Comparison of the Models in Choice Experiments Method Application for Watershed Afforestation in Southern Turkey

Abstract

Aim of study: In this study, the value estimates of the four benefits (preventing floods, reducing soil erosion, increasing dam life and access to quality spring water) created by erosion control activities are compared according to different models and the determination of these models.Area of study: The Çakıt Stream Watershed was selected as a study area.Material and methods: In this study, the data collected by Deniz (2012) to apply Choice Experiments Method were used. While Deniz (2012) had estimated the marginal values of the benefits based on Multinomial Logit Model, in this study, the estimations were separately made in order to compare the results of Deniz (2012) using both another Multinomial Logit Model with different variables and Mixed Multinomial Logit Model. Main results: Although the marginal values found by using Mixed Multinomial Logit Model are different from the estimations found with Multinomial Logit Model, the relationships among the variables did not change. Determination of Mixed Multinomial Logit Model is bigger (Pseudo-R2=0.548) than Multinomial Logit Models (0.251 and 0.071).Highlights: In the study, it was observed that as the models used to determine the value of the erosion control service changed, the magnitude of the estimated values also changed.