The research philosophy

Published: Last Edited:

This essay has been submitted by a student. This is not an example of the work written by our professional essay writers.

The possibility to categorize a researcher's research philosophy by differentiating between qualitative or quantitative research is contested and I believe Gee's (1999) statement that "there really are no grand categories of research" (Gee 1999). As stated in my literature review I see myself as a modernist qualitative researcher and Schwandt (2001) helps me to define my research philosophy in a more concrete way. Schwandt (2001) describes the relationships between methodologies and methods. According to Schwandt (2001) my 'research methodology' outlined in this study fit somewhere between 'ethnographic and naturalistic' and 'narrative and interpretive interactionist'. In this study I am using elements of both ' research methodologies'. The 'ethnographic and naturalistic methodology' uses in-depth, ethnographic and unstructured interviews; life history interviews and finally participant observation as methods to generate qualitative data (Schwandt 2001). The social or everyday life (the life-world) understood from the actor's perpective, knowledge, experience, intentions, interpretations and so on are the object of understanding and theorizing (Schwandt 2001). The 'narrative and interpretive interactionist methodology' utilizes 'active' and narrative interviews. The object of understanding and theorizing is the dialogical process of communication; the 'exchange process' and the joint construction of accounts of social life in conversation and reflection (Schwandt 2001). In the following chapters the methods and the theoretical understanding that fit in those two approaches of Schwandt's (2001) definition will be introduced. In summary, the methods of both approaches show direct links with this study's use of unstructured experience-history interviews with narrative elements. Furthermore I try to let my conversation partners tell their stories with as little interference as possible based on the applied methods.

Ontology and epistemology

Lincoln and Guba (1985, p. 37) who found the naturalistic inquiry say that "realities are multiple, constructed, and holistic". As already explained in the literature review this description fits my research philosophy. But it also applies to the empirical part of this study because I try to build conclusions after hearing, reading and comparing words from diverse perspectives of each of my 'subjects'. The information I get is processed via my eyes, ears and mind and other researchers will have an own perspective on the information which I interpret in this study. Even my own perspective on the perspectives / realities of the people to be studied is a part of the process and conclusions. At every point the researcher leave his fingerprints, starting from the questions I prepared and way I ask questions based on my experiences and how I react on answers.

I see my epistemological base connected with Lincoln and Guba's (1985) ideas. Even going further, the interactions between researcher and 'subject' or 'conversation partner' (Rubin & Rubin 2005) flow into both directions. The wording of my questions and comments affect the answers I get from my conversation partners (Mishler 1986). Besides the responses of my conversation partners affect my following comments and questions.

Regarding Lincoln and Guba's (1985) axiom of generalizability in qualitative research saying that "the only generalization is: There is no generalization (Lincoln & Guba 1985, p. 110). I believe that the findings of my study will not apply in all cases, and I agree with Lincoln & Guba (1985) that there is a possibility of making limited generalizations that go beyond the specific focus of this study although such possibility is uncertain. This study should try to offer answers that go beyond the world of the people I talked to, outside the specific team setting.

I also agree on Lincoln and Guba's (1985, p. 38) axiom of causal linkages in the direction of naturalists. Whereas positivists may believe that every "action can be explained as the result of a real cause that precedes the effect temporally", they argue that differentiating cause from effect is not possible in a world where both are always changing each other (Lincoln & Guba 1985). As described in the following chapters this study includes parts of narrative research when I ask people about their Sino-German teamwork experiences and decisions. Especially for such an approach the researcher needs to be aware that 'before' and 'after' does not clearly mean 'cause' and 'effect' when listing to the narratives of the subjects (Clandinin & Connelly 2000). Finally I also agree on Lincoln and Guba's (1985) axiom related to the roles of values in research. Whereas a positivistic researcher may believe that research can be value-free, I believe that there are outside values which influence research whether it is qualitative or quantitative.

Besides Lincoln and Guba's (1985) axioms also their term paradigm shift, which means a major change in thinking, is important for me as I am researching, thinking and writing about what people say and do. Lincoln and Guba (1985) further developed Thomas Kuhn's initial ideas. I am following a paradigm towards qualitative research based on my assumption that there is not pure objectivity and that statistics should be read sceptical. However, whether a researcher uses qualitative or quantitative research, he needs to protect himself from blindness to subjectivity. Even open-ended interviews can lead to wrong understanding if the interviewer is more listening to himself than to the interviewees and his report will have the same shape like he has talked to no one.

