|Home | About | Journals | Submit | Contact Us | Français|
This study reports on the prevalence and correlates of perceived discrimination among a national sample of Latinos in the U.S. Understanding the prevalence and correlates of discrimination can help us better address disparities in the healthcare system. We define perceived discrimination as self-reported everyday experiences of unfair treatment.
Logistic regression analyses were used to assess rates of perceived discrimination among Latinos and identify correlates of discrimination. Data came from the National Latino and Asian American Study (NLAAS).
The prevalence of perceived discrimination among Latinos was 30%. Cubans and Latinos with high ethnic identity were less likely to perceive discrimination compared to other Latino subgroups or Latinos with low ethnic identity. U.S.-born Latinos and Latinos arriving to the U.S. at younger ages were more likely to perceive discrimination compared to immigrants arriving at older ages.
Perceived discrimination among Latinos is less prevalent than what has been reported for other minorities. Variations in perceived discrimination are related to sociodemographic and cultural differences across ethnic subgroups.
Experiences of discrimination are relevant to how individuals experience health care and their expectations regarding healthcare and other services1–4. Past research studies have reported racial and ethnic differences in perceived discrimination in the United States5. In particular, African Americans (71.3%) reported higher rates of discrimination (71.3%) compared to non-Hispanic whites (23.7%)6. Limited work has been done on the prevalence of discrimination and the factors associated with variations in perceived discrimination among Latino subgroups.
Sex, age, personal income, and marital status all have an effect on perceptions of discrimination or moderate the relationship between psychological health and discrimination6. Different Latino subgroups have different levels of exposure to U.S. culture and human capital and therefore may vary on degree of perceived discrimination; level of acculturation, ethnic identity and cultural factors that may be important factors influencing variations in perceived discrimination.
Finch and colleagues7 reported differences in self-reports of perceived discrimination among immigrant Mexican Americans in Fresno, California. The authors found that more acculturated Mexican immigrants (as defined by more time in the U.S., English language ability and higher education) reported higher rates of discrimination compared to less acculturated immigrants. Other studies of Latinos living in New York City (Puerto Ricans, Dominicans, Mexicans and Other Latinos) show that the prevalence of self-reports of experiencing any type of discrimination (racial and non-racial) among Latinos and African American was 38% and 53%, respectively8. Yet, these studies on the prevalence of discrimination among Latinos had some limitations. The samples were regional constraining the generalizability of the findings to certain areas of the U.S., with no information disaggregated by Latino subgroups. Furthermore, they focused only on discrimination as a predictor of health outcomes. Few studies offered any information about the socioeconomic and cultural correlates associated with the rates of discrimination or how the discrimination rates varied with time in the U.S. or nativity. Previous studies have shown that socioeconomic5,9, and cultural factors 6,7,10,11 (ethnic subgroup, U.S. or foreign nativity, English language proficiency and ethnic identity) may be associated with perceived discrimination7,12. Language isolation potentially protects Spanish-only speaking Latinos from personally perceiving racist comments though not necessarily from experiencing physical discrimination7,13. In addition, generational status is associated with varying degrees of acculturation14–16. This paper evaluates the prevalence and correlates of everyday discrimination among Latinos and across Latino subgroups, including socio-demographic (gender, age, marital status), socioeconomic (education level and household income) and cultural (proficiency in English, nativity, ethnic identity, generation in the U.S.) factors associated with self-perceived everyday discrimination. We hypothesize that: Latinos living in the U.S. for longer periods of time will report higher rates of discrimination; higher level of education and income will be correlated with increased perceptions of exposure to discrimination; and less acculturated Latinos will report lower rates of discrimination than highly acculturated Latinos. We hypothesize that strong ethnic identity will be correlated with higher rates of perceived discrimination. We expect that Cubans will report lower rates of discrimination than other Latino subgroups due to their concentration in ethnic enclaves17.
