Southerners and their Swamps:The View from Middle Georgia
Given that wetlands have become a major environmental issue in the United States, it is important for people to understand the functions and values provided by wetlands so they are in a position to make informed decisions regarding public policy. This study is an attempt to document people's perceptions of wetlands and to determine if college students (who can be surveyed via the Internet at low cost) are representative of the general population. Although many people have at least a fair understanding of wetlands, it would appear that there is room for improvement because many of those surveyed are not aware of some important wetland benefits. Moreover, the results of this study suggest there are broad similarities of opinion between college students and residents of Baldwin County, Georgia. Finally, it would appear that the potential exists to develop a representative sample of the nation's college students (via the Internet), and that such a sample might be used in conjunction with conventionally obtained survey data to determine relationships between demographic characteristics of respondents and their views of wetland functions, values, and public policy.
wetlands, environmental perception, environmental policy, Georgia
Once despised as environments whose full potential could be realized only if they were converted to some other use, wetlands have now become a major environmental issue in the United States (Vileisis 1997; see also the July 1990 issue of Audubon and the November/December 1995 issue of the Journal of Soil and Water Conservation, both of which are devoted entirely to wetlands). Indicative of this shift has been the development of a substantial literature regarding many aspects of wetlands including public policy, agriculture, wetland loss, and wetland ecology (Hook 1988; Williams 1990; Tzoumis 1998; Shabman et al. 1998; Mitsch and Gosselink 2000; Dahl 2000). Over the past several decades, the scientific community has discovered a host of benefits provided by many wetlands such as flood control, water purification, groundwater recharge, habitat for endangered and other important species of plants and animals, and many direct and indirect economic benefits to people. It is important for people to have an accurate impression of these benefits so they will be in a position to make informed decisions regarding whether or not to support policies that encourage wetland conservation. Assessing people's perceptions of wetlands will reveal the relative effectiveness of wetland education efforts to date, and it will identify issues that require further attention. Accordingly, this paper has three major objectives: [End Page 74] to gauge what people know about wetlands (especially ecological functions and resulting values to people); to determine what they think about these environments (especially in terms of their stance on wetland public policy issues); and to learn if college students (who can be less expensively surveyed via the Internet) are representative of the general population in terms of their understanding of wetlands.
The first step in this study consisted of developing a wetland survey that could be duplicated and distributed by regular mail, as well as an Internet version of the same survey (with an associated project website) that could be completed in 10 minutes or less (www.geocities.com/websurvey_geo/wetsurvey.html). Some of the survey questions were intended to provide basic demographic data, others asked participants about their understanding of basic ecological functions in wetlands (and related values to people and wildlife), and still other questions asked for people's opinions of preservation priorities and wetland public policy issues. Although wetlands are a diverse collection of landscapes loosely connected by periodic flooding as well as soils shaped by that flooding and vegetation adapted to waterlogged soils, the survey implied no differences among wetlands. The questions asked respondents their opinion of "most wetlands." Attention was focused on surveying people in Middle Georgia in part because nearly half of the nation's wetlands are in the southeast U.S. (Dahl 1990), and in part because I wanted to engage in a smaller scale pilot study in preparation for a more comprehensive effort in the future. Located at the edge of the Georgia Piedmont (Fig. 1), Baldwin County's topography is dominated by rolling hills which preclude the development of extensive wetlands. Most of the county's few wetlands are riparian swamps associated with rivers and creeks.
Upon visiting the Baldwin County courthouse in Milledgeville, Georgia during the fall of 2002, a list of 800 randomly selected names and addresses of property owners scattered throughout the county was compiled, and a hard copy of the survey instrument was mailed to each of these 800 households. Similarly, freshman seminar instructors at Georgia College & State University (GC&SU, also in Milledgeville) were encouraged to have their students take the wetland survey via the Internet. Given the expense of doing traditional survey work, I wanted to determine if using the Internet was an effective tool for learning about people's perceptions of wetlands and if college students would be representative of the general population. When someone completes the Internet version of the survey, responses are dumped anonymously into an e-mail account that provided easy access to the raw data. Of the 1,012 GC&SU students in freshman seminars during the fall 2002 term, 168 responded (16.6% response rate); and of the 800 print versions of the survey distributed to Baldwin County residents, 147 were returned complete or very nearly complete (18.4% response rate).
