Erina L MacGeorge. 21st Century Communication: A Reference Handbook. Editor: William F Eadie. Sage Publication. 2009.
If you have a situation that’s upsetting you, you might have a friend who listens attentively and makes you feel better by conveying understanding and sympathy. If you have a problem and can’t figure out how to solve it, a coworker might offer some advice that enables you to improve the situation. And if your car has broken down and you need help getting places until it’s fixed, a family member might offer to drive you around. Communication scholars would describe the friends, coworkers, and family members in these scenarios as providing you with social support, where this term refers to assistance provided by family, friends, or acquaintances rather than professionals (e.g., counselors or therapists). Most of us not only appreciate but come to rely on social support and are providers of support in return. As the Beatles put it, we “get by with a little help from our friends.”
Communication researchers were not the first to focus attention on social support; that distinction belongs to scholars in sociology and epidemiology. (For a historical overview, see Burleson & MacGeorge, 2002.) Since the pioneering research in the early 1970s, social support has been defined and studied in a variety of ways, depending in part on the academic disciplines in which different scholars are trained. Sociologists, for example, tend to conceptualize social support as social integration—that is, the extent to which individuals are embedded in relationships and groups, participate in social activities, or experience a sense of community and belonging. Sociological research has demonstrated that people with greater social integration tend to be healthier and live longer. Psychologists have typically taken a somewhat different approach by focusing on perceived support availability—that is, the individual’s perception that caring behavior from others will be accessible if the need arises. Psychological research on social support has shown that high levels of perceived support availability can “buffer” or prevent some of the negative influences that stress can have on health.
Not surprisingly, when communication scholars began to study social support in the 1980s, they introduced a unique perspective in which social support was conceptualized as supportive communication. In a seminal book that defined this emergent perspective, Burleson, Albrecht, Goldsmith, and Sarason (1994) argued that “social support should be studied as communication because it is ultimately conveyed through messages directed by one individual to another in the context of a relationship that is created and sustained through interaction” (p. xviii). Correspondingly, communication scholars have focused much of their research on “verbal and nonverbal behaviors intended to provide or seek help” (Burleson & MacGeorge, 2002, p. 384). These scholars have a special interest in identifying what kinds of supportive communication have a more (or less) positive effect on recipients; why certain behaviors are more effective than others, either in general or for specific types of people, problems, relationships, or situations; and why some people are more (or less) likely to provide others with high-quality supportive communication. In contrast to sociological and psychological perspectives on social support, communication scholars are not content with knowing that social relationships are beneficial (the sociological perspective) or that feeling loved and cared for is good for you (the psychological perspective); they want to know about the details of supportive behavior—that is, what works best and why, and who provides the best support.
This chapter provides an overview of theory and research on supportive communication. It begins by discussing different types of supportive communication and what is known about the features that make comforting and advice giving more or less effective. This is followed by an overview of theory that helps explain how and why social support can produce positive outcomes for recipients. Subsequent sections address the question of who provides effective support and how researchers study supportive communication. The concluding section considers the future of research on supportive communication.
What Makes Supportive Communication Effective?
Types of Supportive Communication
As suggested by the opening paragraph in this chapter, social support scholars frequently distinguish between different types of supportive communication. Perhaps the broadest distinction that can be made is between communication that is intended to be supportive and communication that is not. Most people have had the unpleasant experience of seeking support from another person and receiving a response that conveyed overt rejection or disinterest. This type of communication is most probably not intended to provide support, nor is it likely to be perceived as supportive by its recipient. However, it is important to note that there can be consequential differences between the intentions of support providers and the perceptions of recipients. For example, support providers sometimes say things that downplay or minimize the importance or seriousness of a problem, perhaps with the intention of making the other person feel better (“It’s not that bad. Life will go on.”). In many cases, these minimizing behaviors are viewed negatively by support recipients even when they recognize the supportive intention behind the message (Burleson & MacGeorge, 2002).
