The Limitations of Language

Abstract

The semistructured, open-ended interview has become the gold standard for qualitative health research. Despite its strengths, the long interview is not well suited for studying topics that participants find difficult to discuss, or for working with those who have limited verbal communication skills. A lack of emotional expression among male research participants has repeatedly been described as a significant and pervasive challenge by health researchers in a variety of different fields. This article explores several prominent theories for men’s emotional inexpression and relates them to qualitative health research. The authors argue that investigators studying emotionally sensitive topics with men should look beyond the long interview to methods that incorporate other modes of emotional expression. This article concludes with a discussion of several such photo-based methods, namely, Photovoice, Photo Elicitation, and Visual Storytelling.

Introduction

The semistructured, open-ended interview has become the “gold standard” for qualitative health research (Haines-Saah & Oliffe, 2012; Silverman, 1998). There are various explanations for its popularity. First, for many researchers, the semistructured or long interview has become the foundation for the motivation of qualitative research to give a “voice to the voiceless” (Atkinson & Silverman, 1997, p. 311). By encouraging the participant to directly participate in the research process, vulnerable populations can share in knowledge creation. In so doing, misrepresentation and exploitation of vulnerable populations by researchers can be mitigated. As Sandelowski (2002) writes, “The interview has become the politically correct method to redress the wrongs of both positivism (which ostensibly give no voice and no entrée to the private or authentic) and prejudice (which it does not care to give)” (p. 105). Another possible explanation for the long interview’s popularity is its accessibility. For many researchers, the semistructured interview is, perhaps idealistically, considered a conversation between equals, and is therefore often erroneously seen as being open to anyone willing and able to engage in conversation (Sandelowski, 2002). Unlike quantitative research, which involves a variety of abilities more removed from everyday experience and is clearly something that needs to be learned and mastered, the long interview can appear to some as being readily accessible, and requiring little preparation on the part of the researcher. In this manner, interviews are “deceptively simple” (Kvale, 1996, p. 12).

Despite the strengths of the semistructured interview, it is not well suited for studying topics that participants find difficult to speak about, or for working with participants that lack verbal communication skills. There is a danger that because of the perceived simplicity of the interview, and the appeal of having the direct voice of the participant included in the research, these limitations can be overlooked and interviews will be used to study topics and populations that may be better accessed through other methods (Sandelowski, 2002).

Gender Bias in Qualitative Health Research

Research has found that qualitative health investigators who examine emotionally complex topics have tended to overlook men’s experiences. In a systematic review of pediatric palliative care literature, for example, Macdonald, Chilibeck, Affleck, and Cadell (2010) found that bereaved fathers were significantly underrepresented in parental bereavement literature. In 23 qualitative studies, fathers only constituted 18% of parents studied compared with 82% mothers (Macdonald et al., 2010). Likewise, a study examining gender bias in health research reported similar numbers. In 74 qualitative health studies, from a variety of methodologies, only 20% of research samples were men, compared with 80% women. The gender bias was higher in studies using grounded theory and phenomenology (Polit & Beck, 2008).

There are several possible explanations for this sampling bias. Polit and Beck (2008) suggest that it may reflect an overall greater interest in women’s health issues and/or that it may reflect a desire to redress discrimination by prior generations of health researchers who favored White males (Polit & Beck, 2008). It is also widely recognized that men are more difficult to recruit to health studies than women (Oliffe & Thorne, 2007), which Macdonald et al. (2010) suggest could be the result of researchers using data collection methods that are more women friendly, and may dissuade men from participating.

A lack of emotional expression among some male research participants has been described as a significant and pervasive challenge by numerous qualitative researchers in a variety of health fields, including nursing (White & Johnson, 1998), cancer research (Pituskin, Williams, Au, & Martin-Macdonald, 2007), health promotion (Smith, Braunack-Mayer, & Wittert, 2006), and mental health research (Addis & Mahalik, 2003). Researchers often struggle with a lack of satisfactory data when some male participants appear unforthcoming, laconic, and terse about their emotional experiences (Duncombe & Marsden, 1993; Presser, 2007). Emotions play a central role in the experience of illness and other health conditions, as these conditions often evoke feelings of fear, shame, sadness, guilt, and loneliness (Bowman, 2001). Understanding the emotions that are associated with an illness or a health condition is fundamental for understanding the experience of the participant. A shortage of quality data on emotional experience can significantly affect both the nature and the quality of the research.

