As our communities grapple with critical water resource management issues—including drought response, water conservation, water quality impairment, climate change, agricultural and landscape irrigation, and environmental flows and water rights—it is increasingly important to provide key water resource information to citizens, enabling them to make informed water-use decisions (Floress et al., 2009; Rosenzweig et al., 2007; Willis et al., 2011). Based on state-level needs assessments that identified water resource status and management as a priority for extension programs, extension services and land-grant institutions (LGIs) have directed human and financial resources toward providing education for the public regarding water use efficiency, water quality, and environmentally sustainable practices.
The number of people reached by a program is one of the indicators widely used to evaluate outreach program impacts (Bennett, 1975; Diem, 2003). Although information regarding the number of people attending specific types of educational activities or accessing specific websites is collected at the local level, rarely is this information collected and reported on regional or national levels. However, regional data are required to answer questions such as: What percentage of the population in the region is reached by an outreach organization? What sources of information are most successful in reaching the public with water resource information messaging? How do preferences for information sources and methods of information delivery depend on community size and other socio-demographic characteristics?
A nationwide survey of public attitudes and perceptions related to water issues (Mahler et al., 2013) was developed to provide answers to these questions and evaluate the impact of outreach programs at the regional and national levels. This article uses the nationwide survey to examine the information sources that frequently deliver water resource information to populations in communities ranging from small towns of less than 3,500 to large cities of more than 100,000 people. Particular attention is given to cooperative extension services, which historically focused their efforts in agricultural and rural areas and are now increasing their efforts to also serve populations in large cities (Ilvento, 1997; Bull et al., 2004; West et al., 2009).
The objectives of this article are to:
Assess the number of people reached by various water resource information sources.
Characterize audiences currently receiving their water resource information through extension services and LGI programs.
Evaluate audiences’ preferences for additional information on particular water resource topics.
Evaluate audiences’ preferences for the medium used to deliver water resource information.
Materials and Methods
A national random sample survey, stratified by state, was conducted in 41 U.S. states, as described in Mahler et al. (2013). In this article, we focus on respondents from the Southern, Mid-Atlantic, and Northeast regions, specifically the states of Alabama, Arkansas, Delaware, Florida, Georgia, Louisiana, Maryland, Mississippi, New Jersey, New York, Oklahoma, Pennsylvania, Texas, Tennessee, Virginia, and West Virginia, which were surveyed in 2006 through 2010. Other states were not included in the present analysis because they were surveyed before 2006 or after 2010, outside the time window selected. Several other states were also excluded from the analysis because their survey questionnaires did not use the same wording for questions discussed in this article.
State survey questionnaires were based on the template developed by state extension water quality coordinators in the Pacific Northwest region (Mahler et al., 2010) and modified through discussions with LGI water quality coordinators for the Southern, Mid-Atlantic, and Northeast and Caribbean Island Regional Water Programs. The questionnaires included 54 to 59 questions that addressed water resource, water quality, and other environmental issues, with some questions tailored to specific concerns in each state.
As discussed in Mahler et al. (2013), the nationwide survey was designed following the tailored survey design method (Dillman, 2000). The survey was administered using a four-step mail procedure to a randomly selected sample of residential addresses. For the 16 states discussed in this article, the overall survey response rate was 46% (total number of responses was 5,010; margin of error is ±3%). State response rates and survey dates are shown in Table 1.