Previous research has established the benefits and challenges of developing and using Internet-based surveys (e.g., Fisher and Margolis 1996; Schmidt 1997; Couper 2000; Madge and O'Connor 2002). My experience appears to mirror these findings. On the positive side, Internet surveys cost [End Page 75] nothing to implement—no small consideration when one considers the expense of conducting such surveys in person, on the telephone, or by conventional mail. Furthermore, on-line questionnaires are easy to update or modify if one discovers errors. Moreover, since much of the process is automated, it requires less time and energy on the part of the researcher, and the results can be sent or downloaded to an easily accessible database.
On the negative side, five potential problems associated with using the Internet to conduct surveys soon appeared. To begin with, because Internet access is unequal, surveying a true cross-section of the [End Page 76] population is impossible (Warf 2001). Furthermore, simply putting a survey on the web is no guarantee that Internet users will find and complete the instrument (Coomer 1997). Although I initially tried to publicize the survey by attempting to convince administrators of selected web sites to place a link to the wetland survey on their web sites, I received little if any cooperation. Of course, even this strategy would have yielded data from only the relatively small universe of Internet users, a situation that has the potential to support the views of those privileged with computer access (Madge and O'Connor 2002). Although college students were ultimately targeted, this strategy is heavily dependent upon the ability of fellow college professors to incorporate such a survey into their classes, and the response rate of the students suggests that this is (understandably) not a given. In addition, web surveys can be subject to technical problems, either with the computer server hosting the survey or with the respondent's computer. Finally, there is no way to prevent repeat submissions while maintaining the anonymity of respondents. Indeed, there is evidence that many respondents made the honest mistake of double clicking the submit button, thereby sending their results twice. These (apparently) duplicate submissions were identified and discarded.
Despite these problems, using the Internet to conduct research has potential. This potential should not be exaggerated because as Madge and O'Connor (2002, 100) conclude, "many of the issues and problems of conventional research methods still apply in the virtual venue." For example, we can never be certain that responses are accurate reflections of respondents' thoughts (or just guesses); nor can we be certain that respondents will behave as they claim they would in responding to certain questions. Nevertheless, although it may not be possible to overcome some of these problems, the savings in time and money associated with Internet surveys suggests that this method can be used to develop significant databases.
Finally, I compared the responses of the two groups—the property owners who were surveyed by regular mail and the college students who were surveyed via the Internet. Analysis focused on comparing the percentage of each group that responded a certain way to each question. For the purpose of this discussion, "students" will refer to those who responded to the Internet version of the survey and "locals" will refer to Baldwin County residents who responded to the regular mail version of the survey. Although it is possible that some of those who responded to the Internet version of the survey were also contacted by mail (and possibly surveyed twice), this is extremely unlikely given the fact that I did not advertise the Internet survey and most GC&SU freshman are eighteen or nineteen years old.
In some ways the populations of the two groups surveyed (GC&SU students and Baldwin County local residents) are similar, and in other ways they are different. Seventy-one percent of the student respondents were female, as were 45% of the local respondents, figures that correspond closely to the demographics of GC&SU and Baldwin County, respectively. Students (all college freshmen) were not asked their level of educational attainment, but locals were, and nearly half reported [End Page 77] that they had at least a bachelor's degree, 34% reported some education beyond high school, and 15% reported having various levels of education up through and including high school graduation (or its equivalent). Several students and locals alike did not report their family income. However, as suggested in Figure 2, roughly half of both groups of respondents live in households that earn more than $55,000 per year. Similar percentages of both groups are represented in lower income categories except that 11% of the students claimed they came from families that earned $25,000 or less while only 5% of all locals reported this level of family income. Most of the students surveyed (91%) claimed to be less than 25 years of age, but just over half of the locals responded that they are more than 55 years old; another 35% of the locals claimed to be between 40 and 55. Finally, students were asked whether they came from a large city (10% checked this option), sub-urban area (27%), small town (32%) or rural area (13%). Oddly, nearly 20% of the students left this question blank. Locals were not asked this question. Baldwin County has fewer than 45,000 people, and it was assumed that nearly all of the locals live either in or near the small town of Milledgeville (population 19,000) or in outlying rural areas.
