Mixed Methods Research to understand Gender Inequality in Tourism through a Feminist Post-constructionist Lens

Abstract

The purpose of this article is to analyze how mixed methods research can be framed in feminist studies. Contentions in mixed methods research are approached, and the claims of the advocators and opponents to mixed methods research are analyzed. Contentions in feminist approach are also analyzed. Although feminist theorizing has been reluctant to the use of mixed methods research, it is claimed that combining quantitative and qualitative methodologies can contribute to the feminist goals of social justice and equality. Feminist post-constructionism is proposed as a good thinking tool for framing mixed methods research in feminist studies. By simultaneously embracing and transgressing previous theorizing, it overcomes the irreconcilability of quantitative and qualitative approaches. Finally, an example of a feminist mixed methods research design is analyzed. Using a mixed methods approach, this study analyzed how gendering processes influence the career and life paths of women who reach top-level management positions in the Portuguese tourism sector.

Headnote

Abstract

The purpose of this article is to analyze how mixed methods research can be framed in feminist studies. Contentions in mixed methods research are approached, and the claims of the advocators and opponents to mixed methods research are analyzed. Contentions in feminist approach are also analyzed. Although feminist theorizing has been reluctant to the use of mixed methods research, it is claimed that combining quantitative and qualitative methodologies can contribute to the feminist goals of social justice and equality. Feminist post-constructionism is proposed as a good thinking tool for framing mixed methods research in feminist studies. By simultaneously embracing and transgressing previous theorizing, it overcomes the irreconcilability of quantitative and qualitative approaches. Finally, an example of a feminist mixed methods research design is analyzed. Using a mixed methods approach, this study analyzed how gendering processes influence the career and life paths of women who reach top-level management positions in the Portuguese tourism sector.

Keywords: Mixed methods research, Feminist post-construcionism, Feminist methodologies, Gender and tourism, Research design

Introduction

Feminist research can be defined as research that brings the experiences of women to the center stage, and seeks to expose and change gender inequalities, to empower women and to improve their status or material reality. According to Letherby (2003), feminist research has "a political commitment to produce useful knowledge that will make a difference in women's lives through social and individual change" (Letherby, 2003, p. 4). Although the combination of both quantitative and qualitative approaches may contribute to more powerful studies on gender inequality (Hodgkin, 2008), mixed method studies in feminist research are still scarce (e.g. Hesse-Biber, 2012).

Mixed methods research has been defined as the third major research approach, along with qualitative and quantitative research (Johnson & Onwuegbuzie, 2004; Johnson et al., 2007). According to some authors, combining quantitative and qualitative methods may contribute to expand one's understanding of research problem, thus leading to superior research (Johnson & Onwuegbuzie, 2004; Johnson et al., 2007). However, the literature suggests that there is a 'paradigm war' concerning mixed methods research (Denzin & Lincoln, 2011b). Both qualitative and quantitative purists have criticized mixed methods research, claiming that it mixes viewpoints that are irreconcilable (Johnson & Onwuegbuzie, 2004).

In feminist research, the ontological and epistemological beliefs of some feminist strands-i.e. beliefs in the nature of reality and in the relationship between the researcher and that being researched-clash with the materialities of mixed methods research. Hence, this article aims to analyze how mixed methods research can be used in feminist studies. Feminist post-constructionism is proposed as a good 'thinking tool' in feminist mixed methods research, dvin order to overcome the apparent irreconcilability of quantitative and qualitative methods. In the first section of this article, contentions in mixed methods research are approached, and the claims of the advocators and opponents to mixed methods research are analyzed. The second section focuses on contentions in feminist research. Although feminist theorizing has been reluctant to the use of mixed methods research, it is claimed that combining quantitative and qualitative methodologies can contribute to achieve feminist goals. In the third section, I propose that feminist postconstructionism, due to its transgressing and embracing of different feminist theories, may be a good thinking tool for framing mixed methods research in feminist studies. Finally, an example of a feminist mixed methods research design is analyzed, and the advantages of combining qualitative and quantitative methods are highlighted. This study analyses gendering processes in the context of top-level management in the Portuguese tourism sector. Gendering processes are implicit attitudes, behaviors, values, organizing processes, and even internalized gender identities that enact patters of female submission and male dominance, and that impact men and women differently (Hearn, 2000).

Contentions in mixed methods research

Some authors believe that the combination of quantitative and qualitative approaches (i.e. mixed methods research) may provide a better understanding of research problems and a more efficient answer to the research questions than if either approach is used alone (Creswell, 2011; Teddlie & Tashakkori, 2011). For them, mixed methods research is characterized by an emphasis on the research question, by a rejection of dichotomies and by paradigm pluralism (Denzin, 2012). Some authors consider that mixed methods research is more than just the mixing of methods-it is a methodology, which involves both methods of inquiry and philosophical assumptions that guide data collection and analysis.