Those who are learning by interviewing others can be described in two ways according to Kvale (1996). The researcher can be described as a miner who dug the ore of knowledge from under the ground in people's words, refines it from oral to written form while maintaining it purity, and offers the refined knowledge without any contamination from the researcher's efforts by leading questions. Some miners seek objective facts to be quantified and others seek nuggets of essential meaning out of a subject's pure experience. However, all miners believe that knowledge is waiting to be uncovered uncontaminated by them (Kvale 1996). In a second way the researcher can be described by an interviewer as a traveller who is on a trip to strange lands, preparing a story to tell others when he returns. The traveller can be methodological or planless but he will anyhow learn from his conversations with those he meets on in the strange lands (Kvale 1996). What the traveller hears and sees expressed qualitatively and build as stories to be told to the people of the interviewer's origin and maybe other who joined the conversations. The potentialities of meanings in the original stories are differentiated and unfolded through the traveller's interpretations. The tales are shaped into new narratives which are validated trough their impact upon the listeners (Kvale 1996). These two metaphors represent different philosophical concepts. The positivist miner picture represents a common understanding in modern social sciences of knowledge as 'given'. The traveller picture represents a post-modern constructive approach which uses a conversational approach to social research (Kvale 1996). As already recognizable in the prior chapters and literature review is see myself more as a traveller than as a miner

Research purpose

The research questions show that the identified factors which influence the effectiveness of multinational teamwork can only be influenced by human beings whether as an individual or group. Researchers like Polkinghorne (1988) believe that the solutions for such human problems, or applied here for a negative human influence on the effectiveness of teamwork, cannot be created by developing sophisticated creative applications of the natural science model. Polkinghorne (1988) rather believes that by developing additional complementary approaches, which are especially sensitive to the unique characteristics of humans, solutions can be provided. I agree with this statement based on my assumption that asking people who are directly involved helps to understand certain phenomena. Bruner (1996), one of the founders of modern cognitive psychology, also agrees that scientific methods are often unable to describe the basis on which ordinary people make sense of their and other's actions. The search for the causes of human sense itself makes little sense for him because the way people make sense of their experiences and themselves may be not testable. Therefore Bruner (1996) argues that the 'logico-scientific' mode of conducting research requires a complementary mode that searches for reasons rather than causes. In summary, both researchers (Polkinghorne 1988, Bruner 1996) state that answers to human problems and sense are more likely to be developed from new ways of asking questions involving human participation rather than from using traditional tools of scientific investigation in more sophisticated ways.

Polkinghorne (1988) suggests a complementary approach based on his assumption that the traditional research model, adapted from natural science, is limited when applied on research on human beings. As a researcher in psychology, Polkinghorne (1998) suggests that practitioners should work with human data and people's stories rather than using more scientific positivistic approaches. The result of such an approach would be a better understanding of the way things are and what needs to be done (Polkinghorne, 1998). I agree with his idea acknowledging that for certain areas statistical data and numbers can be useful but rather not for this study. This study requires an approach in which the researcher asks experts, members of Sino-German teams in international companies, how they achieve successful teamwork and what it makes less successful. The researcher's target is to learn from them rather than analysing them as numerical data, or in other words, talk with them in order to learn how people understand their world and their life, as expressed by Kvale (1996). The idea to learn from those who are interviewed is also represented by 'ethnography', ethnographers seek to learn from people instead of collecting data about people (Spradley 1979). Ethnographers adopt certain stance towards people with whom they work and are always ready to learn from their research 'subjects'.

Research design and instruments

Based on my assumption, that the researcher should learn from its 'subjects' rather than analysing them as numerical data, research interviews seem to fit my research approach. The type of interview I propose to use has been defined in many ways such as "an interview whose purpose is to obtain descriptions of the life world of the interviewee with respect to interpreting the meaning of the described phenomena" (Kvale, 1996, p. 6). The interviews are based on my ideas and my curiosity about their world as starting points whereas the following part is as open-ended as possible in order that my interviewees are free to explain their world to me. Miles and Huberman (1994, as cited in Bahk-Halberg 2007) state that it is generally accepted that structured design is kept to a minimum before fieldwork begins. The interview should be a central element to make sense of the interviewees' life (Silverman 2000). I try to provide an atmosphere which let the interviewees be as free as possible to talk about themselves but with a clear idea of the areas I want to learn about and the possible interview questions I would ask. This interview questions will be prepared out of research questions. This implies variability in how I ask questions which is seen as a key to good interviewing in research interviewing (Mishler 1986) and interesting responses may follow due to such a flexibility (Agar 1994). Therefore the interviews are nearly informal dialogues but it has to be kept in mind that such a 'conversation' is not between equal partners, because the researchers defines and controls the situation by his topic introduction and questions which are searching for more or detailed meaning (Kvale, 1996).