|Category||Demographics (state abbreviation; survey year; survey response rate)||% of Survey respondents||% of Adult population >25 in represented states|
|State of residence (n = 5,010)||Alabama (AL; 2009; 48%)||6%||4%‡|
|Arkansas (AR; 2008; 60%)||5%||2%‡|
|Delaware (DE; 2006; 41% )||2%||1%‡|
|Florida (FL; 2008; 45%)||10%||14%‡|
|Georgia (GA; 2010; 47% )||10%||7%‡|
|Louisiana (LA; 2008; 42%)||5%||3%‡|
|Maryland (MD; 2006; 43%)||5%||4%‡|
|Mississippi (MS; 2009; 64%)||6%||2%‡|
|New Jersey (NJ; 2006; 37% )||6%||6%‡|
|New York (NY; 2006; 38% )||8%||15%‡|
|Oklahoma (OK; 2008; 52%)||5%||3%‡|
|Pennsylvania (PA; 2006; 45%)||8%||10%‡|
|Tennessee (TN; 2008; 51%)||7%||5%‡|
|Texas (TX; 2008; 33%)||8%||18%‡|
|Virginia (VA; 2006; 38%)||5%||6%‡|
|West Virginia (WV; 2006; 44%)||3%||1%‡|
|Size of residence community (n = 4,536)†||Greater than 100,000 residents||30%||na§|
|Between 25,000 and 100,000 residents||29%||na|
|Between 7,000 and 25,000 residents||19%||na|
|Between 3,500 and 7,000 residents||10%||na|
|Less than 3,500 residents||12%||na|
|Duration of residence in the state (n = 4,855)†||Respondents living in state less than 5 years||5%||na|
|Respondents living in state 5 to 10 years||7%||na|
|Respondents living in state 10+ years, but not all life||42%||na|
|Respondents living in state all their lives||46%||na|
|Age (n = 4,768)†||25–34 years old||6%||20%‡|
|35–49 years old||22%||31%‡|
|50–64 years old||33%||29%‡|
|65 years old or older||39%||20%‡|
|Gender (n = 4,851)†||Female||36%||52%‡|
|Educational level (n = 4,825)†||High school education or less||24%||44%¶|
|Some college or college degree||54%||45%¶|
Each survey included questions about demographics; the following three questions are the focus of this article:
Have you received water resources information from the following sources? Thirteen potential sources were listed requiring “yes” or “no” for each source. For this analysis, responses about cable, network, and public television were aggregated, as were responses about local and major newspapers.
If you had the following kinds of learning opportunities to learn more about water issues, which would you be most likely to take advantage of? (Check up to 3 items). Twelve information delivery methods were offered as answer choices.
Would you like to learn more about any of the following water quality issue areas? (Check all that interest you). Answer choices included 16 issues.
JMP 9.0.0 statistical software (SAS Institute, 2008) procedures were used to examine the frequency of survey responses and conduct multinomial logistic regression analyses (Kenney, 2008). Specifically, for each question, the frequency of responses in each answer category was cross-tabulated with respondents’ self-identified residence (inside or outside city limits) and/or community size category. The null hypothesis that the answer frequencies are the same for the various residence categories was tested using likelihood ratio chi-square tests (SAS Institute, 2012a).
To account for the effects of various socio-demographic characteristics, we used multinomial logistic regression analysis as described by Borisova et al. (2013); likelihood ratio chi-square test results are reported for those characteristics that were identified as statistically significant in the multinomial logistic regression model. The likelihood ratio chi-square tests are used to test the hypothesis that a particular variable does not have a statistically significant effect, and therefore, can be eliminated from the regression model. The statistic is part of the standard output of logistic regression analysis in JMP 9.0.0. It “is calculated as twice the difference between the log-likelihoods of the full model and the model constrained by the hypothesis to be tested (the model without the effect)” (SAS Institute, 2012b).
Table 1 summarizes the demographics of the survey respondents. Although mailing addresses were selected at random in each state and balanced for indications of male and female names (no names with only first initial were included), our respondents were somewhat older, better educated, and more likely to be male than the population of the nation (Mahler et al., 2013) or the individual states (as shown in Table 1).
The majority of respondents (59%) resided in communities of 25,000 or more people. Twenty-two percent of respondents resided in small communities of 7,000 or fewer people. A large majority (88%) had lived in their states for more than 10 years.
