Abstract
The research method in suicide studies has been primarily quantitative, and suicide remains without an adequate or accepted general theory that incorporates multiple disciplines and perspectives. Dependence on quantitative research limits an understanding of the complexity of suicide. This article argues for the use of mixed methods for suicide research. Three key topics in suicide research are highlighted: risk factors for suicide, efficacy of suicide prevention, and cultural factors in suicide and suicide prevention. Mixed methods will expand knowledge of suicide by integrating theory-based variables and subjectivity as objects of inquiry. Mixed methods will allow for a broadening of research questions, more substantive understanding, and are necessary for a multidimensional and multidisciplinary suicidology.
Keywords
Unresolved Issues
In spite of more than 100 years of study, since the time of Durkheim’s seminal research, the study of suicide has failed to make progress on many central issues. Although various risk factors for suicide have been identified, it still is impossible to predict suicide and clinicians have little evidence with which to make day-to-day decisions regarding risk (Simon, 2006). A second issue relates to the need to demonstrate the effectiveness of suicide prevention efforts in spite of the recognition of suicide as a global public health problem. Suicide is acknowledged to account for more deaths than violence or war (World Health Organization, 2002), yet proven suicide prevention strategies are almost nonexistent. Thompson (2005) has argued, “In spite of some encouraging work and some isolated and delimited projects, however, it can be safely said that no one has demonstrated an enduring causal relationship between purposeful interventions and reduced suicide rates” (p. 509). De Leo (2002) reflected, “suicide prevention remains essentially a land of hopes and promises but not of certainties” (p. 373). Finally, the issue of the relationship between suicide and culture has been repeatedly observed, yet there is little research that informs our understanding of this relationship. In their review of ethnocultural aspects of suicide in young people, Colucci and Martin (2007a) concluded, “Culture, instead of being a complex aspect [in suicide research] . . . becomes very often the answer to one (demographic) question” (p. 198). While examining culture as an independent variable can be highly informative about differences between nations or cultures on certain topics of interest, a more anthropological view of culture as webs of significance or meaning (Geertz, 1973) and the co-constitution of personal experience with social structures (Ortner, 2006) informs an alternative view than the one typical in suicide studies.
Research in suicidology is almost all quantitative. While demographic, psychological, and psychiatric variables continue to be studied, subjectivity is typically not the focus of research. Research participants are rarely asked to talk about their lives, or about their suicidal states, their motivations for suicide, and their experiences of recovery. Interdisciplinary and mixed methods research on suicide is rare (De Leo, 2002), and there has been a recent call to add qualitative research to this area (Lakeman & Fitzgerald, 2008). Relatively little emphasis has been given to the subjective experiences of suicidal people, but this is necessary to inform the care and help provided to individuals. It is also necessary for research knowledge about suicide. Theory in suicidology is currently insubstantial (Joiner, 2005; Knizek & Hjelmeland, 2007), as research findings have left many unanswered questions. Goldney (2002) has called for a “critical integration” of quantitative and qualitative methods, toward what he refers to as a consilient perspective in suicidology (p. 69). It is time for suicide research to take a next step forward.
Purpose
Our purpose is to argue that convergent theory and research across perspectives is a necessary step in the further development of suicidology. This has not been done in spite of much agreement that suicide is a “multidimensional malaise” (Shneidman, 1985). We believe that such a mixing of research questions and methods will improve the state of theory in suicidology. This article begins with a review of mixed methods research, integration of quantitative and qualitative research, and its relevance to the field of suicidology. Second, the article addresses unresolved issues concerning risk factors, prevention, and culture in suicide, and provides three exemplars in these areas of how mixed methods approaches, quantitative and qualitative, can enhance our substantive understanding of suicide. These are central topics in suicidology, yet the understanding of them is limited. Suicide risk factor research has been correlational, and there is a poor understanding of how these factors contribute directly to suicide, whether these factors are demographic or psychological. Most people with an accumulation of these risk factors will never kill themselves. Suicide prevention research has largely focused on outcomes, such as symptoms or suicide rates, again somewhat distant from understanding their direct relationship to being suicidal. There is a need to study the experiences of suicidal people to see more closely if and how an intervention has worked. On culture, the little research that has been done in suicidology has used culture as an independent variable, such as ethnicity, gender, and sexual orientation. A cultural approach to research will examine collective meanings of suicide and its multiple contexts (Kral, 2008). There is a need to add qualitative research to the study of suicide in order to include subjectivity and collective meaning as research questions. We conclude by highlighting recommended directions for future mixed methods inquiry in suicide-related studies. We argue for a pluralist, pragmatic theoretical framework to accompany mixed methods.