Another distinction is often made between emotional support and instrumental support, with the former understood to be focused on relieving a recipient’s emotional distress and the latter focused on improving a recipient’s problem-solving abilities. This is both an important difference and an oversimplification of how supportive communication really looks and functions. Support providers often attempt to provide both types of support, even in the course of a single message or interaction (MacGeorge, Graves, Feng, Gillihan, & Burleson, 2004). Furthermore, we can distinguish different types of emotional and instrumental support. For example, some scholars have found it useful to classify messages of caring, concern, and sympathy (emotional support) separately from those that convey affection and respect (esteem support), while others may group these types of support together under the heading of emotional support. Again, the effect of a particular type of support may be different from what is intended. For example, advice about what to do could help someone feel better without having any real effect on resolving the problem.
Considering the broad categories of emotional and informational support, research suggests some generalizations about the effects of these support types. Overall, recipients often respond more positively to support providers’ emotional-support efforts than to instrumental support efforts, perhaps because emotional support is relevant to a broader range of circumstances. For example, instrumental support is often perceived as less applicable to problems that cannot be solved (e.g., bereavement; see Servaty-Seib & Burleson, 2007). However, as you may have experienced for yourself, there is considerable variation in the quality of support, whether emotional or instrumental. Some support providers in some situations say things that do an outstanding job of helping reduce distress, promote coping, and resolve problems, whereas others’ support efforts have little or no impact and may even make things worse. Correspondingly, supportive-communication researchers have focused considerable attention on the features or characteristics of emotional support messages that lead to more (and fewer) positive responses and outcomes for recipients. A smaller but growing area of research has examined the same issues with respect to instrumental support, especially advice giving.
Communication of Emotional Support
In the supportive-communication literature, the communication of emotional support is often referred to as comforting. Comforting can take a variety of forms, as evidenced by the comforting efforts of toddlers, who may direct simple behaviors (hugging, offering a toy, saying “I’m sorry”) toward peers who are distressed. As children mature into adulthood, comforting retains important nonverbal elements but becomes increasingly verbal (Clark, MacGeorge, & Robinson, 2008).
The most influential research on the effects of comforting communication has been conducted by Brant Burleson and colleagues, including Wendy Samter and Suzanne Jones. In this paradigm, comforting messages are described as varying in person centeredness, defined as the extent to which messages reflect “an awareness of and an adaptation to the subjective, affective, and relational aspects of communicative contexts” (Burleson & Caplan, 1998, p. 249). Comforting messages low in person centeredness deny another person’s feelings and perspective by criticizing the feelings, challenging their legitimacy, or telling the other how he or she should act or feel. Comforting messages moderate in person centeredness implicitly recognize another’s feelings by attempting to distract the other’s attention from the distressing situation, offering expressions of sympathy, or presenting explanations of the situation that might function to reduce distress. Comforting messages high in person centeredness explicitly acknowledge and legitimate another’s feelings by helping articulate those feelings, elaborating on reasons why the other might feel that way, and trying to place the feelings within a broader perspective. Overall, highly person-centered messages tend to be the most listener centered, emotion focused, and non-evaluative, with moderate- and low-person-centered messages exhibiting fewer of these qualities (see Burleson, 1994, for more details on the distinctions between different levels of person centeredness).
A very substantial body of research indicates that comforting messages high in person-centeredness are perceived as highest in quality (e.g., more sensitive and effective); messages of moderate person centeredness are rated somewhat lower and messages low in person centeredness, considerably lower (Burleson, 2003). Furthermore, recipients of highly person-centered messages (as opposed to moderate- or low-person-centered messages) actually experience greater improvement in their emotional states (Jones & Guerrero, 2001). Consistent with these effects of highly person-centered comforting, research also indicates that individuals who use this type of comforting are seen as more likable and attractive and are more accepted by their peers (Burleson, 2003).
