Abstract
Two studies examined Hispanic individuals’ preferences for using ten different bias reduction strategies when interacting with a doctor whose beliefs about their group were either ambiguous or clearly biased. Consistent with predictions, participants who imagined interacting with a doctor whose beliefs were ambiguous preferred strategies that facilitate positive doctor-patient interactions, whereas participants whose doctor explicitly endorsed negative stereotypes about their group preferred strategies that address stereotype content. The results also revealed that, regardless of whether the doctor's beliefs were ambiguous or clearly biased, stigma consciousness predicted participants’ preferences for using strategies that address stereotype content. These findings suggest that both doctors’ behavior and individual-level factors influence how minority individuals choose to behave in a healthcare setting.
A great deal of research in psychology has examined the effectiveness of strategies aimed at reducing stereotyping and prejudice, but very little of this work has examined which strategies targets of prejudice choose to use to decrease others’ biases. In order to gain a comprehensive understanding of the ways in which stigmatized targets can effectively address and attenuate stereotyping and prejudice, it is necessary to first understand when stigmatized targets are likely to implement specific bias reduction strategies. The present work extends research on bias reduction by exploring the strategies that minority group members choose to implement during interactions with out-group members who either have or have not demonstrated explicit stereotyping of their group.
This research explores Hispanic individuals’ preferences for using bias reduction strategies in one domain where racial and ethnic minorities are particularly likely to face disparate treatment compared with Whites—the healthcare system. Hispanics often report feeling that their healthcare providers are prejudiced (Casagrande, Gary, LaVeist, Gaskin, & Cooper, 2007; Livingston, Minushkin, & Cohn, 2008), and they are typically less satisfied than are Whites with the healthcare they receive (Ku & Waidmann, 2003). Further, research suggests that doctors and other healthcare providers behave differently during interactions with Hispanic patients compared with White patients. For example, providers often neglect to address important information about behaviors crucial to long-term health, such as smoking cessation (Lopez-Quintero, Crum, & Neumark, 2006; National Healthcare Disparities Report, 2008), regular cancer screening (Mead, Cartwright-Smith, Jones, Ramos, Woods, & Siegel, 2006), and weight loss (National Healthcare Disparities Report, 2008).
In order to reduce racial disparities in health outcomes, it is crucial to document the behaviors that both healthcare professionals and minority patients can implement to lessen the effects of bias on providers’ decision making. Although minority patients should not be burdened by having to reduce the biases of healthcare providers, they are often in the best position to observe bias directed at them and their group in a healthcare setting. The present work explores the strategies that Hispanic individuals wish to use during first-time interactions with a doctor who either has or has not indicated that he holds negative stereotypes about Hispanic patients.
Strategies for Reducing Bias During Intergroup Interactions
When stigmatized targets interact with out-group members who might have biases against their group, they may avoid addressing the bias altogether by ignoring the bias or leaving the interaction entirely, or they may choose to engage in strategies aimed at addressing the bias. In the present research, we examine Hispanic individuals’ preferences for using ten different behavioral strategies during an interaction with a healthcare provider who has or has not expressed that he holds negative stereotypes about their group. Doing nothing (ignoring the bias) is appealing when the perceived costs of trying to address the bias outweigh the perceived benefits (Kowalski, 1996) or when it is unclear that an out-group member's behavior is motivated by bias (Ellemers & Barreto, 2009). Members of stigmatized groups may leave or avoid an intergroup contact altogether when they are concerned about being treated negatively due to their group membership (Shelton & Richeson, 2005). However, when targets are motivated to reduce others’ biases, they might intuit a number of influence strategies that have been shown to reduce bias effectively, such as discrepancy strategies, threat reduction strategies, self-expansion strategies, and individuation strategies (see Schmader, Croft, Whitehead, & Stone, 2013 for a comprehensive review).
Discrepancy strategies like interpersonal confrontation (Czopp, Monteith, & Mark, 2006; Monteith, 1993) and drawing attention to value discrepancies (Rokeach, 1968) operate by making perceivers aware of discrepancies between importantly held egalitarian values or standards and biased attitudes or behavior. These discrepancies arouse feelings of guilt or discomfort that motivate changes in attitudes, beliefs, and behavior toward target group members (Ball-Rokeach, Rokeach, & Grube, 1984; Monteith, 1993; Rokeach, 1968; Son Hing, Li, & Zanna, 2002). The use of confrontation, however, is risky for targets because it can cause backlash (Czopp et al., 2006). Thus, discrepancy strategies may not be the most preferred or effective approach for stigmatized individuals unless the perceived benefits of addressing the bias outweigh the costs. Indeed, research by Rattan and Dweck (2010) suggests that targets who believe that people can change are more likely to confront perpetrators of prejudice than are targets who believe that people have fixed traits and personalities.
Threat reduction strategies address the threat and anxiety that biased individuals experience during interactions with stigmatized group members (Fein & Spencer, 1997; Sinclair & Kunda, 1997). Threat reduction strategies, like providing positive feedback (ingratiation) or asking self-affirming questions, have been shown to reduce bias when implemented by targets, leading to less activation of negative stereotypes among perceivers (Sinclair & Kunda, 1999) as well as greater willingness among prejudiced perceivers to engage in future intergroup interactions (Stone, Whitehead, Schmader, & Focella, 2011). Schmader and colleagues (2013) report that asking self-affirming questions is an appealing strategy to gay, lesbian, and bisexual participants who imagine interacting with a straight coworker, and, therefore, may be an appealing strategy to members of other stigmatized groups.
Self-expansion strategies include encouraging perceivers to focus on common identities that they share with stigmatized targets (Brewer, 2000; Dovidio, Kawakami, & Gaertner, 2000) and promoting perspective taking about the experiences of out-group members (Galinsky & Moskowitz, 2000). These strategies cause perceivers to view an out-group target as part of the self, which leads to increased positive regard and prosocial behavior (Aron et al., 2004; Tropp & Wright, 2001). Although self-expansion strategies have the potential to backfire when used by a target (e.g., Stone et al., 2011), they have been shown to reduce stereotyping and prejudice against both a specific target and the group as a whole (Galinsky & Moskowitz, 2000).
Finally, individuation strategies induce perceivers to think about targets as individuals rather than as stereotypic members of a stigmatized social group (Blair & Banaji, 1996). Members of stigmatized groups can use individuation to reduce prejudice by providing perceivers with information about their unique traits and qualities (Singletary & Hebl, 2009). Perceivers also show reduced stereotype activation after they are primed with a counterstereotypic target (Dasgupta & Greenwald, 2001) or are told to expect counterstereotypic information while forming an impression of a target (Blair & Banaji, 1996). Recent research suggests that targets are likely to employ counterstereotyping strategies that address common negative perceptions of their group. Specifically, Neel, Neufeld, and Neuberg (2013) found that obese people, who are stereotyped as unclean, prioritize wearing clean clothing to counteract this negative stereotype, whereas Black men, who are stereotyped as angry and aggressive, take care to smile during intergroup interactions.
To summarize, the current prejudice reduction literature sheds light on several strategies that stigmatized targets may decide to use to reduce bias during intergroup interactions. However, more research is needed to elucidate the conditions under which members of stigmatized groups prefer to use these strategies during intergroup interactions.
Potential Moderators of Bias Reduction Strategy Preferences
The present research examines what bias reduction strategies Hispanic individuals prefer to use when they are the potential targets of negative stereotyping in a doctor's office. Two studies examine, first, whether Hispanic patients’ preferences for using the bias reduction strategies outlined above differ as a function of whether they are interacting with a doctor who has unambiguously expressed negative stereotypes about their group, compared with a doctor whose beliefs about them are more ambiguous. Participants were asked to imagine that they were about to interact with a new doctor whom they overheard making a negative remark that either did or did not endorse a negative stereotype about Hispanic patients (i.e., bias was either unambiguous or ambiguous) and were then asked to rate the likelihood of ignoring the remark, leaving the office before the appointment, or using discrepancy strategies (pointing out a value discrepancy, confronting the doctor about his remark), threat reduction strategies (ingratiating the doctor, asking self-affirming questions), self-expansion strategies (highlighting common identities, inducing perspective taking), or individuation strategies (individuating themselves, providing counterstereotypic information about themselves).
