Grantee Research Project Results
2004 Progress Report: The Stability of Values for Ecosystem Services: Tools for Evaluating the Potential for Benefits Transfers
EPA Grant Number: R831597Title: The Stability of Values for Ecosystem Services: Tools for Evaluating the Potential for Benefits Transfers
Investigators: Hoehn, John P. , Lupi, Frank , Kaplowitz, Michael D.
Institution: Michigan State University
EPA Project Officer: Hahn, Intaek
Project Period: May 1, 2004 through April 30, 2007
Project Period Covered by this Report: May 1, 2004 through April 30, 2005
Project Amount: $235,772
RFA: Valuation for Environmental Policy (2003) RFA Text | Recipients Lists
Research Category: Environmental Justice
Objective:
The overall objective of this research project is to develop empirical procedures for understanding what people in different regions value in terms of wetland ecosystem services. These procedures are used to test hypotheses regarding the prospects for successful benefit transfers. The research builds on stated preference (SP) research conducted in Michigan that used qualitative research methods to develop a questionnaire for eliciting the values of ecosystem services. The present research tests if the SP questionnaire and values elicited in the Michigan research are transferable to other states and regions. Transferability is evaluated at three different levels: (1) whether respondents in different regions share the same perceptions of ecosystem services that the SP questionnaire is designed to elicit; (2) whether the SP questionnaire itself may be transferred to different regions; and (3) whether the values elicited with the SP questionnaire are similar and vary systematically with demographic and other predetermined variables.
The specific objectives of this research project are to: (1) use group interviews to determine if key wetland ecosystem service categories, as perceived and valued by respondents, vary across regions of the United States; (2) use cognitive individual interviews to evaluate the effectiveness of the SP instrument for benefit transfer of values associated with wetland ecosystems; (3) implement a Web-based stated choice experiment in selected regions to test transferability; (4) assess the effect of using regional, secondary data about resource scarcity, income, and demographics on the estimation of the benefits transfer equation; and (5) assess the influence of survey modes (e.g., mail, Web) on marginal attribute values and within-region benefits transfer.
Progress Summary:
The research develops tools and leverages existing studies to improve the accuracy and stability of benefit transfer across regions. The research combines qualitative and quantitative methods to identify pertinent ecological service bundles and to estimate transferable value functions. Qualitative tools are used to shed light on people’s preferences for wetland ecological services and help establish the utility-theoretic basis for benefit transfers. Quantitative tools are used to estimate service values as well as make calibrated adjustments for regional differences.
It is hypothesized that value transfer functions for ecosystem benefits may be specified in terms of general ecosystem service variables and variables derived from secondary data sources, such as the Census and federal natural resource inventories. Some candidate control variable categories include the spatial extent, quality, and scarcity of ecosystem services; characteristics of land cover and land use; and population demographics such as income, education, and age.
The core of the quantitative analysis consists of a Web-based stated preference questionnaire to comparable samples of respondents from selected states and regions. The choice data obtained from these sample sites will be used to quantitatively test the research hypotheses concerning benefit transfer functions (e.g., the form and completeness of the benefit transfer function). Furthermore, the survey data will allow for the testing of hypotheses concerning the extent of the relevant benefit transfer market and the use of control variables. Possible control variables will be developed based on readily available secondary data sources. Alternative specifications of the transfer function will also be tested (e.g., errors and biases).
Hypotheses
The project has seven research hypotheses (H) related to the five objectives (O). The hypotheses address technical and policy issues relevant for benefit transfer at a regional and national scale. These research objectives and hypotheses are:
O1. Determine whether key wetland ecosystem service categories, that are highly valued by respondents, vary across regions of the United States.
O2. Use cognitive individual interviews to evaluate the effectiveness of the SP instrument for benefit transfer of values associated with wetland ecosystems.
O3. Implement a Web-based stated choice experiment in selected regions to test transferability.
O4. Assess the effect of using regional, secondary data about resource scarcity, income, and demographics on estimation of the benefits transfer equation.
O5. Assess the influence of survey modes (e.g., mail, Web) on marginal attribute values and within-region benefits transfer.
Year 1 Progress
The research worked toward implementing the qualitative research based on group interviews and the qualitative pretesting of the stated preference questionnaire based on individual interviews. In addition, the research collected secondary data to support the selection of research sites and the eventual comparison of quantitative survey responses. Manuscripts were prepared relating to wetlands valuation and benefits transfer and presentations were made at a number of different professional meetings, including two U.S. Environmental Protection Agency workshops. Manuscripts and presentations are listed at the end of this report.
Qualitative Research Design
The transfer of ecosystem values across different populations and regions requires an understanding of the ecosystem services and attributes that influence those values. Populations that share similar perceptions of ecosystem attributes and face similar relative ecosystem scarcities are likely to hold similar values. Conversely, if perceptions and relative scarcities vary a great deal across populations, those populations are likely to hold different values and perhaps even different value functions.
