Article Figures & Data
Figures
- Figure 1
An Example Choice Situation Shown to Respondents in the Experimental Instructions
Note: Light gray corresponds to yellow in the instructions; medium gray corresponds to green in the instructions; dark gray corresponds to red in the instructions.
- Figure 2
An Example Choice Situation Shown to Respondents in the Experimental Instructions
Note: Light gray corresponds to yellow in the instructions; medium gray corresponds to green in the instructions; dark gray corresponds to red in the instructions.
Tables
- Table 1
Synthesis of Contributions to the Discrete Choice Experiment Literature by Discipline
Discipline Reliability Convergent Validity Criterion Validity Environmental economics Bliem, Getzner, and Rodiga-Laßnig 2012; Schaafsma et al. 2014; Czajkowski, Barczak, and Budziñski 2016; Rakotonarivo, Schaafsma, and Hockley 2016; Brouwer, Logar, and Sheremet 2017; Matthews, Scarpa, and Marsh 2017 Hanley et al. 1998; Boyle et al. 2001; Caparros, Oviedo, and Campos 2008; Christie, and Azevedo 2009 Bateman et al. 2009; Vossler, Doyon, and Rondeau 2012; Czajkowski et al. 2017; Howard et al. 2017; Jacquemet et al. 2017; Carson et al. 2020 Agri-food economics Mørkbak and Olsen 2015; Rigby, Burton, and Pluske 2016 Asioli et al. 2016; Yangui et al. 2019 Carlsson, Frykblom, and Lagerkvist 2005; de-Magistris, Gracia, and Nayga 2013; de-Magistris and Pascucci 2014; Bello and Abdulai 2016; Lewis, Grebitus, and Nayga 2017; Kemper, Popp, and Nayga 2017; Wuepper, Clemm, and Wree 2019; Fang et al. 2020 Health economics Bryan et al. 2000; Ryan et al. 2006; Skjoldborg, Lauridsen, and Junker 2009; Janssen et al. 2017; Price, Dupont, and Adamowicz 2017 Van der Pol et al. 2008; Ryan and Watson 2009 Özdemir, Johnson, and Hauber 2009 Energy economics Liebe, Meyerhoff, and Hartje 2012 McNair, Bennett, and Hensher 2011 Oehlmann and Meyerhoff 2017; Zawojska, Bartczak, and Czajkowski 2019 Transportation economics Börjesson 2014 Raffaelli et al. 2021 Fifer, Rose, and Greaves 2014; Haghani and Sarvi 2019; Mamkhezri et al. 2020 Cross-sector studies Johnston et al. 2017; Bishop and Boyle 2019; Mariel et al. 2021 Holmes, Adamowicz, and Carlsson 2017; Johnston et al. 2017; Bishop and Boyle 2019; Lloyd-Smith, Zawojska, and Adamowicz 2021; Mariel et al. 2021 List and Gallet 2001; Murphy et al. 2005; Johnston et al. 2017; Penn and Hu 2018; Bishop and Boyle 2019; Mariel et al. 2021 Note: This is far from an exhaustive list of works in the selected areas of research. The “Criterion Validity” column includes all the literature related to hypothetical bias and ex ante correction methods.
Discrete Choice Experiment (DCE) Treatment Choice Matching Treatment Step Description Step Description 1 Prequestionnaire: respondents answer questions about how hungry and how full they feel 1 Prequestionnaire: respondents answer questions about how hungry and how full they feel N/A 2 Practice: respondents take part in a practice preference task and belief task 2 Preference task: respondents make choices in the 12 choice situations of the DCE (direct questioning, individual-level choices elicited) 3 Preference task: respondents make choices in the 12 choice situations of the DCE (direct questioning, individual-level choices elicited) — N/A 4 Belief task: respondents make predictions about other respondents’ choice behavior (indirect questioning, population-level choices elicited) 3 Postquestionnaire: final questionnaire 5 Postquestionnaire: final questionnaire — N/A 6 Payoff disclosure: respondents are informed regarding their additional payoff — N/A 7 Payment respondents are given additional payment (if any) Validity Reliability Null Hypothesis (Ho) Interpretation Null Hypothesis interpretation ωOPT_OUT, MEAN, DCE≥ωOPT_OUT, MEAN, CMa Rejecting H0 means that mWTP for the opt-out alternative is higher in CMa than in the DCE; CMa reduces HB when compared with the DCE ωOPT_OUT, SD, DCE≤ωOPT_OUT, SD, CMa Rejecting H0 means that standard deviations of price and nonprice coefficients are lower in CMa than in the DCE; CMa provides less dispersed coefficient values than the DCE ωFAT_A, MEAN, DCE≤ωFAT_A, MEAN, CMa Rejecting H0 means that mWTP for the nonprice attributes are greater in the DCE than in CMa; CMa reduces HB when compared with the DCE ωFAT_A, SD, DCE≤ωFAT_A, SD, CMa ωFAT_G, MEAN, DCE≤ωFAT_G, MEAN, CMa ωFAT_G, SD, DCE≤ωFAT_G, SD, CMa ωSALT_A, MEAN, DCE≤ωSALT_A, MEAN, CMa ωSALT_A, SD, DCE≤ωSALT_A, SD, CMa ωSALT_G, MEAN, DCE≤ωSALT_G, MEAN, CMa ωSALT_G, SD, DCE≤ωSALT_G, SD, CMa ωPRICE, MEAN, DCE≥ωPRICE, MEAN, CMa Rejecting H0 means that the impact of price on respondents’ decisions is lower in the DCE than in CMa; CMa reduces HB when compared with the DCE ωPRICE, SD, DCE≤ωPRICE, SD, CMa τDCE≥τCMa Rejecting H0 means that error variance in CMa is lower than in the DCE; CMa provides more deterministic choices than the DCE Note: CMa, choice matching; DCE, discrete choice experiment; HB, hypothetical bias.