Research Across Perspectives: Mixed Methods Research
Mixed methods research is coming to be seen as the third methodological model or paradigm (Johnson & Onwuegbuzie, 2004; see Morgan, 2007), following quantitative and qualitative research methods. It is essentially a mixing together within the same study of quantitative methods, which Shweder (1996) called quanta, asking for data that are consistent, replicable, and that can be compared, “pointing, sampling, counting, measuring, calculating, and abstracting” (p. 177), and qualitative methods, which Shweder called qualia, where the objects of inquiry are subjectivity, meaning, and signification, or open-ended information acquired by speaking with someone. Johnson, Onwuegbuzie, and Turner (2007) indicate that mixed methods research has historically been conducted primarily in anthropology and sociology, but it is now being adapted by many disciplines. These authors see research along a continuum from quantitative to qualitative methods, with mixed methods researchers situated in the middle. They define mixed methods as “an intellectual and practical synthesis based on qualitative and quantitative research” (Johnson et al., 2007, p. 129), one that subscribes to a philosophical pragmatism and that is likely to provide superior research findings given its breadth of vision. Pragmatism emphasizes problem solving, practical consequences, research questions over methods, and combining deductive and inductive ressoning (Hanson, Creswell, Plano Clark, Petska, & Creswell, 2005; Morgan, 2007; Reason, 2003; Zimmerman, 2006). The coming together of the two methodological traditions of quanta and qualia can be conceptually challenging, however mixed methods is gaining considerable force in the social sciences (Alastalo, 2008).
Interdisciplinarity is the old idea that more heads are better than one, and that multiple perspectives and methods will provide a superior understanding of complex issues. This bringing together of disciplines and methods will produce a meeting, sometimes a clash, of epistemologies. On the purist side, quantitative approaches see observations as reflecting reality, the observer as removed from the object of study, and time and context free generalizations as a goal, while qualitative researchers see reality as constructed, the observer as influencing the observed, and value the local and timely (Johnson & Onwuegbuzie, 2004). Positivism and hermeneutics, often framed as objective and subjective frames and methods, reflect the two academic cultures of science and the humanities (Snow, 1959/1993). Some, like Shweder (1996), believe that these differences are real and generate tension given their contrasting objects of inquiry. Yet the mixing of epistemologies in the study of a complex problem can incorporate objective and subjective inquiries, quantitative and qualitative methods, in what Greene (2007, 2008) calls a conversation. The current move toward multiple perspectives, convergence of quantitative and qualitative research methods, and the wedding of universal and particular attributes of the person to enhance understanding, is a needed direction in suicidology. Suicide is a very complex phenomenon in need of expanded knowledge and communication between multiple perspectives.
How Does One Mix Quantitative and Qualitative Methods?
Creswell and Plano Clark (2007) identify five types of mixed methods designs. In a (a) sequential design, a component of one study is quantitative while another is qualitative. A (b) triangulation design involves interpreting data on the same topic across different studies using different methods. In (c) embedded design, qualitative data is usually secondary to a quantitative study (or vice versa). This design can be more time-efficient and manageable. An (d) explanatory design, a type of sequential design, adds a qualitative component to answer questions raised by quantitative data. Finally, an (e) exploratory design, similar to the explanatory design, involves the development of a quantitative method based on the results of qualitative data. We will show how such mixed methods designs can enhance suicidology research in some of this field’s biggest questions: Who is at risk? Can suicide be prevented? How does culture influence suicide and its prevention? Table 1 outlines our presentation of these three topics with research examples used and implications suggested in this article. Studies were selected that exemplify each of these areas.
