|Home | About | Journals | Submit | Contact Us | Français|
This article describes the influences of social context on women’s health behavior through illustration of the powerful influences of social capital (the benefits and challenges that accrue from participation in social networks and groups) on experiences and perceptions of self-efficacy. The authors conducted inductive interviews with Latino and Filipino academics and social service providers and with U.S.-born and immigrant Latinas and Filipinas to explore direct and indirect influences of social context on health behaviors such as mammography screening. Iterative thematic analysis identified themes (meanings of efficacy, spheres of efficacy, constraints on efficacy, sources of social capital, and differential access to and quality of social capital) that link the domain of social capital with the behavioral construct perceived self-efficacy. The authors conclude that social capital addresses aspects of social context absent in the current self-efficacy construct and that these aspects have important implications for scholars’ and practitioners’ understandings of health behavior and intervention development.
What does it mean when a woman says, “Yes, I feel confident that I can get a mammogram in the next year”? Is that really powerful, in and of itself—what meaning does that have?–Study Key Informant 9
Perceived self-efficacy, measured by confidence in one’s ability to complete a task such as getting a mammogram, has long been used as a predictor of and explanation for health behavior and as the target of health behavior interventions. Intervention studies have shown that changing a person’s confidence in her or his ability to perform a particular behavior—her or his perceived self-efficacy—will likely change the behavior itself. However, there is little understanding of what composes this confidence, of how this construct works for different cultural and ethnic groups, and of whether it operates as indicated in theory in diverse contexts. Thus, we know neither how cultural variations in meanings of perceived self-efficacy influence health behavior nor the extent to which construct measures are valid and comparable across groups. In addition, we know little about the influence of powerful social contexts such as poverty or migration on the performance of this construct.
Social context, the conceptual framework for this study, is defined as the sociocultural forces that shape people’s day-to-day experiences and that directly and indirectly affect health and behavior (Burke et al., 2009; Pasick & Burke, 2008). These forces include historical, political, and legal structures and processes (e.g., colonialism and migration), organizations and institutions (e.g., schools and health care clinics), and individual and personal trajectories (e.g., family, interpersonal relationships). Notably, these forces are co-constitutive, meaning they are formed in relation to each other. Our approach to understanding social context is inductive and is based in the social science disciplines of anthropology and sociology. Burke et al. (2009) describe this framework in detail and show how it complements and differs from dominant models in the study of behavior and health outcomes such as social learning theory and social ecological models.
Methodologically, most health behavior research is informed by theories developed and tested in White, urban, middle-class populations (often college students; Ajzen, 1991; Emmons, 2000; also see Burke et al., 2009) and that focus on cognitive constructs as predictors of behavior. Underlying these theories is the assumption of a shared “norm”: that of White, urban, middle-class Americans. In addition, these theories assume a shared rationality—that if given appropriate information, people will choose to perform the recommended health practice. In practice, this results in the application of inappropriate theories to ethnically and economically diverse populations within multicultural societies such as the United States (Pasick & Burke, 2008; also see Pasick, Burke, et al., 2009).
An alternate approach to the study of health behavior is inductive research using multiple qualitative methods to generate rich, in-depth, and detailed data on the daily lives of communities of interest from their own perspectives (Fernandez & Herzfeld, 1998; Schweizer, 1998). These methods are useful for explaining the social and cultural contexts of health behaviors (Kiefer, 2007; Lambert & McKevitt, 2002), showing the relevance of these contexts to behaviors and providing an understanding of why people do what they do. For example, a woman may say that she is confident that she can get a mammogram but not follow through and actually obtain the exam. Inductive research illuminates more fully the meaning of a claim of “confidence” in light of the many competing priorities she faces—the social, cultural, and practical considerations, importantly not only those that are conscious but also those outside of conscious awareness—in her daily life and that influence her behavior.
The 5-year multimethod study described in this volume and on which this article is based (Behavioral Constructs and Culture in Cancer Screening [3Cs]) arose from a series of trials (e.g., Hiatt et al., 1996) to increase mammography use in low-income, multiethnic communities. The results of the trials raised questions about the cross-cultural appropriateness of concepts and measures associated with the five behavioral theory constructs most commonly used to explain and influence cancer screening: perceived self-efficacy, perceived susceptibility, perceived benefit, intention, and subjective norms. The 3Cs study was designed to assess the cultural appropriateness of these constructs related to mammography screening among Filipinas and Latinas and to explore and describe the meanings and social context associated with each construct through inductive research methods. Methods for the 3Cs study were chosen to foster a multidimensional understanding of the social context of women’s lives and their orientation to health and preventive practices such as mammography screening (Kiefer, 2007; Lambert & McKevitt, 2002).