However, mixed methods research has also been criticized. Firstly, some individuals argue that methods are attached to paradigms and that it is not possible to reconcile such competing discourses in an unproblematic way-the 'incompatibility thesis' (Howe, 2004). The advocators of this view see paradigms as 'having discrete and impermeable boundaries' (Creswell, 2011, p. 75), and argue that mixed methods research attempts to mix viewpoints that are irreconcilable. Hence, they link quantitative methodologies to paradigms that believe in the scientific method (positivism and post-positivism), and qualitative methodologies to paradigms that reject the scientific method and believe that knowledge is a social construction (postmodernism, poststructuralism, interpretivism, constructionism).

A criticism addressed to the traditional mixed methods movement is that it assigns qualitative methods to the exploration phase of the research process, thus subordinating them to confirmatory quantitative methods. In fact, mixed methods have been more frequently used in deductive research, where theories are tested and the purpose of qualitative data is only to enhance quantitative results (Hesse-Biber, 2012). In such mixed methods approaches, the transformative and dialogical dimensions of qualitative research are decreased, since stakeholders are excluded from active participation and dialogue in the research (Howe, 2004).

However, other authors contend that the combination of quantitative and qualitative methods is not incompatible either at the practical or at the epistemological levels. Guba and Lincoln (2005) claim that methods can be to a certain extent 'delinked' from paradigms, and that elements from paradigms might be cautiously blended together in a study. For Johnson and Onwuegbuzie (2004), a paradigm does not impose specific methods for data collection and analysis. In fact, there has been growing support by scholars for mixed method approaches (Creswell, 2011; Ghiara, 2020; Hodgkin, 2008; Teddlie & Tashakkori, 2011).

Moreover, different types of mixed-method designs are possible, depending on whether qualitative and quantitative paradigms are granted equal status or one of them is granted a dominant status, and whether the quantitative and qualitative phases occur sequentially or are undertaken concurrently (Johnson & Onwuegbuzie, 2004). It is possible for the qualitative approach to be dominant in mixed methods research, instead of being subordinated to the quantitative part of a study. Johnson, Onwuegbuzie, and Turner (2007) provide the following definition for qualitative dominant mixed methods research:

'Qualitative dominant mixed methods research is the type of mixed research in which one relies on a qualitative, constructivistpoststructuralist-critical view of the research process, while concurrently recognizing that the addition of quantitative data and approaches are likely to benefit most research projects' (p. 124).

The theorizing developed in this article concerns qualitative dominant mixed methods research.

According to Bryman (2007), quantitative and qualitative findings should be mutually informative, in order to construct a negotiated account of what they mean together. While the triangulation metaphor has tended to concentrate on whether findings are mutually reinforcing or irreconcilable, mixed methods research is not a mere exercise to test findings against each other. It tries to bring together to the debate both components, while negotiating their findings (Bryman, 2007). In this sense, Freshwater (2007) advocates that mixed method researchers adopt a 'sense of incompleteness' and acknowledge the messiness of mixed methods research instead of focusing on 'fixing meaning' and 'moving towards incontestability' (p. 277-278).

Contentions in feminist research

Feminists have been somewhat reluctant to quantitative methods, given that feminism has been traditionally more closely related to the postmodern paradigm (Hodgkin, 2008). Postmodern approaches to knowledge emphasize how reality is constituted in language. Hence, postmodern feminism contends that the final truth about what is good or just is unattainable (Lykke 2010a). It seeks to deconstruct established truths and rejects a 'grand narrative' capable of explaining power relations universally (Aitchison, 2005a; Harding, 1986). In this perspective, 'objectivity is a chimera: a mythological creature that never existed, save in the imaginations of those who believe that knowing can be separated from the knower' (Lincoln et al., 2011, p. 122). Therefore, postmodern feminism rejects traditional science.

This criticism against traditional positivist science has not been exclusively leveled by feminism. Different paradigms and strands of thought have addressed several criticisms to traditional science. Post-positivists criticize the positivist belief that there is a single reality separated from the knower, that the researcher is totally separated from the research, and that total objectivity or universal knowledge are attainable (Clark, 1998; Kuhn, 1970). Interpretivists and constructionists also criticize positivism for ignoring the interpretations and social constructions of the social world. They contend that there is no objective knowledge that is independent of thinking (Grbich, 2013). Influenced by these views, postmodern and poststructuralist thinkers have challenged traditional science by problematizing and deconstructing the 'apparently stable and secure foundations of scientific knowledge production' (Lykke, 2010a, p. 131). They have decentered the knowing subject and interrogated it as the core of knowledge production, and deemed absolute truth and objective knowledge unattainable (Denzin & Lincoln, 2011b; Ramazanoglu & Holland, 2009). Moreover, poststructuralists see reality as produced in language (Lykke, 2010a). This notion has implied a shift from the analysis of reality to the analysis and deconstruction of the ways we think about what exists (Ramazanoglu & Holland, 2009).