Kvale (1996) argues that control of the researcher/ interviewer, or in other words subjectivity, is not a disadvantage but rather a strength because humans are uniquely capable of catching nuances of meaning and differences of opinion in how the interviewees see a phenomenon and communicating it (Kvale 1996). Kvale (1996) even goes further by saying that the interviewer is the research instrument. Other researcher note that the subjectivity in terms of variations within the questions, or the way questions are asked, can lead to a different outcomes regarding the answers that will be received (Mishler 1986). From my experience as a member in Sino-German teams I agree on this restriction of subjective interviewing as I have experienced different outcomes when tone, facial expression and other factors lead other team members to specific answers on my questions, even to answers I was looking for. In this study such an example would destroy the purpose of my research project but I believe that objectivity or neutrality cannot be achieved by human beings. Being isolated from experiments and the world in order to produce an objective description (Lincoln & Guba 1985) does not represent my assumption. A qualitative researcher should be aware and even use his subjectivities to understand the interviewees' stories (Lincoln & Guba 1985). Agar (1996) suggests that the researcher should first assess himself regarding biases before analysing others. Both the awareness of existence of my biases and the self assessment are part of the interview process, keeping in mind that the final results should be based mainly on empirical evidence.

Research data from Narrative inquiry and responsive interviewing

The proposed way to conduct the interview is a form of qualitative research called narrative inquiry which is here represented by Polkinghorne's (1988) definitions who focused on finding meaning from other's stories of their experience, means here in regard to their participation and influence as members in Sino-German teams. Polkinghorne (1988) describes narrative as a form of meaning making. A form which is complex and expressing "itself by drawing together descriptions of states of affairs contained in individual sentences into a particular type of discourse" (Polkinghore, 1988, p. 36). Polkinghorne (1988) also expressed that narrative acknowledges the importance of individual experiences by paying attention how they function as parts in a whole. Therefore narrative inquiry is especially suitable for human actions and events that affect human beings and which lead combined, based on the roles these actions and events play, to a whole (Polkinghorne 1988).

There are more reasons why a narrative inquiry can be used for the interviews. As found in the systematic literature review much research on multinational teamwork looked at outcomes and by using quantitative research disregarded the impact of experience itself. Narrative enables the researcher to understand experience and the people's lives is in focus (Bell 2002). By collecting people's stories and analysing them, deeply hidden assumptions can be recognized (Bell 2002). The researcher gets information that the interviewees may not know by themselves (Bell 2002) and cannot be answered by them via surveys, for example. As people may change their understanding narrative offers a temporal description of experience (Bell 2002), based on the interviewees context, rather than only answers on a set of pre-defined questions. Such a temporal description stands also for qualitative research (Sandelowski, Docherty, & Emden, 1997).

Critics of the narrative inquiry or researchers of a more positivistic persuasion see a lack in reliability, validity and objectivity but as Altheide & Johnson (1998, as cited in Bahk-Halberg 2007) state their standards of the quantitative paradigm cannot be applied to qualitative research. However, there are criteria which may lead to good narrative research in interviews, offered by Runyuan (1984) (see Appendix X) and Clandinin and Connelly (2000). Clandinin and Connelly (2000) mention the criteria "explanatory, invitational quality, authenticity, adequacy and plausibility" (2000, p. 185). However, Clandinin and Connelly (2000) go further searching for criteria for a good overall narrative inquiry. They believe that wakefulness is the most important criteria for following a good narrative inquiry (Clandinin and Connelly, 2000). Wakefulness can be is described as paying close and continuous attention on the interviewees and their answers. However such criteria will never meet the critic of researchers of a more positivistic persuasion. However, keeping the criteria of Runyuan (1984) and Clandinin and Connelly (2000) in mind when conducting the interviews is one of the major tasks of this research project.