Among the findings from the 3Cs study were three overarching domains that pervaded daily life: relational culture (the processes of interdependence and interconnectedness among individuals and groups), transculturation (cultural change processes in which relationships are sustained across national boundaries), and social capital (the benefits and challenges that accrue from participation in social networks and groups). Although these domains are inextricably interconnected in complex ways, with implications for all the constructs, for reporting purposes we identified the most significant context–construct linkages and conducted analyses to explore intention and subjective norms vis-à-vis relational culture (Pasick, Barker, et al., 2009), perceived susceptibility and benefits in the context of transculturation (Joseph, Burke, Tuason, Barker, & Pasick, 2009), and the meanings of perceived self-efficacy in the context of social capital, the subject of this report. Figure 1 outlines the methods and phases of the 3Cs study.
We first summarize the primary components of perceived self-efficacy and the assumptions that undergird this construct in the major health behavior theories. Next is a brief review of the concept of social capital. We summarize the data collection methods of this study and lay out our analysis of perceived self-efficacy in the context of social capital. Our conclusions address the relevance and meaning of the self-efficacy construct vis-à-vis mammography for immigrant and U.S.-born Filipinas and Latinas.
Perceived self-efficacy is a leading concept in Albert Bandura’s (1986) social cognitive theory (SCT), which argues that cognitive, self-regulatory, and self-reflective processes are central to human adaptation and change. Self-efficacy is a key construct in other leading behavioral theories, particularly the transtheoretical model (Prochaska, Norcross, Fowler, Follick, & Abrams, 1992) and the theory of planned behavior (Ajzen, 1991). Perceived self-efficacy refers to “beliefs in one’s capabilities to organize and execute the courses of action required to manage prospective situations” (Bandura, 1995, p. 2). Self-efficacy beliefs are said to provide the foundation for human motivation, well-being, and personal accomplishment (Bandura, 1977). According to Graham and Weiner (1996), perceived self-efficacy is a more consistent predictor of behavioral outcomes than any other motivational construct.
SCT states that self-efficacy beliefs regulate human functioning through cognition (enables people to predict and develop ways to control events that affect their lives), motivation (contributes to self-regulation), affect (influences emotional reactions to difficult situations as well as anxiety arousal), and selection (influences types of activities or environments chosen). Self-efficacy beliefs, following this theory, are developed through interpretation of information from four sources: (a) mastery experience, (b) vicarious experience and modeling, (c) social persuasion, and (d) physical and emotional states. Positive experiences related to each of these areas are believed to increase the likelihood of high levels of perceived self-efficacy (Bandura, 1977, 1986).
One underlying assumption of the current self-efficacy construct is that an individual’s ability to perform the target behavior rests mainly on psychological factors, such as motivation, willingness, perseverance, commitment propensity, expectation of success, and a “positive” attitude. Social structural factors—the social context—are regarded as background factors, modifiers of behavior, but not directly influential. Although Bandura (1977, 1986) described a “reciprocal determinism” among behavior, the environment, and personal factors (i.e., cognitive, affective, and biological factors) thereby explicitly acknowledging the complexity of human behavior, his theory does not allow for unconscious as well as direct influences of social and cultural context (for a detailed discussion, see Burke et al., 2009). Similarly, interventions that use self-efficacy share the assumption (based in a model of rational action) that people will take action to prevent or control illness if they (a) have the necessary information and so know how to prevent or control the illness and (b) feel capable of taking the requisite health action. This assumption does not allow for culturally divergent interpretations of the information, variations in rationality or shifts over time and place depending on context (Kiefer, 2007; Kleinman, 1981).
Like the also widely used construct of intention from the theory of reasoned action/planned behavior, often only a single situation-specific item is used to measure perceived self-efficacy (Luszczynska & Schwarzer, 2005; e.g., “I am confident I can get a mammogram in the next 6 months even if I don’t have insurance”). Part of the 3Cs study, Stewart, Rakowski, and Pasick (2009) report findings from a multiethnic, multilingual longitudinal study of mammography promotion in which the baseline survey asked, “Do you think that you could get a mammogram every year?” (yes or no). The self-efficacy question was worded this way because the term confident could not be comparably translated in all four languages (Spanish, Chinese, Tagalog, and English). Similarly, a multicategory scaled response could not be used in this telephone survey of women from a range of educational backgrounds. The self-efficacy construct significantly predicted recent screening as measured in the follow-up survey 2 years later for White women but did not significantly predict screening for African American, Chinese, Filipina, and Latina women. No significant race/ethnicity interaction was found. However, because there was limited variability in response (i.e., the vast majority of women, regardless of ethnicity, answered yes), it was difficult to detect a statistically significant race/ethnic interaction, which would have been stronger evidence that this construct better predicts screening for White women compared with other groups.