Several of the survey questions deal with people's knowledge and understanding of wetland environments and their ecological functions (many of which provide values to people and wildlife). Since scientists and environmentalists have pushed wetland conservation to a high place on the American environmental agenda, it was expected that large percentages of respondents would be familiar with basic wetland functions and issues. For instance, according to a widely cited estimate by Dahl (1990), between the 1780s and the 1980s, the U.S. lost a little over half of its original wetland base. Figure [End Page 78] 3 shows that relatively few students and locals correctly believe that the conterminous U.S. has lost between 41% and 60% of its wetlands; more than a quarter of each group thought that the U.S. has lost more than 60% of its original wetland base. Very few in each group surveyed thought the U.S. had lost less than 40% of its wetlands since the 1780s. Although such responses may simply be guesses (roughly one-third of students and locals admitted they were not sure what percentage of wetlands the U.S. has lost over the last two centuries), these results suggest that many people realize that wetland losses in the U.S. since colonial times have been substantial.
In terms of wetland hydrology, most wetlands are actually dry for at least a few weeks (and in many cases for several months) during most years. Almost two-thirds of the locals surveyed recognized this, as did one-third of the students (Fig. 4). At the same time, however, 14% of locals and nearly a quarter of the students thought most wetlands were covered with water all year—while even larger percentages of locals and students admitted that they were not sure about water levels in wetlands during most years. In a related matter, among the most celebrated of wetland benefits to people is that of flood control. Because wetlands represent somewhat lower lying spots on the landscape, they are capable of storing water that might otherwise run across the landscape and contribute to flooding elsewhere (Belt 1975; Novitzki 1979; O'Brien and Motts 1980; Tobin 1986; Doyle 1986; Hubbard and Linder 1986; Carter 1996). As is true of other proclaimed wetland benefits, however, this literature suggests that while many wetlands are capable of temporarily storing much excess water on the landscape, [End Page 79] not all wetlands provide the same level of flood control. Seventy-eight percent of the locals and 47% of the students surveyed thought that most wetlands were usually capable of providing some degree of water storage and flood protection, but 18% of locals and over half the students were not sure.
Although many wetlands (such as Central Florida's Green Swamp or the Nebraska Sand Hills) contribute to groundwater recharge, the literature suggests that many wetlands play the opposite role (Stone and Stone 1994). In other words, water often appears at the surface because the base of the wetland intersects the water table and groundwater is being discharged to the surface. Moreover, some wetlands have little or no connection to groundwater systems (perhaps due to impermeable soils) and still others have groundwater oozing to the surface at one end of the wetland while this same water sinks back into the ground at the other end. Only 40% of locals and just 16% of students thought that at least some wetlands display each of these relationships with groundwater systems, and nearly one-third of the locals and fully two-thirds of the students admitted they were not sure about the possible relationship(s) between wetlands and groundwater systems. There is an extensive literature regarding wetland functions and their value to fish and wildlife (see Meyer 1994; Mitsch and Gosselink 2000). Most scientists agree that wetlands provide habitat for birds, waterfowl, several endangered species of plants and animals, and they serve as breeding grounds for many species of fish and shellfish (Stewart and Kantrud 1974; Peters et al. 1979; Boesch and Turner 1984; Sugden 1984; Moler and Franz 1987; Stewart 1996). Seventy-three percent of the students and 91% of the locals thought that most wetlands were home to at least some threatened or endangered species of plants or animals; although only [End Page 80] 5% of all locals were unsure about this, nearly a quarter of the students were unsure. Meanwhile, more than 60% of the students and 85% of the locals agreed that wetlands provide breeding and wintering habitat for many species of birds and waterfowl; again only a small percentage of the locals were unsure, but one-third of all students surveyed were not sure about the importance of wetlands to many species of birds and waterfowl. Finally, over half the students and three-quarters of locals believed that coastal wetlands provide breeding grounds for several commercially valuable species of fish and shellfish. At the same time, one out of five locals surveyed (and twice that many students) were unsure if there existed any relationship between wetlands and many species of fish and shellfish.