Feminism has generated considerable synergies together with postmodernism and poststructuralism, which influenced feminist thought to such an extent that 'feminist postmodernism' is one part of Harding's (1986) tripartite classification of feminist epistemology (Lykke, 2010b). Like feminists, postmodernists and poststructuralists also deny 'grand narratives' and focus on 'small stories' and specific local contexts instead of making claims to universal knowledge (Lyotard, 1979). Postmodernism and poststructuralism also explore diverse, multiple and fragmented identities, while interrogating essential and unproblematized identities such as man/woman, black/white or gay/straight. They have opened up new ways of thinking about gender free from modern thought and the scientific method. Therefore, qualitative approaches are privileged in postmodern and poststructuralist research, while mixed methods and triangulation are not very common in feminist research (Hesse-Biber, 2012).

Despite the influence of postmodernism and poststructuralism on feminist thought, feminism has also transgressed them in its criticisms to traditional science (Collis & Hussey, 2005; Ramazanoglu & Holland, 2009). The criticisms addressed to positivism by feminism are related with ideals of social justice, according to which neutral 'value-free' science is neither possible nor desirable (Denzin, 2012). On the one hand, feminism is engaged with the claim that women should have the same rights as men at the political, economic and social levels and, on the other hand, it claims that all findings have political implications. Theory, epistemology and ethics have an explicit political positioning, which enables the researcher to explore the relations between knowledge and power, to address and question the established 'truths', as well as the commitment of accountability to a community of 'women with moral and political interests in common' (Ramazanoglu & Holland, 2009, p. 14). In addition, a feminist approach brings women's voices to the forefront (Hodgkin, 2008; Letherby, 2003).

Despite the importance of postmodernist and constructionist feminist thought, postmodernism's emphasis on language, discursivity, as well as political and moral relativism, collides with the explicit political positioning of feminism, and with feminist ideas of liberation, emancipation and justice (Lykke, 2010a). Hence, certain viewpoints of these two paradigms are in contention with feminism.

Action against injustices and power imbalances needs to be based on some kind of valid knowledge, accuracy and evidence. Feminist ethics, political action and emancipation are inconsistent with an exclusively relativist position, otherwise feminist knowledge will not lead to action (Ramazanoglu & Holland, 2009, p. 41). Gender is not exclusively confined in language. The connection between knowledge, experience and reality needs to be addressed (Ramazanoglu & Holland, 2009).

In this regard, feminist research can derive many advantages from the combination of both quantitative and qualitative approaches. If the aim of feminist research is to achieve social change towards the improvement of women's lives, quantitative methods used in combination with qualitative ones can give a more powerful voice to women's experiences - as contended by Hodgkin (2008, p. 297), 'the big picture accompanied by the personal story can bring both depth and texture to a study'. While a qualitative approach gives a voice to a group of women and their experiences, quantitative methods indicate the extent and pattern of the inequalities (Brannen, 1992).

Quantitative data in mixed methods research is more likely to be regarded as reliable and valid, and thus be heard in the policy arena, hence contributing to convince non-feminist decision makers more easily. Mixed methods research is also more likely to overcome the shortcomings of both QUAN-only and QUAL-only approaches, since in large quantitative studies women's voices remain unheard, and in qualitative studies there are often problems with poor representation and the tendency to overgeneralize (Hodgkin, 2008).

Feminist post-constructionism

It was concluded in the previous section that mixed methods research can contribute to achieve feminist goals. Nonetheless, it remains problematic to frame mixed methods research in terms of feminist paradigms and epistemologies, given that the prevailing feminist postmodern paradigm privileges qualitative approaches, due to its ontological and epistemological beliefs. Hence, this section proposes feminist post-constructionism, as a good 'thinking tool' in feminist mixed methods research to overcome the seeming irreconcilability of quantitative and qualitative approaches.

Nina Lykke suggests 'post-constructionism' as an umbrella term and a 'temporarily useful framework' (p. 134) to bring together converging trends in feminist thought that address both continuities and discontinuities with feminist constructionism and postmodernism, and that neither fall 'back into the traps of biologic determinism or cultural essentialism, nor (...) leave bodily matter and biologies behind, in a critically untheorized limbo' (Lykke, 2010b, p. 132). The theories that Lykke (2010b) clusters under the name 'post-constructionism' have as a common denominator the transgression of postmodern (anti-)epistemologies, a strong commitment to ethics and a belief that there is a link between ontology, epistemology and ethics. In addition, there is an insistence on both constructedness and objectivity (Barad, 2007; Lykke, 2010b).

