The Significance of Gender in Phenomenological Nursing Research

The aim of this paper is to discuss in the light of phenomenological philosophy, whether it can be argued that men and women have different lifeworlds and how this may legitimize the segregation of men and women in empirical nursing research. We analyzed peer-reviewed papers from 2003-2012 and scrutinized the arguments used for dividing men and women into separate groups in empirical nursing studies based on phenomenology. We identified 24 studies using gender segregation and posed the following questions: 1. What is the investigated phenomenon as explicated by the authors? 2. What arguments do the authors use when dividing participants into gender specific groups? The analysis showed that a variety of phenomena were investigated that were all related to a specific medical condition. None appeared to be gender-specific, though the authors argued for a sole focus on either women or men. The most common argument for segregating men and women were reference to earlier studies. A few studies had references to methodology and/or philosophy as argument for a segregation of men and women. Arguments for gender segregation in empirical nursing studies based on a phenomenological approach tend to build on the conviction that experiences of health related phenomena are gendered. However, it seems to be difficult to identify conclusive arguments for this division within phenomenological philosophy. Therefore we recommend that segregation should be used with caution. Otherwise other research approaches may be more suitable.


Introduction
It has become common to accept that men and women have different worlds. This conviction gathered momentum with the publication in 1992 of 'Men are from Mars, Women are from Venus,' written by the American author and relationship counsellor John Gray (Gray, 1992). During the last 20 years, it has become widely accepted in the Western world that differences in gender are significant, and that they justify participation in gender-specific activities. From early childhood, there is a tendency to assume that even small children should be divided into genderspecific groups so they can live out their inner boy or girl in their games or in their learning. In the adult world, the assumed differences between the two sexes remain unchallenged, resulting in a growing number of gender-specific activities such as travels, courses, festivals, and TV programs specifically arranged for the one sex only. Due to men and women's assumed different values, interests, feelings and experiences, it seems natural that more and more activities are addressed to a specific gender. Also, in empirical research, men and women are commonly separated into specific gender groups. This may be well founded as there are clearly biological differences. However, the tendency seems to have entered phenomenological research and this may conflict with the philosophical beliefs of phenomenology that human experiences are not gender specific. While some researchers provide a rationale for dividing groups along gender lines, others do not appear to question or provide justification for this in their research. So, the aim of this paper is to discuss if it is relevant or possible to transfer the assumed differences between genders from our everyday life into the world of phenomenological research.

Paying Attention to the Arguments Legitimising Segregation
The notion of gender can be considered from different viewpoints. First and foremost, there are biological differences between men and women. For instance, the female body contains more fat than the male body, whereas the male body contains more muscular tissue than the female body (Nielsen & Bojsen-Moller, 2012). Also, men are generally bigger and stronger than women. Sexual characteristics are other self-evident differences leading to a natural segregation between men and women. Thus, from a biological point of view, the human body comes in two basic variations and the differences may make it imperative that the sexes are divided into two groups in biological research. However, from a constructivist perspective, gender may also be considered an acquired performance of gendered behaviour, which is what we commonly consider as femininity and masculinity. For Butler (1990), gender is an illusion since there is no gendered self prior to its acts. Rather, gender is a socially constructed relationship, which is produced and reproduced through people's actions (Butler, 1990). Gender, therefore, is not something we 'are' but rather something we 'do' and may consist of multiple forms of masculinities and femininities. In a classical phenomenological text, Young (1980) describes different modalities of the feminine bodily existence and claims that there is no specific feminine essence that all women have by virtue of being biologically female. Femininity is, rather, a set of structures, which delimit the typical situation of being a woman in a particular society, as well as the typical way this situation is lived by the women themselves. Defined as such, it is not necessary that there are distinctive structures and behaviour typical of women's situation (Young, 1980). The above understandings suggest that using gender, as a sampling criterium in scientific studies requires researchers to precisely explicate their understanding of the notion of gender.