Emotional inexpression is not limited to men. Qualitative researchers studying the emotional experiences of women, children, and adolescents also face issues of inexpression, and a number of strategies for encouraging participants to open up and speak about their experiences have been offered. These techniques include speaking about emotionally difficult experiences in the third person, opening up about the researcher’s own emotional experiences, and using humor, ribaldry, and self-deprecation to attempt to make the participant more comfortable (Adler & Adler, 2001; Gokah, 2006; Liamputtong, 2010).

Along with these techniques, qualitative researchers working with men have put forth a number of their own strategies to better elicit male participant’s emotional experiences (Wong & Rochlen, 2005). These techniques focus on making the participant feel powerful and in control. Strategies include encouraging male participants to pick the time and location of the interview, letting the participant ask the first question, challenging the participant to “take charge” as both expert and teacher, and emphasizing the strength and risk taking behavior of the participant (Schwalbe & Wolkomir, 2002). Even direct flattery and flirting with the participant have been discussed (Oliffe, 2010; Schwalbe & Wolkomir, 2002). These techniques rely on the explanation that those men who are emotionally reticent during the research interview do so out of a fear of being perceived as feminine, vulnerable, and not in control. As Schwalbe and Wolkomir (2002) write,

The interview situation . . . is potentially threatening to the masculine self because the interviewee relinquishes control, the exercise of which is a basic way in which masculinity is signified . . . minimizing can arise from a desire to protect the masculine self by maintaining control or by revealing no vulnerabilities or uncertainties. (p. 61)

Research has emerged from the field of clinical psychology that challenges this explanation, finding that whereas some men withhold information about their emotional experiences out of a fear of revealing their vulnerabilities, others may be unable to adequately verbally articulate them (Levant, Hall, Williams, & Hasan, 2009). These findings shed new light on the nuance and complexity of emotional inexpression, and have implications for researchers studying emotionally sensitive issues with men. However, to date this research has been largely overlooked by those attempting to explain and mitigate issues of men’s emotional inexpression in the research interview. In the following paragraphs, three theories of men’s emotional inexpression will be discussed that each propose a different etiology: the first to sex/biology, the second to gender socialization, and the third to social constructionism.

Theories of Emotional Inexpression

Sex/Biology

The study of emotional expression has a long and complex history, with roots as far back as Charles Darwin (1872). Darwin hypothesized that emotional expression facilitated intergroup relations, and as such played a key role in the evolutionary process. Darwin’s work inspired many scientists to investigate the biological and evolutionary role of emotional expression. A result of this research was the discrete emotions model, first suggested by Tomkins (1962), and then expanded by Izard (1977) and Ekman (1992). Proponents of this model theorize that patterns of expression developed as a survival tactic when our species was still in the hunter/gatherer phase of development. Because of the historically male role of protector/provider, and the concomitant possibility of injury or death as well as separation from family, psychologically our male ancestors evolved to maintain an emotional distance from kin. This entailed a more practical, action-driven development including dissociation from emotions (Rossi, 1984). As a result, males’ ability to verbally express their emotions did not develop a high sensitivity. This can be juxtaposed to the evolutionary psychology of females, who, because of their biological role as mothers and nurturers, evolved a more acute sense of emotion and a heightened ability to verbally express emotional experiences. Although biological explanations can offer valuable insights into the nature of emotional expression, from what is now understood about the social influences on gender, and gender behavior, relying solely on these explanations may be problematic.

Gender Socialization

As mentioned above, new research on male emotional expression has emerged from the field of clinical psychology, which has found that whereas most men do not have difficulty expressing their emotional experiences, others lack the emotional vocabulary that is necessary to adequately do so. This research relies on a theory of normative male alexithymia (NMA), which holds that some men have been so heavily socialized, by their parents, community, and peers, to conform to masculine gender norms that they have lost, or failed to initially develop an adequate emotional vocabulary.