Although many questions remain regarding measurement of the construct across ethnicity and language, the combination of specificity and dominance of cognitive factors renders this construct difficult to reconcile with the concepts of culture and social context, specifically because some influences on behavior are outside of conscious awareness, rooted in historical and institutional precedents (Bourdieu, 1990; also see Burke et al., 2009). The meanings that people hold for practices such as mammography and concepts such as health and illness can vary dramatically across populations and contexts, shaping cognition and logic (Geertz, 1973; Kleinman, 1981). We explore consistencies and inconsistencies between the conception and specificity of the perceived self-efficacy construct as operationalized in research on mammography behavior as well as demonstrate how and why the concept of social capital illuminates dimensions of self-efficacy.
We understand social capital to encompass both the benefits and challenges that women face as members of families, groups, and communities and the ways they employ and deploy resources accrued from participation in these groups. It is important to note that social capital is characteristic of the structure of social relations, not the individual, even although individuals benefit from social capital (Bourdieu, 1985; Portes, 1998). This aspect has been lost in much of the debate over the concept; in conceptualization and measurement, it has been stretched from being a property of relationships among individuals and families to a feature of communities, cities, and nations (Coleman, 1988; Kreuter & Lezin, 2002; Muntaner, Lynch, & Smith, 2001; Portes, 1998; Putnam, 1995). Much of the public health literature emphasizes the positive attributes of social capital—civic engagement and norms of reciprocity and trust—and utilizes them to explain population health (Kawachi & Berkman, 2000; Skrabski, Kopp, & Kawachi, 2004). The “downside” of social capital—the potential for exploitation and inequality (i.e., differently positioned individuals experience associational life differently and some benefit at the expense of others)—has been largely neglected. This has limited the usefulness of the concept, especially when applied to impoverished and marginalized communities within a larger, dominant society (Magalong & Takahashi, 2006; Mitchell & LaGory, 2002; Veenstra, 2000).
We chose to use Bourdieu’s conception of social capital for our framework because it attends to this potential “downside,” to the influences of power and social structures on access to and ability to deploy social capital resources (Bourdieu, 1985). Bourdieu distinguished between the social networks to which people belong and the outcomes of those social relationships (i.e., the extent to which one may obtain resources by virtue of membership in different social structures vs. the resources themselves; Bourdieu, 1985; Portes, 1998). He noted that although individuals benefit from participation in groups, benefits and costs of participation are not equally distributed but rather on the basis of one’s social position. Such power differentials constitute and are constituted by the opportunities and constraints offered various individuals by social institutions (i.e., the social structure) that influence both the actions that individuals take and the contexts in which they act. These influences inform one’s ability to make choices and to activate one’s self-efficacy, both consciously and outside of conscious awareness (Bourdieu, 1990; Burke et al., 2009).
A “bridging” concept between self-efficacy and social capital is collective efficacy. Collective efficacy has been defined as “the aggregate of individual group members’ perception of group capability” (Goddard, Hoy, & Hoy, 2004, p. 7) and “the collective belief in undertaking coordinated action” (Sampson, 2003; Skrabski et al., 2004, p. 341). It is also variously described as a domain or measure of social capital (Skrabski et al., 2004) and as a concept building on and expanding the idea of social capital (Frankenberg, 2004). Bandura (1997) noted that “perceived personal and collective efficacy differ in the unit of agency, but in both forms efficacy beliefs have similar sources, serve similar functions, and operate through similar processes” (p. 478).
Frankenberg (2004) focused on a behavioral manifestation of collective efficacy (rather than beliefs) and defined collective efficacy as the “shared willingness among neighbors to intervene on the behalf of the common good” (p. 1). Distinguishing between collective efficacy and social capital, she argued that the social cohesion and networks that form the basis of social capital create the conditions necessary for the emergence of collective efficacy (expectations for engagement in purposive social action). The concept of collective efficacy as employed in these studies begins to address social context through its attention to neighborhood environs (Frankenberg, 2004; Sampson, 2003; Skrabski et al., 2004). As a nascent construct, collective efficacy has the potential to better capture the social context of communities; however, the influences of social and cultural contexts, the focus of this study, on individual health behavior remain inadequately theorized.
This study employed inductive qualitative methods and was designed to describe, from multiple perspectives, the broad web of direct and indirect influences in the lives of immigrant and U.S.-born Filipinas and Latinas living in the San Francisco Bay area. This study does not attempt to characterize or to compare Filipinas and Latinas. Rather, it explores sociocultural context in two immigrant populations that differ in many ways from the U.S. European American middle-class mainstream but whose similarities, in terms of social context and social capital (e.g., social connections; social and economic constraints linked to immigration and poverty; local, transnational, and extrafamilial relationships; and gendered role expectations), are more relevant than the differences with regard to the appropriateness of the behavioral constructs examined in the 3Cs study. We focused on Latinas and Filipinas because (a) they have low rates of breast cancer screening (Jacobs, 2005; Kagawa-Singer et al., 2007), (b) they were included in our quantitative study already under way and we had the most prior data collection and intervention experience with them (see Stewart et al., 2009), and (c) they are well represented in the San Francisco Bay area (4.8% Filipino and 19.4% Latino; Bay Area Census, 2000). Although some aspects of the social context we describe may be specific to Latinas and Filipinas, others likely apply more widely.