Mixed Methods Studies for Studies of Risk, Prevention, and Culture in Suicidology
Next Step in Suicide-Related Research: More Substantive Understanding
The complexity of social and psychological phenomena is seeing a new direction in the development of novel research designs and analyses (Denzin, Lincoln, & Smith, 2008; Hesse-Biber & Leavy, 2008; Holstein & Gubrium, 2008). We hope to illustrate that adopting mixed methods approaches can provide a more substantive understanding of suicide and we illustrate this point discussing research on risk factors, prevention, and culture regarding suicide. The studies presented here are ones conducted by the authors, and they are selected because they are unusual in their use of qualitative or mixed methods in suicide research.
Risk Factors for Suicide
A recent example of the need for the mixed methods approach arises from concerns about the limits of suicide risk factors research and the call for the development of warning signs for suicide (Rudd et al., 2006). Rudd et al. (2006) demonstrated that many risk factors for suicide have been identified. Yet these factors are evaluated for the assessment of long term, generalized risk, and do not capture the emergence of a suicide crisis. However, Rudd et al. (2006) defined warning signs as near-term detectable signs within minutes, hours or days of suicidal behavior. They noted that the person’s current psychological state has to be understood and that this understanding must be specific to that individual. Warning signs are related to the person’s imminent risk and their current life context. In addition, warning signs involve a constellation of features that define the immediate crisis and as a result are fluid in nature. Utilizing these features, Rudd et al. (2006) differentiated warning signs from risk factors for suicide. Although Rudd and colleagues call for the empirical examination of their warning signs developed by expert consensus, progress in this area cannot be achieved without melding the objective with the subjective.
In an article on the value of suicide warning signs in clinical practice, Rudd (2008) indicated how they arise subjectively from the clinical interview. He explained that warning signs essentially function by understanding the “observable markers of intent to die.” This includes assessing levels of perturbation, a subjective experience one can only know by attending closely to the words and behaviors of individuals. Warning signs will emerge from asking a person how much she or he hurts (Shneidman, 1993a, 1993b). Based on his clinical and qualitative work with suicidal people, Shneidman developed a psychache scale for measuring psychological pain. If this had been done in one study, moving from qualitative interview data to a quantifiable scale that can be examined for reliability and validity, it would be an example of an exploratory design. According to Jacobs (1989), to assess risk the clinician needs “to be with the patient, in a sense to allow him or her to feel suicidal” (p. 331). Inquiry about subjectivity and intent in this case requires the person to feel safe, where empathic rapport is essential. Words and feelings become warning signs through dialogue. This integrating of the objective and observable risk for suicide with the subjective and inner experience of the person stands as a central issue in suicide research that requires qualitative and quantitative subquestions and mixed methods approaches.
Mixed methods research on warning signs might serve a number of purposes in advancing our understanding of the risk of suicide. Drawing from Greene, Caracelli, and Graham (1989), this research may fulfill a developmental purpose, where one method informs another, or an initiation purpose, where one method contributes new perspectives to the other. As Rudd et al. (2006) offered, the study of warning signs might foster the development of new measures for the assessment of imminent suicide risk. This would be a sequential design, moving from qualitative to quantitative data serving a developmental purpose. A focus on warning signs research offers greater depth of inquiry by building in immediacy, fluidity, and one’s inner experience into our understanding; all of this is in keeping with the initiation purpose of mixed methods research.
