Secondary Qualitative Analysis of Interviews. A Method Used for Gaining Insight into the Work/Life Balance of Middle Managers in Germany¹⁾

Abstract: This paper outlines a secondary analysis of interviews from an application-oriented perspective. We conducted a secondary analysis in a project on "work/life balance of managers". The original data derived from an investigation of "interest orientations of middle managers", in which the author herself participated. I will describe the methodological steps as well as the benefits and limitations we experienced in applying this approach.

Secondary analysis allows us to use an existing pool of data to pursue a research interest which is different from that of the original work. We used this approach in an investigation into "Work/life Balance of male Managers" (see NOTZ, 2001, 2004). In this qualitative research project, German managers and their spouses, as well as personnel officers, were interviewed. But before collecting primary data, we conducted the secondary analysis. The data derive from an earlier research project on "Interest Orientations of Middle Managers" (e.g. DEUTSCHMANN, FAUST, JAUCH & NOTZ, 1995; FAUST, JAUCH & NOTZ, 2000), a project in which the author herself participated. [1]

The main purpose of this article is to describe the motives for conducting the secondary analysis (2), some general considerations (3) and the concrete steps of the analysis. Finally, I will present some examples in order to outline potential benefits and limitations of this approach (4). [2]

The idea of conducting a study on work/life balance of male managers occurred to us when working on "Interest Orientations of Managers". Besides exploring questions on the changing working conditions of managers, we looked at conflicting work-family demands, although the latter were not our main concern. [3]

Nevertheless, managers' responses to home-work conflicts attracted our attention. The executives, faced with demanding working conditions, found it difficult to meet the family's expectations. In order to manage family tensions, they employed peace-keeping strategies towards spouses and children. A notable result was that the majority of managers thought it natural for them to be relieved of housework and child care responsibilities. Given social changes in gender roles, the question arose as to whether wives felt satisfied with the family situation and the division of labour within the family. [4]

When we started to explore questions on "work/life balance of middle managers" we first conducted a secondary analysis of these existing interviews. We wanted to help develop the current study by obtaining some prior information about how managers perceive the home-work situation. Firstly, the contents analysis served to identify executives' problems with their work/life balance. Secondly, the interviews provided an insight into managers' views of their different social roles, as company man on the one hand and marital partner and father on the other. Thirdly, the ideas found in the interviews helped to design the interview guides for the managers and their spouses as well as for the personnel officers. Fourthly, the secondary analysis contributed to the interview process by providing knowledge about the work situation of middle managers. Finally, the secondary analysis helped to make up for a lack of literature on this topic. [5]

Before conducting a secondary analysis, it is productive to reflect on the research interest and data collection of the original work (THORNE, 1994, pp.269ff.; HEATON, 1998). This is because existing data, collected for an earlier study, are to be used to explore a distinct research interest. Before describing the focus and methodological approach of the project "Interest Orientations of Managers", I will first give a general outline of these topics. [6]

An important issue is how well the topics of the original research project match those of the present project. Do the focal points of the earlier investigation fit in well with the present questions, and if so, how well? What kind of findings do we hope to gain from the secondary analysis? In which ways are we limited by the existing data base? If new data are to be collected, we will need to determine whether original data and new data are to be compared later, and whether it is possible to transfer the previous interview guide to the new project. [7]

If only parts of the sample are important for the present research interest, we need to define which elements of the existing material are of relevance to us. Data selection might be based on respondent characteristics (e.g. age, sex, company), specific contents (e.g. various batteries of questions) or features of the interviews (e.g. time and location of the interview). Analysing material of a research project one participated in personally facilitates the selection process, because the investigator is familiar with the group of respondents and the circumstances of the interviews. If this is not the case, the selection relies on the material being adequately archived.[8]

Furthermore, the interview technique itself needs to be examined. What kind of methodological approach was chosen for the conversations (e.g. expert interviews, narrative interviews), how were the data recorded, transcribed and analysed; that is: what kind of material (e.g. interviews, fieldnotes) was produced from them, what might be of interest for the current investigation and how can it be used? The existing database not only determines analysis options and limitations of a secondary analysis, but also—if new collection of material is intended—how the original and new databases are related. Thus it is not possible to connect data produced by meaning-seeking methods with data that were gathered for conducting a content analysis. [9]

