Grantee Research Project Results
Final Report: Perceptions of and Exposure to Arsenic in Private and Public Drinking Water Among Households
EPA Grant Number: R832235Title: Perceptions of and Exposure to Arsenic in Private and Public Drinking Water Among Households
Investigators: Shaw, W. Douglass , Moeltner, Klaus , Walker, Mark , Riddel, Mary , Jakus, Paul
Institution: Texas A & M University , Utah State University , University of Southern California , University of Nevada - Reno
Current Institution: Texas A & M University , University of Nevada - Reno , University of Southern California , Utah State University
EPA Project Officer: Hahn, Intaek
Project Period: April 18, 2005 through October 15, 2007
Project Amount: $310,017
RFA: Valuation for Environmental Policy (2003) RFA Text | Recipients Lists
Research Category: Environmental Justice
Objective:
The project involves implementation of a survey questionnaire to households that are in potential violation of the new Safe Drinking Water Act standard for arsenic in drinking water, of 10 parts per billion (ppb). The key parts of the study relate to eliciting risk perceptions associated with arsenic. These perceived risks are used to explore:
- Whether and why people in the study areas drink water contaminated with arsenic.
- Reasons risk perceptions vary across households
- Relationships between perceived risks and maximum willingness to pay (WTP) to reduce risks
- Whether household heads act in ways that protect not only themselves, but their children (when present).
Summary/Accomplishments (Outputs/Outcomes):
Provide an executive summary of project results, emphasize findings and their significance to the field, their relationship to the goals of the project including an explanation of how the research adds to our understanding of, or solutions for, environmental problems or is otherwise of benefit to protection of the environment and human health.
Response rates appear low, but the scheme used to recruit respondents (initial telephone and survey, mail booklet, follow-up mail survey) allows estimation of a model of survey participation. Of various variables used in estimating the probability a person would participate in the study, years of residence had a negative and significant effect and the importance of environmental quality had a positive and significant effect (see Table 1).
Table 1: Participation (in the study) Model | |
---|---|
Variable Constant Years in House Importance of Environmental Quality Education
|
Coefficient (White’s t statistics*) 0.074 (1.68) -0.016 (-2.56) 0.253 (2.95) -0.814 (-1.54) |
* The log likelihood value at convergence indicated continued improvement over specifications with fewer explanatory variables. White’s t statistics are robust to specification errors. Environmental quality was rated on a scale as low (1) to high (5).
The estimating sample of about 350 households does, however, not appear to substantially differ from the populations in the study areas, and we find that in assessing behaviors, the potential sample bias factors do not appear to influence results. We managed to obtain many households with one or more children, which was important in studying risk averting behavior to protect children. Though risk elicitation is quite complicated, the survey and booklet led to over half of the respondents being able to offer a point estimate of risk (single mark on the risk ladder) and almost 100 were able to offer a range of risks (bounded by two marks on the ladder).
Table 2 Final Estimating Sample Characteristics and Behaviors | |
---|---|
Variable | = Frequency (N); Mean indicated |
Gender Age Have private or public insurance? Homeowner (=1 if yes, 0 = No) Have smoked in the past? Self-rating Health Status
Number of Children Public Water system (=1, = 0 private) Initial Marks on risk ladder (own risks)
|
Male = 198, female = 145 (10 missing) Mean = 51.3 (of 342 reporting) Yes = 336, No = 14 1 = 320, 0 = 21 (12 missing) Yes = 161, No = 192 1 Excellent = 97 Zero = 217 1 = 45 Mean = $69,190 (minimum 7,500) 242 Public, 111 Private One = 181
|
Data is on 353 participants in follow up. Ten of these provided no screener data, as they are from the pretest. |
We conclude that a mailed information booklet can be effective in communicating complex ideas about risks that households face to respondents, especially when visual aids such as risk ladders are used. This is very important in future attempts by agencies to explain risks to households.
Table 3 reports summary statistics on water consumption, treatment and respondent concerns. The study echoed studies by others that show that learning about water consumption and drinking water behavior is not a simple matter. Many sample respondents, though often first stating that they do NOT consume tap water with arsenic in it, often do. This is first because they are unaware of behaviors relating to tap water consumption such as making coffee or juice, or filling an ice tray. Over half of the respondents treat their water, though only 220 out of 353 had heard about arsenic in their drinking water before they were contacted by us.
Table 3 Water Quality and Consumption, Treatment, and Arsenic Concerns | |
---|---|
Variable Is all the water you drink from the tap? Do you or other family drink bottled water at home? Do you use a water treatment device? Why do you use one?
