the interpretative accounts of the group and MEPs regarding their strategies and discourses (Kantola & Lombardo, 2021; Yanow, 2006: 9)—core to our research questions (for more on the replicability of qualitative research see Ritchie et al., 2013; Seale, 1999). Finally, we did not approach debates by paying attention to gestures, tone, facial expressions and physical environment but rather as a written text channelling a political discourse (see for instance, Holm, 2020 for visual methods).

Data Storage and Archiving

As previously noted, all interviews were recorded (if the participant agreed) and all data (including notes and transcripts) were stored on the protected drive of the host university. Following the pilot study, we decided to name all our files according to a jointly established system. First, the data files were sorted into the following five categories: MEP, APA, SG, PG staff, and PRESS. Then, all file names followed the prefix format: GROUP; POSITION; GENDER; DATE of interview (sometimes we included a NUMBER, if there were more than one interview within the same group on a given day). In addition, all ethnographic field note file names ended in _FN and post-interview notes in _IN. We also created several ‘metadata’ lists and tables that included information on nationalities, genders and the specific position of the interviewees.

A local transcription company recommended by the university completed all the transcriptions of Finnish and English interviews. One team member anonymised and uploaded the transcripts, submitted all the recordings and kept track of the ongoing cataloguing and archiving of data. This made the process more reliable and the cataloguing more consistent. Individual interviewers arranged transcriptions of interviews in French, German and Polish from national transcription services on a case-by-case basis.

During the pilot study, we attempted to translate some French interviews into English so that all team members could access them equally for analysis, but the quality was unsatisfactory and we felt that a lot of the nuance was lost in translation. For that reason, we transcribed, stored and later coded the interviews in their original languages, still maintaining the same cataloguing system (just adding language suffixes to name files, e.g. _pl for Polish or _de for German).

Conclusion: Qualitative Assessment of the Data

This chapter presented a thorough overview of the various types of data collected in the research project, including a description of our interviews (including post-interview notes), ethnographic field notes and the wide range of document data related to the EP and its political groups that we collated. Whilst we sought to undertake 170 interviews when designing the data-gathering period, we managed to get 140 interviews (still maintaining a gender balance in the dataset). Despite not reaching the numbers we expected, we were more than satisfied with the quality and content of the data, which provided us with a considerable tranche of material for analysis. The following chapter will cover in greater detail the logistical challenges we faced, suffice to say, ‘objective’ obstacles aside (e.g. the Covid-19 pandemic) the scheduling, rescheduling and ‘juggling’ of numerous interviews within weeks, and sometimes days, proved challenging. It imposed greater limits on our research than we would have liked.

In retrospect, our major takeaway from handling the data might seem trivial, but proved to be essential—we learned the hard way the importance of keeping track of the data and the locations of the files. Only after the pilot study were we forced to rethink and adapt older files; later we developed our archiving and labelling system for data files which was time well spent. Since we had files in various formats (audio, video, text, etc.), we needed a system that enabled us to be in control of their numbers and names, as well as our protocols for archiving, whilst maintaining participant anonymity and privacy. The sheer volume of data coming in during fieldwork would have been challenging to keep track of, had we not systematically maintained and regularly updated lists and tables of data files with information on gender, nationality, and positions in the EP. Concomitantly, writing up post-interview notes and ethnographic field notes proved easiest and most effective immediately after the events or interviews, despite the time constraints imposed by continuing fieldwork.