Using mixed methods approaches to study families and relationships

Abstract

Mixed methods research—methodologies that synthesize qualitative and quantitative approaches in the design, collection, analysis, and dissemination of research related to a specific topic or aim—is increasingly common, offering innovative empirical insight into families and relationships. We first elaborate on our definition of mixed methods research, emphasizing that there is significant heterogeneity within mixed methods approaches to studying families and relationships. Second, we discuss benefits of mixed methods projects within family and relationship research, including theory building and innovation. Third, we provide practical suggestions for designing and implementing a mixed methods project, highlighting useful resources for researchers as they develop research questions, plan designs, collect and analyze data, and disseminate findings. We emphasize the unique opportunities from abductive analytic approaches for mixed methods researchers and point to the need for reflexivity. Fourth, we consider common obstacles associated with disseminating mixed methods research and explain why family researchers need “mixed methods literacy” regardless of their research paradigm. Finally, we identify key areas of future growth for mixed methods researchers. We advocate that understanding mixed methods research has practical benefits, even for researchers not using these approaches. To cohesively build—and critique—our knowledge of families and relationships, family and relationship researchers across paradigms should be familiar with the basic tenets, strengths, and limitations of qualitative, quantitative, and mixed methods research.

INTRODUCTION
Researchers use a diverse set of methodologies and epistemological approaches to study families and relationships, including exploring how social forces and contexts shape family life and how families and relationships matter for multiple types of outcomes (Adamson et al., 2022). As one example, both quantitative and qualitative methodological approaches have yielded critical knowledge about the role that families play in transmitting socioeconomic advantage across generations. Statistical analysis of a large longitudinal dataset identifies that parental financial investments in children are a driver of widening class gaps across generations (Schneider et al., 2018), and ethnographic research on working-class and middle-class families shows how parenting methods, themselves products of cultural and social capital disparities, are linked to these growing economic divides (Lareau, 2011). Scientific understandings of families benefit from layering varied and diverse methods.

To draw on the unique insights gained from both qualitative and quantitative approaches, family researchers sometimes use a mixed methods approach (Suitor & Gilligan, 2022). We define mixed methods research as any study that synthesizes qualitative and quantitative approaches in the design, collection, analysis, and dissemination of research regarding specific research aims. Our review of family and relationship research articles within the Journal of Marriage and Family (JMF), Journal of Family Issues, Journal of Family Psychology, and Family Process demonstrates that mixed method articles are relatively uncommon in social science journals focusing on families and relationships. For example, JMF published only 15 mixed methods articles between 2010 and 2023. In an Online Appendix, we share a table of these articles, including research topic, quantitative and qualitative approaches, and justification of the use of mixed methods. We suggest that the relatively low number of mixed methods publications within family social science journals fails to represent the growing number of mixed methods research designs within family and relationship studies. This growth is evidenced by a recent systematic review of the number of research projects integrating aspects of qualitative and quantitative design throughout the research process (see Timans et al., 2019, although this review not specific to family and relationship studies), even if not resulting in mixed methods publications (e.g., a single journal publication or book involving analysis of both qualitative and quantitative methods).

We argue that an increase in mixed methods approaches—alongside the continued use of independent qualitative and quantitative methods—would positively impact family and relationship research. Our aim is to describe these benefits and articulate why they matter for our understandings of families and relationships, as well as to provide recommendations regarding conducting mixed methods research. We first elaborate on our definition of mixed methods research, including discussing the diversity of approaches beyond both the “survey plus in-depth interview” design and the deductive and positivistic orientations that are most common, while pushing back against assumptions that qualitative and quantitative methods are necessarily epistemologically opposed. Second, we discuss the main benefits of mixed methods research within family studies, such as its usefulness for theory building and its ability to triangulate differing methodologies. Third, we provide practical steps to design and implement mixed methods projects, including strategies for developing research questions and for research dissemination, and point specifically to the usefulness of an abductive approach when moving back and forth between quantitative and qualitative approaches. Fourth, we discuss obstacles commonly encountered by family and relationship researchers when trying to disseminate mixed methods research, and we promote the development of “mixed methods literacy,” useful for family researchers regardless of their own research stage or paradigm. Finally, we provide ideas and reflections for the future of mixed methods research within family and relationship studies, including suggestions for critical theory building and open science. Within each section, we draw on insights from a diverse selection of mixed methods empirical studies focused on families and relationships with a goal of improving knowledge of, literacy around, and approaches to mixed methods research.

DEFINING MIXED METHODS RESEARCH
Mixed methods research describes projects oriented toward a cohesive set of research questions, topics, or aims conducted by either an individual researcher or a clearly delineated research team that integrates qualitative and quantitative approaches throughout the research process (Johnson et al., 2007). This term is often distinguished from multimethod research, typically defined as “research using more than one approach within a qualitative or quantitative paradigm” (Salmons, 2015, p. 522, emphasis ours)—for example, an ethnographic approach integrated with qualitative analysis of in-depth interviews. Notably, there is some debate about the exact meaning of multimethod research or even the specific distinction between qualitative and quantitative approaches (Anguera et al., 2018). In contrast, the general consensus is that mixed methods research requires the intentional combination of quantitative and qualitative research within the same study (Creswell & Clark, 2017; Johnson et al., 2007).

Specific methodological approaches
Within research on family and relationships, mixed methods approaches are characterized by a considerable amount of heterogeneity, reflecting the specific methodologies that can be represented in mixed methods projects (e.g., experiments, content analysis, ethnographies, surveys, in-depth interviews) and researchers’ different epistemological approaches. As we emphasize throughout this review, there is no one “right” way to conduct mixed methods research, and the social scientific community benefits when researchers draw on a plurality of methods and epistemologies.