Research on doctor-patient interactions suggests that patients are motivated to establish positive relationships with their doctors to increase the likelihood that their doctor listens to and addresses their health concerns (Aron, Mashek, & Aron, 2004; Rid, Shaw, Lewis & Salisbury, 2010). Thus, in the case of interactions with a doctor who has not expressed clear bias against their group, we predict that Hispanic patients are likely to prefer strategies that will promote a positive interpersonal interaction with their doctor, like self-expansion and threat-reduction strategies.
When their doctor expresses clear, unambiguous negative stereotypes about Hispanic patients, on the other hand, the patients’ goals may shift from developing a positive doctor-patient relationship to mitigating the effect of their provider's bias on their health outcomes. Previous research suggests that, before targets choose to confront a perpetrator, they must first interpret an event as bias or discrimination (Ashburn-Nardo, Morris, & Goodwin, 2008). When targets construe an individual's behavior as blatantly biased, they experience more anger than after exposure to more subtle forms of bias (Bennett, Merrit, Edwards, & Sollers, 2004), and feelings of anger are associated with greater endorsement of confrontational strategies (Matheson & Anisman, 2009). Thus, compared with when the doctor's comment is more ambiguous, we expect that participants in the unambiguous bias condition will exhibit greater preference for strategies that directly address bias, like individuation strategies, which serve to distance them from negative group-based stereotypes, and confrontational strategies, which induce guilt in perceivers.
In the present research, we also examined whether stigma consciousness, or the extent to which members of stigmatized groups expect to be stereotyped by others (Pinel, 1999; Pinel, 2004), moderates strategy preferences. According to Stigma Consciousness Theory (Pinel, 1999), individuals high in stigma consciousness believe that stereotypes about their group permeate how they are treated when they interact with out-group members. As a result, they are more likely to feel disrespected during interactions with out-group members, suffer lower self-worth, attribute their outcomes to discrimination, and, over time, disengage from and avoid situations in which they feel like the target of stereotyping (Pinel, 1999).
In the present study, it was expected that the higher participants were in stigma consciousness, the more bias they would perceive in the doctor. Further, because individuals high in stigma consciousness are more likely to attribute ambiguous outcomes to discrimination than are individuals lower in stigma consciousness (Pinel, 2004), the relationship between stigma consciousness and perceived bias was expected to be particularly strong when a doctor's beliefs about their group were ambiguous. Thus, we expected that stigma consciousness would moderate perceptions of bias in the ambiguous bias condition (Pinel, 2004).
Individuals high in stigma consciousness also tend to avoid situations where they are likely to be the target of stereotyping (Pinel, 1996; Pinel & Paulin, 2005; Son & Shelton, 2011), so it was expected that, across levels of bias, participants high in stigma consciousness would express more desire to leave the doctor's office than participants low in stigma consciousness. Finally, individuals high in stigma consciousness are often preoccupied with confirming negative stereotypes about their group and, therefore, try to manage their identities in a way that disconfirms stereotypes (Shelton, Richeson, & Salvatore, 2005). Thus, participants high in stigma consciousness were expected to show greater preference for identity management strategies like counterstereotyping or individuation, compared with participants lower in stigma consciousness. Because stigma consciousness represents a general orientation that guides stigmatized individuals’ behavior in situations where they perceive they may be the targets of prejudice (Pinel, 1999), we did not predict that stigma consciousness would moderate strategy preferences across conditions.
Study 1
Methods
Participants
Sixty-five Hispanic undergraduate students (36 female, 29 male; M age = 18.7, SD = 1.67) from a university in the Southwestern United States participated in exchange for partial course credit. Participant gender was not found to differ by bias condition, X2(1) = .84, p = .46.
Materials
Health Stigma Consciousness was measured using a modified version of the Stigma Consciousness Scale (Pinel, 1999). The items were tailored to assess concerns about being negatively stereotyped by doctors (e.g., “When interacting with doctors, I feel like they interpret all of my behaviors in terms of the fact I am Hispanic”). On a scale from 1 (strongly disagree) to 5 (strongly agree), participants rated the degree to which they agreed with the items. One item was found not to correlate highly with the others, so it was excluded from analyses, leaving a 9-item measure (α = .74). Health stigma consciousness did not differ across bias conditions, p = .65, or as a function of gender, p = .34.
Doctor's Office Scenario and Bias Manipulation
Participants were randomly assigned to read a scenario in which a doctor expressed an explicit negative stereotype about Hispanic patients (unambiguous bias condition) or no explicit belief about Hispanic patients (ambiguous bias condition). Specifically, all participants read the following text:
Imagine that you have just moved to a new city. You hurt your back lifting heavy boxes and really need to see a doctor to have it checked out. You decide to visit the nearest doctor's office that you can afford. Upon arriving at the doctor's office and checking in, you are escorted to the examination room. As you sit there waiting to be seen, you overhear Dr. Smith say to the receptionist that he does not want to help you.
The subsequent text differed between conditions to manipulate the perception that the doctor held a negative stereotype toward Hispanic patients. In the unambiguous bias condition, the doctor was described as saying, “Hispanics just never follow my medical advice.” In the ambiguous bias condition, the doctor was described as saying, “I'm just tired and I want to go home.” Thus, he expressed reluctance to treat them in both conditions, but provided different reasons for his reluctance.
Bias Reduction Strategies
The bias reduction strategies were presented as specific actions that participants could perform either before or during their meeting with the doctor. First, participants were provided avoidance responses of doing nothing (i.e., “I would ignore what he said”) and leaving the office before the appointment (i.e., “I would leave the doctor's office”). They were next provided with strategies that could be used during the interaction. There were examples of confrontation (e.g., “I would directly confront the doctor about his bias”), inducing self-affirmation (e.g., “I would ask the doctor to talk positively about himself”), inducing perspective-taking (e.g., “I would try to do things that would get the doctor to see the situation from my perspective”), providing counterstereotypic information (e.g., “I would try to behave in ways that highlight positive aspects about me that are Hispanic”), providing individuating information (e.g., “I would volunteer personal information about myself so that the doctor could get to know me as an individual, not just a patient”), highlighting a common identity (e.g., “I would try to get the doctor to see us as a team to make our visit go more smoothly”), inducing value discrepancy (e.g., “I'd ask the doctor if he values diversity and then remind him of ways in which he might unfairly stereotype people of my ethnicity”), and ingratiation (i.e., “I would find a way to compliment the doctor or agree with his opinion on something”). The order was held constant across conditions.
On scales of 1 (definitely not) to 5 (definitely), participants rated the extent to which they would engage in the strategy (i.e., intention), they could engage in the strategy (i.e., implementation efficacy), and the extent to which they thought the strategy would work (i.e., bias reduction efficacy). For each strategy, these ratings were highly correlated, so they were combined to form an index of their desire to use the strategy (α's ≥ .66). See Appendix A for full list of strategies and individual alpha values.
Manipulation Check
Participants completed four items that assessed the strength of the bias manipulation. On a scale from 1 (strongly disagree) to 7 (strongly agree), participants rated the extent to which they agreed with two statements measuring perceived bias (“I think the doctor holds negative stereotypes about Hispanics” and “I think the doctor does not like Hispanics”) and two items that measured attributions that the doctor's comment was due to factors other than bias (“I think the doctor was probably tired from a long day at work” and “I think the doctor was probably too busy to see patients today”). The perceived bias items were averaged, as were the non-bias attribution items.
Demographic Questionnaire
Participants were asked to provide their ethnicity, gender, parental education, and family income.
Procedure
Participants were recruited for a study on “reactions to different situations when people visit the doctor.” When they arrived at the lab, participants first provided consent and then were informed that they were participating in a study examining how people respond to different types of healthcare situations. Next, they completed the health stigma consciousness scale, read either the unambiguous bias or ambiguous bias doctor's office scenario, and then completed the bias reduction strategy measure and the demographic questionnaire. They were then debriefed, given course credit, and thanked for their time.