The term alexithymia was first coined by the psychologist Peter Sifneos (1967, 1972) to describe an observed condition whereby patients with psychosomatic complaints, alcoholism, and posttraumatic stress disorder had difficulty defining and describing their emotional experiences. The term alexithymia is derived from Greek and Latin roots and literally means without (a) words (lexus) for emotions (thymos). Since the term was first coined, a large amount of research has been conducted on alexithymia. Psychologists believe that alexithymia is a dimensional personality trait that varies from person to person. Though it is not considered a mental disorder by the Diagnostic and Statistical Manual of Mental Disorders, fourth edition, alexithymia has been linked to numerous negative mental health outcomes including alcoholism (Loas, Otmani, Lecercle, & Jouvent, 2000), higher rates of psychosomatic illness (Cohen, Auld, & Brooker, 1994), and eating disorders (Speranza, Loas, Wallier, & Corcos, 2007).

Beginning in the 1980s and 90s, higher rates of alexithymia were observed among male clients in psychological research and clinical practice (Levant & Kelly, 1989; Levant & Kopecky, 1995). Both meta-analysis (Levant et al., 2006) and empirical studies examining emotional expression (Levant et al., 2009) have since substantiated these observations. To explain the consistently higher alexithymia scores among men, Levant and colleagues theorize that a mild form of alexithymia exists, which results from intense gender socialization:

Male gender role socialization, through the combined influences of mothers, fathers, and peer groups, suppressed natural male emotional expressivity. Thus boys grow up to be men who cannot readily put their feelings into words. They are therefore normatively alexithymic. (Levant, 2003, p. 177)

Although recognizing the existence of NMA, investigators remain unclear about who exactly may be susceptible. Studies have found that there is a cultural component, with higher rates of NMA in populations that support traditional masculine ideology, such as some immigrant and minority communities (Levant et al., 2006). Class and educational opportunity have also been found to correlate with higher rates of NMA, with men from working-class communities having higher scores on the NMA scales than American college students (Levant et al., 2006).

These results must be treated with caution, however, as NMA, although socially influenced, is an individual trait, and though it is higher among certain cultural groups, the extent that individuals’ internalize cultural beliefs and expectations differ. Regardless of the varying degrees of NMA among individuals, Levant and colleagues believe that mild-moderate NMA is “very common and widespread among men” (Levant, 2003, p. 177). Considering this research, it seems probable that NMA could be a contributing factor in emotional inexpression among some men during the research interview.

Social Constructionism

Despite the empirical research on NMA, the predominant explanation remains that men who are emotionally inexpressive during the research interview do so out of a fear of being perceived as feminine, vulnerable, and not in control. This explanation is largely built on the work of Jack Satel (1976). According to Sattel, as a result of what was later termed Gender Role Socialization (Connell, 1987), men learn to use emotional inexpression as a strategy to mask their vulnerabilities, allowing them to be perceived as rational, efficient, and nonemotional, which provides a negotiating advantage in intimate situations where they feel emotionally fragile. For Sattel, some men’s refusal to disclose their emotional experiences is political, as it feeds directly into the patriarchal nature of our society. As he explains, “to reveal their emotional vulnerability would be to undermine the very power structure that maintains men’s privileged position of power. This position of power relies on the perception of invulnerability” (Sattel, 1976, p. 470).

Issues of Emotional Inexpression and Qualitative Health Research

While the jury is still out over whether some men choose to be emotionally reticent, or if a percentage of men are inadequately able to articulate their emotional experiences, emotional inexpression during the research interview can affect the amount and the quality of data collected. Using research methods that rely heavily on verbal articulation may also dissuade some men from participating, which in turn may affect the number of men recruited into health studies (Macdonald et al., 2010). With this in mind, qualitative health researchers may consider, both for issues of recruitment and accuracy, whether the in-depth semistructured interview is the best means of examining emotionally sensitive topics with men. In the section that follows, several photo-based methods are presented, which deemphasize formal verbal articulation and are well suited, either on their own or in conjunction with the long interview, to the study of emotional subjects with men.

Photographic and Visual Methods

Photographic and visual research methods have long been used in the social sciences, especially in the arenas of feminist, community-based, and participatory action research, and are increasingly being used in health research. These methods are particularly useful for accessing the experiences of those who have difficulty with language, or who may be more engaged by visual as opposed to verbal forms of expression. To this end, visual methods have been used in studies of children (Epstein, Stevens, McKeever, & Baruchel, 2006), immigrants (Streng et al., 2004), and adults with intellectual disabilities (Booth & Booth, 2003; Jurkowski & Paul-Ward, 2007).