For the concrete design and technique Rubin and Rubin's (2005) 'responsive interviewing' model is applied which fits in the assumptions of this research project. In other words it relies on the interpretive constructionist philosophy to find out how the 'conversational partners' understand what they have experienced (Rubin & Rubin 2005). Responsive interviewing elicits narratives among others from the interviewees (Rubin & Rubin 2005). It emphasize that the interviewer and interviewee are both human beings which form a relationship during the interview (Rubin & Rubin 2005). The model explicitly recognizes that both the interviewer and the interviewee have feelings, personality and so on, so that it cannot be expected that they neutral and that they do not affect the interview (Rubin & Rubin 2005). Rubin and Rubin (2005) also agree on my goal that the interviewer should be self-aware regarding his biases and expectations that may influence the interviewee. They even suggest that the researcher should sensitize himself to these biases and learn to compensate them (Rubin & Rubin 2005). I agree on this suggestion by trying to be reflective of my questioning and evaluating behaviour before, during and after the interviews.

Besides responsive interviewing shares the goal to generate depth of understanding rather than breadth and it also agrees on my plan to remain the design of research flexible throughout the project (Rubin & Rubin 2005). Responsive interviewing tries to understand the studies' subjects solid and deeply also by asking for narratives and stories. However, the individual interpretations of experiences of the interviewees are not seen as right or wrong, rather as different perspectives on what happened in order that the researcher puts them together to construct his understanding of what has occurred (Rubin & Rubin 2005). The depth is reached by "going after the context; dealing with the complexity of multiple, overlapping, and sometimes conflicting themes; and paying attention to the specifics of meanings, situations, and history" (Rubin & Rubin 2005, p.35). Such depth can be achieved by following up and asking more questions about what the interviewer initially heard (Rubin & Rubin 2005). The interview can be seen as a conversation where the interviewer keeps the conversation going, staying on the topic until it is covered by the interviewee and only after that clearly shifting to a new topic (Rubin & Rubin 2005). Concrete illustrations may help the interviewees to give answer regarding experience which provide "nuance and precision, context, and evidence all at the same time" (Rubin & Rubin, p. 37). Misunderstandings can be solved by rephrasing but without any assessment. The underlying research design requires being therefore flexible because questioning remains flexible to adapt to the interviewees answers which may include different depth of information, experiences and exceptions (Rubin & Rubin 2005). But also new discoveries out of the interviews may ask for a redesign of the interview process and differs from the intended questions the researcher wanted to ask (Rubin & Rubin 2005). In other words, the researcher creates future questions based on what he has heard. A precondition for such a proceeding is that the researcher analyzes the interview throughout the interview process and not just at the end (Rubin & Rubin 2005). As not being a positivistic researcher who often begins and ends with questions which have been developed in advance, I agree that many of the questions emerge only during the course of the interview in order to reach depth and solidity underlying the answers (Rubin & Rubin 2005). I agree with Rubin & Rubin (2005) that questions cannot be fully worked out in advance, therefore pauses for reflection are built into my interview design in order that I can compare what I asked with "what they should have asked and what requires more depth, and alter questions accordingly" (Rubin & Rubin 2005, p.37). I even argue further, like Spradley (1979), that a good way to find out what questions to ask people is to ask them for help for creating questions. Spradley (1979, p. 84) give examples like "What is an interesting question about ?".

This technique is applied to collect the research data. It acknowledges that this study is influenced by my past experiences with Sino-German teams I worked in. It treats these experiences as an enrichment and knowledge base when interviewing the 'conversation partners' (Rubin & Rubin 2005). The term 'conversation partners' fits the mentioned assumptions because interviewee and interviewer are in a relationship which there is mutual influence. For example, the researcher initially establishes the general directions of the interview expressed in a broad way there as the interviewee set the more specific path by telling his experiences which suggest the interviewer what to tackle and what to ignore next (Rubin & Rubin 2005).

Methods and procedures

In winter of 2010, I will talk to individuals who I know and who have been working in Sino-German teams within an international company. Following the systematic literature review such teams, they worked in, have to contain only two nationalities (Chinese and German) and are or had been real existing (not virtual teams). Further I should not be or had been a member of the team because this study is not written with an emic view from the inside of the team member's experience but rather from the perspective of an interested outsider who has observed what is going on by himself. The interviewees' experience in Sino-German teams in terms of membership duration and job level may vary to cover a broad experience in Sino-German teamwork. Because responsive interviewing is about learning what people think about their experience and rules under which they operate, the model ask for people who have such a specific experience (Rubin & Rubin 2005). In the first talk I will only ask them for referrals and whether they are themselves willing to participate in the research project. All these people will receive an email or telephone call in order to set up the interview sessions.