Results
Preliminary analyses of the data indicated that there were no main effects of interactions involving participant gender. Thus, gender was dropped from the analyses. Parental education and family income were invariant and significantly skewed, p's < .001, so they were excluded from further analysis.
Manipulation Checks
The averaged non-bias attribution items were subjected to a 2 (condition: ambiguous bias (coded as 0) v. unambiguous bias (coded as 1)) X health stigma consciousness (continuous, centered) regression analysis. Results revealed a significant main effect of condition, β = –2.22, t(60) = −7.09, p < .001, indicating that participants in the ambiguous bias condition were significantly more likely than participants in the unambiguous bias condition to attribute the doctor's comments to being busy or tired. Neither the main effect of health stigma consciousness nor the condition by health stigma consciousness interaction reached statistical significance, p's > .46.
Regression analyses were then conducted on the average perceived bias items to examine the 2 (condition: ambiguous bias (coded as 0) v. unambiguous bias (coded as 1)) X health stigma consciousness (continuous, centered) design. Results revealed a marginally significant interaction between condition and health stigma consciousness for the measure of perceived bias, β = −.81, t(60) = −1.94, p = .058. Follow-up analyses indicated a significant effect of condition at both one standard deviation below and above the mean for stigma consciousness, β = 3.23, t(60) = 6.21, p < .001, β = 1.80, t(60) = 3.52, p < .001, respectively. Specifically, as predicted, participants in the unambiguous bias condition perceived the doctor as more biased than participants in the ambiguous bias condition. Also, whereas stigma consciousness was not related to perceived bias in the unambiguous bias condition, β = −.14, t(60) = −.44, p = .66, there was a significant positive relationship between stigma consciousness and perceived bias in the ambiguous bias condition, β = .67, t(60) = 2.43, p = .02. Participants higher in stigma consciousness inferred more bias from the doctor in the ambiguous bias condition.
These results indicate that our manipulation of bias was successful: participants in the unambiguous bias condition perceived the doctor as more biased than participants in the ambiguous bias condition. Further, consistent with previous research (Pinel, 2004), participants in the ambiguous bias condition were more likely to perceive the doctor's comment was due to bias the higher they were in stigma consciousness.
Strategy Use
Inspection of the correlations among the ratings of the strategies indicated that only ratings on the confrontation and value discrepancy items were highly correlated, r = .51, p < .001. Due to the similar mechanisms through which these strategies reduce bias, they were combined into one item labeled “confrontation.” No other correlations between items reached conventionally accepted levels of “high” correlation, r's < .47.
The SAS GLM package was used to examine the 2 (level of bias: ambiguous v. unambiguous) X 9 (strategy: ignore v. leave v. confront v. perspective-taking v. common identity v. self-affirmation v. counterstereotyping v. individuation v. ingratiation) X health stigma consciousness (continuous, centered) mixed-model design. Bias condition and health stigma consciousness served as between-subjects variables and strategy served as a within-subjects variable. Results revealed a significant main effect of strategy, F(8, 488) = 9.02, p < .001, which was qualified by the predicted interaction between bias condition and strategy, F(8, 488) = 2.54, p = .01. No other effects reached statistical significance, F's < 1.65, p's > .10, suggesting that health stigma consciousness did not moderate the relationship ambiguity of bias and the preferences that participants reported for how to respond to the doctor's comment.
To unpack the ambiguity of bias by strategy interaction, ratings for implementing the nine strategies were compared between the bias conditions and within each bias condition (see Table 1 for a summary of means and standard deviations by condition). Consistent with hypotheses, comparisons between the conditions revealed that participants were more interested in using a confrontational strategy when the doctor's bias was unambiguous compared with ambiguous, F(1, 63) = 5.55, p = .02. Alternatively, and supporting our hypotheses, participants in the ambiguous bias condition were more interested than participants in the unambiguous bias condition in using perspective taking, F(1, 63) = 5.53, p = .02, and affirmation, F(1, 63) = 4.69, p = .03. The use of the other strategies did not vary significantly between the conditions, F's < 2.90, p's > .09.
Table 1.
Study 1: Means and standard deviations of strategy preference ratings by condition.
| Unambiguous Bias Condition | Ambiguous Bias Condition | |||
|---|---|---|---|---|
| Strategy | M | SD | M | SD |
| Ignore | 2.91 | 1.05 | 3.06 | 1.02 |
| Leave | 3.28 | 1.03 | 3.03 | .98 |
| Confront | 3.06 | .84 | 2.61 | .62 |
| Perspective Taking | 3.17 | .95 | 3.62 | .86 |
| Common Identity | 2.73 | .83 | 2.87 | .68 |
| Affirmation | 2.12 | .72 | 2.54 | .68 |
| Counterstereotyping | 2.92 | .71 | 2.86 | .78 |
| Individuate | 2.72 | .69 | 2.78 | .78 |
| Ingratiate | 2.75 | 1.17 | 3.08 | .85 |
Within the unambiguous bias condition, preferences for using the different strategies varied significantly, F(8, 264) = 6.46, p < .001 (see Table 2 for a summary of comparisons). Consistent with hypotheses, when the doctor's bias was clear, participants expressed significantly greater desire to confront the doctor than to use ingratiation, common identity, and affirmation, p's < .03. Also consistent with hypotheses, participants were somewhat, albeit not significantly, more interested in using a counterstereotyping strategy than using ingratiation and common identity, and they were significantly more interested in using counterstereotyping than affirmation, p < .001. Indeed, participants reported significantly less desire to use affirmation than any other strategy, p's < .001. Participants were also significantly more interested in leaving the doctor's office than using any other strategy except confronting the doctor or using a perspective taking strategy, p's < .02. Inconsistent with hypotheses, however, there was no difference in how interested participants were in confronting the doctor and using a perspective taking strategy, p = .44, and participants did not exhibit greater interest in using individuation than ingratiation or common identity, p's > .13.
Table 2.
Study 1: F values for comparisons between strategy preferences in the unambiguous bias condition. Bolded values represent contrasts that reached statistical significance.
| Strategy | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 |
|---|---|---|---|---|---|---|---|---|
| Ignore | 7.78 | 1.31 | 3.67 | 1.93 | 35.2 | .01 | 2.14 | 1.53 |
| Leave (2) | - | 2.71 | .77 | 17.5 | 76.0 | 7.35 | 18.1 | 16.2 |
| Confront (3) | - | .60 | 6.41 | 50.0 | 1.13 | 6.78 | 5.67 | |
| Perspective Taking (4) | - | 10.9 | 61.5 | 3.37 | 11.4 | 9.94 | ||
| Common Identity (5) | - | 20.6 | 2.16 | .01 | .02 | |||
| Affirmation (6) | - | 36.1 | 20.0 | 22.0 | ||||
| Counterstereotyping (7) | - | 2.37 | 1.73 | |||||
| Individuate (8) | - | .05 | ||||||
| Ingratiate (9) | - |
Preferences also varied significantly among participants in the ambiguous bias condition, F(8, 240) = 5.22, p < .001 (see Table 3 for a summary of comparisons). Consistent with hypotheses, participants expressed more desire to use perspective taking than any other strategy, p's < .001, and they were significantly more interested in ingratiating the doctor than either confronting him about his remark or using an individuation strategy, p's < .05. Participants were not more interested in using a common identity strategy than any of the strategies aimed at directly addressing the content of the doctor's negative stereotype, p's > .54, however, and they were significantly less interested in using affirmation than every strategy except individuation and confrontation, p's < .03.
Table 3.
