Evaluating the Impact of Race and Gender on Environmental Risk Perceptions in the Houston Neighborhood of Manchester

Introduction

Prior research in the United States has demonstrated differences in risk perceptions of environmental exposures by race and gender. ^{1 , 2} The largest differences in perceptions of environmental harm have been seen with white males, who typically view the potential health risks of environmental exposure as lower than women or nonwhite racial groups. ^{3 , 4 , 5 , 6} This has been attributed to male predominance in positions of political, social, technological, and economic power. ⁷ However, there is increasing evidence that gender and race may play a smaller role in determining an individual's perceptions of potential environmental risks compared with the community and generational experiences with one's lived environment. ^{8 , 9} One possible explanation hypothesized here is that women and racial minorities who live in areas characterized by poor environmental conditions create cultural norms that make them less likely to recognize potential environmental harm. Poor environmental conditions are experienced more by minority communities than majority populations. To address the inconsistencies of these findings, a cross-sectional study was conducted in the neighborhood of Manchester in Houston, TX.

Background

Risk perception is a principal variable in understanding how individuals and communities respond and react to potential environmental hazards. Owing to the importance of perceptions on potential environmental harm, researchers have struggled to understand the underlining influences that persuade individual environmental risk sensitivities. Lindel and Hwang surveyed 321 households in Harris County, TX, to assess risk perceptions on flooding, hurricanes, and toxic chemical releases. ¹⁰ They found that individuals' hazard experience, gender, and income played a large role in determining how they viewed the threat of potential hazards. It was also shown that proximity to hazards and ethnicity of the respondents affected their perceived personal risk of environmental threats. Similarly, researchers Lachlan et al., after utilizing 937 surveys from Hurricane Katrina displacees, found a clear difference in perceptions of hazard risk and outrage as related to the respondent's race. ¹¹

Prior research has shown different perceptions of potential harm from environmental exposures, with white males tending to have the lowest perception of risk. For example, differences in perceived risk between races, as well as acceptance of living in communities that are exposed to industrial pollutants, have been documented by researchers in studies within the urban settings of Philadelphia, PA. ¹² This study surveyed potential jurors from the city on perceptions of risk to air pollution, finding that nonwhite females reported the highest concern. Similar differences in perceived risk have been found in rural northern California with use of herbicides to manage invasive weed species. ¹³ This phenomenon has been termed the white male effect. ¹⁴

The predominant explanation for this difference is that white males within the United States have more power over their environment and are involved to a greater degree in creation, management, and benefits of technology and industrial pursuits and therefore are more likely to see benefits rather than health risks. ¹⁵ Another explanation for differences in risk perceptions may be cultural differences between groups. Due to white male's risk skepticism and hierarchical and individualistic nature, when activities that are integral to their identity are threatened, they will underestimate the inherent risk. Cultural protective cognition, according to Kahan et al., is what explains this difference because white male cultural identity is inherently linked to technocratic endeavors in ways that the identities of women and minority populations are not. ¹⁶

While this phenomenon has been studied in multiple settings, the majority of research is set in large geographic areas, such as at the national, state, or county level, which do not take into account the actual environmental conditions in which the individuals answering the questions live. More recently, there has been some evidence to suggest that differences in perceptions of environmental harm are more closely related to the specific area in which individuals live rather than the race or ethnicity of the respondents. For example, in a study of Swedish society in which disproportionate exposures to industrial pollutants with regard to differing racial groups are not as distinct as in the United States, Olofsson and Rashid found that differences in perceived environmental risks mostly vanished. ¹⁷ Within the United States, when surveying deep south coastal residents living in an area characterized by industrial water pollution, Marshal et al. found that the difference between groups was less pronounced when both minority and majority groups experienced similar environmental pollution. ¹⁸ Furthermore, according to Satterfield et al., when asked if hazardous facilitates are more common in minority communities, only 50.4% of white males agreed, while 66.5% of nonwhite males agreed. ¹⁹ This division is shown in gender as well, with 71.6% of nonwhite females agreeing with the previous statement compared with slightly above 50% of males (Johnson, 2002). This gap between beliefs and empirical evidence further demonstrates that these different groups are living in different environmental conditions. ²⁰ Generational conditions of living in areas characterized by poor environmental conditions may create expectational norms among minority groups that accept poorer conditions than their white counterparts due to experiencing these conditions for longer periods of time.

The use of race as an etiologic quantity within medical research has been criticized as failing to account for fine-grain differences between individuals and communities. ^{21 , 22} Residual confounding when using race as a surrogate for other variables, such as socioeconomic status, as well as potential societal harm has been voiced by public health practitioners. ²³ However, when utilizing racial stratification as a substitute for environmental, social, and political experiences of these communities, it has been shown to be helpful in understanding communal and individual perspectives. Researchers Kaufman and Cooper (2001) ²⁴ concluded that using race to predict behavioral decisions or aspects of social interactions is valid, interpretable, and important. The norms created within communities through generational exposures to poor environmental conditions are not fully understood within public health research.

To better understand the effects of gender and race on risk perception, data were collected in a location in which all individuals experience a similar relationship with their environmental conditions regardless of race. A goal of this study was to determine whether perceived risks are a continuation of environmental justice issues that have permeated the racial minority experience within the United States for decades or if the white male effect would be shown to exist in this community.