The National Latino and Asian American Study (NLAAS), based on a stratified area probability sample design, has been described elsewhere18–20. The original study makes use of both national Latino and Asian samples; however, we only report on the Latino sample here consisting of 2,554 Latinos (577 were Cuban, 495 were Puerto Rican, 868 were Mexican, and 614 were other Latino), ages 18 and older from the non-institutionalized population of the coterminous U.S. The interviews were conducted in both English and Spanish by trained interviewers at the University of Michigan’s Institute for Social Research (ISR) between May 2002 and November 2003. The final weighted response rate for the NLAAS Latino sample was 75.5%18.
The Internal Review Board Committees of Cambridge Health Alliance, the University of Washington, and the University of Michigan approved all recruitment, consent, and interviewing procedures21.
Detailed description and reliability results of all non-diagnostic measures used in the NLAAS have been described elsewhere19. In the current study, we use measures of sociodemographics and personal characteristics, psychiatric disorders, and measures of acculturation. Respondents were asked to indicate how often in their day-to-day life they experienced any of nine discriminatory situations taken from the Detroit Area Study (DAS)22,23,(e.g., being treated with less courtesy than other people; being treated with less respect than other people; receiving poorer service than other people at restaurants or stores; people acting as if they think the respondent is not smart; people acting as if they are afraid of the respondent; people acting as if they think respondent is dishonest; people acting as if they think the respondent is not as good as they are; being called names or insulted; and being threatened or harassed). The six response categories ranges from “never”  to “daily” . The items in the scale had a standardized Cronbach alpha score of .91 for the Latino sample.
Because the distribution in the discrimination scale was bimodal, we dichotomized the measure. Those respondents who indicated that they experienced any item in the scale a few times a year or more were coded as experiencing moderate to high levels of everyday discrimination as compared to those that indicated never experiencing those events or experiencing those events less than once a year (no or low perceived discrimination) following Mays and Cochran’s approach24.
The socio-demographic variables of age, gender and marital status and the socioeconomic variables of household income and education were used in the analyses. The income variable had nearly 300 missing values; which were imputed using the hotdeck module in STATA 925.
Cultural factors included ethnic subgroup (Puerto Rican, Cuban, Mexican and Other Latino group), English language proficiency (ability to speak, read and write English), nativity (born in U.S. mainland or foreign-born), age of arrival to the U.S., generation status (whether respondent and his/her parents were born in the U.S.26–28, and ethnic identity (identified closeness to ethnic group). Ethnic subgroup information was based on respondent self-report.
We first estimated the prevalence of everyday perceived discrimination. We then examined the association between sociodemographic, socioeconomic and cultural variables with perceived discrimination. We tested for differences in the age and gender adjusted rates of perceived discrimination across socio-demographic, socioeconomic and cultural factors using the chi-square test. We stratified by socioeconomic and cultural factors to see if sub-ethnic differences in the rates of discrimination remained. We also examined correlates for perceived discrimination across Latino subgroups. Using logistic regression we tested for the socio-demographic, socioeconomic, and cultural factors associated with reports of everyday discrimination. We used multivariate analysis because many variables are correlated with discrimination but they are also related to each other and the regression allowed us to assess associations separate from socio-demographic, socio-economic and cultural factors. The three sets of adjustor categories were entered sequentially in three blocks, socio demographic variables were entered first and cultural factors were added last. We compared the regression results using the three progressively inclusive specifications to test whether adding socio-economic and cultural factors modified the significance of the coefficients for the explanatory variables in the earlier step. Analyses were corrected for survey design and weights using STATA statistical software29.
Table 1 summarizes the distributions of the demographic variables of the Latino sample. For those respondents who did report discrimination, significantly more men than women reported discrimination (p<.001). Younger Latinos were significantly more likely than older Latinos to report any discrimination as shown by an almost linear relationship (p<.0001) (see Figure 1). Unmarried Latinos more so than married Latinos (p<.001), those with more education (26.4% at less than a high school degree, 43.1% at college or more; p<.001), and those with higher incomes, (30% living in households with less than 15K per year, 40% earning more than 75K) reported any experience of discrimination.