Even as we highlight the strength of diversity within mixed methods research on families and relationships, we do note that most mixed methods projects—including those typically published in JMF (see our Online Appendix) and other family and relationship journals—involve the integration of quantitative analysis of survey data and qualitative analysis of in-depth interviews and often expect that the qualitative and quantitative components will be complementary. For example, a mixed method longitudinal study using data from the Within-Family Difference Study focused on dynamics within adult child–parent relationships, examining why some adult children become estranged from their mothers (Gilligan et al., 2015). They state that their quantitative analysis (survey data from 2,013 mother–adult child pairs) identifies statistical patterns of association between variables, whereas the qualitative analysis (in-depth interview data with 74 of those mothers who were estranged from their children) provides information regarding the processes undergirding these statistical patterns. A key quantitative finding in their study is that mother–adult child dyads with greater degrees of value dissimilarity are more likely to be estranged, and the qualitative analysis reveals narratives in which mothers describe how conflicts with their children about topics such as religion and finances lead to eventual estrangement. There are other ways of combining survey and in-depth interview data in mixed methods studies focused on families and relationships, as shown in the Online Appendix, and discussed below, but this practice of using the qualitative analysis to provide additional context to statistical findings is arguably the most common.

Mixed methods researchers also make use of other combinations of qualitative and quantitative data collection and analysis beyond surveys and in-depth interviews, drawing on the unique strengths of different methodological tools. For example, Mistry et al. (2008) apply a mixed methods approach that includes ethnographic fieldwork to understand how low-income mothers navigate financial precarity and how it shapes family well-being. Through two qualitative approaches (ethnographic fieldwork and in-depth interviews), the study provides insight into financial pressures experienced by mothers. The team pairs these qualitative methods with statistical analysis of a nationally representative survey of mothers to examine how these financial pressures shape family functioning. In their design, they argue that the quantitative analysis extends the qualitative analysis, showing how the processes from the qualitative findings in turn impact parent and child outcomes. Methodological approaches are typically specifically chosen to capitalize on the complementary strengths of both qualitative and quantitative methods and achieve the broader goal of comprehensively addressing specific research topics.

Specific epistemological paradigms
Epistemological orientations shape how mixed methods projects are employed, including choices regarding what methods used and prioritized, timing and combination of methods, and specific implementation of methods. Based on our review of social science research currently published in family and relationship journals, mixed methods research often (implicitly) uses a positivistic paradigm, which assumes that social scientific knowledge is observable and revealed through systematic methods in an unbiased manner. This paradigm can sometimes lead to a privileging of quantitative methods within mixed methods designs. Although positivistic paradigms are prevalent within mixed methods family and relationship studies, they are not the only orientation currently used. Other paradigms, such as interpretivism or constructivism, point to how knowledge is subjective and socially constructed by multiple actors and are increasingly seen within social science research on family and relationships, yet more typical in qualitative studies rather than quantitative or mixed methods studies (McCall, 2005; Misra et al., 2021). We argue that this does not have to be the case, and there is opportunity for mixed methods studies to incorporate constructivist lenses more often into their research. Notably, critical theories—such as intersectionality and Black Feminist Thought—bring important critiques to positivistic lenses by pushing scholars, including mixed methods researchers, to note their positionality when designing and implementing protocols, interpreting results, and disseminating findings (Bowleg, 2008). For example, Black feminist scholars push against the notion that there is a universal way of constructing knowledge, instead affirming that researchers and their projects are shaped by their position within society (Brantley, 2023; Collins, 1986).

Traditionally, quantitative methods have been associated with a positivist and deductive approach to science and qualitative with a constructivist and inductive approach (Salmons, 2015), but combining the two into a single unified mixed methods project requires a more innovative and flexible epistemology and presents unique challenges and opportunities (Timans et al., 2019). We argue that mixed methods research is uniquely suited to combining these lenses, as mixed methods researchers increasingly dismiss the idea that there is inherent tension between qualitative and quantitative approaches or that these approaches are even necessarily two distinct research paradigms (Pearce, 2015; Timans et al., 2019). Rather, there is a push to embrace the benefits of mixed methods research by blending different methods with attention to the broader philosophical and practical positions of the research team (not each method individually) and with the goal of providing needed empirical support for the advancement of theory (Pearce, 2015). We advocate below for an abductive approach (Timmermans & Tavory, 2012), which can help reconcile tensions in qualitative and quantitative approaches in a mixed methods design by moving back and forth between inductive and deductive reasoning as part of the overall mixed methods project goal. We highlight below that the use of both deductive and inductive approaches may give way to more abductive mixed methods projects that can explore current concepts and patterns, while also utilized to build and expand theories in family research.

BENEFITS OF MIXED METHODS RESEARCH
We emphasize multiple benefits of mixed methods research and provide examples from social science research on families and relationships that demonstrate these benefits. Ultimately, we argue that mixed methods projects have the potential of providing novel and necessary perspectives on families and relationships, assuming researchers are using an intentional study design that builds on the strengths of all methodological components. Family and relationship scholars advocate for mixed methods research across subject areas, recognizing its multi-faceted benefits. For example, in the 2020 decade in review in JMF, authors of several articles suggested that mixed methods approaches have the potential to advance our understandings across a range of topics, including cohabitation and marriage (Sassler & Lichter, 2020), fatherhood (Schoppe-Sullivan & Fagan, 2020), ethnic-racial socialization in families (Umaña-Taylor & Hill, 2020), and families and health (Umberson & Thomeer, 2020).