Nevertheless, this claim about gender as biology, a construction or a modality of the situation often seems to be ignored within qualitative, empirical research where men and women may be divided into gender-specific groups without any explicit reason. In some studies, it is evident that men and women must be separated into two groups. Biological differences between the sexes imply that some issues are tied to a specific gender and, therefore, it is only possible to include one sex in studies related to these issues. A prime example is a study of the re-constructing of masculinity following radical prostatectomy for prostate cancer by Gannon, K., Guerro-Blanco, M., Patel, A., & Abel, P.(2010). Also, some diseases are caused by a genetic disorder and therefore only exist in one gender. In a study by Dreyer, P. S., Steffensen, B. F., & Pedersen, B. D. (2010) it was only possible to include men because Duchenne muscular dystrophy does not exist among women. Not only biological differences can substantiate arguments for solely including one sex. In some cases, biology creates the conditions for human experiences of a particular issue and it becomes the law of nature that the issue can only be investigated within one sex. For instance, Armstrong (2001) investigated fathers' experiences of pregnancy after a prior perinatal loss referring to the mothers' and fathers' different attachment to the unborn baby. Also, Hall (2004) investigated the experience of being a grandmother in a separate study. Another example of a study dealing with gender-specific experiences is a study of women's experiences of legal abortion ( van Dijk, M. G., Arellano Mendoza, L. J., Arangure Peraza, A. G., Toriz Prado, A. A., Krumholz, A., & Yam, E. A., 2011) where it is obvious that men cannot be included. The sexes may also be segregated with reference to gender-specific characteristics making the selected gender particularly suitable as participants in the study. A study of complementary and alternative medicine only focused on women due to their relatively high use of this kind of treatment (Lindenmeyer, Jamie, Griffiths, & Legare, 2011). Another reason to focus solely on one gender is, when a given disease is prevalent in one sex for which reason the opposite sex may be excluded. Since the majority of people with osteoporosis are women, Emmett, C. L., Redmond, N. M., Peters, T. J., Clarke, S., Shepstone, L., Lenaghan, E., & Shaw, A. R. (2012) interviewed 30 women about their experiences with this illness. The segregation of gender in research may also be motivated by a comparison between the sexes. For example, Leegaard, M., Naden, D., & Fagermoen, M. S. (2008) stated that women might have a lower pain threshold than men, which makes them better participants than men in a study of postoperative pain. In many studies, it is assumed that experiences are gendered. Thus, with reference to the lack of knowledge derived from one sex, researchers argue in favour of including only this sex to get insight into this particular gender's experiences. For instance a number of studies dealing with the experience of different heart diseases exclusively included women (Beal, Stuifbergen, & Volker, 2012;Isaksson, Brulin, Eliasson, Naslund, & Zingmark, 2012;Rolfe, Sutton, Landry, Sternberg, & Price, 2010) referring to a one-sided focus on men's experiences. A higher or increasing mortality among one sex may also be used as an argument for including solely this gender. In a study on information and support needs of Faroese women with an acute coronary syndrome, the researchers argued that more women than men die from this disease why men were excluded Askham, J., Kuhn, L., Frederiksen, K., Davidson, P., Edward, K. L., & Worrall-Carter, L. (2010).. The opposite situation-the selected gender tends to survive the longest-may also be used as an argument for choosing only one-sex as participants. Foster & Neville (2010) describe how women over the age of 85 experience their life in New Zealand, referring to the growing group of old women in this part of the world. Eilertsen, G., Kirkevold, M., & Bjork, I. T. (2010) presents a similar argument in a paper on recovery after stroke among elderly women in Norway.
Other studies declare to explore and compare the experiences of men and women (McCaughan, Prue, Parahoo, McIlfatrick, & McKenna, 2011) though acknowledging that masculinity and femininity are constructed and reconstructed from context to context, which may affect the validity of the gender specific findings (McCaughan, Prue, Parahoo, McIlfatrick, & McKenna, 2011). Galdas, P. M., Johnson, J. L., Percy, M. E., & Ratner, P. A. (2010) segregated men and women and anticipated to find distinctive patterns of help-seeking behaviour for cardiac symptoms, but found no stereotypical masculine or feminine attitudes emerging from their data. Some of the studies do not define sex or gender as an inclusion criterion but despite that, due to other characteristics they end up with a study where only one gender is represented. For instance, Sgorbini, M., O'Brien, L., & Jackson, D. (2009) explored the issues surrounding chronic hepatitis C and do not account for the gender of the participants. In summary, several arguments are presented when men and women are segregated within empirical nursing research, using phenomenological and other approaches. The question we shall pursue in this paper is, if gender segregation within phenomenological research has been critically scrutinised.