The effects of person-centered messages are highly consistent across studies, but this research also indicates that effects vary somewhat as a function of certain demographic, cognitive, and situational factors. Some of the variables that appear to influence responses to person centeredness in comforting are gender, ethnicity, and cognitive complexity (Burleson, 2003); degree of distress has also become a recent focus of attention (Burleson, in press). Studies examining gender effects indicate that women evaluate highly person-centered comforting somewhat more positively than men, whereas men evaluate comforting messages low in person centeredness somewhat more positively than do women (MacGeorge, Graves, et al., 2004). Studies examining the influence of ethnicity or culture have found that Asian Americans, African Americans, and Chinese Americans distinguish somewhat less strongly between messages low, moderate, and high in person centeredness than do European Americans (Burleson & Mortenson, 2003; Samter, Whaley, Mortenson, & Burleson, 1997). Research on interpersonal cognitive complexity, which refers to the sophistication of cognitive schema for thinking about people and social situations, indicates that individuals with higher cognitive complexity evaluate highly person-centered messages as higher in quality (and messages that are low in person centeredness as lower in quality) than do individuals who are lower in cognitive complexity (Burleson, in press). Finally, a moderate level of distress appears to produce more discriminating evaluations of comforting messages than either low or high levels of distress, with moderately distressed individuals rating messages high in person centeredness more favorably (and messages low in person centeredness less favorably) than people with low or high levels of distress (Burleson, in press).
All this research indicates that the effectiveness of highly person-centered messages is moderated by features of persons and contexts. Recently, Burleson and colleagues have advanced a theory to provide a comprehensive explanation for much of the variation in comforting-message evaluation that is linked to these individual and contextual variables (Bodie & Burleson, 2008; Burleson, in press). This theory, labeled a dual-process approach, argues that responses to supportive communication depend to some extent on how much the recipient attends to and thinks about (i.e., “processes”) the actual content of a supportive message. Imagine two support recipients with very different levels of message processing: one who pays very little attention to what a support provider actually says and one who carefully thinks about every word. In this case, Burleson and colleagues argue that the first recipient will perceive less of a distinction between comforting messages representing different levels of person centeredness and be less affected by this aspect of the message. In contrast, the second recipient will discriminate carefully between messages with different levels of person centeredness and will be substantially affected by whether the message is high, moderate, or low in person centeredness. According to this theory, factors such as the level of distress help determine how much processing recipients give to comforting messages and thus influence their evaluations and responses to those messages. For example, having a moderate level of distress should increase both the ability and the motivation to process compared with a high level of distress (which decreases ability) or a low level of distress (which decreases motivation).
A fuller explication of dual-process theory is beyond the scope of this chapter, but recent studies have provided support for many of the theoretical claims (Burleson, in press). Thus, it may soon be possible to explain the hodgepodge of effects due to diverse demographic, cognitive, and situational factors through a more cohesive model. Despite the significance of this theoretical development, it is also important to emphasize that the influence of factors such as gender and distress on the evaluation of comforting messages is generally moderate; overall, research evidence indicates that highly person-centered messages produce superior outcomes across people and situations.
Communication of Advice
Instrumental support has been given less research attention to date than emotional support or comforting. Some research has indicated that instrumental support is not as generally beneficial as emotional support and also runs a higher risk of negative effects, perhaps because information or advice is more likely to be unnecessary, unwanted, or poorly matched to a recipient’s needs (MacGeorge, Feng, & Thompson, 2008). However, there is an increasing body of research on advice, motivated in part by studies indicating that advice may be the most commonly offered form of support (MacGeorge, Graves, et al., 2004) and in part by an interest in identifying the characteristics of advice messages that cause them to be viewed more or less positively.
Focused research on advice as a form of instrumental support was initiated by Daena Goldsmith (1994), who observed that advice has the potential to threaten the face or public self-image of the recipient. Specifically, because advice tells or recommends what another person should do, think, or say (MacGeorge et al., 2008), it may be in conflict with the recipient’s sense of independence or feelings of competence. However, as Goldsmith argued, advice givers can alter the manner or style in which the advice is given and thereby potentially reduce face threat. For example, a recipient’s sense of independence might be better preserved if the advised action is presented as a suggestion (“Maybe you could …”) or, even more indirectly, in a story about what the advice giver did when dealing with a similar problem. Of course, face threat could also be increased by the style in which advice is given; advice can come across as patronizing, critical, or even rejecting of the recipient (“Well, duh. Isn’t it obvious that you should …”). Research by Goldsmith and others indicates that face threat is a salient issue for advice recipients: When advice is perceived as more face threatening, it is viewed as lower in quality and does less to facilitate the recipient’s ability to cope with the problem; advice with a higher degree of perceived face threat is also less likely to be implemented (Goldsmith & MacGeorge, 2000; MacGeorge, Feng, Butler, & Budarz, 2004). The research findings are less clear about the specific linguistic strategies that do the best job of reducing face threat. Some work suggests that strategies may be best when used in combination (MacGeorge, Lichtman, & Pressey, 2002). For example, “I know you will get this worked out. I only have one little suggestion …” contains language that conveys respect for the recipient’s competence and independence.