Further reflections focus on the respondents and the circumstances of the interviews: Who was interviewed, where did the conversation take place, and who conducted the interview? It must be clarified whether we are interested in the interviews of all respondents or only a particular group. Conversation is influenced by the local conditions, that is, whether the interview takes place in public, at home or at work. The environment determines to a certain extent how much a person feels committed to a social role and how frankly he or she expresses his or her opinion. Finally, the sex of the researcher influences the conversation in a subtle way (BEHNKE & MEUSER, 1999, p. 79). Men are allowed to have more professional competence, whereas women are thought to be experts in emotional matters. In sum: given all these issues, what can we expect to get out of the material in terms of the current research interest? [10]

If we want to be clear about the objectives, options and limitations of a secondary analysis, we will need to reflect on these matters in advance. With these issues in mind, I will now describe how we conducted the secondary analysis. [11]

First of all, I will outline the research interest of the earlier study. In the project from which data originated we had investigated organisational restructurisation in large enterprises, and the consequences for middle and lower managers (e.g. DEUTSCHMANN et al., 1995; FAUST et al., 2000). For the very first time executives were exposed to rationalisation measures, experienced a reduction in the number of staff and in the number of management levels. These changes affected them deeply. The interviews served to find out how managers perceive organisational restructurisation and what consequences it had on them personally in their everyday life. They covered topics such as work demands and stress, changes in career, increased responsibilities and sphere of control, interest orientations, as well as life/work-balance. Some of the executives stressed the negative impact the changes had on their position in the hierarchy and career, whereas others considered the transformation as liberating them from bureaucratic ties. The majority reported greater involvement in their work, which turned out to be tangential to maintaining a work/life-balance. They found it became more and more difficult to meet the expectations of spouses and children. It was this topic which we were interested in in the following research project: work/life balance of middle managers. The secondary analysis served to reveal different aspects of this topic. [12]

In the original study we conducted interviews with executives from different levels of the hierarchy in large companies. The interviews were structured by an interview guide and lasted between one and a half hour and two hours. They took place in the offices of the companies. The interviews were conducted by a team of two interviewers, either two male researchers or one female and one male investigator. The conversations were recorded on tape and then transcribed. We aimed at a content analysis of the interviews. The transcriptions were available. [13]

The executives were asked by the company management to participate in an interview. The methods used to select the respondents were unclear to us. Some of them volunteered after being informed about the study, while others were invited to take part in the investigation because of their position, their professional development, or simply because they had no other obligations that day. Since the organisational changes had greatly affected lower and middle managers, the company management was interested in getting general feedback from us on the current atmosphere at work among the executives at these levels. [14]

The managers gave detailed, frank and critical responses to questions on their work and the organisational changes. They used the opportunity to draw attention to negative aspects of organisational restructurisation. They presented alternative solutions or just talked to get anger off their chests. Even if some suspected that the researchers might hand on confidential information to the top management, this didn't prevent the majority from expressing their opinion. [15]

By contrast, the managers were rather vague and reserved in their answers to questions about their private life. Obviously these questions didn't seem appropriate in the workplace. When discussing their private life, respondents mainly addressed the female researcher. [16]

Interviews were conducted with senior, middle and lower-level managers. Only one participant was female. Since we concentrated in the current study on middle managers we selected interviews that showed the required characteristics. Of the total sample, 40 interviews were selected. [17]

As mentioned above the secondary analysis was for a research project focusing on work/life balance of male managers. We wanted to find out whether executives felt more and more squeezed between the demands of work and family. In their opinion, did they manage to satisfy the requirements of both worlds? What kind of beliefs about gender and family were revealed by their comments? To answer these research questions we had intended to conduct interviews with male managers and their spouses as well as with personnel officers in large companies. [18]