Phenomenological philosophy does not work with a clear gender distinction making it self-evident to distinguish men from women. According to Gallagher and Zahavi (2010), the founder of the phenomenological tradition Edmund Husserl (1859Husserl ( -1938 created the basis for an epistemology that cuts through the dichotomy and instead focuses on the relationship between the subject (the scientist) and the object (the "thing" in focus of science). Also, within empirical phenomenological approaches, it is argued that phenomenology is an epistemology that cuts through all old ontological and epistemological dichotomies, such as subjectivism and objectivism, materialism and idealism (Dahlberg, Dahlberg, & Nyström, 2008) The question is if empirical researchers are creating a new dichotomy by gender segregation in phenomenological research and if there are philosophical arguments within the phenomenological tradition that would legitimise the segregation of men and women in empirical phenomenological studies. Based on the assumption that presentations of empirical phenomenological studies in peer-reviewed journals have consequences for the conduct of future research and direct implications for the legitimacy in nursing science; the aim of this paper is twofold. First, we seek to systematically scrutinise if there is a tendency to divide men and women into different groups in empirical nursing studies based on phenomenology. Second, we will discuss whether the arguments for a segregation of men and women in phenomenological research can be legitimised in light of phenomenological philosophy.

Scrutinising Empirical Nursing Research
We analysed peer-reviewed papers from 2003-2012. This was considered a sufficient time frame to ensure our ability to identify a broad variation of arguments used for gender segregation and discuss issues of importance. The search included electronic databases CINAHL, PubMed, Embase, Scopus, PsychInfo, Google Scholar as well as a chain-search. The included studies had nurse researchers listed as the author and were limited to publications in English. Besides that, the only pre-established criterion was that the empirical study be gender-specific and contain the term "phenomenology" in the method paragraph ( Figure 1). We searched using the keywords "phenomenology", "phenomenological approach", "gender", "sex", "women", "men", "nursing research" and "qualitative research". One hundred and seventy-two articles were included for further reading. A flowchart containing the in-and exclusion criteria can be found in Figure 1. Twenty-four papers met the criteria and were included in the analysis where we tried to find answers to the following questions: 1. What is the investigated phenomenon as explicated by the authors? 2. What arguments do the authors use when dividing participants into gender-specific groups? With reference to methodology is was argued that to ensure homogeneity of the sample, as well as to limit the effect of confounding factors (i.e. gender) women who identified themselves as primary caregivers of children with autism were chosen. Furthermore, with reference to philosophy it was argued that it was necessary to connect with those "who have directly experienced the phenomenon of interest". Nordgren L., Asp M. & Fagerberg I. (2008) "Support as experienced by men living with heart failure in middle age: A phenomenological study".
The meaning of support as experienced by men living with heart failure.
With reference to earlier studies men's and women's life-situations differ when living with heart failure. With reference to philosophy it is the essential meanings of a specific phenomenon, i.e. the meanings that define a phenomenon as itself. In the section about purposive sampling it is mentioned that only women are included in the study.

Visekruna, S. (2010)
"Experiences of selfmanagement among young women with type 1 diabetes mellitus". With reference to earlier studies women posses characteristics and experiences unique and different from men. Menstruation, pregnancy, puberty and menopause may present challenges for self-management, a prerequisite for those living with Type 1 diabetes mellitus.

Ware J & Raval H. (2010) "A qualitative investigation of fathers' experiences of looking after a child with a lifelimiting illness, in process and retrospect".
Interpretative phenomenological method as described by Smith. In-depths interviews with 8 fathers.
Experiences of looking after a child with a lifelimiting illness.
With reference to earlier studies men engage differently than women. Father's show less outward distress than mothers and also strive to maintain a feeling of control through their work outside the home. Significant gaps remain in the literature regarding father's perspective. With reference to population based studies older women in Korea makes up 59.3% of the total elderly population and very little is known about the health behaviours of this group.