Materials and Methods

Results

Between December 19 and December 26, 2015, 109 (N = 109) surveys were collected with an overall response rate of 72.7%. Of the 192 homes that were approached, contact was made with 150, and this gave a contact rate of 78.1%. Twenty-eight percent (N = 31) were completed by non-Hispanic white individuals, 62.4% (N = 68) Hispanic or Latino individuals, and 8.3% (N = 9) African American, with one respondent refusing to respond to questions about race. There was an almost even split between male, 49.5% (N = 54), and female, 50.5% (N = 55), respondents. Table 1 provides additional community characteristics.

The majority of all respondents, regardless of race or gender, felt that living near waste facilities and industrial buildings could cause negative health outcomes. The largest differences in perceptions were with road infrastructure, where 27.8% (N = 5) of white males said it was a health concern compared with white females (46.1%; N = 6), nonwhite males (54.3%; N = 19), and nonwhite females (53.7%; N = 23). White males were most concerned with living near waste facilities and experiencing homes that are in need of repairs compared with their counterparts (Fig. 1).

OPEN IN VIEWER

FIG. 1.

Mean risk perception by race and gender. Reports of perceptions of risk were obtained through self-reported questionnaires and stratified by race and gender.

When comparing nonwhite participants with non-Hispanic white individuals, the overall perception of environmental harm was lower for minority populations. Table 2 shows the crude and adjusted odds ratios (ORs) and 95% confidence intervals (95% CIs) for each of the environmental exposures mentioned in the survey. On all exposures, with the exception of poor road conditions, the nonwhite group perceived a lower health risk than non-Hispanic whites. Exposure to standing water associated with flooding was significantly associated with a lower health risk by minority residents, with a statistically significant decrease in risk perception (OR = 0.34; CI = 0.13–0.94). Adjusting for the generational age of participants did not impact the direction of the associations.

Table 3 shows the results of comparing non-Hispanic white males with each gender and racial category. Analyses of reported perceptions of environmental harm to one's personal health or the health of the community indicated that on issues of living near industrial buildings, living near waste facilitates, and living in buildings that need repair, non-Hispanic white males showed the greatest concern. On issues of exposure to standing water, non-Hispanic white females were 57% more likely to shows concern than their male counterparts (OR = 1.57; 95% CI = 0.24–10.22). The main exception was the issue of living with poor road infrastructure, where non-Hispanic white females and nonwhite males and females perceived increases in potential harm. It is important to note that each of the points failed to produce statistically significant differences between any groups, likely due to the limited sample size.

Discussion

Our cross-sectional study was designed to evaluate whether there was evidence of the white male effect in a small neighborhood in Houston, Texas, where the environmental conditions are well known and the lived environmental experience was similar for all racial groups within the geographically compact neighborhood of Manchester. Findings of this research suggest that this effect does not hold in this population. Within the surveyed population in Manchester, minority populations tended to underreport perceived health problems associated with certain environmental exposures compared with non-Hispanic white groups. These results were not attenuated when adjusting for gender; non-Hispanic white males often reported the highest level of concern. These findings lend support to the hypothesis that minority groups are less likely to report a problem with even egregious environmental conditions due to cultural norms surrounding these communities from generations of poor environmental living conditions discussed at length in environmental justice research. ^{37 , 38 , 39 , 40} These results further support findings from researchers Olofsson and Rashid who found that equity, in regard to exposure to environmentally polluted areas, tended to remove differences in perceptions of environmental harm between different racial and ethnic groups. ⁴¹ It should be noted, however, other alternative explanations could exist to explain the disparate perceptions between groups in the study. Information on socioeconomic variables, such as educational attainment or employment data, was not collected. This information could soften the results if controlled for as these have been found to inform perceptions of risk. ⁴²

The strength of this study lies in the well-documented problems with the quality of air, standing surface water, proximity to waste and industrial buildings, and infrastructure conditions in this small neighborhood. Furthermore, partnerships with local students and volunteers to conduct survey data collection ensured that language and cultural boundaries did not pose a problem in receiving quality data. A response rate of 72.7% (109/150), a high proportion within a small neighborhood, further demonstrated the interest within the neighborhood around environmental health issues.

This study has several limitations. Despite the relatively high response rate, a small total number of participants completed the survey, reducing our statistical power and our ability to adjust by multiple confounders. Non-Hispanic whites were overrepresented in our survey responses compared with the U.S. Census data on race and ethnicity of Manchester residents (City of Houston Planning and Development Department, 2014). Non-Hispanic whites were more likely to complete the survey than their nonwhite counterparts and this could have caused selection bias within this study. The use of trained local volunteers and Spanish speakers was an effort to address this potential bias. Data on socioeconomic status were not collected and could account for some confounding.

Conclusion

The results of our small case study suggest that the white male effect may not be seen with different racial or gender groups who experience the same environmental conditions. Furthermore, one plausible explanation for the reduction in perceived risk from minority populations in this study could be accounted for from long-term cultural norms that may inform individuals' opinions about the health risks of various environmental exposures. Gender and racial perceptions of environmental harm will require additional research to better understand how individuals and communities view their lived environment and its impact on their health; until that time, no conclusive statements can be made on the underlining reasons for the findings from this study.