We conducted in-person open-ended interviews of 1 to 3 hours first with Latino and Filipino academics as key informants (n = 11), then with Latino and Filipino social service providers and activists as community gatekeepers (n = 13), followed by Filipina and Mexican American and immigrant women (n = 29). Each set of interviews addressed similar domains, informed the next set of interviews, and provided insights into the meanings of perceived self-efficacy for Latinas and Filipinas contemplating mammography screening in particular and accessing health care in general. For each set of respondents, interviews were continued until we reached data saturation—that is, until no new themes emerged in the narratives or during initial data analysis (Sandelowski, 1995).
Our approach to interviews was inductive in that we did not directly ask informants what they thought about each construct; rather, our questions primarily addressed the issues respondents considered important when thinking about women’s lives in the United States, their health care in general, and, in the final stages of interviews, screening mammography in particular. Also, although questions and probes were developed in advance, respondents were free to take the discussion into a wide range of topics that were pursued to the extent that they met the criteria of having implications for health decisions and practices. This approach allowed for meanings to emerge from participants rather than being imposed by researchers (Fernandez & Herzfeld, 1998; Lambert & McKevitt, 2002). Gradually, a multidimensional picture emerged of the daily relationships and activities of women in the context of family, of community, and of a wide array of sociopolitical, economic, and historical influences and ultimately the impact these contexts have on health decisions and practices. Although the conversations were wide ranging, they were always brought back to questions of decision making and, specifically, mammography (for a full list of questions, see Pasick, Burke, et al., 2009). Informed consent was obtained from all respondents, and study protocols were reviewed and approved by the Committee on Human Research at the University of California, San Francisco. We use the following abbreviations to indicate type of respondent: KI (key informant), GK (gatekeeper), and CW (community woman). A number (e.g., KI5, GK2, CW7) indicates the specific interview participant.
For analysis, all interviews were audiotaped and transcribed verbatim. Interviews in Spanish or Tagalog were translated before being transcribed in English. First, transcripts were coded by three study investigators following a standard inductive anthropological approach (Bernard, 2005). Where discrepancies in coding occurred, two investigators discussed differences until consensus was reached, thus reconciling the coding. We used an iterative and multidimensional analytic process to identify the most salient themes of social capital (Bernard, 2005). A detailed account of study methods including analytic strategies is reported by Pasick, Burke, et al. (2009). The results reported here consist of description of the themes related to understandings of self-efficacy, those underlying the domain of social capital, and our interpretation of the implications of these themes for the construct of perceived self-efficacy.
Our key informants explained, and gatekeepers and lay women illustrated through descriptions of daily life, that women differently experienced and perceived self-efficacy depending on the social context and that these perceptions stem from a dynamic mixture of past and present and from experiences in and connections with the homeland and the United States. This mixture was evident in various spheres of women’s lives, including their experiences with the U.S. health care system, public schools, and taking care of their families in situations of wellness (immunizations) and illness (cancer). Our findings converged under five themes: (a) meanings of self-efficacy, (b) spheres of efficacy, (c) constraints on self-efficacy, (d) sources of social capital (local, transnational, familial, and extrafamilial relationships), and (e) differential access to, and quality of, social capital. A subtheme, social connections, was found to link the five primary themes together.
One of our key informants defined self-efficacy as having to do with “the ability of women to get answers, to get access to care … and to understand—to get information to help them understand what’s going on and what they can expect” (KI5). Another noted,
[The notion of self-efficacy is] built on givens that aren’t givens in a low-income community . . . . In order to have efficacy … you need to know what you’re dealing with . . . . There are tremendous structural barriers that exist that have to be confronted. (KI3)
A third respondent pointed to the need to understand experiences of self-efficacy in realms other than preventive health care, such as the often harrowing immigration experience, to understand how decisions are made about participating in cancer screening such as mammography:
I mean, you’re dealing with women that immigrate to this country from rural villages in Mexico with no education, no language skills … and to survive in this city, you know, you’re not talking about weak people! (KI7)
The more in-depth interviews we conducted with key informants, gatekeepers, and women, the clearer it became that self-efficacy was experienced and perceived differently depending on whether participants were discussing family, home, community, or institutional (school, health care) contexts.