Study 1: F values for comparisons between strategy preferences in the ambiguous bias condition. Bolded values represent contrasts that reached statistical significance.
| Strategy | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 |
|---|---|---|---|---|---|---|---|---|
| Ignore | .05 | 10.6 | 15.9 | 2.00 | 14.1 | 2.10 | 4.14 | .01 |
| Leave (2) | - | 9.19 | 17.8 | 1.42 | 12.5 | 1.51 | 3.28 | .10 |
| Confront (3) | - | 52.5 | 3.38 | .25 | 3.26 | 1.49 | 14.4 | |
| Perspective Taking (4) | - | 29.2 | 60.0 | 29.6 | 36.3 | 15.2 | ||
| Common Identity (5) | - | 5.49 | 0.00 | .38 | 2.25 | |||
| Affirmation (6) | - | 5.33 | 2.97 | 14.8 | ||||
| Counterstereotyping (7) | - | .34 | 2.36 | |||||
| Individuate (8) | - | 4.50 | ||||||
| Ingratiate (9) | - |
Discussion
Many of the results in Study 1 support the hypothesis that Hispanic individuals interacting with a doctor who explicitly endorsed a negative stereotype about their ethnic group would prefer behavioral strategies that directly address the content of the doctor's stereotype, whereas participants who imagined interacting with a doctor whose beliefs were more ambiguous would prefer strategies aimed at fostering a positive interaction with the doctor. Specifically, participants in the unambiguous bias condition expressed more desire than participants in the ambiguous bias condition to confront the doctor, whereas they were significantly less interested than participants in the ambiguous bias condition in affirming the doctor or asking him to take their perspective. Taken together, the results of Study 1 suggest that when they are interacting with a doctor who is not clearly biased against their group, Hispanic patients want to take steps to establish a positive interpersonal relationship with their doctor. When their doctor expresses unambiguous bias against their group, however, Hispanic patients’ efforts shift towards addressing the doctor's negative perceptions of their group.
Consistent with predictions, stigma consciousness influenced perceptions of bias in the ambiguous bias condition. It was not found to moderate participants’ desire to use the bias reduction strategies, however. This pattern suggests that individual differences in concerns about being the target of prejudice in a healthcare setting may not be a primary predictor of minority patients’ interaction strategy preferences. Instead, the presence or absence of explicit bias may be the most important predictor of strategy preferences.
The ratings for some strategies were not as expected. First, although perspective taking was rated more positively in the ambiguous bias condition than in the unambiguous bias condition, it was one of the highest rated strategies among participants in both conditions. Further, participants in both conditions rated perspective taking as a more desirable strategy than highlighting common identities, despite the fact that both of these strategies reduce bias through the same mechanism—increasing perceived self-other overlap. This is likely because most patients are motivated to get their doctor to take their perspective, either about their health concerns (Rid et al., 2010) or about their concerns about being the targets of bias, during an appointment.
Another unexpected effect was that, although participants in the ambiguous bias condition again expressed more desire to use affirmation than participants in the unambiguous bias condition, this strategy was not rated very highly among the sample of participants. These data do not replicate the findings that Schmader and colleagues (2013) reported, and the inconsistent results may be due to a number of differences between the two studies. For example, participants in the study by Schmader and colleagues (2013) imagined interacting with an equal status peer, whereas participants in the present study imagined interacting with a doctor—someone in a higher status position than him or herself. In the present study, affirming a higher-status interaction partner may be an unattractive strategy if it is thought to further elevate a biased individual's positive self-regard and contribute to the power differential.
Study 2
The results of Study 1 indicate that health stigma consciousness was not a significant moderator of participants’ strategy preferences. One reason for this may be that, whereas stigma consciousness may reflect one's attention to bias, it does not measure an individual's perceived efficacy or agency in addressing bias. In Study 2 we added a measure of participants’ feelings of agency during interactions with healthcare providers to clarify the relationship between stigma consciousness and bias reduction strategy preferences.
Research in the health domain indicates that patients who feel a sense of agency regarding their health outcomes have more collaborative relationships with their doctors than patients lower in agency (Faulkner, 2001). Thus, individuals higher in health agency may feel more empowered to engage in strategies that would facilitate a positive relationship with their doctor, like asking their doctor to take their perspective or highlighting identities that they and their doctor share. Individuals higher in agency are also less likely to attribute their outcomes to discrimination than those lower in agency because they tend to feel a greater sense of control over their outcomes (Major & Kaiser, 2005; Meyer 2003). Given this, patients who are higher in health agency may also be less likely to attribute the doctor's comment to bias, particularly in the high bias condition.
Because health stigma consciousness increases perceptions of bias, as found in Study 1, and agency can decrease perceptions of bias, however, these two factors may interact to impact perceptions of bias and the subsequent desire to use the bias reduction strategies. Individuals high in stigma consciousness but low in agency may be less likely to confront the doctor than individuals high in both stigma consciousness and agency, whereas individuals low in stigma consciousness but high in agency may be particularly likely to use strategies that foster a positive interaction with their doctor. Thus, we expected that Hispanic patients would be more likely to use strategies that benefit the doctor-patient relationship the higher they were in health agency, and this is especially likely in the unambiguous bias condition when they are faced with a challenging doctor-patient dynamic.
The same procedures were used as in Study 1 with a few exceptions. First, participants completed the study online instead of in a laboratory. Second, participants completed a measure of agency in healthcare before completing the measure of health stigma consciousness. It was again predicted that participants in the unambiguous bias condition would prefer strategies that directly address bias, like confrontation and providing counterstereotypic information, and that participants in the ambiguous bias condition would prefer strategies that would help to facilitate a positive interaction with the doctor, like self-expansion strategies and threat reduction strategies. Finally, it was predicted that participants higher in agency would be more likely to use strategies that would help their concerns be heard, like perspective-taking, and strategies that help to foster a positive interaction with their healthcare provider, like affirmation, ingratiation, and highlighting common identities.
Methods
Participants
One hundred and five Hispanic undergraduate students (67 female, 37 male, 1 unreported; M age = 19.3, SD = 2.18) from a university in the Southwestern United States participated in exchange for course credit. Participant gender was not found to differ by condition, X2(1) = .73, p = .42.
Materials and Procedure
Health Agency
Participants responded to a 7-item measure of their feelings of agency during interactions with doctors (α = .84). Specifically, on a scale from 1 (strongly disagree) to 5 (strongly agree) participants rated the extent to which they agreed or disagreed with items like, “During a doctor's visit, I feel I play an active role in my healthcare” and “During a doctor's visit, I feel comfortable being assertive and speaking my mind to my doctor.” Agency did not differ across bias conditions, p = .36, nor as a function of gender, p = .70.
Health Stigma Consciousness
The same 9-item measure of health stigma consciousness was used as in Study 1 (α = .84). Health stigma consciousness did not differ across bias conditions, p = .99, nor as a function of gender, p = .67. Health agency and health stigma consciousness were not highly correlated, r = −.19, p = .06.
Doctor's Office Scenario and Bias Manipulation
The same bias manipulation procedures and measures of bias reduction strategy preference were used as in Study 1.1
Manipulation Checks
After they completed the measure of bias strategy use, participants completed five items that assessed the impact of the bias manipulation. In addition to the four items used in Study 1, participants were also asked to rate the degree to which they agreed with the statement “I think the doctor holds negative stereotypes about Hispanics, and that these stereotypes often affect the way he treats Hispanic patients” on a scale from 1 (strongly disagree) to 7 (strongly agree). As in Study 1, the perceived bias items were combined, as were items assessing whether participants attributed the doctor's comment to something other than bias.
Participants then provided demographic information. After completing the study online, participants took part in an in-person debriefing session where they were informed about the goals of the study, given course credit, and thanked for their participation.
Results
Preliminary analyses of the data indicated that there were no main effects of interactions involving participant gender. Thus, gender was dropped from the analyses. Parental education and family income were again invariant and significantly skewed, p's < .001, so they were excluded from further analysis.
Manipulation Checks
The SAS GLM package was used to examine the 2 (condition: ambiguous bias v. unambiguous bias) X health agency (continuous, centered) X health stigma consciousness (continuous, centered) between-subjects design.
Attributions to Factors Other than Bias
Replicating Study 1, results revealed a significant main effect of condition, such that participants in the ambiguous bias condition (M = 4.62, SD = 1.43) were significantly more likely than participants in the unambiguous bias condition (M = 2.71, SD =1.34) to attribute the doctor's comment to his being tired and busy, F(1, 97) = 94.7, p < .001. No other effects reached statistical significance, F's < 6.71, p's > .06.