While the overall rate of discrimination among Latinos was 30%, the rates of everyday discrimination varied by cultural characteristics across different subgroups of Latinos (see Table 2). Cubans were significantly less likely (p<0.05) to report experiences of discrimination than any of the other three Latino groups (data not shown). Nearly half of U.S.-born Latinos reported everyday discrimination compared to only one-quarter of immigrants (p<.001). Exposure to U.S. culture as measured by age of arrival or growing up in the U.S. is positively associated with discrimination. Latinos with a strong ethnic identity are less likely than those with a weak identity to perceive discrimination (p<.05). Interestingly, subethnic group differences in the rate of everyday discrimination remain, even when stratified by socioeconomic and cultural factors (data not shown). Cubans were less likely to report discrimination compared to other Latino subgroups. We also found a clear pattern emerge with regard to generational change in perceptions of discrimination.
Table 3 shows the correlates of everyday discrimination in the multivariate logistic regression models. In model one, we assess the socio-demographic correlates and find that perceptions of discrimination decrease as age increases until age 64 years. The younger to middle-aged cohorts were significantly more likely to report everyday discrimination as compared to the reference group, Latinos aged 65 and over. Men (vs. women) and non-married persons (vs. married) were significantly more likely to report everyday discrimination. In model two, we added the socio-economic variables and find that the association between discrimination and age, gender, and marital status remain. We also find that Latinos with some college education and college graduates had an increased likelihood of reporting everyday discrimination compared to those having some high school or lower education. There were no income effects associated with self-reports of everyday discrimination.
Finally in model three, we added cultural factors and found that age, gender, and education remain significant correlates of discrimination, but marital status is no longer significant. Among the age categories, when compared to the reference category of Latinos 65 years of age and older, only the three youngest age cohorts remained significantly higher (age 18–24, age 25–34, and age 35–44) in perceived everyday discrimination. Cubans were less likely to report discrimination compared to Mexicans (reference category). Latinos who arrived between the age of 7–17 years and those who arrived at age 18–24 years were significantly less likely to report everyday discrimination when compared to U.S.-born Latinos or those who arrived age 6 or younger. There was no statistical difference by generational status of Latinos. Latinos who identified as having a strong ethnic identity were less likely to report everyday discrimination compared to those having a weak ethnic identity.
Our findings are consistent with previous studies of other non-Latino populations which found similar associations between discrimination and socio-demographic, socioeconomic and some of the cultural factors assessed in the present study6. Contrary to previous reports6, we did not find that relationship between income and perceived discrimination in the final model of our regression. Our findings are similar to results by Gary and others9,30,31 who found that higher education and employment status were positively related to increased likelihood of perceived discrimination among African Americans but higher income was not, suggesting that socioeconomic factors may be differentially associated to discrimination by ethnic or racial groups.
The findings suggest that as Latinos achieve higher social status and become more assimilated, they have a greater sensitivity to discrimination compared to their less acculturated counterparts. For example, well educated, young U.S.-born Latinos, or those who arrived age 6 or younger, are more likely to perceive everyday discrimination. This may be a consequence of frustrated expectations within the dominant U.S. culture and institutions. Inversely, lower-educated Latinos may have lower expectations for fair treatment and therefore may not be as vulnerable to perceiving everyday discrimination. As immigrants assimilate they may lose their idealized view of America as the land of equal opportunity and therefore have higher expectations for fair treatment.
The increase in rate of perceived discrimination among the younger male cohorts may also be explained by the potential increase in exposure. Minority men are more vulnerable to negative encounters with social institutions26. Younger Latinos may also have higher expectations of fair treatment than their parents and may define treatment as discriminatory that their parents did not. Contrary to earlier findings of Finch27 we found that U.S.-born Latinos residing in the U.S. were more likely to perceive everyday discrimination compared to their less acculturated counterparts. Other studies found similar increases in the experience of discrimination among immigrant groups correlated with increases in time in U.S.32. Latinos arriving at a younger age may be more likely to intermingle with non-Latinos in multiple settings; this increased exposure to cultures different from their own may lend itself to increase incidents of and sensitivity to discrimination.