In the interviews available for the secondary analysis these topics weren't of central interest. As described above they primarily dealt with the managers' views on organisational changes and the consequences for their work situation. The managers had, however, been asked how the present working conditions influenced their private lives. These interview segments were of particular interest. [19]

Since work and family are mutually overlapping, we also concentrated on the contents of the interview that focused on certain aspects of work. For instance, managers were expected to behave as risk-taking entrepreneurs. This led managers to be increasingly personally involved in "their working responsibilities". It became more and more difficult for them to set aside fixed times of day for the family. We also looked at changes in career paths. If, e.g., career progression is linked with a stay abroad, the lives of wives and children are affected as well. Stress caused by difficult working conditions also impacts on family life. Conversely: marital conflicts or bereavement impact on one's work. Thus, statements about work and career were central to us if they concerned issues that affect private life. To sum up: these topics covered only parts of the whole interview. On the other hand, attitudes towards unions weren't of interest. [20]

Taking these notions into account, we defined criteria to help us to extract these relevant contents from the interview. [21]

Secondary analysis of interviews can be conducted in different ways: we chose to rely on word processing software. It met all requirements for copying and pasting segments of text and formatting significant paragraphs of text as headers. [22]

First, each individual case was analysed, then secondary analysis by topic was performed. This provided data access in two ways: instances of a specific subject could be looked up using secondary analysis by topic. If statements of a specific manager were of interest we searched for these contents using individual case analysis. [23]

As mentioned above, we defined criteria for extracting interesting contents from the interview. They were transformed into categories, such as "career, "working hours", "leisure time" or "family life". Each interview was analysed using these categories. We didn't use the search function of the word processing software. We could thus first explore the material without being influenced by catchwords. We wanted to be sure not to miss important statements. The interview guide that was used for the semi-structured interviews was a helpful tool for this step. Since I had participated in the original research project, I benefited from this implicit knowledge when performing this step of the analysis. The next stage involved copying relevant parts of the data into a separate file. Each paragraph was labelled with a category, followed by a short description of its content. The category and description of the content were formatted as headers. Where necessary the header included reference to further categories. Respondent specifications (name, age, company) were also added, as well as the place where we could find the paragraph in the original data. Managers' names and names of the company had previously been replaced by pseudonyms. Furthermore, each participant was profiled according to his education, career, current position in the company, family situation and his social background. Each profile was posted at the beginning of an individual case, that is, preceding all the extracted statements of a respondent. [24]

"Family: Working constraints and effects on family relations" (Müller (37) / Abay, p.15).

"Yes, okay, workload does affect private life, the family. You need a family that is happy to see you just twice a week. Or, as in my case, I do not have children, just a wife, …". [25]

"Family" is the general category, whereas "working constraints and effects on family relations" represents the content of the text sequence. The manager's name is "Müller" and he works at a company called "Abay". The text can be found on page number 15 of the original report. The two lines were formatted as headers in the word processing program. [26]

The extracted sequences of the forty respondents were saved in a single document. We then created a directory using the word processing program. This allowed us to access the data using an overview of all respondents. [27]

The following step of the secondary analysis was to highlight those categories and content descriptions. [28]

First, all statements of all interviews were pooled by category. Since we had formatted the categories (including the descriptions of the contents of a paragraph) as headers this step was easy to perform. We could then draw on a pool of statements, giving us information on categories such as "working hours" "career" or "family". [29]

The next step focused on statements of the different categories. We added sub-categories to rearrange the material. This was not difficult, because each category was—as outlined above—followed by a short description of the content. Similar statements could be pooled by a sub-category. The material of the category "career" e.g. was subdivided into sub-categories such as "former career paths" "new career paths: limitations and alternatives"; the statements of the category "family" were subdivided into the sub-category "family orientation versus career orientation", "peace policy" or "impacts of working constraints on the family". [30]

Finally, all statements and their categories and sub-categories were saved in one document. Again, we created a directory. The directory provided an overview of the categories and sub-categories and enabled speedy access to data by presenting the specific topics. [31]

A printout was made of the secondary analysis representing the individual case as well as the different topics. With hindsight, this pragmatic approach proved to be entirely adequate for our purposes. [32]