It is contested how many individuals and how many interviews are 'required'. Spradley (1979) needs a series of six to seven one-hour interviews for his approach of ethnographic interviewing. This implies an intensive project in which the researcher may live in the setting and observe people's lives. For this project learning from the interviewees' own interpretations is the focus but less my own observation about their behaviour or lives during their teamwork. Richards (2003) compared the recommendations of other scholars which vary from 2 to 15 hours for interview research. Richards (2003) additionally argues that more than one hour for a single interview may cause tiredness; therefore I plan to talk to each individual in one-hour sessions and the amount of interviewees should be maximum 15. In a first session some basic background information should be obtained by using basic demographic questions about their company, family, language, literacy and teamwork background. The answer may lead to interesting areas for further questions to be asked in the follow-up interviews. One to two follow-up interviews will follow the initial interview until the depth and solidity of the responsive interviewing model may be reached. Spradley (1979) gives also a detailed breakdown of the interview process which is divided into the stages apprehension, exploration, cooperation and participation which is taken as a structure example for each single interview (see Appendix B).

To work out the question sequences, that will bring the answer to such a depth, may need several iterations (Rubin & Rubin 2005). Topics and questions are tried out first and then changed based on the responses the researcher got (Rubin & Rubin 2005). However, the questions should stay close to the interviewees' knowledge and what they are willing to talk about in order that the research outcome stays credible (Rubin & Rubin 2005). Therefore not all individuals may have more than two interviews but this is no problem for the responsive interviewing model which's design is tolerant of mistakes and facilitates the correction of false steps (Rubin & Rubin 2005).

Interview data will be nearly exclusively used in this study, aside from what I know about China and Germany and its environment, the company my interviewees come from, my background knowledge and experiences gained in being a team member in Sino-German teams. The interviews will be recorded with a digital audio recorder. Sections containing useful data will be transcribed by me from the audio files, along pauses which will be only mentioned with brackets in the transcription, following Rubin and Rubin's (2005) approach. Also other non-verbal details will only be added to the transcription if they could help for the analysis (Rubin & Rubin, 2005). The focus of the transcription should be the content of what was said rather than a deep examination how it was said.

During the interviews also notes of important points will be taken but only to a degree to which it does not interrupt the flow of conversation (Richards 2003). Such field notes should not try to contain everything possible but rather help to focus on important points to remember, such as ideas and observations to follow up in further interviews (Agar 1996). After each interview these notes will be reviewed and any reflections about the interview itself will be written in a reflective log. At this stage a flow and connection between what was said will be searched and also answers to the research questions may be already found. Kvale (1996) agrees to look already for such answers in form of first impressions, "based on the interviewer's empathetic access to the meanings communicated [...] provide a valuable context for the later analysis of transcripts" (Kvale 2008, p. 56). The reflective log will also help for analysis as a source for ideas and will give space to describe my role in the research process (Agar 1996). It may contain my reactions to the interviewees and the interview in general.

Data Analysis


The approach to analysis is aimed to reduce the story of each conversation partner to some main meanings, patterns and themes (Spradley 1979, Merriam 1998, Kvale 1996) in order to find concepts which can be compared regarding similarities and differences with other concepts found in the analyzed interviews with others.

Data in a qualitative study should be simultaneously analyzed with data collection (Merriam 1998). At the beginning of its conduction the researcher knows the problem and selects a sample to collect data in order to address the problem (Merriam 1998). At this stage the researcher does not know "what will be discovered, what or whom to concentrate on, or what the final analysis will be like" (Merriam 1998, p. 162). The final analysis will depend on the collected data and the analysis which is simultaneously done with the collection process (Merriam 1998). Merriam (1998) argues for a simultaneous analysis because without such an ongoing analysis the data can be unfocused and repetitious. Therefore I will try to find patterns and connections in the stories of the interviewees from the first to the last interviewee. Spradley (1979) agrees with the advantages of an ethnographic study, like this one, which simultaneously collects data and analyzes it, but he reminds the researcher that such an approach requires constant feedback from one stage to another. Found connections and patterns should be grounded in the data I recorded (Glaser & Strauss 1967). Glaser and Strauss (1967) also agree that after each case of data collection the key concepts that arose from the data should be noted and they focus on constant comparison of such concepts. I will compare the interview data I will have got with the upcoming data in order to note dominating patterns and concepts. To sum up Rubin and Rubin (2005) suggest looking for concepts and themes in the questions asked, ideas frequently mentioned in responses, in indirect communication such as emotions or tone and by comparing interviews to each other. Again, such judgement about dominating patterns or concepts is subjective and just one view of a reality. It also has to kept in mind that the meanings of what people say do not directly lead to a certain reality (Silverman 2000), because meaning is also found in how the words are commonly used as expressed by Wittenstein (as cited in Silverman 2000). As mentioned before such an understood meaning is influenced by both conversation partners and my aim to learn how successful Sino-German teamwork is conducted.

