Publication Type: Conference Paper/Unpublished Manuscript
Review Method: Peer Reviewed
Abstract: Studying relationships between health and environmental toxins usually involves statistical tests and a reliance on 95 percent- significance. But because of inherent scientific uncertainties, small sample sizes, unknown exposure levels, and scientific value judgments, some researchers working with contaminated communities propose the adoption of health-protective significance levels, (e.g., 75 percent-). To explore alternative significance levels, health data collected in a poor Latino neighborhood surrounded by industrial facilities in Phoenix, AZ is utilized to analyze three sets of correlations. First, the negative correlation between health problems and distance from a specific point source polluter was significant at 93 percent- for children. Interestingly, the relationship was less significant and positive for adults. Second, while a respondent's self-reported health status was correlated negatively at 87 percent- with length of residence in the neighborhood, a parallel relationship was not found for breathing problems. Third, the relationship between presence of cockroaches in the home and breathing problems was significant at 95 percent- for adults and children. Overall, this analysis illustrates the complexity and confusion surrounding community-level environmental health research. Contradictory explanations can be provided for both the traditionally "significant" and "insignificant" findings. So while alternative health-protective significance levels may be helpful to policy makers and contaminated communities, they become more powerful when paired with other forms of inquiry. Researchers should report data frequencies, use multi-method approaches, involve community residents, balance research goals with scientific rigor, and encourage communities to focus on political, rather than purely scientific, solutions to their environmental concerns.