Research Example: The Transition for Youth With Repeated Suicide-Related Behaviors
In this study, the quantitative approach used descriptive and nonparametric measures of symptoms, showing a decrease in risk factors for people with recurrent suicide attempts (Bergmans & Links, 2009). The investigators then used a grounded theory qualitative methods design to understand the process(es) to transition away from high risk suicide-related behaviors. Overall, the investigators used a sequential mixed methods design to understand potential relationships among the risk factors noted as changing following exposure to the group intervention (Hjelmeland & Knizek, 2010), i.e., first doing a quantitative study investigating “what” changes occurred followed by a modified grounded theory (GT) study investigating “the process of” said changes.
Results from the quantitative study indicated that there was significant improvement on symptom and deficit measures in the cognitive, affective, and behavioral realms, including problem solving, satisfaction with life, hopelessness, depression, alexithymia, and impulsivity (Bergmans & Links, 2009). The qualitative study also included quantitative data to help characterize the participants. In the GT study 16 youth participated. From the quantitative data, the mean age for the participants’ first suicide-related behavior was 15.1 (SD = 2.36) years of age and a lifetime mean number of suicide-related behaviors was 7.1 (range from 2 to 30+). From the qualitative interviews, we learned that these youth went through a process of their transition from higher to lower risk of suicide and this transition was conceptualized as a pathway involving three major elements: (a) “living to die,” (b) ambivalence and tipping/turning points, and (c) a process of recovery that included small steps or phases toward life (pockets of recovery; Bergmans, Langley, Links, & Lavery, 2009). This example focuses on the first element, the “living to die” stage of the transition, as this element highlights the importance of understanding the persons’ experience for clinicians to effectively engage, assess, and manage the youth’s risk of suicide.
“Suicide was uh (sigh) my mistress (laughing) oh (sigh) as long as I can remember it was just the best idea to end everything, it was my solution for all.” (15:100). Most youth in the group reported believing and wanting to die from young ages, “all through my childhood I always kept saying you know I want to die, I don’t want to be here, um I wish I was dead” (16:464-465) with nothing but death potentially offering any relief from the intensity of their pain.
“The mistress” offered comfort as explained by this participant:
in my day to day life the only little bit of comfort or joy that I was able to get was through interpreting life through death . . . everything was seen from the context of suicide and death and that would kind of make me smile and be like “okay this is this is still possible, this is my way out, this is something that I can still control.” (12:108-111)
For many participants, suicide and death were meant to deal with “the pain” and the purpose of suicide and death was almost exclusively related to the eradication of the emotional intensity or to communicate the intensity they were experiencing:
suicide would be to never feel it again . . . because it becomes so intense you know, it’s like um it’s like a you get this feeling and you just want to burst out of your own body um the emotion gets so high. (2:461)
Suicide was also identified as an identity
I was going to die by suicide either by the time I was sixteen or by the end of high school and when you tell yourself that so many times it becomes so integrated, you know that you not only believe it but it becomes a promise to yourself and so every time I failed an attempt I wasn’t just failing myself I was failing that promise you know which really became my whole purpose and identity. (13:220)
Many participants identified worthlessness or being undeserving as significant contributors to their belief that ending their lives was the preferable, “I saw myself as being completely useless, completely pointless, causing more harm than good, there was no reason for me . . . any of me.” (7:381)
This element of the model was felt to have meaningful implications for clinicians working with suicidal youth and indicated that an individual with repeated suicide-related behavior has a long-standing, intimate relationship with death that serves a purpose or provides the youth with an identity. Dismissing or invalidating the individual’s relationship to death would be ignoring one of the most important relationships the person experiences at that point in his or her life. Understanding the relationship to death is critical for clinicians as they attempt “being with the patient” to effectively assess and manage youth at risk of suicide.