Identifying the Arguments for Segregation
In the analysis of the 24 articles, we identified the phenomenon under investigation and searched for the arguments that were used for gender segregation. In particular, we aimed to identify if or how the authors based their arguments for gender segregation on phenomenological philosophy (table 1). As illustrated in table 1, 10 of the 24 articles were published from 2003 -2009 (table 1:  article 1 , 3, 7, 8, 11, 12, 13, 15, 17 and 20) and 14 from 2010-2012 (table 1: article 2 , 4, 5, 6, 9, 10, 14, 16, 18, 19, 21, 22, 23 and 24)"All studies were conducted in the Western world and reflected a variety of commonly used phenomenological approaches with references to Benner, Colaizzi, Dahlberg et al., Giorgi, Kvale, Lindseth & Norberg, Merleau-Ponty, Ricoeur, Spiegelberg, and Van Manen. A majority referred to Giorgi, Lindseth and Norberg and Van Manen. Most studies argue for the phenomenological method, but do not include references to phenomenological philosophy as part of the argument.
Various phenomena were investigated. They are phrased as either "the meaning of", or as "the lived experience of", for example "self-management", "support", "self-injurious behaviour", or "body and life after a stroke" (table 1). All studies represented phenomena related to different medical conditions and to empirically derived questions. None of the phenomena investigated appear to be gender-specific. However, in all 24 studies the authors argue for a sole focus on either women or men (table 1). Eighteen studies describe female experiences and 6 masculine experiences. The only explanation of the higher number of studies on women is, according to the authors (Juuso, Skar, Olsson, & Soderberg, 2011;Ockander & Timpka, 2003;Olsson, Skar, & Soderberg, 2010;Rutberg & Ohrling, 2012;Thurang, Fagerberg, Palmstierna, & Tops, 2010), that women experience increased rates some diseases than men; for example fibromyalgia, longterm sickness absence, alcohol dependency, Multiple Sclerosis and migraine (table 1). Though acknowledging in the background section that the phenomenon under investigation is shared by women and men, some authors state that the phenomenon contains aspects that may be genderspecific and so a design including only one sex was chosen (Kvigne, & Kirkevold, 2003;Kvigne, Kirkevold, & Gjengedal, 2004;Markoulakis, Fletcher, & Bryden, 2012;Scroggs, Shattell, & Cowling, 2010;Tanyi 2006). A more implicit assumption of gender differences is seen in the study by Visekruna , Edge, & Keeping-Burke, (2010) based on the pre-understanding that women possess characteristics and experiences unique to their sex that may affect the lived experience of self-management of Type 1 Diabetes.
The analysis also shows references to earlier studies are the most common ground for segregating men and women. Within this argument, we found three reasons for the segregation lack of knowledge, higher prevalence in one sex or assumed difference in experience. Some authors (Juuso, Skär, Olsson, & Soderberg, 2011;MacAdam, Huuva, & Bertero, 2011;Olsson, Skar, & Soderberg, 2010;Ransom, Siler, Peters, & Maurer, 2005;Thurang, Fagerberg, Palmstierna, & Tops, 2010;Ware & Raval, 2007) state that the segregation aims at closing a gap in the literature and, therefore, it is necessary exclusively to focus on the under-researched sex; for example, the study by Ware and Raval (2010), who argue that there is a lack of literature regarding fathers' perspective of looking after a child with a life-limiting illness. In a similar way, Olsson et al. (2010) argues that there is limited knowledge of the experience of feeling well as a female with Multiple Sclerosis. The second common argument refers to a higher prevalence of a diagnosis among one sex in earlier studies (Juuso, Skär, Olsson, & Soderberg, 2011;Ockander & Timpka, 2003;Olsson, Skär, & Soderberg, 2010;Rutberg & Ohrling, 2012;Schoppmann, Schröck, Schnepp, & Büscher, 2007;Thurang, Fagerberg, Palmstierna, & Tops, 2010;Yang & Yang, 2011). For example in the study by Juuso et al. (2011) it is stated that the majority of people with fibromyalgia are women. The third reason for segregating men and women, that was described was the argument of an assumed difference in the lived experience between women and men (Daggett, 2002;Flinck, Astedt-Kurki, & Paavilainen, 2008;Herning, Hansen, Bygbjerg, & Lindhardt, 2011;MacAdam, Huuva, & Bertero, 2011;Nordgren, Asp, & Fagerberg, 2008;Notter & Burnard, 2006;Sundin, Bruce, & Barremo, 2010;Visekruna, Edge, & Keeping-Burke, 2010;Ware & Raval, 2007). As an example Nordgren et al. (2008) state that men and women's life situations differ when living with heart failure. Likewise MacAdam et al. (2011) argue for a difference in men and women's experience of sexuality after having a child.