Another reason for continuous data analyzing is that it protects the researcher to be overwhelmed by the data volume in form of a lot of interview transcript pages (Merriam 1998). However, before the researcher reads such transcripts interpretations already occur. According to Kvale (1996) the interpretive process begins when the interviewees describe their experience with a researcher, continues when the interviewees see meaning in the experience and when the researcher condenses and interprets the meaning of what the interviewees said, and gives based on this a certain answer or following question. Only then the interpretation continues with transcription (Kvale 1996).

Specifics of narrative analysis

As the data out of the interviews can be seen as narratives there are certain methods for the analysis of narrative research. Such methods often look for common themes in the stories of the interviewees (Polkinghorne 1988), and this is I want to find as mentioned above. In other words the stories of current or former team members of Sino-German teams are 'narratively coded' (Clandinin & Connelly 2000). Besides, such analysis fits to cross-cultural-research (Bahk-Halberg 2007). The coding will be achieved during and after the interviews by listening to the records, transcribing them and reviewing the transcriptions. The initial research questions developed from the literature review maybe adjusted and refined by what I am learning from the interviews. Therefore it also has to be admitted that initial patterns are given by these questions. By coding the interview text patterns and relationships can be uncovered (Clandinin & Conelly 2000), but Clandinin and Conelly (2000) suggest to go further by looking into meanings of what the interviewees said. Mishler (1986) aggress on the potential of narratives to express meanings.

As above mentioned approach mainly refers to narratives as a kind of 'content' there are scholars that also suggest analyzing the 'form' (Johnstone 2002). The gain of insights below the surface of the words of my conversation partner, or in other words, of what is said between the lines, is the main objective of such a often called 'discourse analysis' which is also often referred in descriptions of narrative analysis. However, as I am not sticking on the terminology I appreciate its value for this study and include it in the narrative analysis as it may also help to uncover main themes and patterns. Many scholars suggest 'discourse analysis' as well as other methods for analysis of spoken and written communication (Silvermann 2000). The applied discourse analysis in this study draws on Johnstone (2002) which sees the goal of discourse more descriptive than critical. This assumes, like many other scholars, that pure description is possible and desirable (Johnstone 2002) and less critical than in other approaches to discourse analysis (Van Dijk 2001 as cited in Bahk-Halberg 2007). However, critic is applied by critical reading and listening which leads to questioning the status quo (Johnstone 2002). The descriptions in this study's analysis will follow Johnstone's (2002) suggestion how to structure a narrative account. Narrative clauses can be grouped into types depending on how such a narrative clause describes an event: Abstract, orientation, complication action, coda, evaluation, implication, credibility, causality, assigning praise or blame, viewpoint, objectivity and resolution (Johnstone 2002). Such stories will even be structured by the interviewees' themselves, temporally, culturally, in terms of relevance, causality and in terms of their own identity due to the interviewee's idea what is culturally acceptable as narration (Gergen 2001 as cited in Bahk-Halberg 2007). This need to be kept in mind when structuring a narrative account.


Lincoln and Guba (1985) advocate for 'data triangulation' by validating each piece of information against at least two sources or reach methods. Other scholar argue similar by suggesting more than two investigators, sources of data or methods (Merriam 1998). This study was conducted by one investigator and has only one main method to collect data (interviews), but included multiple sources (interviewees). Schwandt (2001) argues for this form of triangulation, because 'different sources or methods of data must necessarily converge on or be aggregated to reveal the truth'. However, others do not agree with this assumption by saying that there are more than three sides from which to see the world (Richardson 2000 as cited in Bahk-Halberg 2007) or saying that the aim of 'validity' sticks to the positivistic philosophy (Janesick 2000 as cited in Bahk-Halberg 2007). This discussion will not be further developed by this study but I will keep in mind that the amount of my interviewees includes different perspectives. Additionally I should analyze my data from more than my own perspective. Other perspectives are provided by my research journal (Bahk-Halberg 2007) and by handing-over my interview transcriptions to my interviewees in order that they check whether they agree with my transcription or whether they want revise their expressions and thinking based on the entire interview or second interview.