Because wetlands typically contain water that is moving slowly, if at all, they allow sediments and other impurities to fall out of the water column, thereby improving water quality in local rivers and streams (Hemond and Benoit 1988; Phillips 1989). Survey respondents seemed even less certain about this valuable ecological function in most wetlands. As seen in Figure 5, just over half of the locals and a much smaller percentage of students surveyed acknowledged this important wetland function, while relatively large percentages of both locals and especially students remained unsure about wetlands and water quality. Still others in both groups thought wetlands either had no impact or a negative impact upon water quality.
Respondents seemed less certain about broader economic values associated with wetlands. There is some literature that suggests wetlands provide economic benefits to human communities and that in [End Page 81] some cases, wetlands have increased local property values (Raphael and Jaworski 1979; Thibodeau and Ostro 1981; Batie and Shabman 1982; Farber and Costanza 1987; Turner 1991; Barbier 1993). Just under half of the locals and only a third of the students thought that most wetlands provided economic benefits to local communities, while the remainder in each group either confessed they were not sure or thought that wetlands provided no economic benefits to human communities. Survey respondents were also asked their opinion of the impact of wetlands upon nearby property values. According to Figure 6, very small percentages of students and locals surveyed thought that wetlands helped increase property values. At the same time, over a quarter of both groups thought wetlands reduced nearby property values, and many more admitted they were not sure about the relationship between wetlands and property values.
The survey also included several questions designed to probe respondents' opinion of wetlands generally, including their thoughts and knowledge of wetland public policy issues. Less than 10% of both student and local respondents claimed to have no interest in the outdoors; large majorities of both students and locals claimed either to enjoy outdoor activities or to be concerned about outdoor issues even if they did not engage in outdoor activities often. Similarly, less than 5% of students (and even fewer locals) thought that most wetlands are ugly; most respondents agreed that at least some wetlands are beautiful landscapes—a view supported by the literature (Niering 1979; Smardon 1979; Fritzell 1979; Reimhold and Hardisky 1979). These questions (and the results) suggest an element of sample bias: a propensity to attract respondents who are predisposed to appreciating wetland environments. [End Page 82]
Meanwhile, respondents were asked if they would "vote in favor of government regulations that would prevent private property owners from converting (developing) any wetlands they own into new homes, businesses, or farmland?" Locals were evenly divided on the issue with similar percentages responding yes, no, and unsure to this question. As Table 1 indicates, students were only a bit more likely to support such hypothetical government regulations. Respondents were then asked their opinion of the following statement: "Private property owners should always be allowed to convert (develop) any wetlands they own into new homes, businesses, or farmland." Relatively small percentages of students and locals either agreed or strongly agreed, while more than half of the students and locals either disagreed or strongly disagreed (Fig. 7).
Much controversy has swirled around the concept of wetland mitigation in recent years (Dennison and Berry 1993; Zedler and Shabman 2001; Scodari and Shabman 2001). Wetland mitigation is the policy action of forcing developers of wetlands to replace destroyed wetlands with either restored or created wetlands not far from the lost wetlands. As Table 1 indicates, well under half of locals and even fewer students were familiar with the concept of wetland mitigation. Similarly, small percentages of both students and locals claimed familiarity with the concept of wetland banking (see Table 1), a policy option where developers are allowed to build credits toward future wetland destruction by restoring or creating new wetlands in the present (Kusler 1992; Dennison and Berry 1993; Hoagland et al. 1996).
Finally, in an effort to gauge the relative importance of wetlands when compared with several other environmental issues, I asked survey takers to place in rank order their preservation priorities for six different environments: grasslands, wetlands, forests, coasts, deserts, and open space in urban areas (Fig. 8).1 As one may have predicted, both locals and students (all from Georgia) put preservation of desert environments as their lowest priority. At the opposite end of the scale, both students and locals more often listed forests as their leading preservation priority. On average, wetlands ranked as the second most [End Page 83] important preservation priority among locals (just ahead of coastal environments), but students ranked wetlands as their number three preservation priority (behind coastal environments), perhaps reflecting students' greater use of beaches for recreation. Grasslands and open space had average ranks ahead of deserts but well behind forests, coasts and wetlands.