The basic beliefs of feminist post-constructionism are briefly presented in Error! Reference source not found.. Since feminist post-constructionism is not a definition 'carved in stone' (Lykke, 2010a, p. 134), the table should be read as an attempt to group together the beliefs underlying some of the authors' writings that Lykke labelled as 'feminist postconstructionist', i.e. Haraway (1991) and Barad (2007), and of Lykke's own writings on feminist post-constructionism as a temporary thinking tool.

Some of the aspects of feminist post-constructionism outlined in Error! Reference source not found. are not directly relevant for my argument of why feminist postconstructionism can be a suitable framework for mixed methods research. Therefore, the subsequent analysis carried out will focus instead on the aspects that are more central to my argument.

Firstly, feminist post-constructionism recognizes that there is a 'real' and irreducible world 'out there', as long as the subjective position of the subject is accounted for (Haraway, 1991; Lykke, 2010b). This also allows for partial objectivity. Whereas postmodernism contend that the researcher cannot be separated from what he or she knows, the belief of feminist postconstructionist theorizing is that, although both entities are not separable, it is possible to make a momentary cut between the knower and the known. Haraway (1991) reclaims the concept of scientific objectivity, albeit an objectivity that is only possible through partial perspective. Due to this difference in philosophical beliefs, feminist post-constructionist ontological and epistemological stance becomes much more compatible with mixed methods research than postmodern or poststructuralist feminist theorizing.

Since only a 'momentary cut' is possible, feminist post-constructionism remains a more adequate framework for qualitative dominant than for quantitative dominant mixed methods research. While inspired by postmodern and constructionist approaches to science, Donna Haraway also transgresses them when she claims that it is possible to speak about the 'real world' as long as the situated and embodied position of the subject researcher is accounted for (Lykke, 2010b). The concept of 'situated knowledge' advocates a focus on specific local contexts, instead of the universalization and construction of grand narratives on women, such as the universalizing narrative of 'women's path to emancipation' (Haraway, 1991; Lykke, 2010a). Hence, for Haraway (1991), feminist objectivity should be partial and should result in embodied and 'situated knowledge'. This emphasis on embodied and situated knowledge highlights the relevance of qualitative approaches, e.g., through narratives and interviews.

Nevertheless, the explicit rejection of the dichotomy between quantitative and qualitative methods allows for the integration of quantitative approaches. Since it is possible to make a momentary cut between the knower and the known, it is not unsuited to integrate a postpositivist research paradigm as a frame for the quantitative part of a study within a broader feminist framework. Despite the fundamental differences between post-positivism and feminist post-constructionism, post-positivism also rejects the dichotomy between qualitative and quantitative paradigms, believes in the attainment of some form of truth and that the researcher cannot totally detach herself/himself from the research process-although she or he tries to attempt to gain some objectivity by doing that. Hence, the similarity between this post-positivist idea with the 'momentary cut' proposed by Barad (2007) is clear.

Moreover, feminist post-constructionist trends support the importance of rethinking the relationship between materiality and discourse, and mixed methods research may be an appropriate tool for bringing both together by combining quantitative and qualitative approaches. It is enriching for a study to compare the materiality of women's living conditions with their discourses on their living conditions, and how they make sense of their lives. For Aitchison (2005b) and Aitchison (2005a), material constraints should not be left unanalyzed by critical theorizing.

Barad (2003) denied the separation between ontology and epistemology, since 'knowing is a matter of part of the world making itself intelligible to another part' (p. 829). This way, the thinking human subject is part of the world rather than over it. In addition, ontoepistemological practices are always onto-ethico-epistemological because they have ethical consequences. Therefore, in feminist post-constructionist theorizing, social justice, political and ethical concerns are not separate from research. For Haraway (1988) and Lykke (2010a), the researcher should be accountable and justify why some pictures and accounts of reality are better than others. For Barad (2007), not all knowledge should be treated as equally false. This way, if mixed methods are used in transformative research for social justice, they may contribute to the purpose of achieving feminist goals.

To sum up, feminist post-constructionism as a methodological tool can both embrace and transgress previous theorizing, namely social constructionism, postmodernism and poststructuralism (Lykke, 2010a). It can be a good 'thinking tool' for framing mixed methods research and overcoming the apparent irreconcilability of quantitative and qualitative methods. In addition, by not separating ethics from epistemology, its ethico-onto-epistemology allows for partial objectivity and situated knowledge.

Lykke claimed that it is important to resource to several analytical methods, both qualitative and quantitative, from within a broad range of disciplines, ranging from arts and humanities to biology and medicine: it is essential to 'mobilize a maximum of unorthodox creativity' and combine this 'with a rigorous, scholarly endeavor to seek partial objectivity and moral accountability' (Lykke, 2010a, p. 161). Therefore, feminist post-constructionism is not only suited to mixed methods research, but it may also be appropriate and flexible for investigation that relies on the combination of different and multidisciplinary methods (Lykke, 2010a).