Results of this study identified concordance with previous understandings of suicide risk factors (Joiner, 2005) and concordance with previously identified theories within suicidology (Joiner, 2005; Knizek & Hjelmeland, 2007). Clients did not speak of their changes in terms of symptoms and deficits. Rather, as a process this work furthered our understanding beyond “what” changes occurred to understanding the first factor regarding “how” change takes place. This learning provided a client-based language, which clinicians could use within the assessment process that might be more in keeping with how suicidal individuals express and understand their own experiences of suicidality, thus potentially assisting clients being/feeling understood by clinicians and strengthening the empathic rapport. This language could further assist in developing assessment tools identifying the many meanings of the use of the word “suicide,” thus enabling a more focused and meaningful intervention plan. Furthermore, mixed methods studies could investigate the use of standardized measures and the qualitative interview investigating the concordance of the two methods of risk assessment with individuals as they present to an emergency or psychiatric service as a result of suicidal ideation.
Suicide Prevention Program Evaluation
The evidence for suicide prevention is limited, with no significant demonstrated efficacy (Gould & Kramer, 2001; Gunnell & Frankel, 1994; Lester, 1993). Although there is more recent optimism about confirming such efficacy (Goldney, 2004), in their review Mann et al. (2005) indicated “knowledge deficits about the effectiveness of some common key components” in suicide prevention (p. 2070). They identified as most promising the areas of physician education to increase psychiatric diagnosis and treatment, means restriction of popular suicide methods, and gatekeeper education such as in the military. Why has the demonstration of the prevention of suicide been so elusive? De Leo (2002) indicated that skepticism about the effectiveness of prevention efforts is based on suicide being both complex and rare. It is often impossible to adequately demonstrate if interventions truly protect against suicide. De Leo (2002) indicated that the multidetermined nature of suicide makes it difficult to track the suicidal process.
De Leo (2002) concluded that the answer to studying suicide prevention efforts is to “think big,” calling for major investments, multidisciplinary teams, and integrated approaches. We agree with De Leo and insist the study of suicide prevention requires integrated approaches. Mixed methods research is well suited for complex issues that call for comprehensive investigation. Again, mixed methods research has grown out the need for pragmatism. Greene et al. (1989) noted that for the pragmatist, practical demands will lead to the best solution for a problem. For example, true experiments will never be available to understand the cultural and social determinants for preventing suicide; however, mixed methods research provides an important option forward. Mixed methods research has found an important place in evaluation research, and Rallis and Rossman (2003) argued that the mixed methods perspective is well suited to address three characteristics essential to the field of evaluation: “description” of a program’s workability and merit, “comparison” or judging a program’s merit by use of standards, and “prediction” of a program outcomes by using both quantitative and qualitative data. The growing role of mixed methods research for program evaluation makes it well suited for adoption by suicide prevention programs.
Research Example: Gatekeeper Education Program to Prevent Subway Suicides
We participated in a developmental–sequential mixed methods program evaluation of a gatekeeper education program with a large urban transit commission. The objectives of the gatekeeper education program were to (a) improve positive attitudes toward suicide prevention and the suicidal patron, (b) increase factual knowledge about suicide and suicide-risk warning signs, and (c) enhance risk assessment and intervention skills of staff directly intervening with distressed patrons. Although suicides in the subway were an uncommon event, the transit commission was committed to implementing and evaluating prevention programs that were supported by suicide prevention experts. From the program’s inception, the transit commission understood that demonstrating significant reductions of suicides or suicide attempts per year, given the low base rate, was likely to never be demonstrated. However, the transit commission was very aware of the significant impact of suicides on the subway to patrons and staff.
The rationale for a sequential mixed methods approach to program evaluation that included individual interviews with staff following the intervention was enthusiastically supported. The purpose of the qualitative substudy was to provide the commission information about how the program affected staff and to obtain in-depth information about the perspectives of the staff regarding the training effects on their knowledge, skills, attitudes, and intervention abilities. The quantitative evaluation involved a repeated measures design testing the participant staff’s knowledge, skills, and attitudes pre- to posttraining and at 3-month follow-up (the measures of knowledge, skills, and attitudes are available from the authors on request). Eligible employees were those staff who attended the workshops during the evaluation period. A total of 307 of 360 attendees (85.3%) of the education program completed the pre- and postquestionnaires; 150 of 347 (43.2%) attendees completed the follow-up questionnaires. Using a purposive convenience sampling strategy, 30 staff participated in the individual qualitative interviews. A semistructured interview asked the participants following their participation about their ability to identify distressed patrons at risk for suicide, ability to engage patrons in a direct and open talk about suicide and their comfort and confidence to intervene.