Women talked about efficacy in spheres they found particularly significant, for example, in terms of their ability to keep their family healthy and together or in navigating unfamiliar systems. In other words, conceptualizations of self-efficacy that women shared with us were highly gendered and culturally informed. Women’s efficacy was valued and expected in the familial sphere, but less so when faced with bureaucratic structures and institutional contexts, notably in regard to their own health care. One of our key informants noted that this conception posed a “source of strain on women,” which “often means a set of priorities for partners and children, before women’s own health-care needs” (KI5). The director of a small nonprofit serving Latino youth and mother of four told us,
There’s a saying—everyone will tell you: “Oh, hay tanto que hacer” [“There’s so much to do”]. And it’ll be the weekend: “Oh, hay tanto que hacer.” I mean, that says so much. It’s like there’s always so much for me to do that I can’t do anything else because I have to do all of this first . . . . Yeah, I have my housekeeper but I always have a lot to do because I guess I’m just trained to think there’s always something else I have to do where that takes my time away from taking care of myself medically. (GK11)
Another gatekeeper respondent characterized this prioritization of the family and domestic sphere as a cultural value for many of the women with whom she works.
I think the value of control of the household—that you know what food is cooked, and when and where, where things are put, how things are decorated . . . . I think it’s just having a certain pride in your home and your home life. Like in U.S. culture, there’s much more of an emphasis on what you do for a living as defining who you are. And I think for a lot of Latina women it’s having that sense of your family, knowing your family well, knowing your family’s whereabouts. I think a lot of merit is put on that. (GK10)
That poor women put their families first has become axiomatic in the health behavior literature, often in rather pejorative ways. But the dimensions and meanings of this tendency have not been explored. Women we interviewed talked with pride and determination about enrolling their children in school, monitoring their progress, getting them in to see the doctor, and making sure they had everything they needed. It was often an issue with a child or family member that would push a woman to enter a realm in which she was insecure and uncomfortable, such as the educational or health care system. A teacher at a state university and health educator told us she had not been to the doctor for herself, other than prenatal care, in 15 years. For her children, however, she never missed an immunization shot. She stated,
Well, okay, I think I’m a fairly educated woman myself, and I leave myself last. So why do I do that? Other than, part of it is probably a little bit of fear. Until my friend died [from cervical cancer], I had not been, aside from the prenatal care . . . . I hadn’t had a physical exam since [in] 15 years. And it wasn’t until then, kind of like the slap of reality. Why? Because I was always too busy. But I always was there when the kids were sick. It was the first thing I went to. Never missed an immunization shot. (GK7)
Another woman related how she somehow figured out how to access the health care system and to communicate with providers without interpreters when she took her daughter to the emergency room.
When your child is sick, I don’t know how you manage, but you do. You may use sign language to communicate, but … you do it because … it’s like being afraid of something but then suddenly you forget you were ever afraid. You see your child suffering. I would go anywhere and do anything for her. (CW14)
Women also reported acting on their own behalf in health care encounters, often after having learned to do so from interactions for children and other family members.
Social, cultural, and economic considerations and expectations (unpredictable employment, inflexible work schedules, gendered role responsibilities toward family members and friends) framed women’s experiences and perceptions of self-efficacy. As one gatekeeper told us, about 10% of the students in her class at the state university—women who matriculated in attempts to leave behind house cleaning and service sector jobs—drop out because “they’ve got sick kids, their work schedule is not giving them the flexibility, they had to take another job, their husbands wouldn’t let them. And so that plays into why they don’t seek services” (GK7). Poor women, she continued,
have all these other issues that it’s like, oh yeah, breast cancer is a bad thing, but I have all these other problems, and that’s not high on their priorities. So a health issue is not always a high priority. It’s trying to get food on the table, and that particular health issue is at the bottom. (GK7)
To “make it” and access what they and their families need when living in a context of poverty, many women rely on each other and creatively share resources. As one of our key informants stated,
There is a kind of dynamic where that’s what you need to survive in very poor communities because if you were doing equity theory you couldn’t survive. You always share resources. How you share them is very [important], it’s a very intricate dance. Reciprocity—who it is you choose to share resources with, who it is that you choose to befriend and become part of your family and your community. So it’s a very complicated decision making that then transfers to all spheres of life, including health behavior. (KI7)
For many women we interviewed, family relationships (both local and transnational), especially those involving mothers, daughters, and siblings, provided a major source of social support and social capital, especially when accessing the health care system (also see Washington, Burke, Joseph, Guerra, & Pasick, 2009). Daughters brought their mothers in for appointments, sons and husbands provided transportation and support, and bilingual children often served as interpreters and cultural brokers. A key informant noted that there is “a strong sense that you value family input before institutional … family members help you figure where to go for resources, give you emotional support in relation to problems” (KI5).
