Grantee Research Project Results
2007 Progress Report: Meta-Regression Analysis of Recreation Valuation and Demand Elasticities: Identifying and Correcting Publication Selection Bias to Improve Benefit Transfer
EPA Grant Number: R832421Title: Meta-Regression Analysis of Recreation Valuation and Demand Elasticities: Identifying and Correcting Publication Selection Bias to Improve Benefit Transfer
Investigators: Rosenberger, Randall S. , Stanley, Tom D.
Institution: Oregon State University
EPA Project Officer: Hahn, Intaek
Project Period: October 1, 2005 through September 30, 2008
Project Period Covered by this Report: October 1, 2006 through September 30, 2007
Project Amount: $409,947
RFA: Valuation for Environmental Policy (2004) RFA Text | Recipients Lists
Research Category: Environmental Justice
Objective:
In order to improve the accuracy of meta-analysis for benefit transfer, this research develops meta-regression analysis models for identifying, measuring and correcting publication selection bias in the recreation valuation literature; in particular, estimates of economic value and demand elasticities. These models are used to test several hypotheses across a variety of publication bias sources and to correct for publication bias. Monte Carlo simulations are used to evaluate how alternative corrective methods perform under realistic research conditions.
The three objectives of this research are to (1) test for publication bias by publication type (e.g., journal vs. report, or publications that make methodological contributions vs. those that provide new estimates of values); (2) test for publication bias of recreation demand elasticities, including (i) on the distribution of estimate elasticities, (ii) on the generalizability of elasticity estimates, (iii) to measure the effect of moderator variables, and (iv) provide publication bias corrected elasticity measures; and (3) evaluate and improve methods for publication bias detection and correction.
Progress Summary:
Data collection is complete; the recreation use values database consists of 329 documents that jointly provide 2,705 estimates of recreation use values ranging from studies published from 1958 to 2006. Initial analyses of publication type (thesis, dissertation, journal article, government report, etc.) using binary and meta-regression methods indicate that publication types have significant statistical differences in mean recreation values from each other. Initial analyses on the effect of publication motivations (introducing efficient estimators, testing for biases, and introducing new estimates of value) using binary and meta-regression methods show that studies whose primary contribution is introducing efficient estimators are statistically different from studies whose primary contribution is introducing new estimates of value. These preliminary results have important implications for benefit transfers by showing that the type of publication and its primary contribution systematically affect empirical estimates of values reported by these different documents. Preliminary results from publication bias detection methods have revealed the clear and strong signal of publication selection bias. This is true for both the data as a whole and when disaggregated by specific recreational activities. Correcting for publication selection, based on these preliminary results, would reduce estimates by one-half or more, although the distributional properties of recreation use values are likely contributing to this result.
Future Activities:
Additional analyses will be conducted, including testing for publication selection bias effects using more rigorous meta-regression based methods in recreation values and price elasticities. Simulations of publication selection models mirroring the distributional properties of our data will be conducted to evaluate the efficiency of the models. The models will then be calibrated based on the simulation results and applied to the recreation use value metadata.
Journal Articles on this Report : 2 Displayed | Download in RIS Format
Other project views: | All 11 publications | 2 publications in selected types | All 2 journal articles |
---|
Type | Citation | ||
---|---|---|---|
|
Rosenberger RS, Stanley TD. Measurement, generalization, and publication: sources of error in benefit transfers and their management. Ecological Economics 2006;60(2):372-378. |
R832421 (2007) |
Exit Exit Exit |
|
Stanley TD. Meta-regression methods for detecting and estimating empirical effects in the presence of publication selection. Oxford Bulletin of Economics and Statistics 2008;70(1):103-127. |
R832421 (2007) |
Exit |
Supplemental Keywords:
methodological improvements, extent of the market, data collection methods, efficient bid designs, demand function, benefit function, empirical evidence, applied research, research synthesis, public policy, decision making, non-market valuation, contingent valuation, travel cost method, preferences, public good, socio-economic, willingness-to-pay, environmental assets, social science, modeling, measurement methods, econometric analysis, Economic, Social, & Behavioral Science Research Program, RFA, Scientific Discipline, Ecology and Ecosystems, Economics & Decision Making, Social Science, decision-making, Monte Carlo study, benefits transfer, bias, decision analysis, decision making, demand elasticities, econometric analysis, ecosystem valuation, environmental decision making, meta-regression analysis models, public policy, public values, recreational value,, RFA, Economic, Social, & Behavioral Science Research Program, Scientific Discipline, decision-making, Ecology and Ecosystems, Economics & Decision Making, Social Science, benefits transfer, ecosystem valuation, valuation, demand elasticities, decision analysis, decision making, environmental decision making, recreational value, bias, meta-regression analysis models, Monte Carlo study, public values, public policyRelevant Websites:
Recreation Use Valuation Literature Search http://www.cof.orst.edu/cof/fr/research/ruvd/Recreation_welcome_page.html Exit
Recreation Use Values Database http://recvaluation.forestry.oregonstate.edu/ Exit
Progress 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.