The quantitative results demonstrated significant changes in participant staff’s positive attitudes toward suicide prevention pre- to posttraining and at 3-month follow-up (effect size post = 0.46; follow-up = 0.38; p < .001). The staff’s knowledge of suicide procedures showed significant gains from prior to after the intervention and at 3-month follow-up (effect size post = 0.28; follow-up = 0.46; p < .001). The staff’s general knowledge about suicide and suicide risk also significantly increased from prior to after the intervention and at 3-month follow-up (effect size post = 1.02; follow-up = 0.70; p < .001). Finally, the staff responsible to intervene with distressed patrons demonstrated enhanced intervention responses comparing pre- with posttraining and at 3-month follow-up (effect size post = 0.58; follow-up = 0.92; p < .001).
The subsequent follow-up interviews added crucial depth to the understanding of the program’s value for staff participants. In terms of increased knowledge, one staff responsible for interventions commented on his ability to directly talk about suicide with patrons:
Just in the way I approach someone, knowing that I am not going to instigate a suicide attempt just by bringing it up in the conversation. I was always trained that [when confronting a person] you just asked ‘is everything okay, you are not thinking of doing anything, are you?’ The training taught me to ask directly: ‘Are you here to commit suicide? (Constable)
Another staff member spoke about the changes in his attitudes and his increased awareness:
It made me understand the reasons, why they may be contemplating suicide. Prior to the course, inside of me, I was angry that they would be wanting this, to disrupt so many people’s lives, to interrupt my job, and make my job more difficult. How could they! I would still do what I had to do, but now I tend to be more sympathetic, more concerned for their well being. (Supervisor)
One staff member discussed the effects of the training to heighten his alertness when on duty:
You are paying attention a bit more . . . now I walk the platform and if I see anything out of the ordinary . . . Well just this morning I had a lady standing here for fifteen minutes I approached her and asked her if she was okay, if there is anything that I can help her with . . . I may not have done that prior to the suicide training.
The use of a mixed methods approach to the program evaluation provided significant and meaningful evidence of the value of the gatekeeper education program. In particular, administrators for the transit commission appeared persuaded about the program’s impact by the staff’s description of their increased suicide prevention awareness. The transit commission was able to provide ongoing support for the program, in spite of knowing that the actual rate of suicides may or may not be affected. Although the rates of suicide will be analyzed and examined over the longer term, the mixed methods evaluation approach was successful in providing proximal feedback to judge the program’s value. With increased awareness of the issue of suicide prevention throughout the organization, the transit commission has moved to actively look at installing barrier doorways on subway platforms to prevent patrons from accessing the subway tracks.
In the future, sequential mixed methods designs can be used in the evaluation of gatekeeper education programs to enhance our understanding of the outcomes and impacts of these preventive interventions. Suicide will always be an elusive outcome to demonstrate; however, quantitative followed by qualitative inquiry can be used to explain and deepen our understanding of the impacts of suicide prevention strategies. Exploratory studies using qualitative methods can be carried out to develop more proximal and meaningful outcomes; for example, the current study suggested that significant changes in attitudes and confidence led to greater alertness and awareness. These findings may lead to the development of quantitative measures to capture these changes in participants.