Photographic methods also have a long history in research addressing gender issues in relation to broader social determinants of health. These methods have been used, for example, in studies looking at women’s experiences of homelessness (Klitzing, 2004; Walsh, Rutherford, & Kuzmak, 2010), maternal and child health (Jainghong, 2004), and aboriginal women’s experience of breast cancer (Poudrier & Thomas MacLean, 2009). There are also a small number of studies that have used photographic methods to study men’s health issues that involve a high level of emotionality (Haines-Saah & Oliffe, 2012). It has been used, for example, to study men’s experiences of living with prostate cancer (Oliffe, 2003; Oliffe & Bottorff, 2007), smoking (Oliffe, Bottorff, Kelly, & Halpin, 2008), grief (Creighton & Oliffe, 2010), and transgendered men’s access to health services (Hussey, 2006).

Photographic research methods vary. In Photovoice, the best known of the visual methods, participants take photographs that represent their experience and write an explanatory text-based caption about the image describing why it represents their experience. The photographs themselves can also be analyzed as data, used as a prompt to elicit discussion within focus groups, or displayed in an exhibition to inform policy makers or other key decision makers (Wang & Burris, 1994, 1997). In a similar method known as Photo Elicitation, photographs taken by participants are used to facilitate discussion within the research interview (Epstein et al., 2006). Finally, in the recently developed method of Visual Storytelling, Photovoice, and Photo Elicitation are combined. Photographs are used to both relay ideas and to facilitate discussion within the interview (Drew, Duncan, & Sawyer, 2010).

There are many benefits that make photographic methods well suited for studying emotionally sensitive topics with men. First, as researchers meet the participant several weeks before the interview to give them the camera and explain the research process, photographic methods provide an opportunity to “break the ice” and build a rapport between the investigator and the participant well before the interview itself. This may help to ease the anxiety that some men may feel about discussing their intimate and emotional experiences with strangers.

Another key benefit of photo-based methods is that they give the participant more control over the direction of the interview. Unlike the long interview, where the researcher holds the power to direct the discussion (Schwalbe & Wolkomir, 2002), photographic methods allow the participant to regulate what is seen, and hence what is discussed. Photo-based methods allow the participant take charge as experts and teachers, while researchers are relegated to the role of active spectators rather than interviewers (Flick, 2002). Furthermore, the independent nature of the data collection, and the time required to take the photographs, allows the participant a greater opportunity to reflect on the photos, collect their thoughts, and formulate their discussion points without feeling rushed to do so. This too may help to ease the anxiety that some men feel during the face-to-face interview. In fact, researchers have reported that some male participants, who may otherwise be anxious about the interview, grew to anticipate the meeting with the researcher so that they could discuss their photos as a kind of “show and tell” project (Oliffe & Bottorff, 2007, p. 851). For some participants, this not only eased their anxiety but also proved to be a therapeutic exercise. Taking photos of their experience enabled them to think deeply about the treatments they were undertaking, and see, through the series of photographs of themselves, the progress they had made in their treatment (Oliffe & Bottorff, 2007).

Another appealing benefit of photo-based methods is that they can make the research process easier for participants who may be uncomfortable directly discussing their emotional experiences; participants can divert the conversation from their own personal experience, and focus on the photographs they have taken (Jurkowski & Paul-Ward, 2007). This can be useful when researching men who may be uncomfortable with high levels of emotional self-disclosure. As Oliffe and Bottorff (2007) discuss,

Highly emotive, ordinarily private details were shared; however, when the content or level of self-disclosure became uncomfortable, participants also found respite and refuge in their photographs . . . photographs facilitated self-disclosure, but also provided anonymity and distance from sensitive issues. (p. 853)

Removing some of the discomfort that some men may have with self-disclosure can result in a more fluid, relaxed, and conversational interview, which in turn may result in richer, thicker data (Geertz, 1973).

While giving more control to the participant and helping ease anxiety, photo-based methods also have a structure that can help those men who may have difficulty verbally articulating their experiences. By using photographs as metaphors, participants can convey complex and intense emotional experiences that may otherwise be difficult to express because of the limitations of emotional vocabulary. A good example of this is found in Ornelas et al.’s (2009) study of African American men’s perceptions of social inequality. In this study the authors describe how a participant used a photograph to help explain his experience: “One of the men took a photo of an abandoned car, which he stated was a metaphor for ‘irresponsible Black men’: ‘It’s like an abandoned car, something valuable with so much potential being wasted’” (Ornelas et al., 2009, p. 561). As many men are predominately visual learners, using photos to help discuss their experience may enable them to see and say more with fewer words (Oliffe & Bottorff, 2007).