Erina MacGeorge, Bo Feng, and colleagues (2004) noted that advice recipients were likely to be influenced by the content of the advice message as well as its style and have undertaken research focused on recipients’ perceptions of advice content (i.e., what they think about the action being advised). Advice recipients are strongly influenced by what is termed response efficacy—that is, the perceived likelihood that the advised action will actually help resolve the problem (Feng & Burleson, 2008). Specifically, when response efficacy is perceived to be high, advice recipients rate the advice more favorably, feel better able to cope, and are more likely to implement the action. Other content factors that also influence advice recipients include the perceived feasibility of the advised action (“Can I do it?”) and the potential limitations or drawbacks of the action (“Will doing this create more problems?”) (MacGeorge, Feng, et al., 2004). Research on advice content also shows that advice givers can influence recipients’ perceptions of advice content, such as response efficacy or limitations. In one recent study, advice messages that contained arguments that supported response efficacy (e.g., “You should do [advised action]. I know this will help you because I did this when I had the problem with …”) caused advice recipients to perceive more response efficacy and rate the advised action more positively than messages that stated the advised action but did not make any further argument in its favor (Feng & Burleson, 2008).
A handful of studies have examined whether demographic or situational factors influence how advice recipients respond. Gender appears to have relatively little influence, with men and women liking advice to about the same degree and being affected to the same extent by style and content factors (MacGeorge, Graves, et al., 2004). Other factors appear to be more powerful influences: A collection of studies show that advice is evaluated more favorably when it comes from advice givers who are seen as experts, who evince confidence in their opinions, and who are in closer relationships with the recipients (MacGeorge et al., 2008). In addition, several studies indicate that the interactional sequencing of advice can have important influences on its outcome. Two studies have indicated that advice is viewed as less face threatening when it is requested, directly or indirectly, than when it is given without being requested (Goldsmith, 2000; Goldsmith & Fitch, 1997). In addition, advice is preferred when it is presented subsequent to efforts at emotional support and problem analysis, indicating that there are complex relationships between comforting, advice, and discussion of the problem (Feng, 2009).
There is relatively little focused research on supportive communication that is not classifiable as either comforting or advice; the study just mentioned (Feng, in press) is one exception. Another exception is Goldsmith (1994), who noted that offers of tangible help (e.g., “Can I help you study for the next exam?”) can create many of the same face concerns as advice. However, offers have not received much attention as a type of support. Given that many supportive interactions include behaviors apart from comforting or advice, additional research is needed to understand what features cause these behaviors to have a positive or negative impact on recipients.
How and Why Does Supportive Communication Help?
As may be evident from the prior section, research on supportive communication has been strongly focused on determining “what works” when people engage in supportive communication. In the process of answering this question, scholars have drawn on a potpourri of theoretical perspectives, and it is probably fair to say that the questions of “how” and “why” have often been placed backstage. Nonetheless, as the study of supportive communication has matured, the importance of theory has become increasingly evident. Without theory to explain how supportive messages affect recipients or why some supportive messages have better or different outcomes than others, it is difficult to build a comprehensive understanding of supportive interactions, or to help people become better support providers.