Suicide and Culture
A cultural perspective on suicide is emerging; however, it is still having difficulty finding a place at suicidology’s table (e.g., Canetto & Lester, 1998; Eckersley & Dear, 2002; Leach, 2006). Cultural research is primarily qualitative in the investigation of meaning, and is often focused on the bidirectional relationship between individuals and their social worlds (Herzfeld, 2001; Quinn, 2005). There has been very little anthropological research on suicide (see Cátedra, 1988/1992; Counts, 1980, 1991; Tousignant, 1995), and when culture has been included in research on suicide it has typically appeared as an independent variable, e.g., an ethnic demographic (Colucci & Martin 2007a, 2007b). This may be because the research questions in suicidology have been primarily universalist and timeless, where methods under the quanta umbrella come into play. While clinical work and case studies are informative, general knowledge gleaned from this perspective remains marginal in suicidology. Notable exceptions include Shneidman (1996, 2004), Maltsberger (1988, 2004), and others whose valuable insights into suicide have come not only from empirical research but also from clinical work. Yet studies of particular individuals in the context of communities and cultures can inform about suicide more broadly. It is here that theories can be tested, refined, and developed. This is the case in Indigenous North America.
Research Example: Prevention of Indigenous Suicides
Indigenous North Americans have a suicide rate 5 to 10 times the national averages in the United States and Canada (EchoHawk, 1997; Kirmayer, 1994), and many years and dollars spent on applying Western approaches to prevention have failed. This is true of mental health services in Native North American communities more generally (Gone, 2008). There is growing evidence that suicide prevention is possible through local community action in Native communities. In Nunavut, Arctic Canada, home of Inuit for a great many centuries, suicide is an epidemic among youth with a rate 10 times the national suicide rate (Kral & Idlout, 2009). A wide range of suicide rates is observed across communities from zero to extraordinarily high numbers. In recent years, the suicide rates within some high-rate communities have fallen suddenly without any outside intervention taking place. Qualitative and ethnographic research in these communities is showing that activities and programs promoting the well-being of youth, often organizing by the youth themselves with community support, are what Inuit in these communities attribute to suicides stopping while these programs are in place (Kral & Idlout, 2009). Other positive outcomes of these activities and programs are seen such as crime prevention and increased high school attendance. This research is finding that rather than there being a common program or intervention that is responsible for these positive outcomes, the common feature across successful communities is that they have created and run the programs themselves. This is an example of collective agency at work. These findings are being discovered using on-the-ground, qualitative methods.
Quantitative research is also showing these same results regarding community responsibility and suicide prevention in Indigenous communities, focusing on ownership or control of community resources. Chandler and Lalonde (2009) report a strong relationship between suicide and community control over local resources and services, including education, health, police and fire services, land title, and, importantly, self-government in First Nations communities in British Columbia, Canada. Another variable tied to lower suicide rates was having women in positions of leadership. Communities having control over all six variables under study had no suicides during the time period under investigation, while those with no control over any had the highest suicide rates, more than 10 times the national average. These findings offer a powerful explanation of between-community differences in suicide rates, centered on the concepts of reclamation, decolonization, and the taking of personal responsibility for healing and well-being. Although different research projects, these qualitative and quantitative studies of Inuit and First Nations complement each other.
There is no reason that such research cannot be combined into one mixed methods investigation, as one limit to combining the above studies is that they have been with different Indigenous peoples in different places. Based on our knowledge of this literature from previous work, we were unable to identify a mixed methods study on suicide and Indigenous peoples. Yet the findings are, not surprisingly, similar. These and other studies have assisted the federal government of Canada, Health Canada, to launch a National Aboriginal Youth Suicide Prevention Strategy in 2005, whereby Indigenous communities and organizations receive funding to develop their own suicide prevention activities/programs from their own points of view (Kral et al., 2009). Such research is thus valuable. A mixed methods study incorporating the two studies described above could, for example, have an explanatory design where the quantitative data on community control would lead to qualitative inquiry about the identified variables linked to reduced suicide rates in the communities. Here the people’s narratives about these variables such as community control over essential services, women in positions of leadership, or self-governance would provide important knowledge about how these things work in the communities, how they are valued, and the differences they make in people’s lives. A sequential or triangulation design would be a quantitative and qualitative study asking perhaps different questions, such as the two studies described here, where the findings are complementary. In an embedded design, either the quantitative or qualitative component of the study would be primary, dependent on the goals of the researcher. Finally, an exploratory design could be set up where the narratives of the people in the community would identify variables of interest, what the people believe are related to the prevention of suicide, that could then be quantified and compared with suicide rates. These might be next steps in this particular area of research.