Dependent Variable Model 1 (CMa) Model 2 (DCE) Choice Choice Coefficient Coefficient ωOPT_OUT, MEAN 1 342*** 1.865*** (0.391) (0.477) ωFAT_A, MEAN 1.320*** 1.569*** (0.198) (0.353) ωFAT_G, MEAN 2.406*** 2.584*** (0.311) (0.464) ωSALT_A, MEAN 0.783*** 0.459* (0.188) (0.251) ωSALT_G, MEAN 1.571*** 1.633*** (0.188) (0.316) ωANT, MEAN 0.861*** 0.879*** (0.141) (0.179) ωMEAN −0.981*** −1.344** (0.135) (0.566) ωOPT_OUT, SD 2.297*** 4.698*** (0.384) (0.862) ωFAT_A, SD 0.352* 0.653*** (0.194) (0.224) ωFAT_G, SD 0.673** 1.970*** (0.257) (0.389) ωSALT_A, SD 0.881*** 1.602*** (0.223) (0.328) ωSALT_G, SD 0.490 2.345*** (0.345) (0.450) ωANT, SD 0.919*** 2.579*** (0.206) (0.504) αSD 0.800** 1.741 (0.398) (1.326) τ 0.611*** 0.993*** (0.160) (0.232) Subjects 66 64 Observations 2,376 2,304 Log-likelihood −648.447 −645.453 BIC 1,413.492 1,407.042 - Table 5
Comparison of Distributional Means and Standard Deviation of Estimated Coefficients across Treatments
Coefficients Choice Matching Discrete Choice Experiment H0 (Null Hypothesis) p-value ωOPT_OUT,MEAN 1.345
(0.677; 1.975)1.875
(1.037; 2.648)ωOPT_OUT,MEAN,DCE≥ωOPT_OUT,MEAN,CMa 0.802 ωFAT_A,MEAN 1.317
(1.004; 1.636)1.584
(1.002; 2.157)ωFAT_A,MEAN,DCE≤ωFAT_A,MEAN,CMa 0.252 ωFAT_G,MEAN 2.408
(1.897; 2.915)2.585
(1.795; 3.343)ωFAT_G,MEAN,DCE≤ωFAT_G,MEAN,CMa 0.376 ωSALT_A,MEAN 0.788
(0.479; 1.107)0.465
(0.048; 0.874)ωSALT_A,MEAN,DCE≤ωSALT_A,MEAN,CMa 0.845 ωSALTT_A,MEAN 1.573
(1.253; 1.880)1.645
(1.089; 2.170)ωSALTT_A,MEAN,DCE≤ωSALT_A,MEAN,CMa 0.419 ωANT,MEAN 0.866
(0.636; 1.093)0.879
(0.599; 1.185)ωANT,MEAN,DCE≤ωANT,MEAN,CMa 0.479 ωPRICE,MEAN −1.033
(−1.540; −0.674)−1.556
(−2.988; −0.704)ωPRICE,MEAN,DCE≤ωPRICE,MEAN,CMa 0.754 ωOPT_OUT,SD 2.300
(1.674; 2.934)4.722
(3.184; 6.119)ωOPT_OUT,SD,DCE≤ωOPT_OUT,SD,CMa 0.007 ωFAT_A,SD 0.358
(0.031; 0.686)0.662
(0.289; 1.033)ωFAT_A,SD,DCE≤ωFAT_A,SD,CMa 0.156 ωFT_G,SD 0.671
(0.221; 1.144)1.975
(1.305; 2.625)ωFT_G,SD,DCE≤ωFAT_G,SD,CMa 0.005 ωSALT_A,SD 0.886
(0.506; 1.240)1.613
(1.050; 2.154)ωSALT_A,SD,DCE≤ωSALT_A,SD,CMa 0.363 ωSALT_G,SD 0.469
(0.106; 1.045)2.355
(1.603; 3.121)ωSALT_G,SD,DCE≤ωSALT_G,SD,CMa 0.001 ωANT,SD 0.922
(0.554; 1.727)2.603
(1.691; 3.416)ωANT,SD,DCE≤ωANT,SD,CMa 0.002 ωPRICE,SD 0.847
(0.470; 1.439)1.840
(0.533; 4.756)ωPRICE,SD,DCE≤ωPRICE,SD,CMa 0.203 τ 0.721
(0.462; 0.974)1.010
(0.627; 1.417)τDCE≤τCMa 0.845 Note: The comparison uses the test in Poe, Giraud, and Loomis (2005); 5% and 95% percentiles are in parentheses.
In this issue
Jump to section
Related Articles
Cited By...
- No citing articles found.