Short-Term and Long-Term Effects of Social Support
To address the question of how social support is helpful (or unhelpful), it is useful to make a general distinction between short-term and long-term, or proximal and distal, effects (Burleson & MacGeorge, 2002). Supportive communication can—and does—have effects that occur either immediately or within a relatively short time after interacting with a support provider. These effects can include improvement in emotional state, increased capacity for coping, an intention to undertake an advised behavior, and feeling respected. Other short-term effects may include conversational satisfaction, a liking for and attraction to the support provider, and satisfaction with the relationship. Of course, short-term effects are not necessarily positive; to the extent that supportive communication is insensitive and ineffective, there can easily be increased distress, diminished capacity for coping, negative perceptions of the support provider, and similar undesirable consequences (Burleson & MacGeorge, 2002).
In research on supportive communication, these short-term personal and relationship outcomes have typically been emphasized. In fact, the recognition that supportive behavior has important, and relatively immediate, consequences is one of the distinguishing characteristics of the communication perspective on social support (Burleson & MacGeorge, 2002).
In contrast, psychologists, sociologists, and others who study social support have typically emphasized longer-term outcomes of social support, such as physical and psychological health (Cohen, Gottlieb, & Underwood, 2000). This emphasis is quite consistent with conceptualizing social support as social integration or perceived support availability. People tend to have relatively consistent levels of social integration and perceptions of support availability over time (unless significantly disrupted by life transitions or crises), so that it makes sense to examine how different levels of social integration or perceived support availability might affect physical or psychological health over time. In contrast, any single supportive message or interaction is probably less likely to have a detectable influence on health—though one can imagine exceptions, such as advice that prompts someone to immediately stop an unhealthy behavior. It is more likely that the quality of supportive communication could have a cumulative long-term effect on psychological or physical health such that those who receive high-quality supportive messages across time (perhaps from multiple individuals) are healthier than those who receive lower-quality messages. Research is needed to examine this (and similar) possibilities.
Explaining the Benefits of Social Support
The distinction between the short-term and long-term effects of social support continues to be important when considering why social support has positive (or negative) effects because different kinds of theoretical explanations are relevant to these different kinds of effects. To explain the differences in the short-term effects of supportive communication, theorists have offered a variety of explanations, including (a) perceived supportive intent, (b) change in face (i.e., public self-image), (c) information and motivation, and (d) reappraisal (Burleson & MacGeorge, 2002).
One explanation for why supportive communication succeeds or fails at improving affect, coping, and so forth is that it expresses the supportive intention of the provider. To the extent that a support recipient recognizes a provider as communicating with the intention to be supportive, the recipient is probably more likely to feel cared for, liked, and valued than if this perception is absent. These feelings should, in turn, affect their emotional state and capacity to cope. In some cases, support recipients may recognize and benefit from the supportive intention even when the communication itself is less than ideal. On the other hand, the perceived supportive intention may not “override” the impact of a poorly conceived message.
A second explanation for the differing outcomes of supportive communication draws from the theory of face, which has informed much work on advice. Having a problem or needing assistance prevents a person from presenting a public self-image of complete independence or competence; thus, it is potentially face threatening. In addition, many provider behaviors can themselves threaten face, even when the supportive intention is strong (Goldsmith, 1994). Asking questions about the problem can invade privacy, offering advice can come across as critical, and offering emotional support can suggest that the recipient is (overly) dependent on others (Burleson & MacGeorge, 2002). Perceptions of face threat are closely tied to negative feelings such as embarrassment, shame, guilt, and anger. Thus, supportive communication may benefit recipients to the extent that it “rebuilds” any damage to face that occurs, either from the original problem or as a result of the support provider’s actions.
A third explanation for why supportive communication does (or does not) help has to do with the informative content of support messages. This idea is easy to illustrate with respect to advice because the information being given is about some action that the recipient could take to resolve the problem: To the extent that an advised action is perceived as efficacious or feasible and is implemented by the recipient, the information may affect the recipient’s outcomes. In addition, the information in advice messages may affect self-efficacy or motivation. Information can also come from other kinds of support messages: Even comforting messages contain certain kinds of information, such as ideas about the normality or legitimacy of the feelings being experienced by the recipient. These feelings could affect recipients’ perceptions of their problems and their subsequent actions (Burleson & MacGeorge, 2002).