An Example of a Feminist Mixed Methods Research Design

In this section, an example of a feminist mixed methods research design is analyzed. This research relied on qualitative dominant mixed methods research. This type of mixedmethod research is symbolized by Johnson and Onwuegbuzie (2004) as quan^QUAL and by Johnson et al. (2007) as QUAL +quan. This methodological approach was chosen because it allows for the integration of elements of a different nature, which is believed to result in a wider and deeper understanding of the phenomena under analysis. It was considered that a qualitative dominant research design was the most appropriate answer to the overarching feminist research question of the empirical study (Johnson & Onwuegbuzie, 2004; Johnson et al., 2007): How do gendering processes influence the career and life paths of women who reach top-level management positions in the Portuguese tourism sector?

The quantitative and qualitative approaches adopted in this study are outlined in .

The qualitative approach gave voice to research participants and contributed to a deeper understanding of their experiences. The quantitative approach was complementary and, by representing the population, it indicated the extent and pattern of the inequalities (Brannen, 1992). This implied that there were different paradigms underlying each of these methodological approaches. While the qualitative approach had a post-constructionist paradigm, the quantitative part had a post-positivist paradigm. However, as analyzed in the previous section, they are not incompatible or mutually exclusive. Since a study can integrate various paradigms related to different phases of the research design (Creswell, 2011), this study began with a quantitative part reflecting a post-positivist leaning, and in the second part it shifted to a post-constructionist paradigm and used qualitative methods, i.e. interviews and thematic analysis with narrative elements. This way, the broader feminist perspective of this study was underpinned by combining a quantitative and a qualitative approach.

In the quantitative part of the study, secondary data were used, namely an official longitudinal matched employer-employee micro-data set (GEE/MEE, 2012). The aim of this analysis was to characterize tourism businesses and gendered patterns in tourism employment, namely in senior management. Exploratory and confirmatory statistical analyzes were carried out with these data with the softwares SPSS (Statistical Package for the Social Sciences) and Stata.

The qualitative part of the study is the core of this investigation. The researcher carried out semi-structured in-depth interviews with female top-level managers in travel agencies, tour operators and hotels in Portugal, and performed thematic analysis combined with narrative elements with the primary data obtained. During thematic analysis, there was a mix of deductive and inductive approaches. This analysis had a constructionist leaning, since it regarded meanings not as inherent in the experiences, but as socially constructed (Braun & Clarke, 2006). Despite the emphasis on constructedness, there was also a belief in the possibility of 'partial objectivity' and 'situated knowledge'. Some elements of narrative analysis were also introduced during the interpretation of the interviews, in order to capture the fluidity and variability of individual experiences (F. Clark, Carlson, & Polkinghorne, 1997; Czarniawska, 2004).

The research had seven sub-questions of investigation:

Q1. What are the gendered patterns of employment and gender inequalities observed in the tourism sector, namely among senior managers, and how widespread are they?

Q2. How do women top managers themselves articulate the circumstances under which they have reached the top?

Q3. How do gendering processes affect women in tourism organizations and how can the identification of gendering processes contribute to a better understanding of gender in the tourism sector?

Q4. What is the influence of the family context and the traditional family ideology on women's construction as mothers and as economically active beings?

Q5. How do women do, undo or re-do gender in the different contexts of their lives?

Q6. How do women accept or challenge the gender order in the different contexts of their lives?

Q7. To what extent is the tourism sector women-friendly vs. reinforcing of gendered patterns of employment?

The purpose of the first sub-question (Q1) was to provide a broader picture of the context of investigation. The quantitative approach was used in order to test the hypotheses developed and to answer the research question.

The next five sub-questions were primarily related with the qualitative approach. In the second research question (Q2), women's interpretations of their own career paths were brought to the center stage, and in the fifth sub-question (Q5) it was also women's perceptions of how they do, undo or re-do gender differently in the different contexts of their lives that were the focus of analysis. In the third, fourth and sixth sub-questions (Q3,4,6), women's discourses and views of their career and life paths were interpreted from a gender-aware, feminist and poststructuralist perspective. Acker's (1990, 1992, 2012) theorizing on the gendered organization and gendered processes framed the analysis carried out in Q3. In the fourth subquestion (Q4), women's discourses and narratives about their family situation and the gendered division of tasks at home were the focus of analysis. It was investigated to what extent the traditional family ideology permeated women's discourses and their own internal gender constructions, or whether they were critical of it.

According to Bryman (2007), mixed methods research studies often treat the quantitative and qualitative components as separate domains, instead of bringing their findings together. However, bringing these findings together may offer important insights (Bryman, 2007). Hence, the last sub-question (Q7) was a concluding overall research question that intertwined the conclusions from both the quantitative and the qualitative approaches.