There are a great variety of goals one can pursue with secondary analysis. To mention some examples, existing qualitative data can be used as an inspiration for creating a research project, to compare the results of two surveys, or—because data collection is time-consuming—to save time and money and, finally, they can be utilised for teaching purposes in method application (THORNE 1994, pp. 266.; THOMPSON 2000). [33]

As mentioned above, the secondary analysis described here was used to help develop the current study. We were interested in getting information that helped to create and conduct the new research project. [34]

Firstly, the contents analysed served to identify executives' problems with work/life balance. The analysis provided an insight into how executives viewed themselves in their role as successful company men. Work demands are high and managers continuously feel they are being tested to see whether they are meeting the requirements. Mistakes are seen as harmful to professional identity. "You have to take care that you don't make a mistake. Because this will incur costs, but besides the costs, you are afraid to lose face". The notion emerged of a manager trapped in the organisational system. Since the role model for managers expects executives to keep on rising within the corporate hierarchy, stepping back for a while to have more time for the family isn't allowed. "If you are part of the system, there is no way back, or you are out". Managers complained that wives didn't understand the pressure the system put them under. "Or when she is accusing you for not having called her, although you promised, and you try to explain that you didn't have the time, because there were so many calls all day long, she won't listen. This happens every day. You permanently suffer from a guilty conscience". [35]

Secondly, the interviews gave a general idea of managers' beliefs about the divergent social roles of being a company man on the one hand and marital partner and father on the other. The professional and marital demands were often seen as competing. Typically, the respondents felt strongly committed to the company and to "their responsibilities". When the manager cited above mentions his guilty conscience towards his wife and the family, we also get an idea of how priorities were set. Here the secondary analysis revealed how managers assuaged their feelings by adopting peace-keeping strategies towards their spouses, and showed what kind of underlying beliefs about family roles guided their strategies. For instance, occasionally spouses were invited to accompany their husbands on business trips abroad. Implicitly, this served as a sort of compensation for the many days of absence from home. At the same time it provided a perfect opportunity to demonstrate to their wives that these journeys abroad weren't to be seen as pleasure trips. "Yes, twice a year I tried to take my wife with me, so that she could see I wasn't living as a king abroad". It was implicitly seen as natural that men's jobs were more important than their spouses' jobs. If compromises were to be made within the relationship, it would mean something different for both partners: the wife had to accept that her husband's work was extremely absorbing, whereas the manager was to reduce his working hours once in a while for his wife. "If both partners were working under this kind of condition, the relationship would probably suffer greatly". [36]

Thirdly, based on these findings, secondary analysis was helpful when designing the interview guides for the interviews with the couples (manager and wife were to be interviewed separately) as well as for the interviews with personnel officers. We wanted to use the interview guide to stimulate couples to talk and reveal their beliefs about the importance of their professional work within the relationship. We wanted to find out whether the interviewed couples tended to prefer traditional marriage, such as the findings of the secondary analysis had indicated. Thus the interview guide contained questions covering three temporal phases: past, present and future. We first asked the respondents to describe the story of their relationship: how they got to know each other, how they managed living together in terms of domestic division of labour and professional obligations. After this we concentrated on the present situation in the relationship. Once again we were interested in how the couple organised everyday life, how they shared the housework, how they combined work and family. Finally, we asked the couples to think about their hopes and wishes for the future. The temporal component of the interview guide helped to investigate, for example, whether managers became more emotionally involved with work over the years and how they reconciled this involvement with the—maybe also changing—marital demands. Furthermore, the temporal component served to analyse a couple's gendered ideas of marital roles and the division of labour. The secondary analysis had highlighted gendered conceptions of relationships as well as peace-keeping strategies to alleviate marital demands, based on gendered role models. The interview guide for the personnel officers focused, for instance, on executives' opportunities to work part time. This question was also inspired by the findings of the secondary analysis, the notion of the manager trapped in the organisational system as described above when identifying problems in managers' life/work-balance. [37]