A fourth explanation, based on the idea of reappraisal, has been given the most formal presentation as a theory. Burleson and Goldsmith (1998) advanced the idea that supportive communication works to reduce emotional distress when it encourages recipients to produce detailed, emotion-focused narratives about their problems. Doing this, in turn, helps recipients make sense of their problems and their feelings and, ultimately, helps them “reappraise” the situation in a way that reduces emotional distress. Burleson and Goldsmith argued that this reappraisal process is most likely to result from highly person-centered messages, which legitimate and encourage emotional expression. However, they also connected reappraisal with a larger conversational process that incorporates problem analysis and advice giving.
Although these theoretical ideas are all consistent with at least some existing research evidence, there has been little direct testing. One recent exception is a study testing Burleson and Goldsmith’s reappraisal theory (1998). In this study, Jones & Wirtz (2006) asked research participants to disclose actual problems and to participate in interactions with confederates (research assistants pretending to be other participants). The confederates were trained to provide responses that were high, moderate, or low in person centeredness. In this study, participants who received highly person-centered comforting used the most “emotion words” to describe their experiences (i.e., words such as sad, angry, or frustrated) during the interaction and showed more improvement in their emotional states and a greater change in perspective on the problem (i.e., reappraisal) after the interaction. In addition, there was evidence that these variables formed a causal chain that begins with receiving highly person-centered comforting, continues with the use of emotion words, leads to reappraisal, and then results in emotional change.
Not surprisingly, theory about the long-term effects of social support has developed primarily within the psychological and sociological traditions and, consequently, has focused on social support as social integration or perceived support availability rather than supportive communication. In these traditions, theories of support effects can be loosely classified into main-effect models and buffering-effect models (Cohen et al., 2000). In main-effect models, increased social support is theorized to have a positive, direct impact on health regardless of stress. Research support for these models has come primarily from studies in which social support is conceptualized and measured as social integration. Thus, it appears that being more fully integrated in a network of social relationships and activities leads to health benefits, perhaps because tangible support, information, and social influence combine to produce healthier behavior. In buffering-effect models, increased social support is theorized to “buffer” or prevent the deleterious effects of stress on health, such that when stressors occur, those who feel that they have more social support available are less likely to experience negative health effects (or they experience fewer negative health effects) than those who have less social support. This is thought to occur because perceived availability of support influences how people view or “appraise” their problems, so that those who have greater availability of support see their problems as more manageable, experience less stress, and consequently suffer fewer of the physical and emotional consequences of stress.
Theory on the short-term effects of supportive communication and theory on the longer-term effects of social integration or perceived support availability are yet to be systematically integrated. However, it is not difficult to see how integration of this type could proceed, especially with respect to buffering effects. It is possible, for example, that high-quality supportive communication, provided in the wake of a stressful event or a series of stressful events, leads to stronger perceptions of support availability (among other effects), which in turn, cause these recipients to be less susceptible to the noxious effects of stress on health when subsequent events arise. Efforts to develop theories that provide integrated accounts of the short-term and long-term effects of supportive communication are likely to emerge as a particularly important research area.
Who Provides Effective Supportive Communication, and Why?
For most people, supportive communication is exchanged in the context of close relationships, such as marriage, dating relationships, friendships, and family relationships. Support provision also occurs between work associates, though this is probably more likely when there is a friendship that extends beyond work-based roles. Less frequently, supportive communication may come from what sociologists call “weak ties,” including acquaintances and people engaged in service roles that bring about consistent contact with clients (e.g., hairdressers or restaurant employees with their “regulars”). Although it is also possible to conceive of social support coming from medical and mental health professionals (e.g., doctors or psychologists), most research on social support focuses on nonprofessional helpers.