The scientific community is still a long way from fully understanding the precise nature of all ecological functions in wetland environments. Moreover, it is important to keep in mind that most wetlands do not perform all of the functions attributed to the universe of wetlands. Yet enough is known to suggest that many swamps, marshes and bogs perform functions that provide values to people, fish and wildlife. This study suggests that although people are aware of some wetland functions and values (and probably more so than in the past), there is room for more widespread understanding of such functions and values. Indeed, Hugh Prince (1997) documented significant shifts in people's perception of Midwestern wetlands from the late nineteenth to early twentieth centuries, so it would seem that there is the potential for further changes of attitude. For example, significant percentages of survey respondents admitted they were unsure of selected wetland functions and values, and many others signaled their ignorance by selecting answers to questions that are not supported by the literature. Again, understanding wetland environments is important if people are to make intelligent decisions regarding their support [End Page 84] (or lack thereof) for certain wetland public policies.
Moreover, the results of this study suggest there are broad similarities of opinion between GC&SU college students and residents of Baldwin County, Georgia. For most questions regarding ecological functions in wetlands, somewhat smaller percentages of locals (most of whom are older and have completed more formal education) responded that they were unsure, and larger percentages of locals consistently provided responses supported by the scientific literature. With regard to wetlands and public policy, locals (all of whom were, by definition, property owners) appeared a bit more knowledgeable of emerging policy tools such as wetland banking and mitigation—but they were almost as divided as the students regarding their opinion of the proper course of public policy vis-à-vis wetlands.
The relative comparability of responses between students and locals is an important issue because it was originally hoped that the responses of the two groups would be similar enough to justify a nationwide survey of college students that might be representative of a large percentage of the U.S. population, especially since surveying college students via the Internet is much less expensive. On this count the results of this study are inconclusive. For instance, it is not clear if the two groups would have produced more similar responses had all of GC&SU's student body (particularly juniors and seniors) been asked to participate. Furthermore, the sample of locals who responded to the mail survey (mostly older, more affluent and more educated property owners) is not as representative of the Baldwin County population as it should be in order for the results to be valid. Finally, given the diversity of wetland environments, it will probably be necessary to survey people (students and locals) [End Page 85] from a broader range of places to ensure that people somewhat familiar with different wetland types—and from different environments (e.g., urban vs. rural)—are included. Indeed, perhaps a telephone survey of people across an entire state could be compared with an Internet survey of students (from freshman to graduate students) from several locations in that state, which might provide results that could be used to justify a national survey.
The goals of this project were twofold. First, I wanted to engage in a local pilot study intended to determine public understanding of wetlands and wetland public policy issues. In terms of this aspect of the study, one might draw two tentative conclusions. To begin with, it does appear that many people have at least a fair understanding of wetlands. This is important because if people understand the ecological functions and values provided by wetlands, they are more likely to make better decisions regarding which wetland public policies to support. At the same time, it would appear that there is room for improvement because some of those surveyed are not aware of some important wetland benefits or policy options. Others may incorrectly believe that all wetlands provide all the functions and values attributed to wetlands. It is important for environmental educators to continue to disseminate accurate information regarding wetland functions and values and to make clear that different wetlands provide different suites of benefits.
The study's second goal dealt with the idea of determining if college students (who can be surveyed at lower cost via the Internet) would be representative of the general population in terms of their understanding of wetlands. It would appear that the potential exists to develop a somewhat broader and more representative sample of the nation's college students and that such a (less expensive) sample might be used in conjunction with (more expensive) conventionally obtained survey data. Future research might include a similar study in which college students from an entire state are invited to take a slightly revised wetland survey via the Internet, while at the same time efforts are made to survey a more broadly representative sample of non-students via telephone. With an additional study based on a larger sample size of both students and others, one could eventually justify a national study and conduct a more refined analysis including comparison of different demographic characteristics with different questions to determine if there are any relationships between variables.
Christopher F. Meindl is Assistant Professor of Interdisciplinary Social Sciences in the College of Arts & Sciences at the University of South Florida, St. Petersburg, Florida 33701-5016. Email: firstname.lastname@example.org. His research interests include wetlands, environmental perception and environmental policy.
My heartfelt thanks go to: Georgia College and State University's Faculty Development and Research Committee for providing funding for this pilot study; Linda Machlus for spending many hours in the Baldwin County courthouse selecting and recording the names and addresses of over 800 local property owners who were mailed surveys; and to Andrew Thomason and Joshua Crowe, who painstakingly prepared paper copies of surveys for postal distribution and who engaged in much data preparation. Without their help, this project would not have happened. [End Page 86]