Along with facilitating emotional expression among men, there are other more general benefits of using photo-based methods. First, using photographs as conversational prompts can provoke memories that the participant may have forgotten or repressed. Researchers describe these as “fresh” details, which provide new and intricate details in the interview (Wang & Redwood-Jones, 2001, p. 562). These intricate details can increase the richness or thickness of the interview and the quality of the research.

In the case of ethnography, which relies in part on participant observation, photo-based methods also grant researchers access to situations and contexts that would otherwise not be available to them. Researchers cannot always be present at those moments and functions that the participant deems significant, such as group meetings, doctor’s visits, and meetings with significant people such as friends or past acquaintances. By taking photos of these events participants can help researchers to better understanding their experience.

Last, photographs are a powerful form of communication. They can put a human face on the research, evoke empathy for the participant, and can greatly enhance knowledge transfer between the researcher and their intended audience. Using participant-produced photographs as visual aids in the classroom, or at academic conferences, for example, can be more effective than simply sharing illustrative participant quotes (Oliffe & Bottorff, 2007). Photos also tend to spark a great deal of discussion and debate among students, audiences, as well as different members of the research team who may have their own interpretations of the data. Due to the social constructivist nature of qualitative research, these different interpretations and the discussions that follow can result in new and important insights for the research team, and can further enrich the data analysis (Oliffe et al., 2008).

Conclusion

There are many strengths of semistructured interviews: They allow the voice of the participant to be heard within the research context; they allow the investigator to pick up on social cues such as voice, innovation, and body language; and they can result in highly personalized data. However, the semistructured interview may not be the best method for working with participants who lack verbal communication skills, or for studying topics that participants have difficulty discussing. There is a danger that these strengths, as well as the perceived simplicity of conducting interviews, can result in the use of interviews for topics or participants that might be better served by other methods. Investigators must remember that it is the research question and the population under study that determines the method, and not vice versa.

Considering the different theories of emotional inexpression discussed in this article, it is likely that the level of emotional discussion required by long interview may be uncomfortable and perhaps intimidating for some men. There is a danger that using research methods that rely solely on verbal articulation may result in a lack of accurate data on these participants’ emotional experience, and dissuade some men from participating who may be uncomfortable with the prospect of discussing their emotional experiences at length.

With this in mind, both novice and experienced researchers who study emotionally sensitive topics with men may wish to rethink the use of the long interview as their primary method of data collection. Photo-based methods, when used either on their own or in conjunction with in-depth interviews, can help ease the anxiety some men may have with emotional expression, and assist those without adequate emotional language to participate in the research process. In so doing, these methods can help researchers to better understand the nuance and complexity of the individual’s experience and make it easier for some men to participate in qualitative health studies.

Although recruitment issues may play a role in men’s underrepresentation in qualitative health research, it is likely, as Polit and Beck (2008) suggest, that there are also ideological factors at play. First, there may be some qualitative health researchers, who wish to redress the exclusion of women and minorities by health researchers in the past. Second, considering men’s privileged position of power, there may be some researchers who are less inclined to recognize men’s vulnerability or to consider men’s experience as a worthy area of study. By putting a human face on the research participant, photo-based methods can highlight the individuality of the participant and evoke empathy for their experience. This can help those researchers who may be disinclined to recognize men’s vulnerability to do so, and can also help to distinguish the individual’s experience from the larger sociopolitical context in which it exists. Along with decreasing the discomfort that some men have with discussing their emotional experiences, this may also help address the underrepresentation of men in qualitative health research.

Finally, it should not be forgotten that apart from the many benefits offered by photo-based methods, relying on a variety of data sources is simply good research practice. It has been argued that no single empirical method can ever fully capture a phenomenon (Patton, 2002). To use a single method is to rely on what some researchers have referred to as the “precariousness of the one-legged stool” (Hall & Rist, 1999, p. 291). Using a variety of methods, especially from other qualitative data sources (e.g., observations, documents, and artifacts) is important for increasing the robustness and rigor of the research design, and subsequently, for achieving deeper insight into the phenomenon under study.