From both theoretical and practical perspectives, the question of who provides support is probably less important than discovering who provides good support. Answering this question has been an important part of the research on supportive communication, though virtually all the studies have focused on predictors of the person centeredness of comforting, leaving the question largely unaddressed with respect to advice. The most-studied predictors of skill at person-centered comforting are age, sex, and cognitive complexity. In general, the person centeredness of comforting increases with age through adolescence, reflecting developments in the cognitive, emotional, and communicative capabilities necessary to support sophisticated verbal comforting efforts (Burleson, 2003). Gender appears to be a significant predictor of comforting quality throughout the life span, with ample and consistent evidence that girls and women, on average, provide comforting with a higher level of person centeredness than do boys and men (MacGeorge, Graves, et al., 2004). Cultural or social factors probably explain much of this difference: In most cultures and societies, women’s gender roles include the expectation of nurturing and comforting others, and the traditional activities of girls and women (e.g., child care) provide greater opportunity for practice at comforting.
Complementing this research on demographic factors are studies that have examined cognitive complexity as an influence on comforting behavior. As previously discussed, cognitive complexity refers to the sophistication of cognitive schema for thinking about people and social situations (Burleson & Caplan, 1998). Several studies indicate that individuals who are higher in cognitive complexity typically produce comforting messages that are higher in person-centeredness. Furthermore, these studies indicate that differences in cognitive complexity help explain the previously reviewed differences in person-centeredness due to age and gender. Specifically, there is evidence of a causal chain in which being female (vs. male) predicts higher cognitive complexity, which in turn, predicts the use of comforting messages with greater person-centeredness. The same is true for age, though only between early childhood to adolescence. Thus, cognitive complexity not only predicts the person-centeredness of comforting but also helps explain the effects of the other variables.
How is Supportive Communication Studied?
Researchers have used a variety of methods to conduct research on supportive communication. Most research methods can be classified into one of four types or “paradigms”: (1) naturalistic, (2) interaction analysis, (3) message perception, and (4) experimental (Burleson & MacGeorge, 2002). Each of these has contributed to the current understanding of supportive communication, and each has strengths and weaknesses that are complemented by the other paradigms.
The naturalistic paradigm includes research studies with participants who have actually experienced a specific kind of stressful event, either chronic (e.g., an ongoing medical condition) or acute (e.g., the loss of a job). Through interviews or questionnaires, these participants provide reports regarding “helpful” and “unhelpful” communication (and other behaviors) they have received from others. A classic study in this paradigm was conducted by Lehman, Ellard, and Wortman (1986), who interviewed bereaved spouses and parents to determine the types of support they had received, who provided the support, and how these types of support were evaluated. The primary strength of this paradigm is what is termed ecological validity—that is, it examines supportive communication in the “real world,” as actually experienced by people experiencing problems. However, there are important limitations to this paradigm as well. People are not especially good at remembering what others actually say to them, especially when weeks or months have passed, so that participants in these studies often provide general descriptions of intentions and outcomes (“She really tried to be helpful; she made me feel cared about”) rather than the details of what was actually said. This can make it very difficult to determine what would be helpful or unhelpful for other support providers to say in the same or similar situations.
The interaction analysis paradigm includes studies with pairs of participants in a laboratory context; their interactions are audio or video recorded and subsequently transcribed. The transcriptions are then coded (i.e., classified) for different types of supportive communication (e.g., emotional or instrumental) so that the frequency of these support behaviors can be used to predict outcomes of interest (e.g., reduced distress, improved coping, conversational satisfaction). For example, the author of this chapter is currently conducting a study in which college students talk with friends about a current personal problem; the conversations are both videotaped and audiotaped. One obvious advantage of this paradigm is that the actual supportive behaviors are “captured” for subsequent analysis. In addition, researchers have the opportunity to measure (usually via a questionnaire) multiple outcomes of the interaction as well as predictors of what happens during the interactions. In the author’s study, for example, questionnaires are used to measure variables such as the closeness of the relationship between the friends and the intention to implement any advice that was offered during the interaction. The major disadvantage of this paradigm is that participants may react to the laboratory setting by, for example, providing or responding to supportive communication in ways they ordinarily wouldn’t. The quality of studies in this paradigm also depends significantly on what the researchers choose to code; in general, large-scale classification of behavior (e.g., emotional vs. instrumental) does not yield as much useful information as more fine-grained classification (e.g., coding messages for person-centeredness).