A key element evidenced in our study is that these familial relationships are conditioned by the migration experience and are not limited to current location but rather often span national boundaries. These transnational ties add another layer of complexity to the social and cultural contexts of the women in our study (also see Joseph et al., 2009). Several researchers have noted the disruptive influence of migration on family life (Flores-Ortiz, 2004; Sluzki, 1979). Gender roles, parent–child relationships, and role expectations among family members are all affected “as the process of migration challenges the cultural, economic, and psychological resources of individuals and families” (Flores-Ortiz, 2004, p. 272). In this immigrant context, women form new communities that often remain connected to those in the homeland (for a detailed discussion of the effects of these connective ties to health behavior, see Joseph et al., 2009). In addition, roles and definitions of family and knowledgeable resources shift and change in the new context and new acquaintances fill what were formerly familial roles. Having left behind mothers, sisters, and lifelong friends in the home community, women establish “extended family” relationships with coworkers and neighbors who fulfill sibling and other kinship roles in the new place.
The practice of “connecting with someone who has been here longer [and therefore who knows the system] and will help” was repeated throughout the interviews we conducted. This practice yielded essential social capital in the form of health care access, social support, and negotiating skills. Some women drew on social networks to address the need for information and the difficulty they experienced getting it. Providing support for her mother during an illness taught one woman that
you have to have a team of people working with you, to understand what the doctor’s even telling you. And to be your sounding board and your advocate when real problems happen . . . . Or you have to bring people with you that know something about medicine. (CW29)
Another told us of how she looked to other Latinas to learn how to navigate the school and health care systems.
When you go to the schools, especially when you are an immigrant or when we do not know anyone or anything, you look for help from other Latinas, other mothers, who are in the school or at the doctor’s office. You think, well maybe she knows something more than I do since she has spent more time than me here. You look for how to communicate, how to learn more. (CW5)
These experiences of support, like those of other women advocating for their children in health care interactions, changed this woman’s exercise of self-efficacy. As she stated, “What happens to me, now … I confront the situation … now I defend my rights more. I know more about the services. I know more about the laws” (CW5). Other women found this support as members of religious communities—whether Catholic, Protestant, Jehovah’s Witness, or other. One woman recounted her use of social connections through her congregation to ensure quality medical care during a recent hospitalization.
When people realize that you have someone who cares about you, they always treat you differently. In this case Jehovah’s Witnesses are very united . . . . [The hospital staff] admire this and would treat me very well because they noticed how many Jehovah’s Witnesses, how I had so many people and visitors. (CW11)
Another reflected on how her access to resources in the form of social relationships changed over time after her arrival in the United States. When she first arrived, she found that “sometimes friends help more than family” (CW5). She went on to explain,
I feel very fortunate that I found people who were angels in my life … because when I took the children to register at school and everything they told me that they [the children] would need a physical … and I said, “What is that? A physical?” A physical for us was physical education . . . . When the lady from the school told me that and I was embarrassed to ask her what a physical was, I called my friend from school and asked, “Do you know what a physical is?” And since she was a little more involved in those things, she said, “That is to give them an exam, a physical from the doctor. A check-up . . . . Call the number and make an appointment . . . . I will take you.” (CW5)
The social capital women gained from strong local and transnational familial relationships, coupled with the sense of empowerment associated with “knowing their families well” and controlling their households, brought with it a strong sense of self-efficacy and responsibility for those others. Thus, the familial sphere may be both a source of efficacy and strain. A gatekeeper respondent shared concern about telling her sisters that she had found a lump on her breast and needed to go through a mammogram, ultrasound, and biopsy.
Part of it for me internally was I didn’t want to have to tell my sisters. You know what I’m saying? Because it’s like when you’re so intertwined in the family you have to take so much responsibility for everybody else’s feelings and how things are dealt with. (GK10)
Although many women recounted the importance of family and social networks, especially in an immigrant context, their narratives also showed that these networks took time to recreate and were not always easy to establish. Access to the benefits of network-based resources or social capital often depended on where and when one migrated. Migration changed women’s “connectedness” and thus perceptions of self-efficacy in relation to the loss or transformation of social capital. Time is a critical influence here. Without social connections, or when in the process of establishing them, women told us they felt lost. As a gatekeeper told us,
Once [new immigrants] get here, “Okay, I did you a favor, I got you here, you’re on your own,” you know? So, they’re on their own. They have no support and they have to figure out a way to learn how to navigate the system here. And even finding out where something is doesn’t mean that you really understand how it works. (GK4)
Another woman, living with her son and daughter-in-law, did not know how to access health care, despite having been efficacious in the management of a major health problem before immigrating. While in Mexico, she was getting regular cancer screening tests.
The cancer exam, the Pap smear, I usually have every year. Every year. Not here, because I don’t even know where or how or how much or, really. I had it done the day before I came to the U.S. (CW1)
Although she said she would like to get a mammogram here, the facts that she did not know where to go or how much it would cost and that she was dependant on her daughter-in-law (with whom she had a strained relationship) for transportation kept her from doing so. In other words, her loss of social capital was detrimental to her preventive health behavior. She no longer evinced self-efficacy. Although in Mexico she knew where to go, what to do, and who to talk to—including a long-term, trusting relationship with her doctor—here she was at a loss. In her new social context, she was willing but, from her perspective, unable to get a mammogram.