Perceived Bias
Results revealed a significant main effect of condition, F(1, 97) = 57.4, p < .001, such that participants in the unambiguous bias condition (M = 4.78, SD = 1.58) perceived the doctor as more biased than participants in the ambiguous bias condition (M = 2.62, SD = 1.61). Results also revealed a significant main effect of health stigma consciousness, F(1, 97) = 11.7, p < .001, that was qualified by a significant interaction between condition and stigma consciousness, F(1, 97) = 12.2, p < .001. Follow-up analyses revealed that there was a significant effect of condition at both one standard deviation below and above the mean for stigma consciousness, β = 3.11, t(101) = 4.43, p < .001, β = 1.21, t(101) = 2.95, p = .004. Further, whereas there was no relationship between health stigma consciousness and perceived bias in the unambiguous bias condition, β = .05, t(101) = .20, p = .84, there was a significant positive relationship between health stigma consciousness and perceived bias in the ambiguous bias condition, β = 1.43, t(101) = 4.43, p < .001. Replicating Study 1, participants perceived the doctor as more biased in the unambiguous bias condition than in the ambiguous bias condition, but among participants in the ambiguous bias condition, greater health stigma consciousness was associated with greater perceptions of bias.
There was also a significant interaction between health agency and condition, F(1, 97) = 6.97, p = .01. Follow-up analyses revealed that there was a significant effect of condition at both one standard deviation below and above the mean for health agency, β = 2.70, t(101) = 6.20, p < .001, β = 1.71, t(101) = 3.84, p < .001, respectively. Further, whereas there was no effect of agency in the ambiguous bias condition, β = .01, t(101) = .04, p = .97, there was a significant negative relationship between health agency and perceived prejudice in the unambiguous bias condition, β = −.63, t(101) = −2.41, p = .02. While the doctor was again perceived as more biased in the unambiguous bias condition than in the ambiguous bias condition, the higher participants were in health agency, the less likely they were to perceive the doctor as biased in the unambiguous bias condition. No other effects reached significance, F's < .46, p's > .50.
The manipulation checks again revealed that the doctor was perceived as more biased in the unambiguous bias condition than in the ambiguous bias condition, suggesting that our manipulation of bias again was successful. Health stigma consciousness increased the likelihood that the doctor was perceived as biased in the ambiguous bias condition, consistent with previous research (Pinel, 2004) and the results of Study 1. Finally, feelings of agency during doctor-patient interactions were negatively associated with perceived bias in the unambiguous bias condition, suggesting that when the doctor expressed a negative stereotype about Hispanics, participants higher in agency were less likely to attribute the doctor's comment to bias than individuals lower in agency, likely because they feel a greater sense of control over their outcomes during the interaction than participants lower in agency (Major & Kaiser, 2005; Meyer 2003).
Strategy Use
As in Study 1, participants’ ratings for the confrontation and value discrepancy responses were highly correlated, r = .75, p < .001, so they were combined into one confrontation measure.
The SAS GLM package was used to examine the 2 (condition: ambiguous bias v. unambiguous bias) X 9 (strategy: ignore v. leave v. confront v. perspective-taking v. common identity v. self-affirmation v. counterstereotyping v. individuation v. ingratiation) X health agency (continuous, centered) X health stigma consciousness (continuous, centered) mixed-model design. Bias condition, health agency, and health stigma consciousness served as between-subjects variables and strategy served as a within-subjects variable.
Replicating Study 1, results revealed a significant main effect of strategy, F(8, 776) = 9.57, p < .001. Results also revealed a main effect of health agency, F(1, 97) = 12.2, p < .001. The main effects of condition and health stigma consciousness were marginally significant, F(1, 97) = 3.37, p = .07, F(1, 97) = 3.50, p = .06, respectively.
These main effects were qualified by significant two-way interactions between strategy and condition, F(8, 776) = 7.90, p < .001, strategy and agency, F(8, 776) = 2.52, p = .04, agency and condition, F(1, 97) = 6.40, p = .01, and health stigma consciousness and condition, F(1, 97) = .31, p < .01. The interaction between strategy and health stigma consciousness was marginally significant, F(8, 776) = 1.76, p = .08. No other higher order interactions were significant, F's ≤ 1.30, p's > .23.
Strategy Preference Within and Between Conditions
Preference ratings for using the nine bias reduction strategies were compared between the bias conditions and within each bias condition (see Table 4 for a summary of means and standard deviations by condition). Across conditions, and consistent with hypotheses, participants in the unambiguous bias condition reported more desire to confront the doctor than participants in the ambiguous bias condition, F(1, 103) = 28.7, p < .001. Participants were also more interested in using a counterstereotyping strategy when bias was unambiguous compared with ambiguous, F(1, 103) = 4.44, p < .05. However, when bias was ambiguous compared with unambiguous, participants expressed more desire to ignore the comment, F(1, 103) = 15.8, p < .001, and affirm and ingratiate the doctor, F(1, 103) = 4.53, p < .05, F(1, 103) = 16.1, p < .001, respectively, again supporting our hypotheses. Preferences for using the other strategies did not differ as a function of condition.
Table 4.
Study 2: Means and standard deviations of strategy preference ratings by condition.
| Unambiguous Bias Condition | Ambiguous Bias Condition | |||
|---|---|---|---|---|
| Strategy | M | SD | M | SD |
| Ignore | 2.37 | 1.11 | 3.00 | 1.04 |
| Leave | 2.71 | 1.24 | 2.52 | 1.06 |
| Confront | 2.96 | .78 | 2.16 | .78 |
| Perspective Taking | 2.83 | .96 | 2.96 | 1.09 |
| Common Identity | 2.57 | .80 | 2.74 | .87 |
| Affirmation | 1.86 | .57 | 2.20 | .77 |
| Counterstereotyping | 2.57 | .79 | 2.23 | .78 |
| Individuate | 2.30 | .88 | 2.43 | .88 |
| Ingratiate | 2.10 | .90 | 2.73 | 1.02 |
Strategy preferences again varied significantly and were quite similar to those found in Study 1 within the unambiguous bias condition, F(8, 440) = 9.67, p < .001 (see Table 5 for a summary of contrasts). Consistent with hypotheses, participants expressed significantly more desire to confront the doctor than to use ingratiation, common identity, and affirmation strategies, p's < .001, but there was no difference in how interested they were in using confrontation and perspective taking, p = .25. Similarly, participants were more interested in using counterstereotyping than ingratiation or affirmation, p's < .02. Inconsistent with hypotheses but consistent with Study 1, individuation was not rated more favorably than the threat-reduction and self-expansion strategies. Finally, consistent with Study 1, participants expressed less desire to use affirmation than any other strategy, p's < .04, and they were more interested in leaving the doctor's office than ignoring the doctor's remark, or using individuation, ingratiation, or affirmation strategies, p's < .003.
Table 5.
Study 2: F values for comparisons between strategy preferences in the unambiguous bias condition. Bolded values represent contrasts that reached statistical significance.
| Strategy | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 |
|---|---|---|---|---|---|---|---|---|
| Ignore | 10.1 | 29.2 | 18.0 | 3.26 | 21.6 | 3.26 | .40 | 6.34 |
| Leave (2) | - | 4.95 | 1.14 | 1.87 | 61.3 | 1.87 | 14.5 | 32.4 |
| Confront (3) | - | 1.34 | 12.9 | 101.0 | 12.9 | 36.3 | 62.7 | |
| Perspective Taking (4) | - | 5.93 | 79.1 | 5.93 | 23.7 | 45.7 | ||
| Common Identity (5) | - | 41.7 | 0.00 | 5.93 | 18.7 | |||
| Affirmation (6) | - | 41.7 | 16.2 | 4.56 | ||||
| Counterstereotyping (7) | - | 5.93 | 18.7 | |||||
| Individuate (8) | - | 3.57 | ||||||
| Ingratiate (9) | - |
Preferences for using the different strategies also varied significantly in the ambiguous bias condition, F(8, 384) = 7.73, p < .001, and the patterns were again similar to those found in Study1 (see Table 6 for a summary of contrasts). Consistent with hypotheses, participants were significantly more interested in using perspective taking, common identity, and ingratiation strategies than they were in confronting the doctor or using counterstereotyping or individuation strategies, p's < .02. Unlike in Study 1 but consistent with predictions, however, participants were more interested in ignoring the doctor's remark than leaving the interaction, p < .001. Indeed, they were significantly more interested in ignoring the remark than using any strategy except perspective taking, p's < .04.
