DCchoice: a package for analyzing dichotomous choice contingent valuation data

Share:

Description

The package provides functions for analyzing single-, one-and-one-half-, and double-bounded dichotomous choice contingent valuation (CV) data

Details

This package provides functions for analyzing single-, one-and-one-half-, and double-bounded dichotomous choice contingent valuation (CV) data.

In the single-bounded dichotomous choice (SBDC) CV that was first proposed by Bishop and Heberlein (1979) respondents are requested to answer a question like the following:

If the environmental policy burdens you with USD X annually, would you agree or disagree to it?

This format, respondents only states "yes (I agree)" or "no (I disagree)," meaning whether their willingness to pay (WTP) is greater or lower than the bid (USD X) they are offered.

The double-bounded dichotomous choice (DBDC) CV was proposed by Hanemann (1985) and Carson (1985) to improve the efficiency of SBDC-CV. In the CV format, respondents are requested to answer the second (follow-up) question just after they answer the SBDC-CV style question (the first/initial question). An example of DBDC-CV questions is as follows (Xl < X < Xh):

First question
If the environmental policy burdens you with USD X annually, would you agree or disagree to it?

Second question for the respondents who agree to the policy in the first question
If the amount of tax is USD Xh, would you agree or disagree to it?

Second question for the respondents who disagree to the policy in the first question
If the amount of tax is USD Xl, would you agree or disagree to it?

In the DBDC-CV question, there are four possible response outcomes: (yes, yes); (yes, no); (no, yes); and (no, no). If the respondent i's answer is (yes, yes), the analyst can tell WTPi > Xh (WTPi is the WTP of the respondent i). Similarly, (yes, no) means X < WTPi < Xh, (no, yes) means Xl < WTPi < X, and (no, no) means WTPi <Xl.

One-and-one-half-bound dichotomous choice (OOHBDC) CV, which was developed by Cooper et al. (2002), is an intermediate CV format between SBDC-CV format and DBDC-CV format.

In the OOHBDC-CV survey, after answering the first SBDC-CV style question (the first stage), only respondents who satisfy certain conditions are requested to answer an additional SBDC-CV style question (the second stage). Details in the OOHBDC-CV survey are as follows:

Step 1) A list of bid ranges [BLj, BHj] (j = 1, 2, ..., J), where BLj < BHj, are decided: i.e., [BL1, BH1], [BL2, BH2], ..., and [BLJ, BHJ].

Step 2) One of the bid ranges is randomly presented to respondents (e.g., a bid range of [BL3, BH3] for j = 3).

Step 3) One of the two bids presented to the respondents is selected randomly (i.e., BL3 or BH3 in the case of step 2 example) and then the respondents are asked whether they would be willing to pay the amount of the bid selected (the first stage).

Step 4) The respondents are asked to answer the second stage if they satisfy either condition: (a) their answer in the first stage is "yes" when the lower bid is presented in the first stage, or (b) their answer in the first stage is "no" when the higher bid is presented in the first stage.

Therefore, there are six possible responses to the OOHBDC-CV survey: "no", "yes-no", and "yes-yes" when the lower bid is shown in the first stage; and "yes", "no-yes", and "no-no" when the higher bid is shown in the first stage. Refer to Cooper et al. (2002) for detailed explanation of OOHBDC-CV, including the example CV questions.

There are two ways of estimating WTP from the SBDC-, OOHBDC-, and DBDC-CV: parametric and nonparametric approaches. In this package, the functions sbchoice, oohbchoice, and dbchoice, which are based on the utility difference approach (Hanemann, 1984), are developed for the parametric approach to SBDC, OOHBDC, and DBDC data, respectively.

Confidence intervals for the estimates of WTPs are constructed by two methods. These are the Krinsky and Robb (1986, 1990)'s method and the bootstrap one. The former is implemented by krCI while the latter by bootCI.