The finding that Spanish-proficient U.S.-born Latinos were less likely to perceive discrimination than English-proficient foreign-born Latinos confirms our hypothesis that lower acculturated Latinos report lower rates of perceived discrimination than acculturated Latinos. Linguistic isolation may reduce the perception of discrimination. The more English Latinos speak, the more likely that they will interpret any inter-cultural interactions as discriminatory and understand it when someone discriminates against them. For all subethnic groups, speaking English was associated with twice the rate of reporting everyday discrimination compared to Spanish-speaking Cubans, Mexicans, Puerto Ricans and Other Latinos who may be more socially isolated from U.S. culture and institutions.
Our study shows that Latinos who express strong ethnic identity may be buffered against perceptions of discrimination, a result that has been demonstrated in other ethnic subpopulations including Korean, Filipino and Chinese28. As far as we know, this is the first time that the effect of strong ethnic identity as a protector against everyday discrimination has been found for Latinos. People with high levels of ethnic identity may be more likely to associate with people of their own ethnicity and therefore be less exposed to discrimination28.
Cubans were least likely to report discrimination. Cuban immigrants have arguably the best infrastructure for transition into the U.S. of any other Latino group15. This supportive infrastructure may also be associated with the presence of strong ethnic identity in the context of a politically and socially well-developed enclave. In addition, living in an ethnic enclave may provide protection against the perception of discrimination if not against actual discriminatory acts.
Similar to other studies, we used a subjective measure of everyday discrimination rather than objective measures of discrimination. However, studies on the subject of discrimination have shown that personal assessments of discrimination and their psychological impact are similar to objective discriminatory acts6,33,34. Furthermore, we use a well-established discrimination scale as our outcome measure. While no causal direction of the observed association between discrimination and cultural factors is established, it is evident that cultural factors should precede experiences of discrimination. It may be true; however, that discrimination may play a role in some socioeconomic measures as well as strengthening ethnic identity as a coping mechanism35.
Understanding the prevalence and correlates of discrimination can help us better address disparities in the healthcare system. Reducing discrimination can improve the patient/provider relationship and in turn improves healthcare outcomes. This study reveals the importance of examining the multi-dimensional impact of varying degrees of acculturation. Health researchers attempting to explain Latino differences in health care access, behavior and attitudes need to test their hypotheses disaggregating the cultural elements. The findings presented here suggest that the rate of perceived discrimination is differently associated with several such components including subethnic differences, generation, and ethnic identity. Identifying how particular elements impact the healthcare experience such as English proficiency and generational status also has implications for service delivery. For example, low English proficiency is often cited as an important contributor to healthcare disparities36 but less attention has been given to experiences of discrimination and the impact on patient-provider interactions among English-speaking younger generations.
This paper contributes to the existing literature by providing information about the prevalence of discrimination among Latino subethnic groups. People who care about how the health care system operates should also care about the prevalence of everyday discrimination among vulnerable populations. Our findings suggest that multiple factors need to be accounted for in understanding Latinos perceptions of the world and their interactions with institutions. Additionally, further research is also needed to understand the mechanisms by which perceptions of discrimination change over generations.
The NLAAS data used in this analysis was provided by the Center for Multicultural Mental Health Research at the Cambridge Health Alliance. The project was supported by NIH Research Grant # U01 MH62209 funded by the National Institute of Mental Health as well as the Substance Abuse & Mental Health Services Administration/Center for Mental Health Services and the Office of Behavioral and Social Science Research. This publication was also made possible by Grant # P20 MD000537 from the National Center on Minority Health and Health Disparities (NCMHD) and Grant #P50 MH073469-02 from the National Institute of Mental Health.