A Bridge Between the Qualitative and the Quantitative Data

The purpose of this section is to establish a bridge between some of the findings obtained in both the quantitative and qualitative parts of the study, and highlight how they inform each other. There is analysis of how the mix of quantitative and qualitative approaches contributed both to enrich the study and to achieve feminist goals.

Although both the qualitative and the quantitative data suggested that male senior managers are advantaged in relation to their female counterparts concerning several aspects, with the quantitative data it was possible to have an overview of the extent of some of these inequalities.

The mix of quantitative and qualitative methodologies made it possible to overcome some of the limitations of the present study. One of the problems with the selection of research participants was related to the lack of ethnical diversity. When the snowball sampling technique was attempted, the research participants had difficulties to identify foreign women in top-level positions, and they could not recall practically any foreign women in similar positions, particularly non-whites. In fact, the statistical data confirmed the existence of only a few dozen foreign women at the managerial level in tourism organizations in Portugal, which justified the lack of diversity in the sample.

This lack of diversity is in itself an important result, and it raises the question of whether there is only a very small percentage of foreign women in the tourism workforce, or whether they are simply underrepresented at the top and unevenly distributed through the hierarchy. Even if this question had been asked during the interviews, quantitative data can provide a much more solid answer to this question. Quantitative data revealed that foreign female employees, particularly from former Portuguese colonies, are particularly concentrated in low-skilled jobs, which may be a niche for marginalized workers.

Quantitative data also revealed that the hotel sector is based on a large army of female cleaners and chambermaids. Although there is a marked pay gap between female senior managers and their male counterparts, the pay gap between management in general and lower hierarchical positions is particularly wide. This points to the importance of not leaving class relations unanalyzed when studying the tourism workforce, as well as to the importance of intersectional analysis. In the hotel sector, there is an abundance of low-skilled jobs, which are largely feminized. This confirms the existence of 'pink ghetto' jobs in the sector (Woods & Viehland, 2000). This way, quantitative data reinforced the importance of further intersectional analysis in tourism and gender studies concerning gender, class and ethnicity.

Vertical and horizontal segregation of the tourism sector along gender lines is confirmed in the quantitative data, and reinforced in research participants' discourses. Besides, quantitative analysis revealed that it was in the highest earnings echelons that women were more disproportionately underrepresented, while in the qualitative component it was concluded that the higher the positions reached by women, the more inequalities they observed.

Although the differences between the hotel and the travel sectors were not clear with the qualitative data, the quantitative data showed that class and gender inequalities are more widespread in the hotel industry, while in travel agencies and tour operators, gender inequalities are mostly visible at the top of the pyramid, but not at its bottom.

While both the quantitative and qualitative approaches used revealed that gendered patterns of employment persist in tourism, quantitative results pointed to the existence of a 'male advantage' in the tourism sector. In fact, the decomposition of employees' earnings using quantitative methods (the Oaxaca-Blinder decomposition) revealed that the higher earnings obtained by men seem to more the result of a 'male advantage' (i.e. positive discrimination towards men) rather than the result of women being paid salaries below the average. This means that it is not women who are underpaid, but rather men who receive additional bonus and premia.

It is important to highlight that the purpose of this mixed methods research was not to confirm qualitative interpretations with quantitative results. However, as analyzed in this section, the quantitative results contributed to provide a wider picture of inequalities.

The qualitative approach not only covered a much wider range of topics than the quantitative one, but also contributed to explain certain quantitative results. For example, in the quantitative component of the study, it was observed that the gender pay gap is widest in senior management positions, and that this gap could not be explained by the variables used in the model. This means that the gender pay gap among senior managers cannot be explained by differences between men and women in terms of education, tenure, occupational distribution or age. However, qualitative analysis provided some insights on factors that could explain it. Several interviewees mentioned how some women in top-level management positions are paid less than their male counterparts because salaries are negotiated individually, and that it is common practice to under-reward women. Hence, the individual negotiation of managers' salaries is very likely to be widening the gap between men and women's earnings. As one interviewee underlined, it is a 'cultural issue' to pay male managers more than female managers. Other participants also observed how male managers earn higher salaries than their female counterparts in their own organizations or in other organizations. Hence, discrimination may explain most of the gender pay gap in senior management.

Conclusion

Several criticisms have been addressed to mixed methods research, such as that it attempts to mix viewpoints that cannot be conciliated. Another criticism is that it decreases the transformative dimension of qualitative research because qualitative methods become subordinated to confirmatory quantitative methods in mixed methods studies.

Despite this, there has been growing support to mixed methods research. Some authors have argued that methods can be delinked from paradigms, and that it is possible that qualitative methods take precedence in mixed methods research (Guba and Lincoln, 2005).

In addition, although mixed methods research has not been common in feminist enquiry, it can be useful in transformative investigation for social justice. The combination of both approaches can give a more powerful voice to women's experiences by indicating the extent and pattern of inequalities (Brannen, 1992). This way, this kind of research is more likely to be heard in the policy arena. Hence, mixed methods research can contribute to achieve feminist goals.

