Investigating the Viewpoints of Local Communities in Conservation and Reclamation of Urmia Lake: Application of Bayesian Approach for Analyzing Choice Test Data

مجله اقتصاد و توسعه کشاورزی

Article 3, Volume 31, Issue 4 - Serial Number 38, January 2018, Pages 332-343   

Document Type: Research Article

DOI: 10.22067/jead2.v31i4.66148

Authors

M. Salehnia^{* 1}; B. Hayati¹; M. Molaei²

¹university of Tabriz

²university of Urmia

Abstract

Introduction: The degradation and destruction of natural resources is being considered as an economic issue; because when these resources are destroyed or lost, significant values are destroyed because some of which are irreversible. The major difference between the science of economy and other subjects such as ecology on the definition of the “value” is the emphasis of economy on the preferences. Differing sensitivities are the basis for targeted communication programs and promotions. As consumer preferences and sensitivities become more diverse, it becomes less and less efficient to consider the society at the aggregate level. In this research, we will show how hierarchical Bayesian approach is ideal for these problems as it is possible to produce individual -level parameter estimates. Urmia Lake in the northwestern corner of Iran is one of the largest permanent hyper saline lakes in the world and the largest lake in the Middle East. The lake’s surface area has been estimated to be as large as 5585 km2. However, since 1995 it has declined and was estimated to be only 926 km2 in 2014 based on satellite data. Considering no significant trend in the drought pattern, Urmia Lake's observed physiographic changes may be attributable to the overuse of renewable water resources and unbalanced development of agricultural sector. Therefore this research emphasizes the active role of local communities in the conservation and revitalization of Urmia Lake and analyzes the data from the choice experiment using hierarchical Bayes.
Materials and Methods: Choice-based conjoint (discrete choice) measurement has attracted more attention over the last years. Many researchers assert that choice-based tasks are more realistic for respondents than ratings- or rankings-based conjoint questions. However, choice-based conjoint data does not contain as much information per unit of respondent effort as traditional conjoint analysis. There are different ways to analyze the choice data. Hierarchical Bayes is the newest estimation method. The mathematical specification of these model is a Bayesian hierarchical model in which, broadly speaking, a different vector of utility is defined for each respondent. The distribution of these utilities in the whole population has some specified forms, usually normal. Hierarchical Bayes allows for heterogeneity at a respondent's level by specifying different utilities for each respondent. This leads to a greater improvement in simulation techniques: simulation conducted using aggregate or clustered models often lead to the biased results. Its ability to borrow information from other respondents to stabilize part worth estimation for each individual is particularly valuable for choice data. Applying HB to choice data allow analysts largely to solve IIA problems. Four attributes consist of animal habitat, climate regulation and prevention from salt storms, aesthetic and ecotourism, and education and research were considered in this study. The required data have been collected from 13 districts located in the northwest of Iran and Exogenous stratified random sampling applied as the sampling strategy.
Results and Discussion: Estimating willingness to diagnosed climate regulation and prevention from salt storms as the most important attribute from the view of the respondents. Animal habitat, aesthetic and ecotourism, and education and research were in the next places of people’s willingness to pay priorities. Hence, from the public’s point of view, mentioned attributes in the same order, should have most importance and priority in the management scenarios. Individual-level parameters of the Bayes model showed the highest variance for the full restoration of the climate, which implies the existence of conflicting preferences in this attribute. This indicates although some variables are important, they also fluctuate in a wide range of variations and the probability of their selection is different among people. Certainly, hierarchical bayes provides information far beyond the average utility and applying this information will give experts a better understanding of the distribution of preferences. Another important subject to know is that even with four sets of choices in each questionnaire and the need for people to respond to all of them, there is still some uncertainty about the part worths of the individual level.
Conclusions: An important point about model estimation is the diminutive presence of individual explanatory variables. The Bayesian model is recommended to be based on just those respondents’ features that are directly related to the part worths and preferences in choosing goods. It is also recommended that, in order to attract higher rate of contributions, variables with low variation, such as reviving the current status of climate and full restoration of aesthetic and ecotourism which are generally accepted, should be used.

Keywords

Choice Experiment; ierarchical bayes; Preferences; Urmia Lake

References

1- Acosta L.A., Eugenio E.A., and Enano N.H. 2014. Sustainability trade-offs in bioenergy development in the Philippines: An application of conjoint analysis. Biomass and Bioenergy, 64: 20-41.

2- Allenby G., and Rossi P.E. 2003. Perspectives based on 10 years of HB in marketing research. Sawtooth Software Conference Proceedings, Sequim.