Sources of Water Resource Information
Respondents were asked to identify from which sources they had received water resource information by replying “yes” or “no” for each of the 13 sources listed. Not surprisingly, mass media sources such as newspapers and television reached the greatest number of people (59% and 47%, respectively) (Table 2). Receiving water resource information from television and newspapers was fairly constant and high among communities of all sizes. The exception was very small communities (less than 3,500 people), where the use of these information sources was lower than in communities of other sizes. However, even in very small communities, 42% and 56% of respondents reported receiving water resource information from television and newspapers, respectively.
|Information source||Overall (n = 3698)||City or town less than 3,500 (n = 397)||City or town 3,500 to 7,000 (n = 343)||City or town 7,000 to 25,000 (n = 660)||City or town 25,000 to 100,000 (n = 979)||City larger than 100,000 (n = 1021)|
Among the three key types of organizations that implement public outreach programs— extension, environmental groups, and environmental agencies—environmental agencies was selected most frequently by respondents (31%), followed by environmental groups (29%). However, among the respondents residing in the very small communities, the three types of outreach organizations were almost equally important, reaching 23 to 29% of respondents. As city or town population size increased, the percentage of respondents indicating that they had received water resource information from extension decreased to 10% for very large cities (more than 100,000 people; Likelihood ratio test, p < 0.0001). In contrast, the use of environmental groups or agencies for water resource information increased from 24 and 29%. Although we cannot quantify the amount, it should be recognized that mass media often relies on the information developed by universities and extension services, as well as environmental organizations and agencies. Hence, the results reported here likely underestimate the indirect contributions of these sources. Sixteen percent of respondents received water resource information from universities. Fifteen percent (one in seven households) of the adult population in the Southern, Mid-Atlantic, and Northeast regions recognized having received water resource information from cooperative extension services. This corresponds to about 15.3 million in the 16 states sampled.
Priority Topics for Water Resources Programs
Preferred topics are shown in rank order in Fig. 1. The three water resource topics of greatest interest to respondents (n = 2,112) were Protecting Public Drinking Water Supplies (65%), Watershed Management (51%), and Nutrient and Pesticide Management (50%). Multinomial logistic regression analyses (n = 1,892) of those topics selected by at least 40% of the respondents showed similar responses across demographic categories, with the following exceptions (results in text only): (1) younger respondents (p < 0.0049) and those with less education (p < 0.0032) were more likely than other groups to want to learn about Protecting Public Drinking Water Supplies; (2) males (p < 0.0177) were more likely to want to learn about Watershed Management; (3) females (p < 0.0334) were more likely to want to learn about Nutrient and Pesticide Management; (4) males (p < 0.0051), younger respondents (p < 0.0004), and those with less education (p < 0.0001) were more likely to want to learn about Fish and Wildlife Water Needs; and (5) females (p < 0.0019) and younger respondents (p < 0.0001) were more likely to want to learn about Home and Garden Landscaping.
Six topics were of interest to at least 40% of respondents living in communities of fewer than 3,500 (Table 3). Four of these six water resource topics also were of most interest in larger cities: (1) Protecting Public Drinking Water Supplies, (2) Nutrient and Pesticide Management, (3) Watershed Management, and (4) Fish and Wildlife Water Needs. In contrast, Septic System Management and Private Well Protection were not of great interest in larger communities. Home and Garden Landscaping and Watershed Restoration were of greater interest in large cities than in smaller communities (Table 3).
|Water resource topic||<3,500 (n = 228)||3,500–7,000 (n = 202)||7,000–25,000 (n = 368)||25,000–100,000 (n = 552)|
|Protecting public drinking water||61%||66%||66%||64%||68%|
|Septic system management||50%||42%|
|Private well protection||49%|
|Nutrient and pesticide management||49%||49%||51%||48%||51%|
|Fish and wildlife water needs||45%||50%||48%||43%||40%|
|Home and garden landscaping||41%||40%||45%|
Preferred Educational Methods
Respondents generally preferred to learn about water resources by Reading Printed Fact Sheets, Bulletins, or Brochures (52%); Watching Television (47%); Reading a Newspaper Article (45%); or Visiting a Website (42%) (Fig. 2). Following those preferences, all other choices included in the survey for delivery of water resource information were likely to be used by 16% or fewer of respondents.