Table 6.
Study 2: F values for comparisons between strategy preferences in the ambiguous bias condition. Bolded values represent contrasts that reached statistical significance.
| Strategy | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 |
|---|---|---|---|---|---|---|---|---|
| Ignore | 16.8 | 52.5 | .10 | 4.81 | 47.3 | 43.7 | 24.2 | 5.34 |
| Leave (2) | - | 9.91 | 14.2 | 3.62 | 7.72 | 6.33 | .67 | 3.18 |
| Confront (3) | - | 47.9 | 25.5 | .14 | .40 | 5.43 | 24.3 | |
| Perspective Taking (4) | - | 3.51 | 42.9 | 39.6 | 21.1 | 3.96 | ||
| Common Identity (5) | - | 21.9 | 19.5 | 7.40 | .01 | |||
| Affirmation (6) | - | .07 | 3.84 | 20.8 | ||||
| Counterstereotyping (7) | - | 2.90 | 18.5 | |||||
| Individuate (8) | - | 6.78 | ||||||
| Ingratiate (9) | - |
Health Stigma Consciousness and Strategy Use
To unpack the interaction between health stigma consciousness and strategy preference, individual regression analyses were performed to examine the relationship between stigma consciousness and ratings for each strategy. Results revealed positive relationships between health stigma consciousness and ratings for using a confrontation strategy, β = .35, t(103) = 2.06, p = .04, and for using a counterstereotyping strategy, β = .31, t(103) = 2.69, p < .01. Consistent with our initial hypotheses, the more concerned participants were about being the target of stereotyping in a healthcare environment, the more they wanted to enact strategies that directly addressed stereotypes about their group.
Health Agency and Strategy Use
To unpack the strategy by health agency interaction, individual regression analyses were performed to examine the relationship between agency and ratings for each strategy. Results revealed a negative relationship between health agency and desire to ignore the doctor's remark, β = −.29, t(103) = 2.83, p < .01, and positive relationships between health agency and desire to use a perspective taking strategy, β = .29, t(103) = 2.84, p < .01, and common identity strategies, β = .21, t(103) = 2.04, p < .05. Thus, the more agency participants reported feeling during interactions with doctors, the less likely they were to ignore the doctor's remark and greater interest they expressed in using strategies that are likely to facilitate a positive doctor-patient interaction.
Discussion
Consistent with the results of Study 1, the results of Study 2 support the hypothesis that participants in the unambiguous bias condition would be most interested in directly addressing the doctor's bias, whereas participants in the ambiguous bias condition were most interested in implementing strategies aimed at fostering a positive doctor-patient interaction. Participants reported more desire to confront the doctor and use counterstereotyping when bias was unambiguous compared with ambiguous, but they expressed more desire to ignore the doctor's comment and use affirmation and ingratiation strategies when bias was ambiguous compared with unambiguous. Also replicating Study 1, participants in both the unambiguous and ambiguous bias conditions provided higher ratings for perspective taking than most other strategies and lower ratings for affirmation than for most other strategies, again suggesting that during interactions with their healthcare providers patients are motivated to have their concerns heard.
As in Study 1, however, participants exhibited differing levels of interest in using strategies that reduce bias through similar mechanisms. Participants in the unambiguous bias condition were again more interested in using perspective taking than highlighting common identities, and they were more interested in using counterstereotyping than individuation. Despite the fact that both strategies reduce bias by distancing targets from negative stereotypes associated with a stigmatized group to which they belong, presenting counterstereotypical information may seem like a more direct way to address the doctor's negative comment than trying to be seen in terms of individual traits unrelated to the healthcare context.
As predicted, adding health agency to the statistical model accounted for a proportion of variance that was not accounted for in Study 1, and the results revealed that both health stigma consciousness and health agency predict strategy preferences. However, rather than interacting, stigma consciousness and health agency appear to have independent effects on strategy preferences. Across conditions, participants higher in stigma consciousness were more likely than those lower in stigma consciousness to endorse strategies aimed at directly addressing biased perceptions of their group, either by directly confronting the doctor or providing him with counterstereotypic information about themselves. These results suggest that individuals who enter a doctor's office concerned that they will be the targets of negative stereotyping are more likely to enact strategies during interactions with their healthcare providers that are meant to address negative perceptions of their group, regardless of whether the healthcare provider has expressed clear bias against their group.
Participants higher in health agency, on the other and, expressed more desire than participants lower in health agency to use strategies aimed at fostering a positive doctor-patient relationship, such as getting the doctor to take their perspective or discussing shared identities. Individuals higher in health agency were also less likely to report desire to ignore the doctor's comment, supporting previous research suggesting that patients higher in health agency feel more comfortable addressing issues when they arise during medical appointments than individuals lower in health agency.
General Discussion
Across two studies, the results of the present research suggest that members of stigmatized groups moderate their behavior during intergroup interactions depending on the ambiguity of the bias expressed by their interaction partner. When they imagined interacting with a doctor who unambiguously endorsed a negative stereotype about their group, Hispanic participants expressed more desire to confront the doctor about his remark and present counterstereotypic information about themselves compared with participants who imagined interacting with a doctor who made an ambiguous remark. Participants in the ambiguous bias condition, on the other hand, were more interested in using strategies that are likely to facilitate a positive interaction with their doctor, like affirming or ingratiating him, or ignoring his remark. Further, the results of Study 2 suggest that a doctor's level of bias, health stigma consciousness, and health agency all play distinct and important roles in predicting how Hispanic patients behave during interactions with their healthcare providers.
Across both studies, some of targets’ preferences deviate from previous theory and research on bias reduction. Of note, participants did not always rate strategies that reduce bias through similar mechanisms as being equally appealing. Across both studies, participants exhibited greater preference for perspective taking than highlighting common identities, despite the fact that both of these strategies reduce bias by increasing perceivers’ sense of self-other overlap with a target. In Study 2, participants expressed more desire in using counterstereotyping than individuation—another strategy aimed at distancing a target from negative stereotypes associated with their stigmatized group membership. Thus, it appears that contextual factors may affect targets’ desire to use theoretically similar bias reduction strategies.
Limitations & Future Directions
The present study provides an important contribution to our understanding of when and how members of stigmatized groups choose to use different behavioral strategies when interacting with out-group members. There are some limitations to the present research, however. First, the large number of statistical comparison needed to examine for differences between strategy preferences within and between conditions increases the possibility that some of the significant effects found are due to Type I error. Given the fairly consistent pattern of results across both studies, however, we believe that our findings provide convincing evidence that Hispanic individuals want to enact different behavioral strategies as a function of whether or not their provider has expressed clear levels of bias toward their group.
Next, participants in the present studies reported their intentions for how they behave while interacting with a doctor, but we did not examine how individuals actually choose to behave during real-world interactions. Multiple studies have found that while people imagine that they would confront perpetrators of prejudice, they do not actually do so when they experience prejudice in the real world (Shelton & Stewart, 2004; Swim & Hyers, 1999; Woodzicka & LaFrance, 2001). Nevertheless, few studies examine the implementation of strategies other than confrontation, and it is possible that the various responses in the present research are more indicative of what participants would actually do if the opportunity arose. Future research should examine which of the intentions expressed by participants in the present research are enacted behaviorally in response to a biased remark.
In addition to examining Hispanic patients’ real-world behavior, it is important that future research examine the degree to which implementation of bias reduction strategies influences health care providers’ treatment of Hispanic patients as well as patients’ short-term and long-term health outcomes. It is also important that researchers examine whether some strategies are more likely to lead to positive outcomes in the health domain than others.