3- Allenby G.M., Rossi P.E., and McCulloch R.E. 2005. Hierarchical Bayes Models: A Practitioners Guide.

4- Borghi C. 2009. Discrete Choice Models for Marketing, New Methodologies for Optional Features and Bundles. Ms.c thesis. Mathematisch Instituut, Universiteit Leiden.

5- Chapman C., and Feit E.M. 2015. R for Marketing Research and Analytics. Springer International Publishing, Switzerland.

6- Edelenbosch C.N. 2014. Visualization of choice options on actual choice. Ms.c thesis. Department of Econometrics, Erasmus University, Rotterdam.

7- Eisen-Hecht J., Kramer R., and Huber J. 2004. A hierarchical bayes approach to modeling choice data: a study of wetland restoration programs. American Agricultural Economics Association Annual Meeting, Denver, Colorado, July 1-4.

8- Food and Agriculture Organization of the United Nations (FAO). 2003. Fisheries Management 2. The ecosystem approach to fisheries. FAO technical guidelines for fisheries, Rome.

9- Hansson H., and Lagerkvist C.J. 2016. Dairy farmers’ use and non-use values in animal welfare: Determining the empirical content and structure with anchored best-worst scaling. Journal of Dairy Science, 99 (1): 579-592.

10- Hayati B., Salehnia M., and Molaei M. 2017. Dealing with Heterogeneous Preferences Concerned with Lake Urmia Restoration Using Multilevel Latent Class Model. Journal of Agricultural Economics and Development, 30(4): 285-296. (in Persian with English abstract)

11- Kuhfeld W. F. 2010. Marketing research methods in SAS. SAS institute Inc. Cary, NC, USA.

12- Lagerkvist C.J., Okello J., and Karanja N. 2010. Anchored vs. relative best–worst scaling and latent class vs. hierarchical Bayesian analysis of best–worst choice data: Investigating the importance of food quality attributes in a developing country. Food Quality and Preference, 25: 29-40.

13- Lipton D., Wellman K., Sheifer I.C., and Weiher R. F. 1995. Economic valuation of natural resources, a handbook for coastal resource policymakers. NOAA Coastal Ocean Program.

14- Maliki Esfanjani M. 2012. Estimating Conservation Value of Lake Urmia from the Perspective of the People of Urmia and Tabriz Cities. Ms.c thesis, Faculty of Agriculture, University of Tabriz. (in Persian with English abstract)

15- Nordh H., Alalouch C., and Hartig T. 2011. Assessing restorative components of small urban parks using conjoint methodology. Urban Forestry and Urban Greening, 10: 95-103.

16- Orme B. 2000. Hierarchical Bayes: Why All the Attention? Research Paper Series, Sawtooth Software, Sequim.

17- Orme B. 2009. The CBC/HB system for hierarchical bayes. Technical Paper Series, Sawtooth Software, Sequim.

18- Rose J.M., and Bliemer M. C.J. 2013. Sample size requirements for stated choice experiments. Transportation, 40: 1021-1041.

19- Rossi P.E., Allenby G.M., and McCulloch R.E. 2005. Bayesian Statistics and Marketing. New York, Wiley.

20- Salehnia M. 2017. Investigating Households Heterogeneous Preferences Concerned with Lake Urmia Restoration Using Multilevel Latent Class and Hierarchical Bayes Models. Ph.D thesis, Faculty of Agriculture, University of Tabriz. (in Persian with English abstract)

21- Salehnia M., Hayati B., Ghahremanzadeh M., and Molaei M. 2014. Estimating the value of improvement in Lake Urmia’senvironmental situation: An application of choice experiment. Journal of Agricultural Economics and Development, 27(4): 267-276. (in Persian with English abstract)

22- Tabi A., Hille S.L., and Wustenhagen R. 2014. What makes people seal the green power deal? Customer segmentation based on choice experiment in Germany. Ecological Economics, 107: 206-215.

23- Urmia Lake Recovery Headquarters. 2015. Urmia Lake, Causes of Drought and Possible Threats. (in Persian)

24- Walter S., Ulli-Beer S., and Wokaun A. 2012. Assessing customer preferences for hydrogen-powered street sweepers: a choice experiment. International Journal of Hydrogen Energy, 37: 12003-12014.

25- West Azerbaijan Department of Environment. 2014. Lake Urmia in the past and present, consequences of the crisis. Research studies of the Lake Urmia, University of Urmia. (in Persian)