Multinomial logistic regression analyses (n = 4,028) of the responses regarding learning opportunities most frequently preferred by respondents against demographic information showed consistent responses across demographic categories, with a few exceptions: (1) respondents having more education (p < 0.0001) and those living in their states for more years (p < 0.0354) were more likely than other groups to select Reading Printed Fact Sheets, Bulletins, or Brochures as a preferred learning opportunity; (2) older respondents (p < 0.0001), those living more years in their state (p < 0.0404) and those with less formal education (p < 0.0001) were more likely than other groups to select Watch TV Coverage as a preferred learning opportunity; (3) Reading a Newspaper Article was less likely to be selected by younger respondents (p < 0.0001) and respondents living fewer years in their state (p < 0.0419); and (4) younger respondents (p < 0.0001), those with more education (p < 0.0001), and those having lived fewer years in their state (p < 0.0029) were more likely than other groups to select Visit a Website.
As shown in Table 4, educational material delivery preferences were similar across community sizes for all delivery methods. However, there were significant differences among community sizes for likelihood to Visit a Website (p < 0.001), Look at a Demonstration/display (p < 0.041), and Take a Course for Certification/credit (p < 0.021).
|Information source||<3,500 Residents (n = 228)||3,500–7,000 Residents (n = 202)||7,000–25,000 Residents (n = 368)||25,000–100,000 Residents (n = 552)||>100,000 Residents (n = 591)|
|Read fact sheets, bulletins, or brochures||54%||56%||52%||50%||51%|
|Watch TV coverage||45%||46%||50%||48%||50%|
|Read newspaper article||45%||47%||41%||46%||45%|
|Visit a website***||37%||40%||41%||45%||47%|
|Watch a video of information||17%||14%||16%||15%||16%|
|Ask for water practices assessment||15%||11%||12%||12%||11%|
|Look at demonstration/display*||11%||13%||15%||11%||12%|
|Participate in one-time volunteer activity||12%||10%||11%||13%||12%|
|Attend a fair or festival||12%||12%||11%||11%||11%|
|Attend a short course or workshop||11%||9%||11%||10%||10%|
|Take course for certification/credit*||8%||6%||7%||4%||6%|
|Get trained for regular volunteer position||7%||6%||7%||6%||6%|
Likelihood of Visiting a Website to receive water resource information followed a clear trend, with those living in smaller communities being less likely than those living in larger communities. While no clear trend was apparent for Look at a Demonstration/display responses, preferences for this information delivery method were higher in communities in the range of 7,000 to 25,000 people (15% compared with 11% for very small communities and larger communities of 25,000–100,000 people) (p < 0.041). Similarly, there was no clear trend in the responses for Taking a Course for Certification/credit; however, 8% of respondents in very small communities were interested in this educational opportunity, compared with only 4% in communities of 25,000–100,000 people (p < 0.021).
A large majority (57–58%) of respondents 18 to 44 years old indicated that they would prefer to Visit a Website to learn about water resource issues (Fig. 3). Respondents 18 to 34 years old were also more likely to prefer to Attend a Fair or a Festival (18%) or Take a Course for Certification/credit (15%) (chi-square test, 95% or higher confidence level).
This article evaluates the results of a nationwide survey of public attitudes and perceptions regarding water resources, and focuses on participants’ responses to questions related to preferred water information sources, learning opportunities, and topics. Mass media sources such as newspapers, television, and the Internet reach the greatest number of people. Receiving water resource information from television and newspapers was fairly constant and high among communities of all sizes, as was the case for a similar study conducted in the Pacific Northwest (Mahler et al., 2010).
The reach of three types of organizations that implement public educational outreach programs (i.e., extension, environmental groups, and environmental agencies) lagged behind mass media sources, but reached a significant proportion of the population (15–31%). Compared to extension, environmental groups, and environmental agencies, mass media enjoys larger budgets for designing and delivering programs, a broader base audience, and infrastructure to reach individuals in their homes. Additionally, mass media frequently relies on and delivers information produced by environmental agencies and environmental groups, and universities/extension.