However, epistemological and ontological issues have been marginalized to a significant extent in discussions of mixed methods research. This article aimed to contribute to foster these discussions by introducing feminist post-constructionism as a useful lens from which to consider and analyze mixed methods research in feminist studies.

Although postmodernism and poststructuralism have influenced feminist thought considerably, their emphasis on language, and political and moral relativism collides with feminist ideas of emancipation and social justice (Lykke, 2010a). Despite the collision between these viewpoints, feminist post-constructionism offers a possibility of conciliating them from a philosophical point of view. Lykke (2010a, 2010b) proposed feminist post-constructionism as a useful temporary framework that embraces the anti-foundationalism of postmodern philosophy while simultaneously transgressing it by embracing more material analyzes. Therefore, it can provide a good framework for discussing ontological, epistemological, methodological and axiological aspects in feminist studies that resource to mixed methods.

Feminist post-constructionist trends in feminist theorizing do not regard the researcher and the researched as inseparable, but recognize the possibility of partial objectivity, and of making a momentary cut between the knower and the known. They also reject the dichotomy between qualitative and quantitative methods. Besides, feminist post-constructionist philosophical beliefs are not incompatible with the integration of a post-positivist paradigm in the quantitative part of a mixed methods research study. In addition, by advocating the importance of rethinking the relationship between materiality and discourse, feminist postconstructionism offers more enriching possibilities for ontological and epistemological discussions than pragmatism, which most commonly underpins mixed methods research.

Furthermore, in feminist post-constructionist theorizing, ethical concerns are not separated from ontological and epistemological ones. Therefore, it provides an appropriate lens for framing mixed methods research that is transformative and aims to attain social justice and feminist goals.

An example of a feminist mixed methods research design was presented in the last section of the article, and some theoretical aspects discussed in this article were analyzed in relation to this specific example. It can be concluded that feminist post-constructionist ways of theorizing can contribute to deeper methodological discussions of mixed methods research, and that they can provide a suitable framework for such studies.

Feminist post-constructionist thinking can open up the way for transformative mixed methods research that aims to achieve feminist goals. It allows both for the attainment of 'situated knowledge' and the analysis of a broader setting. By disrupting the dichotomy of quantitative and qualitative methods, it invites a meaningful and dialogical integration of quantitative and qualitative results. We suggest that further discussions of the possibilities and limitations of this approach are discussed in future studies.

References

References

Aitchison, C. (2005a). Feminist and Gender Perspectives in Leisure and Tourism Research. In B. W. Ritchie, P. Burns & C. Palmer (Eds.), Tourism Research Methods: Integrating Theory with Practice (pp. 21-36). Oxfordshire: CABI Publishing.

Aitchison, C. (2005b). Feminist and gender perspectives in Tourism studies. The socialcultural nexus of critical and cultural theories. Tourist Studies, 5(5), 207-224.

Barad, K. (2003). Posthumanist performativity: toward an understanding of how matter comes to matter. Signs, 28(3), 801-831.

Barad, K. (2007). Meeting the Universe Halfway: Quantum Physics and the Entanglement of Matter and Meaning. Durkham & London: Duke University Press.

Brannen, J. (1992). Mixing methods: Qualitative and quantitative research. Aldershot, UK and Brookfield, VT: Avebury.

Braun, V., & Clarke, V. (2006). Using thematic analysis in psychology. Qualitative Research in Psychology, 3(2), 77-101. doi: 10.1191/1478088706qp063oa

Bryman, A. (2007). Barriers to Integrating Quantitative and Qualitative Research. Journal of Mixed Methods Research, 1(8), 8-22.

Clark, A. M. (1998). The qualitative-quantitative debate: moving from positivism and confrontation to post-positivism and reconciliation. Journal of Advanced Nursing, 27(6), 1242-1249. doi: 10.1046/j.1365-2648.1998.00651.x

Clark, F., Carlson, M., & Polkinghorne, D. (1997). The Legitimacy of Life History and Narrative Approaches in the Study of Occupation. The American Journal of Occupational Therapy, 51(4), 313-317.

Collis, J., & Hussey, R. (2005). Pesquisa em Administração. Um guiapráticopara alunos de graduaçao e pós-graduação [Business Research A Practical Guide for Undergraduate and Postgraduate Students]. Porto Alegre: Bookman.

Creswell, J. W. (2011). Controversies in Mixed Method Research. In N. K. Denzin & Y. S. Lincoln (Eds.), The SAGE Handbook of Qualitative Research (pp. 269-284). Thousand Oaks: Sage.

Czarniawska, B. (2004). Narratives in Social Science Research London: Sage Publications.