We discuss three benefits that are particularly relevant for family researchers: (1) mixed methods research is useful for theory building and testing; (2) mixed methods research facilitates triangulation; and (3) mixed methods research is innovative with high potential for advancing knowledge. These benefits depend in part on the epistemological orientation of the research project, and we do not suggest that each of these benefits universally apply to all mixed methods research related to families and relationships. For example, theory building may be more salient for an inductive study but less so for a deductive study, whereas triangulation may be an important goal when using a positivist paradigm but less so when using an interpretivist paradigm. We still suggest that, regardless of epistemology, many of the benefits of mixed methods research are rooted in its ability to draw on the strengths of both qualitative and quantitative research, while potentially minimizing the impact of either methods’ limitation. Johnson and colleagues argue that the purpose of mixed methods research should be “breadth and depth of understanding and corroboration” (2007, p. 123), combining the strengths of both qualitative and quantitative analysis to reach this goal.

Theory building
Mixed methods research can be useful for theory building. Mixed methods researchers typically allow the specific research topic, setting, and population, as well as in some cases relevant theories and previous empirical work, to guide their choice of methods, rather than conforming to a preconceived or commonly used paradigm. For example, a mixed methods approach motivated by an intersectionality framework pushes researchers to consider both the microsystemic and macrosystemic levels of disadvantage, oppression, and privilege experienced by families (Harper, 2011)—a framework difficult to incorporate in solely qualitative or quantitative designs. Hassoun Ayoub et al. (2022) use a mixed methods approach driven by an intersectionality theoretical framework to examine interpersonal and systemic violence for Black mothers during COVID-19 shelter-in-place orders. They integrate their quantitative and qualitative findings to generate and expand current theory on intersectionality within the context of violence against Black women and their families.

Although theory building is often associated with inductive approaches which commonly draw on qualitative designs (Charmaz, 2006), mixed methods research tends to challenge this traditional understanding, often (but of course not always) blending deductive and inductive epistemologies by using the quantitative arm of the study for theory-testing (deductive) and the qualitative arm for theory-building (inductive) (Shim et al., 2021). For example, Agadjanian’s (2020) research on religious organizations and family life in Mozambique combines analysis of two sources of quantitative data (a household-based survey and a census of local religious congregations) and two sources of qualitative data (focus group discussions and individual in-depth interviews) to examine the persistence of polygyny within Christian communities. He develops specific hypotheses, drawing on multiple theories of organized religion and family life within sub-Saharan Africa, and tests those hypotheses using statistical analysis, revealing differences between denominations and between church members and church leaders in the prevalence and acceptance of polygyny. The qualitative analysis is more oriented toward induction, building theory regarding why and how polygyny is either rejected or accepted and legitimized within specific churches in Mozambique.

We suggest that the most successful mixed methods projects—especially those oriented toward theory development and refinement—would move back and forth between qualitative and quantitative approaches. This is a similar strategy as advocated in an abductive approach, which has become increasingly common in qualitative family research, although rarely explicitly used in mixed methods or quantitative studies (Timmermans & Tavory, 2012). An abductive approach starts with noting surprising patterns or findings within the data—evaluating whether something is surprising based on what is known from prior research or theories—and then seeks to develop an explanation or theory of this surprising pattern or finding through analysis of the data. As a qualitative example, Randles (2022) analyzed in-depth interviews with 70 mothers experiencing diaper need and developed a theory on how diaper need is a type of class and racial stratification. As an abductive study, this project drew on existing racial stratification and intersectional family justice theories to guide the study design and qualitative data analysis, going back and forth between the analysis and existing theories to refine and expand these theories.

We suggest that mixed methods family studies are especially well-suited for utilizing an abductive framework. This framework together with a mixed methods approach provides opportunities for theory building within family research. Mixed methods designs already often integrate qualitative and quantitative approaches by moving back and forth between the two sources of data and types of analyses as research questions are addressed (Morgan, 2007). Therefore, it may make sense to also move back and forth between inductive and deductive strategies and between existing theories and qualitative and quantitative data to refine and develop new theories using an abductive orientation. Because abduction is “a creative inferential process aimed at producing new hypotheses and theories based on surprising research evidence” (Timmermans & Tavory, 2012, p. 167), abduction together with mixed methods research increases the likelihood of advancing family research by developing and testing new innovative theories.

Triangulation
A second benefit is that mixed methods research facilitates the triangulation of research methods, which can contribute to more scientifically robust conclusions. Triangulation is an approach DuBois (1899) developed at the end of the 19th century that involves incorporating multiple research approaches into a study design, most typically with the goal of arriving at the most comprehensive conclusions regarding the specific research questions (Heesen et al., 2019). Although triangulation is typically associated with positivistic paradigms, it is also sometimes used within other epistemologies, especially when the goal is to provide multiple ways of knowing in order to shed light on specific positionalities. With triangulation, researchers take advantage of the strengths—and take seriously the weaknesses—of various approaches in the design of their study. By using both quantitative and qualitative methods in a mixed methods design, DuBois (1899, p. 3) advocated that researchers “may perhaps have corrected to some extent the errors of each [method].” For example, in The Negro American Family, DuBois (1908, p. 9) notes the difficulty of studying the family life and history of Black Americans given the “lamentable dearth of material” at the time. Thus, to conduct this family research, DuBois combined several quantitative sources (e.g., Census reports, Bureau of Labor reports) with a qualitative study of 32 families to provide the most comprehensive scientific understanding of Black families within the United States up to that point. As he demonstrated, triangulation is particularly important when examining a poorly understood topic or a topic that is relatively new within family research.