Few of the included studies reference methodology and/or philosophy (de Witt, Ploeg, & Black, 2010;Kvigne, Gjengedal, & Kirkevold, 2003;Kvigne & Kirkevold, 2003;Markoulakis, Fletcher, & Bryden, 2012;Nordgren, Asp, & Fagerberg, 2008) when arguing for a segregation of men and women. Markoulakis et al. (2012) who investigated female caregivers' lived experience of being caregiver to children with autism refer to methodology when they argue for homogeneity of the sample. In the study by de Witt et al. (2010), purposive sampling was used to achieve "information-rich participants" (p. 1700). Gender segregation is not mentioned in the inclusion criteria, but only women were included. When authors base the arguments for segregation on philosophy, they refer to "the phenomenon itself" (Nordgren, Fagerberg 2008, p. 1346. In a similar way Markoulakis et al. (2012) state that it is necessary to connect with those 'who have directly experienced the phenomenon of interest. Kvigne et al. (2004) refer to the philosophers Merleau-Ponty (1962/1994) and de Beauvoir (1974, who state that feminist phenomenological theorists have discussed how the female body's interaction with its surroundings leads to experiences and expectations that are different from male experiences. Therefore, given the body's fundamental position in creating one's lifeworld, there was a need to explore women's experiences of their bodies after a stroke (Kvigne, Gjengedal, & Kirkevold, 2004).

Study Limitations
In the analysis, we found a majority of female experiences. But, we did not identify any references to a feminist research approach. This may be explained by our focus on nursing research. Therefore, it might have been interesting to broaden our inclusion criteria and use more keywords in the search strategy such as "feminist phenomenology." Furthermore, we did not put any emphasis in contextual analysis, i.e. in what cultural, religious and political setting the studies were performed. This could have shed light on the impact of these aspects on the perception of gender.

Gender Segregation Conflicts with the Traditional Philosophical Beliefs of Phenomenology
Our findings show that when the published research consists of one-sex studies, it seems to be natural for researchers to fill the gap with studies of the opposite sex without assessing the nature of the phenomenon or the propriety of the research method. Hence, the authors do not seem to question if it is relevant or possible to segregate men and women in a study with a phenomenological approach. However, in two studies, it is described as incidental that only women participated in the study and the authors consider this one-sided gender perspective a limitation of the study (de Witt, Ploeg, & Black, 2010;Schoppmann, Schröck, Schnepp, & Büscher, 2007). This underlines an ambiguity among nurse researchers regarding the meaning of a gender perspective on the findings in empirical phenomenological research and how empirical studies can account for gender. Although our findings show that the authors use different arguments for the segregation of men and women, the underlying principle as seems to be that men and women have different lifeworlds. This may be true in some cases, but the question is how, when and to what extent this standpoint can be legitimised within phenomenological philosophy. Although Husserl (1954Husserl ( /1970 himself took up ''the problem of the sexes'' as a question for phenomenological investigation, the question of gender has not been given much attention in phenomenological philosophy (Oksala, 2006). Phenomenological philosophy does not argue that men and women have different lifeworlds. Everybody has his or her own individual lifeworld (Gallagher, & Zahavi, 2010). A phenomenological consequence of different lifeworlds among men and women would be a possibility to outline both a male and a female essence. Further, phenomenological philosophy does not work with a "traditional" view of masculinity or femininity as a stable set of attributes distinguishing men from women (Young 1980;Oksala, 2006). As stressed by de Beauvoir (1974), although using the words 'women' or 'feminine,' 'she' does not refer to an archetype or a no changeless essence. Young (1980) claims that Merleau-Ponty's (2004/1945) account of the relation of the lived body to its world applies to any human existence in a general way. At a more specific level, there is a particular style of bodily comportment, which is typically feminine or masculine. This style consists of particular modalities of the structures and conditions of the body's existence in the world. Young (1980) also stresses that some humans transcend the typical situation and definition of their gender in various degrees and respects. Accordingly, we will argue that gender segregation conflicts with the philosophical beliefs of phenomenology. From a classical phenomenological perspective, gender segregation in empirical phenomenological studies relates to the natural attitude. The natural attitude has the unquestionable axiom about gender that human beings come in two genders and that the male-female dichotomy is natural (Hawkesworth, 1997). As