Denzin, N. K. (2012). Triangulation 2.0. Journal of Mixed Methods Research, 6(2), 80-88. doi: 10.1177/1558689812437186

Denzin, N. K., & Lincoln, Y. S. (2011a). Preface. In N. K. Denzin & Y. S. Lincoln (Eds.), The Sage handbook of qualitative research (pp. ix-xvi). Thousand Oaks: Sage.

Denzin, N. K., & Lincoln, Y. S. (2011b). Introduction: The Discipline and Practice of Qualitative Research. In N. K. Denzin & Y. S. Lincoln (Eds.), The SAGE Handbook of Qualitative Research (pp. 1-19). Thousand Oaks: Sage.

Freshwater, D. (2007). Reading mixed methods research: Contexts for criticism. Journal of Mixed Methods Research, 1(2), 134-145.

GEE/MEE. (2012). Relatório Único/Quadros de Pessoal [Single Report/ Staff]. Lisboa: Gabinete de Estratégia e Estudos - Ministério da Economia.

Ghiara, V. (2020). Disambiguating the Role of Paradigms in Mixed Methods Research. Journal of Mixed Methods Research, 14(1), 11-25. https://doi.org/10.1177/1558689818819928

Grbich, C. (2013). Qualitative Data Analysis. London: SAGE.

Guba, E. G., & Lincoln, Y. S. (2005). Paradigmatic controversies, contradictions, and emerging confluences. In N. K. Denzin & Y. S. Lincoln (Eds.), The SAGE handbook of qualitative research (pp. 191-215). Thousand Oaks: Sage.

Haraway, D. (1988). Situated Knowledges: The Science Question in Feminism and the Privilege of Partial Perspective. Feminist Studies, 14(3), 575-599. doi: 10.2307/3178066

Haraway, D. (1991). Simians, cyborgs and women: the reinvention of nature. New York: Routledge.

Hearn, J. (2000). On the Complexity of Feminist Intervention in Organizations. Organization, 7(4), 609-624.

Hesse-Biber, S. (2012). Feminist Approaches to Triangulation: Uncovering Subjugated Knowledge and Fostering Social Change in Mixed Methods Research. Journal of Mixed Methods Research, 6(2), 137-146. doi: 10.1177/1558689812437184

Hodgkin, S. (2008). Telling It All: A Story of Women's Social Capital Using a Mixed Methods Approach. Journal of Mixed Methods Research, 2(4), 296-316. doi: 10.1177/1558689808321641

Howe, K. R. (2004). A critique of experimentalism. Qualitative Inquiry, 10, 42-61.

Jennings, G. R. (2005). Interviewing: a Focus on Qualitative Techniques. In B. W. Ritchie, P. Burns & C. Palmer (Eds.), Tourism Research Methods: Integrating Theory with Practice (pp. 99-117). Oxfordshire: CABI Publishing.

Johnson, R. B., & Onwuegbuzie, A. J. (2004). Mixed Methods Research: A Research Paradigm Whose Time Has Come. Educational Researcher, 33(7), 14-26.

Johnson, R. B., Onwuegbuzie, A. J., & Turner, L. A. (2007). Toward a Definition of Mixed Methods Research. Journal of Mixed Methods Research, 1(2), 112-133.

Kuhn, T. (1970). The Structure of Scientific Revolutions. Chicago: The University of Chicago Press.

Kvale, S., & Brinkmann, S. (2009). InterViews: Learning the Craft of Qualitative Research Interviewing (Second Edition ed.). Thousand Oaks: SAGE.

Letherby, G. (2003). Feminist research in Theory and Practice. Buckingham: Open University Press.

Lincoln, Y. S., Lynham, S. A., & Guba, E. G. (2011). Paradigmatic Controversies, Contradictions, and Emerging Confluencies, revisited In N. K. Denzin & Y. S. Lincoln (Eds.), The SAGE Handbook of Qualitative Research (pp. 97-128). Thousand Oaks: Sage.

Lykke, N. (2010a). Feminist studies. A guide to intersectional theory, methodology and writing. New York: Routledge.

Lykke, N. (2010b). The Timeliness of Post-Constructionism. NORA - Nordic Journal of Feminist and Gender Research, 18(2), 131-136. doi: 10.1080/08038741003757760

Lyotard, J.-F. (1979). The Postmodern Condition: A Report on Knowledge (1984 ed.). Manchester: Manchester University Press.

Ramazanoglu, C., & Holland, J. (2009). Feminist Methodology: Challenges and Choices. London: SAGE.

Teddlie, C., & Tashakkori, A. (2011). Mixed Methods Research: Contemporary Issues in an Emerging Field. In N. K. Denzin & Y. S. Lincoln (Eds.), The SAGE Handbook of Qualitative Research (pp. 285-299). Thousand Oaks: Sage.

Woods, R., & Viehland, D. (2000). Women in hotel management. Cornell Hotel & Restaurant Administration Quarterly, 41(5), 51-54.