The benefits of methodological triangulation continue to motivate family and relationship researchers to use mixed method approaches. Smock and colleagues published a series of studies on how cohabitors make the decision to marry or not, focusing on the role of finances. For the quantitative component, they analyzed survey data to examine how the economic circumstances of both partners in a cohabiting union are associated with whether and when they got married (Smock & Manning, 1997), concluding that men’s economic resources were more important than women’s in shaping whether the couple married. For the qualitative component, analyzing in-depth interviews with 115 cohabiting young adults, they had the specific goal of triangulating “existing quantitative studies with new qualitative data to explore what positive economic circumstances mean to cohabiting men and women and how those meanings shape marriage decisions” (Smock et al., 2005, p. 684). They found that some cohabitors viewed the decision to marry as dependent on whether the man is fulfilling the breadwinner role, supporting the quantitative finding that men’s economic situations more important for the marriage transition than women’s. Triangulation continues to motivate many family and relationship researchers to use a mixed methods approach, integrating the quantitative and qualitative analysis to capitalize on the strengths of each methodological approach and thus contributing to richer and often more nuanced research conclusions.

Innovation
A third benefit of mixed methods research is that it is an innovative research method, as having multiple components within one cohesive study provides useful flexibility that facilitates novel approaches as the study progresses. As mixed methods researchers implement their qualitative and quantitative designs, they may encounter limitations or obstacles with one of the approaches but can carefully and systematically use (and even redesign) the other approaches to best address those limitations. This innovation is often typically seen when mixed methods researchers work to integrate their different methods, especially if they are integrating research methods that are rarely used together. Timans et al. (2019, p. 194), based on their overview of the current state of mixed methods research across disciplines, concluded that the rise of mixed methods research was due to the work of “academic entrepreneurs” who develop and champion new and cutting-edge approaches that build on previous best practices.

Innovation within mixed methods designs affords researchers flexibility when combining general exploration and description of a novel or hard-to-study topic with a push for theoretical depth and development (Creamer, 2021). This flexibility is necessary, particularly for critical scholars who are looking to challenge traditional notions of the scientific method. As Bowleg (2008) discusses, critical scholars have called attention to the importance of contextualizing the research process, which does not mean abandoning rigor but rather considering how the research process and findings are embedded within a larger sociohistorical context. Mixed methods research provides an opportunity to engage in the research process in this way because mixed methods analysis is often suited for research topics that may present as difficult to study through conventional methods or for topics that are relatively new or under-studied, as mentioned above when discussing triangulation. For example, Curington et al.’s (2021) book, The Dating Divide: Race and Desire in the Era of Online Romance, is based on a decade-long mixed methods project combining large-scale data from a popular dating website, archival research, and in-depth interviews to provide the first comprehensive empirical examinations of what they term “digital-sexual racism.” When they began their project, with data from as early as 2003, online dating was just beginning, so utilizing a mixed methods approach allowed for an extensive exploration into a relatively new phenomenon.

Given this innovation potential, a mixed methods approach also provides opportunities to break out of assumptions or existing paradigms that limit our understanding, especially important when pushing against positivistic ways of thinking and drawing on critical perspectives. Moore’s (2011) book, Invisible Families: Gay Identities, Relationships, and Motherhood among Black Women, combines survey methods, participant observations, and in-depth interviews to examine the experiences of Black lesbian stepfamilies. She describes her study as “a mixed methods but primarily qualitative study of the ways race influences lesbian practice and lesbian family formation for middle-class and working-class Black gay women” (Moore, 2012, p. 34), using the quantitative component of the study to provide additional context regarding the themes uncovered in the qualitative analysis. Her mixed methods results challenge previous notions of gender and power within families, which tend to assume that economic independence and equal divisions of housework and childcare characterized lesbian families. Moore instead finds that the partner who controlled the household labor has more relationship power. Her innovative mixed methods design provides a unique but necessary vantage point for these conclusions.

PRACTICAL STEPS FOR MIXED METHODS RESEARCH
We next recommend how to design, implement, and disseminate mixed methods family research, based on insights from past guides and empirical studies. In describing some recommended practices, we emphasize that mixed methods studies of families occur in diverse ways, and we agree with other mixed methods scholars that there need not be a unified mixed methods research methodology (Tashakkori & Teddlie, 2010; Timans et al., 2019). Too much emphasis on “one right way” to conduct mixed methods family research risks being too removed from the important theoretical frameworks, research questions, and epistemologies guiding the specific study, as well as a lack of reflexivity (Timans et al., 2019). Below we highlight some common principles that often guide the selection of mixed methods study elements, recognizing that these may not be applicable to all mixed methods researchers or every research paradigm. Ultimately, we recommend drawing on the strengths of each individual method within the mixed methods project, working to best integrate them to answer the specific research aims and to grow our knowledge and theories regarding families and relationships.

Research question or aim development
The first step for most mixed methods projects is to define the research topic or questions. For mixed methods projects in particular, this means understanding the rationale behind the methodological approach, such as explaining an observed phenomenon, challenging an existing theory, or building new knowledge of an understudied topic. To develop a rationale, the researchers typically first identify research aims to address the topic of study, as these aims often drive the choice of methodological approaches and are central to the entirety of the research process (Clark & Badiee, 2010).

We suggest that successful mixed methods research aims should include clearly interconnected qualitative and quantitative components, building on the strengths of not only quantitative and qualitative approaches but on the mixed methods approach holistically. As one example, Thomeer et al. (2013), in their analysis of couples in which one or both partners had depressive symptoms, asked first, using quantitative survey data, “Does a married person’s depressive symptoms at one point in time influence their spouse’s depressive symptoms at a later time, and does this differ for men and women?” and then, using qualitative in-depth interview data, “What are the interpersonal and emotional processes that occur when one or both spouses are depressed, and how does this compare when the husband is depressed, when the wife is depressed, or when both depressed?” These two questions fed into the comprehensive mixed methods question: “What is the importance of gender in shaping processes around depression within marriage over time?” This set of questions capitalizes on the strengths of the mixed methods design, using the quantitative side to examine broad longitudinal patterns, the qualitative side to investigate specific narratives, meanings, and mechanisms within couples, and the overall mixed methods approach to consider the big picture and develop an empirical understanding of gendered processes of depression within couples.