My tap water is perfectly safe to drink
How concerned are you about negative health problems caused by the level of arsenic in the water you drink?
Prior to speaking with us, had you heard about potential problems your public supply has, or had with arsenic?
Do you know the level of arsenic in your well water? The government probably does not know exactly what the arsenic level is in the water that I drink |
Frequency (N) or Mean Yes = 231, No = 122 Yes = 117, No = 236 Yes = 182, No = 171 To improve the taste (n = 45) Too expensive to treat (n = 31) 1 (Strongly disagree) = 59 1 (not at all concerned) = 70 Yes = 220, No = 133*
Yes = 5 No = 106** 1 (strongly disagree) = 77 |
Total responses may not equal 353 because of item non-response or skipped questions. * Public systems sample only. ** Private well users only. |
We find substantial understanding of the risks that scientists believe pertain to ingesting arsenic, and upon eliciting subjective risks, sample members offer risks that are consistent with science-based risks, though frequently a bit lower (see Table 4). Smokers got the message from the information booklet that they were at greater risk of dying from ingesting arsenic than non-smokers. While average risks are generally lower than the science-based estimate of 1 in 100 that pertains to average people who consume arsenic at 50 parts-per-billion (ppb), we note that the risk information we provided, and the elicitation process led to subjective or perceived risks that are quite close to the same order of magnitude as 0.01. For example, for the 99 who had smoked at some point in their lives (not necessarily now), the average risk reported was about 0.008, or 0.002 different than the science-based estimates.
Table 4: Mean Perceived Arsenic-related Mortality Risks for Smokers and Non-smokers | |
---|---|
Group (number of observations) |
Estimated mean risk |
Full sample (n = 201) Respondents who have never smoked (n = 102) Respondents who have ever smoked (n = 99) Respondents who have quit smoking (n = 71) Respondents who current smokers (n = 28)
|
0.0059 (0.0001) 0.0038 (0.0001) 0.0081 (0.0013) 0.0057 (0.0014) 0.0139 (0.0028) |
Again, the importance of the above is in the area of communicating risks to people, especially as these pertain to drinking water, a more complicated act than one may initially believe. Respondents have to be reminded of the effectiveness of treatment, or the lack thereof (e.g. for simple carbon filters in water pitchers). We find some weak evidence that households will also behave as to pay attention to children’s risks that might be different than their own risks. Again however, special attention has to be paid to these differences in risks.
Conclusions:
Finally, we estimate a complicated model of ambiguous risks that are linked to WTP measures. We show that ambiguous risks can still be used to obtain WTP estimates.
Journal Articles on this Report : 4 Displayed | Download in RIS Format
Other project views: | All 11 publications | 4 publications in selected types | All 4 journal articles |
---|
Type | Citation | ||
---|---|---|---|
|
Jakus PM, Shaw WD, NguyenTN, Walker M. Risk perceptions of arsenic in tap water and consumption of bottled water. Water Resources Research 2009;45: W05405, doi:10.1029/2008WR007427. |
R832235 (Final) |
Exit Exit |
|
Jindapon P, Shaw WD. Option price without expected utility. Economics Letters 2008;100(3):408-410. |
R832235 (Final) |
Exit Exit |
|
Nguyen TN, Jakus PM, Riddel M, Shaw WD. An empirical model of perceived mortality risks for selected U.S. arsenic hot spots. Risk Analysis 2010;30(10):1550-1562. |
R832235 (Final) |
Exit Exit |
|
Shaw WD, Jakus PM, Riddel M. Perceived arsenic-related mortality risks for smokers and non-smokers. Contemporary Economic Policy (early view online, 2012). |
R832235 (Final) |
Exit |
Supplemental Keywords:
Drinking water; risk; human health; contingent valuation; non-market valuation; treatment, RFA, Economic, Social, & Behavioral Science Research Program, Scientific Discipline, INTERNATIONAL COOPERATION, Water, POLLUTANTS/TOXICS, Health Risk Assessment, Arsenic, Biochemistry, decision-making, Environmental Policy, Water Pollutants, Economics & Decision Making, Social Science, children's health, environmental awareness, decision analysis, decision making, drinking water, environmental decision making, morbidity risks, household choice, efficient household framework, mortality risks, arsenic exposure, environmental risk assessmentProgress and Final Reports:
Original AbstractThe perspectives, information and conclusions conveyed in research project abstracts, progress reports, final reports, journal abstracts and journal publications convey the viewpoints of the principal investigator and may not represent the views and policies of ORD and EPA. Conclusions drawn by the principal investigators have not been reviewed by the Agency.