Fourthly, the secondary analysis contributed to the interview process by providing knowledge. For example, managers' narrations about working conditions and demands were easier to understand. Or the knowledge we got from the secondary analysis supported further inquiry when interviewing personnel managers. [38]

Finally, the secondary analysis helped to compensate for the lack of literature on this topic. At that time, middle managers' problems of work-life balance were scarcely investigated in Germany. [39]

Thus, our research project benefited from conducting secondary analysis. Nevertheless, the conclusions we could draw on the topic of managers' work/life balance were limited. Firstly, within the investigation of "interest orientations of managers" questions concerning leisure time and family hadn't been given too much significance. When the topic "family demands" arose during interviews in this earlier project, we usually didn't probe too deeply when we realised that the respondents were hesitating. Secondly, the managers were prepared to give an interview on their working situation but not on their private lives. Questions on work/life balance revealed their notion of strictly separate worlds of family and work. The way managers avoided talking about some issues, the way they skirted around the main subject, the shunning and side-stepping, also provided inspiring insights. The secondary analysis was supportive in identifying constructions of gender. [40]

To sum up: although research interests of the two projects differed, we benefited from conducting the secondary analysis. Whether using existing data to pursue a research question distinct from the original work can be recommended depends on the individual case. In qualitative research, at least, we should be more aware of using the richness of available data. [41]

1) This is the English translation from a German text, published in FQS in January 2005. FQS thanks the author for providing the English version and Louise CORTI for copy-editing, May 2007. <back>

Behnke, Cornelia & Meuser, Michael (1999). Geschlechterforschung und qualitative Methoden. Opladen: Leske und Budrich.

Deutschmann, Christoph; Faust, Michael; Jauch, Peter & Notz, Petra (1995). Veränderungen der Rolle des Managements im Prozess reflexiver Rationalisierung. Zeitschrift für Soziologie, 24(6), 436-450.

Faust, Michael; Jauch, Peter & Notz, Petra (2000). Befreit und entwurzelt. Führungskräfte auf dem Weg zum "internen Unternehmer". München: Rainer Hampp Verlag.

Heaton, Jane (1998). Secondary analysis of qualitative data. Social Research Update, 22, , http://www.soc.surrey.ac.uk/sru/SRU22.htm [Date of Access: Oct. 27th, 2004].

Notz, Petra (2001). Frauen, Manager, Paare. Wer managt die Familie? Die Vereinbarkeit von Beruf und Familie bei Führungskräften. München: Rainer Hampp Verlag.

Notz, Petra (2004). Manager-Ehen. Zwischen Karriere und Familie. Konstanz: UVK-Verlagsgesellschaft.

Thompson, Paul (2000). Re-using qualitative research data: A personal account [48 paragraphs]. Forum Qualitative Sozialforschung/Forum: Qualitative Social Research, 1(3), Art. 27, http://www.qualitative-research.net/fqs-texte/3-00/3-00thompson-e.htm [Date of Access: Oct. 27th, 2004].

Thorne, Sally (1994). Secondary analysis in qualitative research: Issues and implications. In Janice M. Morse (Ed.), Critical issues in qualitative research methods (pp.263-279). Thousand Oaks: Sage Publications.

The research described in this article was conducted at the "Research Institute of Work, Technique and Culture" in Tübingen, of which the Petra NOTZ was scientific employee for many years. Her research interests are in the area of sociology of management, gender issues and vocational education. She has been working at the Institute of vocational education, labour market policy and social policy" in Offenbach since 2004.

Institut für berufliche Bildung, Arbeitsmarkt- und Sozialpolitik (INBAS)
Herrnstraße 53
63065 Offenbach, Germany

Notz, Petra (2007). Secondary Qualitative Analysis of Interviews. A Method Used for Gaining Insight into the Work/Life Balance of Middle Managers in Germany [41 paragraphs]. Forum Qualitative Sozialforschung / Forum: Qualitative Social Research, 6(1), Art. 34, http://nbn-resolving.de/urn:nbn:de:0114-fqs0501347.

Forum Qualitative Sozialforschung / Forum: Qualitative Social Research (FQS)

ISSN 1438-5627

Creative Commons Attribution 4.0 International License