Even when breaking the research question into qualitative and quantitative components, we argue that mixed methods research benefits when there is an overarching mixed methods-oriented question or aim that guides the study, as this facilitates integration of the methods. The research questions should display a clear necessity for multiple integrated methodologies. A key step in developing the overall study design is determining the unique but interrelated roles of both the qualitative component and the quantitative component, as the most successful mixed methods projects have a clear understanding of the goals and benefits of each component. More examples of this are shown in the Online Appendix which provides details on the overall research objectives and the roles of the qualitative and quantitative components of studies published in JMF since 2010.

Data timing and collection
After the development of the research questions, mixed methods researchers must decide the timing of the components of the project, including the ordering of methods. Staggering the qualitative and quantitative methods is a common strategy because researchers can intentionally use the different approaches to inform each other by providing unique perspectives on the same research topic. Creswell and Clark (2017) refer to different ways of and motivations for staggering the data collection and analysis stages as exploratory (qualitative component followed by quantitative) and explanatory (quantitative component followed by qualitative). Timing choices are often shaped by the research goals. For example, some mixed methods projects may first consider patterns within the quantitative findings, then use the qualitative methodologies—such as in-depth interviews—to gain insight into the processes shaping those findings. In Perry-Jenkins and Gerstel (2020, p. 442) review of research on work and family, they state, “mixed method designs that blend quantitative results with qualitative explanations have been and will prove particularly useful for understanding the varied processes undergirding…distinctive patterns.”

This is also referred to as an explanatory mixed methods approach, wherein the quantitative component typically involves identifying larger patterns and the qualitative component provides explanations for those patterns. Reczek et al. (2016) used an explanatory mixed methods design, analyzing survey data to identify how different marital histories were associated with drinking trajectories for men and women then analyzing in-depth interviews with never married, previously married, and currently married adults to understand the mechanisms behind those associations. As another example, a study on the well-being of children in transnational families in Southeast Asia examined survey data from children (ages 9–11) and their caregivers from three countries alongside qualitative interviews (Graham et al., 2015). The survey data identified overall patterns of mental health, specifically which transnational families were most at risk for mental health issues and when. The qualitative analysis then identified why and how this mental health risk was concentrated within specific families.

On the other hand, exploratory mixed methods projects typically first begin with a qualitative approach that uncovers a particular theme via observations or narratives, later integrating quantitative methodology, such as survey data, to examine if the theme is patterned within a population. For example, Luke et al.’s (2014) study of married couples in India first used in-depth interview data to establish which household tasks were viewed as the specific responsibility of women by husbands and wives and then quantitative analysis of survey data to examine how spouses’ relative earnings associated with men’s participation in these tasks. As another example, Bernardi (2011) utilized qualitative analysis of in-depth interviews to develop themes regarding the role of social relationships within families and then quantitative analysis of social network data to quantify the importance of particular relationships.

Whether the approach is exploratory or explanatory, the goal is typically to provide richer context or information by using multiple methodologies to examine the same research topic. We suggest that researchers outline the timeline of the project before beginning data collection, as timing is a key part of the research design that should be driven by the specific research aims as well as practical considerations (e.g., access to research sites, length of time for recruitment, whether the phenomenon they are studying is time sensitive). Additionally, researchers should consider how conclusions may have differed if the methods were done in different order to draw explicit attention to how research findings are shaped by research decisions (Perez et al., 2023).

Choices regarding timing also matter for the recruitment of participants and the collection of data. In some cases, the same people participate in both the qualitative and quantitative arms of the study, or at least the qualitative participants are an intentional subset of the quantitative participants. For example, over the course of 13 months, Hernandez (2013) sequentially enrolled pregnant health care patients as participants in a survey and then invited a stratified-random subset to participate in in-depth interviews. This approach was strategic by using the quantitative data collection to inform the qualitative, while also ensuring that the qualitative interviews represented participants from different social positions. Participants for the qualitative component are often chosen intentionally—for example, they may represent a pattern from the quantitative data that the researchers want to explore further. In de Jong Gierveld and Merz’s (2013) study of living arrangements after divorce or widowhood, the quantitative sample included divorced and widowed adults, whereas the qualitative sample was chosen from those divorced and widowed adults who had a Living Apart Together arrangement to examine the mechanisms behind the choice to enter that specific type of relationship.

Data analysis
We suggest that the more integrated the analysis is between the different data in a mixed methods project, the richer the research conclusions. One way this can occur is through intentional blending of qualitative and quantitative expertise. Pearce (2015) advocates for “thinking outside the Q boxes,” rather than viewing qualitative and quantitative components as distinct. This way of thinking contributes to an important component of analysis within mixed methods design—the ability to draw meta-inferences. A meta-inference is an “overall conclusion, explanation or understanding developed through an integration of the inferences obtained from the qualitative and quantitative strands of a mixed method study” (Tashakkori & Teddlie, 2008, p. 101). Meta-inferences require carefully integrating the qualitative and quantitative design and ensuring that all sources of data and analysis steps rely on best practices within that specific method (Perez et al., 2023). For example, McClintock (2010) conducted statistical analysis of the College Social Life Survey—a survey of college students—alongside qualitative analysis of in-depth interviews with a sample of those students. She stated that, “Using both qualitative and quantitative data allows me to exploit the different strengths of each method,” resulting in meta-inferences regarding “racial differences in the degree of in-group preference/isolation and of the attitudes and processes that result in in-group partner choices” (2010, p. 47).