stated in the introduction, we do not necessarily question differences related to gender in our everyday life. It is part of our un-reflected,shared understanding that men and women may have different needs and values. Our findings show many of the phenomenological empirical studies were built on the pre-understanding. The investigated phenomenon is experienced differently among men and women (e.g. Daggett, 2002;Flinck, Astedt-Kurki, & Paavilainen, 2008;MacAdam, Huuva, & Bertero, 2011). This premise also formed the foundation in a study that aimed to describe differences in help-seeking behaviour between men and women with cardiac symptoms, which has been described in earlier studies (Galdas, Johnson, Percy, & Ratner, 2010). Interestingly, they did not find any stereotypical masculine or feminine attitudes. Thus, gender segregation as a premise in empirical phenomenological research raises questions as to whether researcher openness is taken sufficiently into account. The ability to wonder and the importance of researcher openness is one of the unique characters of phenomenological research (Dahlberg, Dahlberg, & Nyström, 2008;Giorgi, 2009). As explained by Oksala (2006), if a phenomenological researcher tries to understand what a woman is, by only investigating female experiences means already having assumed that which one seeks to explain. Besides researcher openness, variation among participants is also mentioned as a cornerstone in phenomenological research (Dahlberg, Dahlberg, & Nyström, 2008). Hence, it is obvious to ask if gender segregation disregards an important variation of a phenomenon that can be experienced by men as well as women. For instance, our findings show that several studies explored the lived experience of phenomena like "loss of spouse", "support", "intimate partner violence", "pain", etc. Using differences between men and women's perceptions as a starting point for investigating these phenomena, the researchers do not take into account, or seem to ignore, both the importance of researcher openness and variation among participants. Young (1980) acknowledges the gendered modalities, but dismisses the idea that it is possible to identify a gendered essence. Also Oksala (2006) claims that "the focus on the body is simply too limited a framework to support a philosophical understanding of gender…" (p. 233), as the philosophical meaning of gender cannot be reduced to the differences between two types of living bodies. Oksala (2006) suggests a modification of the classical phenomenology -a postphenomenological reading -that is able to deal with the notion of gender. Not to identify a concept of gender or a biological investigation of the body, but to understand the constitution of gender as a gendered experience. To achieve this, Oksala (2006) leaves out the traditional understanding of epoché and replaces it with the hermeneutical circle, "This means understanding epoché not as total, universal and complete, but as an endless, circular and always partial task" (Oksala 2006, p. 238). Starting by reading anthropological and sociological investigations and medical reports etc. while critically reflecting on one's own understanding of these texts may reveal something previously hidden about the constitution of gender. Oksala (2006) stresses that, "The aim is not to find eidetic structures of for instance female experience that characterize all women whether they come from Norway or Nigeria: it is rather to seek the structures that are constitutive of the sense of normal in our homeworld" (p.239). So, it seems that examining lived experiences from a gender perspective is only possible considering the cultural and historical context of the study and taking one's own limited experience into account.

Being Aware of the Nature of the Phenomenon
Classical phenomenological philosophy does not support the existence of gendered experiences. However, it is possible to identify some gendered modalities of being in the world that are closely connected to the concrete situation (e.g. the contemporary society). But, these modalities are not determinative of a human being with a particular gender. A post-phenomenological approach suggested by Oksala (2006), leaving out the notion of epoché, offers a comparable, but yet not conclusive understanding of gender based on existing anthropological, sociological and medical knowledge supplemented by critical reflection on one's own interpretation of the material. Such an interpretive hermeneutic approach makes it possible to investigate gender as an aspect of a given phenomenon.
When empirical phenomenological researchers use references to earlier studies as the most commonly used argument for gender segregation, they risk building their new research on studies that neglect or fail to address phenomenological philosophy's dissociation with dichotomies. Since it is difficult to identify definitive arguments for gender segregation within phenomenological philosophy, segregation should be used with caution in studies based on phenomenological approaches. Focusing exclusively on one gender is suitable only if the feature of the phenomenon of interest calls for it.