Conclusion: Narrowing the Gaps by Widening the Circle
The questions central to suicide research identified in this article remain unanswered because the search has been restricted. The knowledge gaps concerning suicide will be better served by combining quantitative and qualitative studies, addressing quanta and qualia together. Each type of method focuses on a different object of study: the particular or the universal, and first-person or objective points of view. These methods in combination will expand the understanding of suicide. How, then, will mixed methods research be supported in suicidology?
Employing multiple methods has long been seen as enhancing the strength of research. The multimethod matrix was proposed by Campbell and Fiske (1959) as a practical methodology to enhance validity. Convergent validity, the most common form of validation, is assessed by using different methods to measure the same construct, which is similar to triangulation in qualitative research (Hammersley & Atkinson, 1983). Discriminant validity is evaluated when methods are used for two or more constructs to help differentiate them. In mixed methods research, the two methods may converge or diverge the phenomena of interest, and in both cases can tell us more about them. Similarities and differences found by mixing quantitative and qualitative methods on one construct will help us understand the construct in question.
In suicidology, mixed methods will be able to integrate objective and subjective accounts, both quanta and qualia research questions on constructs and subjectivities. Contradictions may emerge, calling for new questions and/or a reexamination of methodological details or previous research questions. Mixed methods can only serve to move the field forward, both theoretically and methodologically.
Interdisciplinarity, which is on the rise and being summoned by both scholars and granting agencies, necessarily brings differing epistemological perspectives and methods together. Greene (2008) has found that researchers make methodological decisions more for practical than philosophical/paradigmatic reasons. Pragmatism, with its emphasis on the practical, is becoming a philosophical stance in mixed methods research that supports eclecticism and pluralism (Creswell & Plano Clark, 2007; Greene, 2007; Johnson & Onwuegbuzie, 2004). Mixed methods will also allow for competing interpretations, even contradictions, which may indeed reflect the field and allow the understanding of suicide to become more nuanced and contextual (see Freshwater, 2007). Adding qualitative to quantitative research will add to knowledge about the influence of culture and social forces on suicide, and interactions between culture and cognition, emotion, motivation, and psychopathology more generally concerning suicide. Quantitative research will add generalizability. The field of suicide studies needs to examine its methodological order. Mixed methods can serve as a procedural link between epistemology and method by adding a new way of how research can proceed (Hesse-Biber & Leavy, 2008). This will include flexibility, openness, and a pluralist view.
It will be important to stimulate mixed methods research on suicide by establishing conversations on the subject more formally through interdisciplinary collaboration, meetings, symposia, and conferences. At this writing, such symposia and conferences have not yet taken place. This article is a call for the creation of such contexts for the practice of mixed methods in suicide studies.
We have highlighted in this article three critical questions in suicidology regarding risk factors, suicide prevention, and culture, and try to show how mixed methods can improve the current state of knowledge in these areas. Adding qualitative to quantitative research provides an epistemology based on first-person points of view that focus on the local, timely, oral, and intersubjective, to join one wedded to objective, generalizing, and replicative goals. Greene (2008) argues that mixed methods thinking
rests on assumptions that there are multiple legitimate approaches to social inquiry and that any given approach to social inquiry is inevitably partial . . ., embraces multiple paradigmatic traditions . . ., [and] offers deep and potentially inspirational and catalytic opportunities to meaningfully engage with the differences that matter in today’s troubled world. (p. 20)
We believe that the time has come for suicidology to embrace a mixed methods approach.
Footnotes
Acknowledgements
Our thanks to Jennifer Greene and Rahel Eynan for comments on an earlier draft of this article.
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: Funding for some of the research reported is from the Wellseley Institute, Ontario, Canada to Yvonne Bergmans and John Langley.