Fifteen percent of the adult population in the Southern, Mid-Atlantic, and Northeast states reported receiving water resource information specifically from cooperative extension services in the 16 states sampled. This corresponds to about 15.3 million people. Cooperative extension services have provided water resource information to about one in four households (23%) in the demographic traditionally targeted by extension, those in very small or small communities. A similar analysis of the portion of this nationwide survey conducted in 2004 in more western and less populated states (South Dakota, Montana, North Dakota, Wyoming, Colorado, and Utah ) showed 32% of that population reported receiving water quality information from extension services (Clay et al., 2007). Mahler et al. (2010) reported that almost 50% of adult Pacific Northwest residents living in rural communities of less than 25,000 people in 2007 cited extension as a water information source.
Even in very large cities, where extension efforts may not have a long history, 1 in 10 households has received water resource information from extension services. Even though extension has reached a smaller percentage of the population in large urban areas, the number of households is larger than that reached in the smaller communities and rural areas. These results are similar to those reported by Borisova et al. (2013), indicating that in Texas 40% of extension’s traditional rural audience, or approximately 1.2 million individuals, reported receiving water resource information from the Texas A&M AgriLife Extension Service, whereas 3.1 million individuals, or 14.5% of city dwellers, reported receiving extension water resource information.
Analyses of those water resource topic areas of interest to at least 40% of respondents suggest some reasons for extension services’ success in expanding programs into large and very large cities where they have not traditionally focused. As shown in Fig. 3, respondents in the larger cities had very similar water resource interests to those of the traditional audience. Both audiences are interested in protecting drinking water supplies, nutrient and pesticide management, and watershed management. Respondents in smaller cities are more likely to have an interest in septic systems and private water wells, probably because they are more likely to have one. Because of these common interests in water resource topics, extension services are well-poised to expand their water resource educational programming into large urban areas.
If outreach organizations aspire to reach broader audiences, focusing on priority topics such as public drinking water supply is essential. Additional needs assessment surveys should be conducted to examine more specific issues and the changes in topics that respondents care about (e.g., indoor and outdoor water conservation, protecting the quality of drinking water sources, and reducing water treatment costs/water bills). Based on the survey, an overwhelming majority of respondents (65%) were interested in learning more about protecting public drinking water supplies. Topics that are traditionally addressed by environmental agencies and groups and extension—watershed management and nutrient and pesticide management—also are primary topics of interest to approximately half of respondents overall. Special outreach programs focused on septic system management and private well protection may attract rural and peri-urban residents, whereas home and garden landscaping topics attract urban residents.
As shown in Fig. 2 and Table 4, educational material delivery preferences were similar across community sizes for these delivery methods. The exception is older respondents living in smaller communities who are less likely to Visit a Website than those who are younger and living in larger communities. This result underscores the increasing importance of having online materials designed to reach urban and younger audiences. Although printed materials will remain important for reaching audiences of all ages in communities of all sizes, as the 18 to 34 year old group ages and children reared with easy access to the Internet enter the 18 to 34 year old age group, a larger proportion of all audiences will likely prefer receiving their information from websites, as well as reading fact sheets, and fewer individuals will read newspapers and watch TV coverage.
Overall, national and regional public surveys are an important tool for evaluating the impacts of outreach programs. Public surveys allow assessment of the proportion of the population reached by outreach programs, and the comparative evaluation of audiences reached by various programs. Such evaluation can help recognize program successes, and identify and address possible challenges in outreach programs. In the future, surveys should focus on specific water resource issues (e.g., water conservation or nutrient water pollution), and examine public reliance on and public trust of information delivered by various information sources. Public use of information during particular time periods should also be evaluated (e.g., the number of times information was received throughout the last year). Finally, regional public surveys can also be used to evaluate public response to various types of informational messages and incentives programs, current level of knowledge about specific water resource issues, or public willingness to be engaged in specific behaviors.