When conducting analysis of mixed methods data, we again point to the usefulness of an abductive approach which merges deductive and inductive orientations, moving back and forth between theory, qualitative data, and quantitative data throughout the research process. This can facilitate the development of meta-inferences and also provide a useful strategy for dealing with seemingly divergent findings. While quantitative research tends to focus on statistical patterns, qualitative research, by emphasizing themes within the data, often operates from a paradigm in which a finding may be infrequent yet important (Charmaz, 2006). This could give the appearance of divergent findings, yet also provides an opportunity for researchers to consider outliers more carefully if using an abductive approach. Divergent findings are especially welcome within abductive approaches and constructivist paradigms as they provide unique insight into important critiques and shortcomings of existing theories and prior assumptions. When encountering seemingly divergent findings within a mixed methods design, researchers should consider if these are driven by data and analysis quality issues or if represent a new and unique finding seen only through the mixed methods approach (Perez et al., 2023).

Dissemination of findings
Mixed methods family researchers publish their research in a multitude of venues, including books, peer-reviewed journal articles, and policy briefs. O’Cathain et al.’s (2008) Good Reporting of a Mixed Methods Study (GRAMMS) tool provides a helpful resource for what to include in mixed methods publications. The GRAMMS steps emphasize that when disseminating mixed methods findings, it is necessary to describe the mixed methods data, analytic process, and results—including details from both the qualitative and quantitative components—with sufficient detail for the reader to evaluate the study and place its results within the strengths and limitations of the design. In Kane et al.’s (2015) study of low-income non-resident fathers, they provided details regarding the sample selection and data collection for the survey and in-depth interview data, including the sampling frame, strategies to recruit a diverse sample, and who conducted the interviews. For the quantitative analysis, they gave detailed information regarding the survey measures (e.g., formal support, relationship between father and mother, hours of visitation per month) and specific analytic approach (e.g., hierarchical linear regression). For the qualitative analysis, they discussed their inductive approach to coding and interpreting interview data, including details regarding how many coders involved, how tasks were divided and validated across coders, and steps taken for codes to be developed into subthemes. The results section was then split between the quantitative and qualitative results.

Additionally, it is imperative that a mixed methods project justifies the mixed methods design, specifically within the context of the research question. Our Online Appendix provides justifications from each of the articles published in JMF since 2010 regarding why they used a mixed methods approach. We suggest that it is especially beneficial for mixed methods studies to provide a clear succinct statement regarding why they chose a mixed methods approach. For example, Gilligan and colleagues stated that their mixed methods approach was “useful in helping to understand the processes that underlie statistical relationships in the study of interpersonal relationships by revealing patterns that could not be identified using quantitative approaches alone” (2015, p. 918). We recommend that these types of statements be placed in the introduction and even the abstract to help orient the reader.

Notably, providing these method and result details require a considerable amount of space within a journal. When there is not enough space due to word count issues to sufficiently describe the qualitative and quantitative components of the study, we suggest that authors propose the inclusion of online appendices to accompany the article where these details can be shared. Journal decision-makers, editors and reviewers can also sometimes present barriers for mixed methods researchers to describe their study, results, and mixed method justification in sufficient detail because of word count limits or resistance to mixed methods publications, as we discuss more below in our section on obstacles. This can hinder mixed methods journal publications that include an integrated analysis of both qualitative and quantitative data, pushing researchers to often limit their published work to include analysis from solely a qualitative or quantitative component of the overall study. Choosing which analysis to include depends on the specific research question being addressed in the publication and the intended audience. Results from mixed methods projects are often uniquely suited to be distributed across multiple audiences, including outside of academia. Drawing on the same mixed methods study, it may be more appropriate and strategic to share only qualitative findings with some specific findings, but only quantitative findings or integrated qualitative and quantitative findings with other audiences.

Mixed methods research provides the innovative flexibility to choose when and how different components of a study are disseminated. For example, the work of Oparah et al. (2021) highlights the importance of researchers, midwives, and mothers in the community cooperating in efforts to mitigate the disparate health outcomes of Black mothers prenatal and postnatal. Working within the Black Women Birthing Justice organization as a predominantly qualitative oriented researcher, Oparah discusses how collaboration with other researchers and community members allowed her to expand the audience for her research. Because there were multiple stakeholders and methods involved, the project had multiple goals in terms of desired audiences when making choices about dissemination of findings. For the audiences focused on creating change at the community level, the qualitative component of the project which emphasized Black women’s voices was most valued, but for policy makers who required generalizable data to create and advocate for policy changes for Black mothers, the quantitative analysis was more useful to disseminate.

As another example of disseminating mixed methods research outside of traditional academic spaces, Desmond’s research on family experiences of housing precarity primarily draws on an extensive ethnographic study of communities in Milwaukee with high rates of eviction alongside a survey of over 1000 tenants in the same city (Desmond & Travis, 2018). This research has been shared in multiple ways—including podcasts, an accessible website, traveling exhibits in museums, books, and peer-reviewed journal articles—with goals of increasing scientific knowledge, creating policy change, and connecting people to community resources. As with Oparah’s research, components of the mixed methods design are leveraged differently depending on the specific audience and goals, a flexibility facilitated with a mixed methods design.

Reflexivity: Key component of all research steps
Reflexivity is a necessary aspect of the mixed methods research process which should be threaded throughout each of these steps above. Reflexivity is the process of continual examination of research decisions, preconceived notions, and/or positionality within the research project, recognizing how the researchers’ values, identities, and social statuses impact the study design, data collection and analysis, and interpretation of results (Timans et al., 2019). Discussions of positionality and reflexivity are too often siloed within qualitative research, yet reflexivity is an important aspect of research within mixed methods designs—as well as quantitative designs—when thinking through particular research questions and how to engage in research that addresses the aims of the research project. For research teams using mixed methods to address a particular research focus, reflexivity means understanding how specific methodological decisions and the relational power between researchers, community members, and other stakeholders shape understandings of and engagement with the project objectives.