In addition to taking time to establish, the existence of and access to social networks—family, hometown associations, groups of friends—in the new place are not equally distributed. As the director of a small nonprofit organization serving Latino youth told us, it often depends on where one migrated from.
I think that Mexicans know somebody that’s been here, even though they may be undocumented, that can navigate them through the health systems and know they won’t get charged—whereas the immigrants from El Salvador and Nicaragua don’t have anyone here . . . . Mexican immigrants already have support. Just by the town that they’ve come from they’ll find other people. (GK11)
Although strong social networks bring benefits such as trust, reciprocity, and feelings of belonging, there are negative effects as well (Bourgois, 1995; Portes, 1998; Portes & Landolt, 1996). Specifically with respect to accessing preventive or diagnostic breast cancer screening services, the quality of social capital may be influential. Social capital in different contexts may work antagonistically, reflecting different interpretations and values, rather than synergistically. The director quoted above told us about the mixed messages she received after discovering a lump in her breast. Her first inclination was to go to her doctor, but after talking with her mother-in-law and friends, she put off going to the hospital.
Even one time I had a little bump under my arm—so of course I freaked out. I’m calling up [the hospital]: “I want a test. What’s going on?” And everyone is like, “Oh, you should go get it checked out. But, you know, it’s probably grasa [grease]” and this and that or, “Maybe it’s a cyst—try to pop it first.” You know, without going to the hospital—it’s like all these remedies that I could do before I get to the hospital . . . . There’s all this advice: “Drink this tea. Do this.” … It’s just that you have that whole village telling you to do something else. You’re more apt to kind of go through that first because of course you don’t want to think that you could possibly have breast cancer either. (GK11)
Although for many women lack of health insurance or cost plays a role in their decision to try alternative remedies first, expense is not the only driving force. One of our gatekeepers explained that alternatives to the hospital are based in large part in a learned mistrust of the medical system. Experiences of mistrust, misunderstanding, and discrimination counter and contextualize women’s perceptions and experiences of self-efficacy as strongly as perceptions and experiences of social support and social capital resources.
I think in the first place people just have an inherent mistrust of medicine. Especially within the Latino culture, people really want to try to embrace other kinds of treatment. For example, when my mother first came here to the United States, she had appendicitis and my dad could not convince her to go to the doctor. She just wanted to drink teas that she had gotten from an older women and she’d just refuse to go to the doctor. I mean, her appendix practically burst inside of her. They had to do emergency surgery. It’s really common to hear from people: “If I go into the hospital I’m never going to come out alive.” (GK10)
Another gatekeeper, a provider at a community health clinic serving primarily Filipino patients, explained the influence these negative experiences have on her ability to talk her patients into mammography screening.
It’s not the experience of the mammogram, but the environment, the process, the navigating the system, the eligibility [worker] telling them they have to pay … not enrolling them in the program they’re supposed to. Getting a bill is one deterrent for them . . . . They get billed or [complain], “Your people are rude!” Then you lose one and it’s really, really hard because you spend so much time with them trying to get them to understand what a mammogram is. (GK5)
A key informant described the U.S. health care as a “system of dominance over poor people with very limited options” (KI3) that continually constrains women. Even those who do feel confident to complete a health action will be thwarted in their attempts.
The sense that you’re going to teach these women negotiating skills and then the institution is allowed to stay the same, is really not realistic if you want long-term change. No matter what negotiating skills you teach them, they just have to have two or three bad experiences and they’ll never use those skills again. (KI3)
In other words, “systems of dominance” directly affect health behaviors, whether women feel confident in their abilities or not.
Emergent themes from our interviews—meanings of self-efficacy, spheres of efficacy, and constraints on self-efficacy—led us to see the multiplicity of ways that the social context of social capital affects women who are contemplating mammography screening. Women’s responses to their feelings of low self-efficacy (particularly in institutional contexts) were often to “look for someone familiar” to accompany or advise them or “find someone who’s been here longer and can help.” In addition, women told us they turned not just to family members but also to larger aggregates of familiar people in senior centers, churches, and religious groups and other community-based and voluntary organizations. Through these connections and resources—and the awareness of their existence—women accessed benefits such as feelings of trust, reciprocity, support, security, and belonging. However, these benefits were not accessible to everyone and depended on their social position, established relationships, and place of origin. Access to these benefits and the ability to deploy them successfully changed over time as well. Although these solutions could be interpreted as examples of collective efficacy (Bandura, 1997), the meanings and effects of these strategies indicate a wider web of social consequences than that inferred in current conceptualizations of collective efficacy.