The qualitative research was designed to evaluate the potential for transferring a stated preference questionnaire and values across different regional populations. Group interviews were included in the qualitative research to elicit data for comparing the ecosystem perceptions of respondents drawn from three different regions. The group interviews were intended to explore whether participants use the same language, perceive the same services, and understand wetland ecosystems in substantially similar ways. Qualitative pretesting of a stated preference questionnaire was included to detect differences in the ecosystem service perceptions of different populations relative to the stimulus of a standardized stated preference questionnaire. With different perceptions and values, respondents drawn from different populations might differ in their understanding of questionnaire narratives, in the plausibility of the ecosystem mitigation scenarios, and in their ability to make the requested mitigation choices.
Qualitative Research Treatments
Each group interview consists of a 1 and one-half to 2 hour discussion among 8 to 10 respondents. The discussion is guided by a professional moderator who follows a discussion guide. The discussion guide is structured to begin the discussion with general perceptions of the environment and then draw the discussion to specific perceptions about the specific ecosystem type being evaluated: wetland ecosystems. The successive stages of the discussion guide include questions to: (1) learn about the natural resources that were important to respondents; (2) explore respondents’ prior knowledge concerning wetlands; (3) gather information concerning respondents’ knowledge of different wetland types; (4) explore respondents’ knowledge of public policies relating to wetlands; (5) learn respondents’ opinions of the importance of certain wetland functions; (6) evaluate how respondents process given wetland definitions/pictures; and (7) examine respondents’ reactions to a particular wetland replacement scenario.
The individual pretests are structured in two stages. In the first stage, a respondent completes a stated preference questionnaire designed to elicit preferences for wetland ecosystem services. In the second stage, each respondent is debriefed about the questionnaire by a trained interviewer who follows a written debriefing questionnaire. The debriefing questionnaire begins with questions about the respondent’s overall impressions of the questionnaire and whether the respondent had any specific difficulties with different parts of the questionnaire. The questionnaire then focuses on each of the five wetland mitigation questions. The respondent is asked to think aloud about each pair of wetlands involved in the mitigation choice and about how the respondent decided to select one as the preferred wetland. The respondent is asked about whether and which wetland services were important to the respondent’s stated choice and which wetland services were unimportant to the choice. The respondents are also asked to describe in their own words various terms used in the questionnaire. The purpose of such descriptions is to determine whether the respondents’ understanding of key terms is consistent with the meaning intended by the researchers. The debriefing questionnaire ends with a small number of questions about specific parts of the questionnaire that are considered difficult to understand by the researchers. The interviewer then closes the interview with the respondent and completes a five– item, postinterview form about the respondent’s cooperation and apparent understanding.
Site Selection
The populations of Ohio, Missouri, and Georgia were selected as the three states for conducting the transferability tests for the Michigan wetland ecosystem questionnaire. Each of the three states have substantial areas of freshwater wetlands. Ohio and Missouri have no coastal, saltwater wetlands, whereas the portion of saltwater wetlands in Georgia is not likely to entirely overwhelm the general public’s interest in freshwater wetlands. Ohio and Missouri contain wetlands that are biogeochemically similar to those of Michigan, whereas Missouri also has wetlands that are biogeochemically distinct from those in Michigan. Georgia’s freshwater wetlands have features that differ in many details from those of Michigan, but it is not clear if these details would limit the transfer of the questionnaire or the elicited values. Taken together, the three states provide a range of conditions for testing transferability, from Ohio, with wetland features that are quite similar to those of Michigan, to Georgia, where wetlands may have services that are distinct from those of Michigan.
Sites for the qualitative research were selected from metropolitan areas in each of the three states: Dayton, Ohio; St. Louis, Missouri; and Atlanta, Georgia. Metropolitan sites were chosen to facilitate the selection of respondents with a representative cross-section of demographic characteristics and to take advantage of the interviewing facilities usually found in metropolitan areas.
Future Activities:
We will proceed as described in the project proposal work plan. The group and cognitive interviews will be completed. The existing Web-based questionnaire will be adapted to the benefit transfer survey. The adapted questionnaire will be set up on a working server, debugged, and tested. Negotiations will be carried out and completed with a subcontractor regarding e-mail access to an Internet panel of respondents. The Internet survey will be carried out and the response database created. The data will be reviewed and choice functions will be estimated. As data collection proceeds, theoretical analysis will be completed to refine the conceptual framework and develop specific hypotheses in terms of the estimated choice function parameters. Estimation results will be used to test the hypotheses. The researchers will continue to develop presentations and manuscripts regarding the research completed and the research in progress. These manuscripts will be submitted, as appropriate, for presentation at professional meetings and for publication in books and/or professional journals.
Journal Articles:
No journal articles submitted with this report: View all 4 publications for this projectSupplemental Keywords:
benefit transfer, ecosystem, wetlands, indicators, restoration, aquatic, habitat, integrated assessment, public policy, cost benefit, conjoint analysis, nonmarket valuation, survey, psychological, preferences, public good, economic, willingness-to-pay, compensation, conservation, environmental assets, social science, surveys, internet, web, midwest, Great Lakes, Michigan, Ohio, Georgia, southeast,, RFA, Scientific Discipline, Economic, Social, & Behavioral Science Research Program, Geographic Area, State, Economics, decision-making, Ecology and Ecosystems, Economics & Decision Making, Social Science, deliberative policy, policy analysis, surveys, decision analysis, web-based methods, environmental values, environmental policy, aquatic ecosystems, public values, public policy, wetlands preservation, stated preference, econometric analysis, economic tradeoffsProgress 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.