Reflexivity within mixed methods research should be incorporated in all research steps as the study develops, such as when revising interview questions or choosing covariates for models, while also being attentive to the predispositions held regarding particular methodological paradigms. Regarding the Black feminist and intersectionality scholars who largely introduced the concepts of reflexivity and positionality to the research process, reflexivity should be thought of as a tool for contextualizing the scientific method (Bowleg, 2008). Those engaging in mixed method projects should be reflexive at every stage of the research process to ensure that no specific methodological paradigm (i.e., qualitative or quantitative approaches) is being favored, while also continually monitoring the aims of the project to ensure the methodology is achieving the research objectives.

Specifically, reflexivity challenges researchers to analyze their data within a broader sociohistorical context and to examine their research process, including the larger purpose of the project and the paradigms that could be shaping the quantitative and qualitative methods chosen (Martens et al., 2016). For example, even when forming the research team for the project, reflexivity requires family scholars to consider who is invited to the team—for example, their race/ethnicity, sexuality, and their experiences with or without families—and whether team members feel empowered to fully participate. It is necessary for researchers and research teams to recognize how their positionality shapes how they approach research questions and analyze data, while also shaping the ability of (or extent to which) researchers can integrate into the communities they are working with. This approach may involve including Community Advisory Boards. In Hernandez et al.’s (2024) ongoing mixed methods work, the research team engages with a paid Community Leadership Board, comprised of representatives from five cities across the state. The team centers the experience of board members and periodically shares preliminary results from their qualitative in-depth interviews and quantitative analysis of health care enrollment and claims data. In an iterative process, the team then relies on board members’ expertise to make research decisions.

OBSTACLES ENCOUNTERED IN DISSEMINATING MIXED METHODS RESEARCH
Researchers often struggle to disseminate their mixed methods studies, especially when contending with traditional peer-reviewed journals with strict word limits and a limited pool of reviewers with proficiency in evaluating mixed methods research. To the first issue—strict word limits—mixed methods researchers rarely have the adequate space needed to describe the qualitative and quantitative arms of the project in sufficient details and must make choices about which part to emphasize more. Although word restrictions are understandable, this often results in epistemological bias where one methodological approach is privileged over others. Specifically, our review of mixed methods articles in JMF (see Online Appendix) indicates that authors often provide much more detail regarding the quantitative data and analysis than the qualitative data and analysis. Word limits may lead to a superficial division of findings across publications that work against the goal of developing meta-inferences, where the results are not presented in tandem but are separated across publications to provide more space to elaborate on the full breadth of the research project.

Regarding the second issue, a common problem encountered in the review process is reviewers who are experts in quantitative or qualitative research—but not mixed methods research—and so may evaluate the mixed methods study using one of those two paradigms. This can occur with reviewers for peer-reviewed articles, books, and grant proposals. Some frustrations often identified are that a reviewer unfamiliar with evaluating mixed methods research may ignore the part of the research that is not their main expertise (e.g., the qualitative component if a quantitative researcher), may evaluate the part of the research that is not their main expertise using erroneous criteria, or may overlook the unique contributions of the mixed methods project and focus on the quantitative and qualitative components as separate entities.

To address these concerns, we advocate for the development of “mixed methods literacy” within family and relationship research—a skillset to be developed by researchers across career stages and research programs. We adapt the idea of “mixed methods literacy” from Small and Calarco’s (2022) concept of “qualitative literacy.” Specifically, we recommend that all family scientists, not just those conducting mixed methods research, have more knowledge surrounding mixed methods approaches. Regarding the issue of page limits specifically, mixed methods literacy includes recognition of the richness mixed methods data afford; it requires editors and publishers to think creatively about how to ensure mixed methods papers are given enough space (e.g., using online appendices). This could in practice include a formal process among family research journals to request page length extensions for mixed methods studies, with authors submitting explanations for what would be gained from the additional pages and including any supplemental material that would assist in the editor’s decision regarding whether to grant these pages.

Regarding the issue of reviewer expertise, mixed methods literacy could help journal editors and grant agencies in developing clear strategies for encouraging and evaluating mixed methods submissions, such as providing instructions for what method details need to be included within the submission and what reviewers should prioritize in their evaluation of mixed methods studies. For example, Guetterman et al. (2023) emphasize the importance of mixed methods researchers reporting on how their methods are integrated to ensure that reviewers and editors can accurately evaluate their work. A mixed methods literacy guide for journals will assist the editor in better engaging in mixed methods research and will also be useful for reviewers in recognizing the extensiveness of mixed methods research so that they are better suited to thoroughly engage in the peer review process. Enhancing mixed methods literacy at funding agencies among grant reviewers will likely facilitate more funding opportunities for mixed methods projects. We also encourage editors to urge reviewers to be forthcoming about their skillsets when reviewing manuscripts to accurately balance comments during the review process. We suggest family and relationship journals consider special issues specifically centered around mixed methods research, as this would provide needed outlets for mixed methods publications as well as produce a resource for researchers to see a wide array of approaches to mixed methods research. We argue that only 15 mixed methods articles over 13 years at JMF suggests the need for institutional changes to encourage the submission and constructive review of these types of articles.

There are currently multiple resources family and relationship researchers could draw on to build mixed methods literacy. For example, the National Institute of Health has developed mixed methods guidelines for health sciences research, including instructions for reviewers of mixed methods grant proposals such as asking whether the application makes the case that the research question is best addressed through the multiple perspectives of mixed methods research (Creswell et al., 2011). Hong et al. (2018) developed the Mixed Methods Appraisal Tool to be used across disciplines, adaptable depending on the specific methods within the mixed methods design. The quality criteria are distributed across three main domains: methodological quality, conceptual quality, and reporting quality. These guidelines provide researchers with specifics on the information that should be included within a mixed methods publication and how a reader or reviewer can evaluate the study. Mixed methods literacy could be a priority of grant review boards and editorial boards to facilitate the continued success and innovation of mixed methods approaches within family and relationship research. Teaching mixed methods literacy to students and junior (and even senior) scholars as they work to evaluate mixed methods research within their own subfields helps them to expand their own knowledge and expertise.