The present study compared participants’ desire to use ten separate bias reduction strategies, but recent research suggests that targets may be more effective at increasing out-group members’ positive perceptions of them when they use bias reduction strategies in combination rather than individually. Specifically, Stone and colleagues (2011) found that individuals high in prejudice toward Arab Americans expressed more positive attitudes toward an Arab American individual who encouraged them to affirm before asking them to take his perspective on being the target of prejudice compared with an Arab American individual who only asked them to take his perspective. It is unclear in the current research if participants considered the use of the various strategies individually or in combination with other strategies in the decision set. Future research can shed light on the degree to which targets believe that some strategies work best in tandem with other strategies, and, if so, which strategies represent the most effective combination for reducing bias against themselves and their group.
Participants in the present study were all college students and, thus, came from a fairly homogeneous socio-demographic background. Indeed, measures of parental income and education were significantly negatively skewed and, thus, were not able to be examined as important predictors of strategy preferences. Future research should examine whether socio-demographic factors, like socioeconomic status, English proficiency, or acculturation affect strategy use preferences among Hispanic individuals. Further, future research should examine whether similar patterns of bias reduction strategy preferences are found among members of other stigmatized groups, because members of different minority groups may choose to address bias differently (Lee, Soto, Swim, & Bernstein, 2012; Neel et al., 2013).
Finally, recent research suggests that the degree to which targets believe individuals’ personalities are fixed or malleable trait affects their willingness to confront a biased individual (Rattan & Dweck, 2010). Future research should also examine whether perceptions of personality as fixed or malleable affects targets’ preferences for using bias reduction strategies other than confrontation.
Implications
Taken together, the findings of the present research suggest that both doctors’ level of explicit bias and patient-level factors influence how minority patients choose to interact with their healthcare providers. Given the striking health disparities that Hispanic Americans face compared with White Americans (Casagrande et al., 2007; Ku & Waidmann, 2003; Livingston et al., 2008; Lopez-Quintero et al., 2006; N Mead et al., 2006; NIH Disparities Report, 2009), it is important to consider how the results of this work can inform interventions that will improve Hispanic patients’ health outcomes. Our research suggests that increasing Hispanic patients’ feelings of agency in healthcare may be particularly beneficial. As shown in Study 2, the higher participants were in health agency, the more likely they were to endorse the use of strategies that are likely to benefit the doctor-patient interaction and the less likely they were to ignore potential problems in the doctor-patient relationship. Thus, although minority patients should not be burdened with having to reduce stereotyping or prejudice against their group, interventions that help Hispanic individuals feel a greater sense of agency during doctor-patient interactions may help them improve the quality of care they receive from their providers.
Conclusions
The present research provides some of the first evidence regarding the specific strategies that targets of bias themselves choose to use during intergroup interactions. It is important that psychological research begin to document how members of stigmatized groups can play a proactive role in addressing prejudice and stereotyping directed at themselves and their groups, as well as what contextual and individual-level factors make bias reduction strategies more or less appealing to members of stigmatized groups.
Acknowledgments
This research was supported by NIH grant #5R01MD005902-02
Appendix A Bias-reduction Strategy Items and Alphas for Each Composite
Confrontation (Study 1 α = .83, Study 2 α = .91)
I would directly confront the doctor about his bias.
I would simply tell the doctor that he should try not to think about me in terms of stereotype.
I would identify the remark as bias and express my outrage/anger/distaste.
Affirmation (Study 1 α = .80, Study 2 α = .86)
I would get the doctor to talk about his best accomplishment/awards/achievements in medicine.
I would ask the doctor to talk positively about himself.
I would get the doctor to describe the most positive aspect of his group's identity.
Ingratiation (Study 1 α = .84, Study 2 α = .82)
I would find a way to compliment the doctor or agree with his opinion on something.
Common Identity (Study 1 α = .80, Study 2 α = .84)
I would try to highlight aspects that the doctor and I share in common.
I would try to get the doctor to see us as a team to make our visits go more smoothly.
Individuation (Study 1 α = .77, Study 2 α = .89)
I would stress to the doctor that we'll probably have a better doctor/patient relationship if we get to know each other.
I would volunteer personal information about myself so that the doctor could get to know me as an individual, not just a patient.
Counterstereotyping (Study 1 α = .86, Study 2 α = .91)
I would highlight things about me that are Hispanic, but I would also emphasize things about me that are different from most people's perceptions of me ethnic group.
I would do things to show the doctor that I'm very different from most people's perceptions of my ethnic group.
I would try to behave in ways that highlight positive aspects about me that are Hispanic.
I would try to behave in ways that deemphasize aspects about me that are Hispanic.
Value Discrepancy (Study 1 α = .80, Study 2 α = .87)
I'd ask the doctor if he values diversity and then remind him of ways in which he might unfairly stereotype people of my ethnicity.
I'd ask the doctor if he values freedom and equality for all and then point out how the stereotype about my group contradicts these values.
Perspective Taking (Study 1 α = .84, Study 2 α = .78)
I would try to do things that would get the doctor to see the situation from my perspective.
Ignore (Study 1 α = .72, Study 2 α = .78)
I would ignore what he said.
Leave (Study 1 α = .66, Study 2 α = .87)
I would leave the doctor's office.
Footnotes
The design of Study 2 originally included two different unambiguous bias conditions. In the unambiguous bias-high probability condition, participants read a scenario indicating that the doctor explicitly endorsed negative stereotypes about Hispanic patients and that the probability of him applying this stereotype to the participant was high. Specifically, after expressing that he does not want to help the patient, the doctor said, “Hispanics just never follow my medical advice. The six Hispanic patients I saw earlier today sure didn't.” In the unambiguous bias-low probability condition, participants read a scenario indicating that the doctor explicitly endorsed negative stereotypes about Hispanic patients, but that the probability of demonstrating this bias was not necessarily high. Specifically, as in Study 1, the doctor said, “Hispanics just never follow my medical advice.” Results on the manipulation check items and strategy use preferences did not differ between these two conditions, however. Thus, they were collapsed into one “unambiguous bias” condition. All subsequent analyses represent comparisons between the combined unambiguous bias condition and the ambiguous bias condition.