In the message perception paradigm, researchers present participants with sets of supportive messages designed or selected to represent features of theoretical interest (e.g., person centeredness or different types of facework). Such messages may be presented as lists or in scripted conversations, on paper or video, or, more recently, via computer. Participants evaluate these messages as third-party observers (e.g., “What do you think of what Jack said to Jane?”) or as hypothetical recipients (e.g., “What would you think if someone said this to you?”). Such evaluations are made on whatever dimensions are of interest to the researcher but typically include one or more factors such as supportiveness, helpfulness, effectiveness, and so on. A recent example of research in this paradigm tested the influence of “argument explicitness” (Feng & Burleson, 2008). In this study, research participants read and evaluated advice messages that either did or did not contain various kinds of arguments in support of the advised action. Because this paradigm allows researchers to focus directly on message qualities of theoretical interest and can be easily carried out with large numbers of participants, it is a particularly popular choice. However, there are obvious differences between actually experiencing a supportive message and making judgments about a message directed at others or imagining oneself as the target of a message in a hypothetical situation. There is also a difference between evaluating a message (e.g., for supportiveness or helpfulness) and assessing its predicted outcome (e.g., reduction in distress), though researchers using this paradigm are increasingly doing both or the latter (MacGeorge, Feng, et al., 2004).
The experimental paradigm encompasses studies in which researchers induce stress or upset (usually mild) in participants and then expose them to supportive messages generated by the researcher or by experimental confederates (i.e., people who are pretending to be ordinary participants but are actually trained by the researcher to do and say specific things). After reading or hearing the messages, participants provide assessments of relevant message evaluations (e.g., helpfulness or sensitivity) and outcomes (e.g., reduction of distress). An important example of work in this paradigm, conducted by Suzanne Jones (Jones & Wirtz, 2006), was described previously as a test of Burleson and Goldsmith’s reappraisal theory. The experimental paradigm brings together the strengths of the interaction analysis and message perception paradigms, permitting the systematic manipulation of specific message features and the evaluation of multiple message effects and outcomes. One obvious challenge in using this paradigm arises from the ethical concerns inherent in inducing upset or stress in research participants; participants cannot be unduly upset, and they must be carefully debriefed (i.e., counseled by the researcher) so that little or no distress remains when the experiment is over. In addition, this research method is typically expensive in terms of time and resources. For this reason, the experimental paradigm is the least used, though its use is on the rise as supportive-communication researchers devise skillful ways to manage ethical issues and train confederates (Jones & Guerrero, 2001).
Collectively, each of these paradigms has been demonstrated to be useful toward building a base of knowledge and will continue to be used as their methods prove relevant to questions being addressed by researchers. To illustrate, although the naturalistic paradigm may seem the most limited, it is often an important starting point for understanding supportive communication in specialized populations, especially if there is reason to believe that the experience or evaluation of support may be somewhat different in those populations than in the general population. Other research methods are also beginning to show promise (Burleson & MacGeorge, 2002).
The Future of Supportive Communication Research and Theory
Although scholarly interest in supportive communication does not have the lengthy history of some other areas in the communication discipline (e.g., persuasion), the past two decades have seen considerable development in theory and research and an increasing numbers of scholars who focus their work in this area. Many, if not all, are motivated by the practical nature of research on social support: Findings have high potential for translation into ways of improving support processes and thereby the lives of all who give and receive support.
As research on supportive communication continues to develop, there are at least three areas that are likely to (and should) receive special attention. First, as discussed in the section on how and why supportive communication is effective, there is a need to move beyond identifying what works to explaining why it works. Part of this endeavor will include making further links between psychological, sociological, and communication perspectives on social support. Second, more research needs to be focused on supportive communication in particularly difficult contexts, such as bereavement (Servaty-Seib & Burleson, 2007), and with specific populations, such as those experiencing major health issues (Goldsmith, 2004). Third, work is needed in the area of training in supportive-communication skills. With all the practical knowledge that is now available about how to provide effective comfort and advice, communication scholars and practitioners need to do considerably more to develop, implement, and evaluate ways of helping people acquire better skills in this area (Burleson, 2003).