FUTURE STEPS
Finally, we highlight important future areas for mixed methods family researchers: (1) utilizing more mixed methods approaches that build on participatory action research (PAR) and similar approaches, (2) developing and refining innovative theories regarding families and relationships by capitalizing on mixed methods’ strengths, and (3) using best practices for open science within mixed methods approaches.

First, PAR is a type of research approach that includes members of the community who are the focus of the research in multiple stages of the research process, including developing research questions, designing research approaches, conducting studies, analyzing data, interpreting results, and disseminating findings with a view toward long-term change (Letiecq, Vesely, & Goodman, 2022). This collaborative approach is most often used in qualitative research designs. For example, in a recent community-based PAR study, undocumented Central American immigrant mothers alongside researchers coproduced a “resistant knowledge project” on the racialized and gendered structural oppressions these mothers experienced (Letiecq, Davis, et al., 2022, p. 1291). Pairing a PAR approach with mixed methods research provides more opportunities for innovation, potentially moving beyond the positivistic lens typically used in the mixed methods studies most often published in family and relationship journals while also opening up a wider range of methodological tools. As Letiecq, Vesely, and Goodman (2022) argue, critical family scholarship together with family-centered PAR efforts provides new insights into intersectional power and oppression within families, and we suggest this may be especially fruitful when using innovative mixed methods approaches in low-income international settings. Given our focus on pragmatic issues, we advise teams working on PAR to carefully construct team guidelines. This helps protect everyone involved, facilitating the communication of norms within different spaces.

As a second area in need of more attention, mixed methods research—together with an abductive approach—has significant potential for theory building and testing within family and relationship research. One topic where this would be particularly beneficial is in applying a more critical and intersectional understanding of families. For example, prior research has found that Black families experience several adverse outcomes in comparison to families of other racial groups, such as higher rates of poverty and increased social stressors (Murry et al., 2018). Previous scholarship often pointed to factors such as family structure as a potential explanation for this racial inequality; however, Williams (2020) emphasizes how these explanations treat constructs, such as race, as innate facts and ignores the racist processes that perpetuate this inequity. In this context, utilizing a mixed methods approach to explore the inequity experienced by Black families provides a chance to examine large patterns, while also integrating an exploration of microlevel processes within and among the families themselves. For example, a mixed method study by Haxton and Harknett (2009) explored differences in kin support between Black and Hispanic couples as it related to racial inequality surrounding economic hardship. Qualitative interviews examined how couples discussed their relationship with their kin networks, while quantitative survey data allowed for more generalizable findings. Through taking a mixed method approach, understandings of race-ethnicity, gender, and social support were expanded and complicated, while also providing insight into the family-level processes that may shape macrolevel trends. More broadly, mixed methods research provides significant insight into topics that are understudied, undertheorized, or even oversimplified. Family scholars asking research questions about marginalized populations specifically, and rarely studied topics more broadly, would benefit from continued use of mixed methods approaches to generate new and expound on existing understandings of social factors and structures and how they shape family outcomes.

As the third future area we highlight, mixed methods research would benefit from developing best practices for open science. We note that open science is often associated with positivistic or deductive paradigms that prize replicability, but these are not required goals of open science nor are they necessarily desirable for many mixed methods studies. Rather, we draw attention to open science’s goals of transparency and other characteristics that facilitate collaboration. This includes data sharing, a practice that is generally straightforward with quantitative data but presents more issues for mixed methods and qualitative studies with important concerns regarding the anonymity or pseudonymity of study participants and unintended consequences of making in-depth interview data, fieldnotes, and other qualitative data sources public (Tsai et al., 2016). In practice, this has meant that qualitative study components are often not publicly shared due to concerns of maintaining participants’ anonymity. Tsai et al. (2016) recommend ways to provide open access qualitative data that prioritizes the protection of participants, such as posting anonymized field notes and sharing codes rather than full transcripts. These sorts of practices would allow the sharing of both qualitative and quantitative data from a mixed methods project, an important step toward improving mixed methods science. Yet even given these recommendations, we acknowledge that there remain serious practical constraints around applying open science principles to qualitative data, thus more work and flexibility is needed to develop reasonable best practices to be applied in mixed methods settings.

CONCLUSION
We advocate that understanding mixed methods research has practical benefits beyond just for mixed methods researchers. To cohesively build—and critique—our knowledge of families and relationships, family researchers across paradigms should be familiar with the basic tenets, strengths, and limitations of qualitative, quantitative, and mixed methods research. Thus, we recommend our article on mixed methods research be read alongside the multiple methods and theory articles in this special issue. Cooper and Pugh (2020, p. 291) recently stated that family research “may be uniquely positioned to make important contributions in part because of the diversity of our methodological toolkits.” Mixed methods research is an important component of that toolkit, particularly well-suited for family and relationship research which considers a broad range of topics and draws on multiple epistemologies and paradigms. We also suggest that many of the points we make here regarding mixed methods research have broad implications beyond family and relationship research and can be used in other topic areas and contexts. Mixed methods research has special relevance for uncovering new and needed scientific information on rarely studied and overlooked topics, as DuBois (1899) highlighted over 100 years ago and numerous other family and relationships researchers have continued to show in the years since. Mixed methods research, alongside qualitative and quantitative research, can advance our diverse and growing knowledge about relationships and families.