References
- Aron AP, Mashek DJ, Aron EN. Closeness as including other in the self. In: Mashek DJ, Aron AP, editors. Handbook of closeness and intimacy. Lawrence Earlbaum Associates; Mahwah, NJ: 2004. pp. 27–41. [Google Scholar]
- Ashburn-Nardo L, Morris KA, Goodwin SA. The confronting prejudiced responses (CPR) model: Applying CPR in organizations. Academy of Management Learning & Education. 2008;7:332–342. [Google Scholar]
- Ball-Rokeach S, Rokeach M, Grube JW. The great American values test: Influencing behavior and belief through television. Free Press; New York, NY: 1984. [Google Scholar]
- Bennett GG, Merrit MM, Edwards CL, Sollers JJ., III Perceived racism and affective responses to ambiguous interpersonal interactions among African American men. American Behavioral Scientist. 2004;47:963–976. [Google Scholar]
- Blair IV, Banaji MR. Automatic and controlled processes in stereotype priming. Journal of Personality and Social Psychology. 1996;70:1142–1163. [Google Scholar]
- Brewer MB. Reducing prejudice through cross-categorization: Effects of multiple social identities. In: Oskpamp S, editor. Reducing prejudice and discrimination. Lawrence Erlbaum Associates; Mahwah, NJ: 2000. pp. 165–183. [Google Scholar]
- Casagrande SS, Gary TL, LaVeist TA, Gaskin DJ, Cooper LA. Perceived discrimination and adherence to medical care in a racially integrated community. Journal of General Internal Medicine. 2007;22(3):389–395. doi: 10.1007/s11606-006-0057-4. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Czopp AM, Monteith MJ, Mark AY. Standing up for a change: Reducing bias through interpersonal confrontation. Journal of Personality and Social Psychology. 2006;90(5):784–803. doi: 10.1037/0022-3514.90.5.784. [DOI] [PubMed] [Google Scholar]
- Dasgupta N, Greenwald AG. On the malleability of automatic attitudes: Combating automatic prejudice with images of admired and disliked individuals. Journal of Personality and Social Psychology. 2001;81(5):800–814. doi: 10.1037//0022-3514.81.5.800. [DOI] [PubMed] [Google Scholar]
- Dovidio JF, Kawakami K, Gaertner SL. Reducing contemporary prejudice: Combating explicit and implicit bias at the individual and intergroup level. In: Oskpamp S, editor. Reducing prejudice and discrimination. Lawrence Erlbaum Associates Publishers; Mahwah, NJ: 2000. pp. 165–183. [Google Scholar]
- Ellemers N, Barreto M. Collective action in modern times: How modern expressions of prejudice prevent collective action. Journal of Social Issues. 2009;65(4):749–768. [Google Scholar]
- Faulkner M. A measure of patient empowerment in hospital environments catering for older people. Journal of Advanced Nursing. 2001;34(5):676–686. doi: 10.1046/j.1365-2648.2001.01797.x. [DOI] [PubMed] [Google Scholar]
- Fein S, Spencer SJ. Prejudice as self-image maintenance: Affirming the self through negative evaluations of others. Journal of Personality and Social Psychology. 1997;69:1013–1027. [Google Scholar]
- Galinsky AD, Moskowitz GB. Perspective-taking: Decreasing stereotype expression, stereotype accessibility, and in-group favoritism. Journal of Personality and Social Psychology. 2000;78(4):708–724. doi: 10.1037//0022-3514.78.4.708. [DOI] [PubMed] [Google Scholar]
- Kaiser CR, Major B. A social psychological perspective on perceiving and reporting discrimination. Law & Social Inquiry. 2006;31(4):801–830. [Google Scholar]
- Kowalski RM. Complaints and complaining: Functions, antecedents, and consequences. Psychological Bulletin. 1996;119(2):179–196. doi: 10.1037/0033-2909.119.2.179. [DOI] [PubMed] [Google Scholar]
- Ku L, Waidmann T. How race/ethnicity, immigration status and language affect health insurance coverage, access to care and quality of care among the low-income population. Henry J. Kaiser Family Foundation; Washington, DC: 2003. [Google Scholar]
- Lee EA, Soto JA, Swim JK, Bernstein MJ. Bitter reproach or sweet revenge: Cultural differences in response to racism. Personality and Social Psychology Bulletin. 2012;38(7):920–932. doi: 10.1177/0146167212440292. [DOI] [PubMed] [Google Scholar]
- Livingston G, Minushkin S, Cohn D. Hispanics and health care in the United States: Access, information and knowledge. Pew Hispanic Center; Washington, DC: 2008. [Google Scholar]
- Lopez-Quintero C, Crum RM, Neumark YD. Racial/ethnic disparities in report of physician-provided smoking cessation advice: Analysis of the 2000 National Health Interview Survey. American Journal of Public Health. 2006;96(12):2235–2239. doi: 10.2105/AJPH.2005.071035. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Matheson K, Anisman H. Anger and shame elicited by discrimination: moderating role of coping on action endorsements and salivary cortisol. European Journal Of Social Psychology. 2009;39(2):163–185. [Google Scholar]
- Mead H, Cartwright-Smith L, Jones K, Ramos C, Woods K, Siegel R. Racial disparities in U.S. healthcare: A chartbook. The Commonwealth Fund; New York, NY: 2008. [Google Scholar]
- Meyer IH. Prejudice, social stress, and mental health in lesbian, gay, and bisexual populations: Conceptual issues and research evidence. Psychological Bulletin. 2003;129(5):674–697. doi: 10.1037/0033-2909.129.5.674. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Monteith MJ. Self-regulation of prejudiced responses: Implications for progress in prejudice-reduction efforts. Journal of Personality and Social Psychology. 1993;65(3):469–485. [Google Scholar]
- National Healthcare Disparities Report, 2009. U.S. Department of Health and Human Services Agency; Rockville, Maryland: [Google Scholar]
- Neel R, Neufeld SL, Neuberg SL. Would an obese person whistle Vivaldi? Targets of prejudice self-present to minimize appearance of specific threats. Psychological Science. Online first publication. 2013 doi: 10.1177/0956797612458807. doi: 10.1177/0956797612458807. [DOI] [PubMed] [Google Scholar]
- Pinel EC. Stigma consciousness: The psychological legacy of social stereotypes. Journal of Personality and Social Psychology. 1999;76:114–128. doi: 10.1037//0022-3514.76.1.114. [DOI] [PubMed] [Google Scholar]
- Pinel EC. You're just saying that because I'm a woman: Stigma consciousness and attributions to discrimination. Self and Identity. 2004;3(1):39–51. [Google Scholar]
- Pinel EC, Paulin N. Stigma consciousness at work. Basic and Applied Social Psychology. 2005;27(4):345–352. [Google Scholar]
- Rattan A, Dweck CS. Who confronts prejudice? The role of implicit theories in the motivation to confront prejudice. Psychological Science. 2010;21(7):952–959. doi: 10.1177/0956797610374740. [DOI] [PubMed] [Google Scholar]
- Rid M, Shaw A, Lewis G, Salisbury C. The patient-doctor relationship: A synthesis of the qualitative literature on patients’ perspectives. The British Journal of General Practice. 2010;59:e116–e133. doi: 10.3399/bjgp09X420248. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Rokeach M. Beliefs, attitudes and values: A theory of organization and change. Jossey-Bass; San Francisco, CA: 1968. [Google Scholar]
- Schmader T, Croft A, Whitehead J, Stone J. A peek inside the targets' toolbox: How stigmatized targets deflect discrimination by invoking a common identity. Basic and Applied Social Psychology. 2013;35(1):141–149. [Google Scholar]
- Shelton JN, Richeson JA. Intergroup contact and pluralistic ignorance. Journal of Personality and Social Psychology. 2005;88(1):91–107. doi: 10.1037/0022-3514.88.1.91. [DOI] [PubMed] [Google Scholar]
- Shelton JN, Richeson JA, Salvatore J. Expecting to be the target of prejudice: Implications for interethnic interactions. Personality and Social Psychology Bulletin. 2005;31(9):1189–1202. doi: 10.1177/0146167205274894. [DOI] [PubMed] [Google Scholar]
- Shelton JN, Stewart RE. Confronting perpetrators of prejudice: The inhibitory effects of social costs. Psychology of Women Quarterly. 2004;28(3):215–223. [Google Scholar]
- Sinclair L, Kunda Z. Reactions to a Black professional: Motivated inhibition and activation of conflicting stereotypes. Journal of Personality and Social Psychology. 1999;77:885–904. doi: 10.1037//0022-3514.77.5.885. [DOI] [PubMed] [Google Scholar]
- Singletary SL, Hebl MR. Compensatory strategies for reducing interpersonal discrimination: The effectiveness of acknowledgments, increased positivity, and individuating information. Journal of Applied Psychology. 2009;94(3):979–805. doi: 10.1037/a0014185. [DOI] [PubMed] [Google Scholar]
- Son D, Shelton JN. Stigma consciousness among Asian Americans. Asian American Journal of Psychology. 2011;2(1):51–60. [Google Scholar]
- Son Hing LS, Li W, Zanna MP. Inducing hypocrisy to reduce prejudicial responses among aversive racists. Journal of Experimental Social Psychology. 2002;38(1):71–78. [Google Scholar]
- Swim JK, Hyers LL. Excuse me—What did you just say?!: Women's public and private responses to sexist remarks. Journal of Experimental Social Psychology. 1999;35(1):68–88. [Google Scholar]
- Tropp LR, Wright SC. Ingroup identification as the inclusion of ingroup in the self. Personality and Social Psychology Bulletin. 2001;27(5):585–600. [Google Scholar]
- Woodzicka JA, LaFrance M. Real versus imagined gender harassment. Journal of Social Issues. 2001;57(1):15–30. [Google Scholar]