Women in our study defined social connections as both local and transnational, framed by immigration histories, family life, and employment and community networks. The migration experience, whether their own or their parents’, provided the backdrop for women’s narratives about health care access, self-efficacy, and network participation, specifically its marginalizing effects. Women who immigrate, whether from the Philippines, Mexico, or elsewhere, face a difficult transition, depending on their migration trajectory and visa type (i.e., whether for professional purposes or family unification or as undocumented immigrants) and experience on arrival in the United States. The upheaval of the move results in the “the falling away of some of the fixed positions of class, family and kinship,” which “means that people are required increasingly to make their own way through social life without these firmer ’identity hooks’ … in a situation of continuing forms of economic inequality” (Williams, Popay, & Oakley, 1999, p. 166).
SCT argues, on one hand, that migration can result in a decrease of one’s sense of self-efficacy after repeated defeats in attempting to “make it” in the new environment. On the other hand, the theory recognizes that migration experiences can increase one’s sense of self-efficacy—that is, “I made it through that, I can do anything” (Jerusalem & Mittag, 1995). Although the application of SCT would lead one to focus on the perception of self-efficacy that immigrants gain or lose based on their migration trajectory, putting this concept in the context of social capital demands analysis be made of the quality and strength of social relations lost, gained, and maintained albeit in altered form. Consideration must be given to change in social relationships during the migration experience and how these network transformations influence access to resources as well as how such transformations influence perceptions of self-efficacy based on access—or lack thereof—to resources. Thus, this perspective raises the question, “How do women from, say, rural Mexico make it in this city?” not only “How do they feel about making it?” or if they feel they can make it or not. And what does this mean? What does this tell us about their ability or willingness to access health care? Or their ability to get a mammogram if they wish to do so? The social capital perspective widens the analytic lens and places perceptions of self-efficacy into a complex context of intertwined material and emotional resources, including reciprocity, exchange, interdependency, trust, indebtedness, and selective relationships.
Although the meanings of self-efficacy expressed by our participants were not necessarily inconsistent with the construct as described in SCT, the complexity they point to and the strong influences of the contexts of poverty and migration highlight the need to address the social and cultural contexts in which self-efficacy is perceived and experienced. Through their discussions of meanings, constraints, and spheres of efficacy, the women we interviewed clearly showed the importance of social capital to perceptions of self-efficacy. They also demonstrated that social capital is not a static or singular concept but is constructed of different forms of network participation, to which access is unequally distributed and in which the quality of social capital varies. Most important, the loss and acquisition of social capital in the context of immigration is a process that commences before leaving a homeland and continues long after arriving in a new location. Social capital addresses aspects of social context absent in the current self-efficacy construct, and these aspects have important implications for our understandings of health behavior and intervention development.
Perceived self-efficacy, as commonly operationalized within an individualistic framework, fails to address the issues the women in this study saw as important and the multiple contexts in which they struggle with their ability to harness resources (variously interpreted as ability to exercise self-efficacy or deploy social capital). This limited interpretation is reflected in the ways in which the construct is measured, often with a single question and, notably, not necessarily in the ways in which it was originally theorized in SCT. But although more complete scales are useful, the specificity of the construct would exclude many varied and important influences. Interventions stemming from this specific and inadequate operationalization (as distinct from the theory) fail to account for the complexities of women’s lives and the processes of prioritization within culturally informed rationalities that influence their decisions about mammography screening.
Just as we elicited multiple understandings of women’s lives to illuminate influences on the ways women acquire and lose “confidence” (e.g., self-efficacy) in their ability to achieve a health action, interventions to enhance this confidence and positively influence behavior should be based on the realities of daily life and all that affect it. As evidenced in our study, “confidence” was often established through relationships and connections with others rather than an individually acquired and assessed attribute. Therefore, understanding how the concept of social capital translates into power and efficacy for individual women will improve and strengthen interventions. Practitioners and researchers should evaluate their own practice structures and procedures (as welcoming, intimidating, etc.) in terms of how these might enhance or decrease interaction to increase trust and opportunities for education. In addition, practitioners can work to facilitate connections between individuals to create effective and supportive networks where social capital might be fostered. This perspective should become fundamental to health promotion and the enhancement of health care consumer skills.
In recognition of the centrality and complexity of social capital, practitioners should be wary of overly simplistic models (e.g., role modeling) and earnestly seek community input at all phases of development. Ideally, this should include community leaders and scholars for multiple perspectives and the identification of existing networks to build community capacity.
Finally, practitioners in community settings can seek out linkages with other organizations such as newcomer services, senior centers, clinic outreach programs, and patient navigation services with the express purpose of increasing social connectivity and thereby expanding interpersonal resources.
This study was funded by the National Cancer Institute (Grant RO1CA81816, R. Pasick, principal investigator). The authors would like to thank the editors and two anonymous reviewers for their thoughtful review and suggestions on earlier versions of this article.
This supplement was supported by an educational grant from the National Cancer Institute, No. HHSN261200900383P.