Both of the methods rely on simulation techniques with different settings. Usually, a bootstrap method takes much longer time than the Krinsky and Robb's method does. It has been pointed out that each method has both advantages and disadvantages, see, for instance, the discussions in Hole (2007) and the references therein.

Functions for nonparametric approaches are also included in the package. kristrom (Kristrom, 1990) and turnbull.sb (Carson and Steinberg, 1990) are designed for analyses for SBDC data, whereas turnbull.oohb and turnbull.db (Carson and Hanemann, 2005) for OOHBDC and DBDC ones, respectively.

Note

DCchoice depends on interval (Fay and Shaw, 2010) package. It further depends on several packages, among which Icens (Gentleman and Vandal, 2011) package may not be installed by the GUI Package Installer accessible from the menu bar (Windows and Mac OS) because it is available only from Bioconductor.

To install DCchoice and other dependent packages simultaneously, use the install.packages function instead. See the Examples section for the code.

References

Aizaki H, Nakatani T, Sato K (2014). Stated Preference Methods Using R. CRC Press, Boca Raton FL.

Bishop RC, Heberlein TA (1979). “Measuring Values of Extra-Market Goods: Are Indirect Measures Biased?” American Journal of Agricultural Economics, 61(5), 926–930.

Carson RT (1985). “Three Essays on Contingent Valuation”. Dissertation, University of California Berkeley.

Carson RT, Hanemann WM (2005). “Contingent Valuation.” in KG M\"aler, JR Vincent (eds.), Handbook of Environmental Economics. Elsevier, New York.

Carson RT, Steinberg D (1990). “Experimental Design for Discrete Choice Voter Preference Surveys.” in 1989 Proceeding of the Survey Methodology Section of the American Statistical Association, 821–822.

Cooper JC, Hanemann M, Signorello G (2002). “One-and-one-half-bound dichotomous choice contingent valuation”, The Review of Economics and Statistics, 84, 742–750.

Croissant Y (2011). Ecdat: Data Sets for Econometrics, R package version 0.1-6.1, http://CRAN.R-project.org/package=Ecdat.

Fay MP, Shaw PA (2010). “Exact and Asymptotic Weighted Logrank Tests for Interval Censored Data: The interval R Package”, Journal of Statistical Software, 36(2), 1-34. http://www.jstatsoft.org/v36/i02/.

Gentleman R, Vandal A (2011). Icens: NPMLE for Censored and Truncated Data. R package version 1.24.0, http://CRAN.R-project.org/package=Icens.

Hanemann, WM (1984). “Welfare Evaluations in Contingent Valuation Experiments with Discrete Responses”, American Journal of Agricultural Economics, 66(2), 332–341.

Hanemann WM (1985). “Some Issues in Continuous- and Discrete-Response Contingent Valuation Studies.” Northeastern Journal of Agricultural Economics, 14, 5–13.

Hole AR (2007). “A Comparison of Approaches to Estimating Confidence Intervals for Willingness to Pay Measure.” Health Economics, 16, 827–840.

Krinsky I, Robb AL (1986). “On Approximating the Statistical Properties of Elasticities.” The Review of Economics and Statistics, 68, 715–719.

Krinsky I, Robb AL (1990). “On Approximating the Statistical Properties of Elasticities: A Correction.” The Review of Economics and Statistics, 72, 189–190.

Kristr\"om B (1990). “A Non-Parametric Approach to the Estimation of Welfare Measures in Discrete Response Valuation Studies.” Land Economics, 66(2), 135–139.

Examples

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
## Installation of DCchoice along with dependent packages is carried out 
## by the following lines of code:
## Not run: 
install.packages("DCchoice", 
  repos = c("@CRAN@", "http://www.bioconductor.org/packages/release/bioc"), 
  dep = TRUE)

## End(Not run)
## You may select a CRAN mirror a few times.
## For Mac and Unix/Linux users, add the option argument 'type="source"'. 

Want to suggest features or report bugs for rdrr.io? Use the GitHub issue tracker.