Modelling attribute non-attendance in choice experiments for rural landscape valuation

Scarpa, Raffaele; Gilbride, Timothy J.; Campbell, Danny; Hensher, David A.

doi:10.1093/erae/jbp012

Cited by 310 publications

(303 citation statements)

References 32 publications

Supporting

Mentioning

274

Contrasting

Unclassified

Order By: Relevance

“…Class 2, which accounts for 28.97 % of the sample has an insignificant parameter estimate for the money variable indicating that these respondents were ignoring, or non-attending, this Standard errors in parentheses * p < 0.10; ** p < 0.05; *** p < 0.01 attribute or may have negligible utility from this attribute (see for example Scarpa et al 2009;Lagarde 2013). Those respondents' whose households have received a negative arsenic test are more likely to be a member of Class 2 than those with a positive result.…”

Section: Latent Class Analysismentioning

confidence: 99%

“…This potential confounding may lead to a misinterpretation of the insignificant parameter estimate as non-attendance when it could be related to low or insignificant preferences. A commonly used method in the literature is to impose a parameter restriction to account for non-attendance of an attribute in the latent class model (see for instance Burton and Rigby 2009;Scarpa et al 2009;Campbell et al 2011;Lagarde 2013). To investigate this issue further we re-estimate the latent class models of Table 11 with the parameters for the money and work attributes constrained to be zero for the MAP and WAP models respectively.…”

Section: Latent Class Analysismentioning

confidence: 99%

See 1 more Smart Citation

Discrete Choice Experiments in Developing Countries: Willingness to Pay Versus Willingness to Work

Gibson

Rigby

Polya

et al. 2015

Environ Resource Econ

View full text Add to dashboard Cite

A concern when conducting stated preference valuation studies in rural developing or very low income contexts is the use of monetary willingness to pay (WTP) estimates. In circumstances where cash incomes are extremely low, a significant proportion of the population are not engaged in waged labour and the exchange of goods or services is augmented through barter or work exchange, the role of money is likely to be different from that within an urban developed setting. As such, ability to pay using money may be impaired and downwardly biased when compared with other mediums of exchange. In recognition of this several studies have used hypothetical labour contributions as payment vehicles and a common finding is that households are more often willing to contribute labour than they are money. In this paper we present the results of a split sample DCE using money and labour contributions as payment vehicles for improved drinking water quality in Kandal Province, Cambodia. We find little differences between the payment vehicles in terms of attribute non-attendance, marginal utilities of attributes or derived welfare values. We argue that this provides support for the use of WTP in rural developing areas where there are functioning labour markets.

show abstract

Section: Latent Class Analysismentioning

confidence: 99%

Section: Latent Class Analysismentioning

confidence: 99%

Discrete Choice Experiments in Developing Countries: Willingness to Pay Versus Willingness to Work

Gibson

Rigby

Polya

et al. 2015

Environ Resource Econ

View full text Add to dashboard Cite

show abstract

“…It is hypothesised that respondents who state that they have considered several or all attributes when stating their choices have taken more time for this task. Further, an equality-constrained latent class model is used to infer attribute attendance from choice data and test the influence of response time on class membership (Scarpa et al 2009(Scarpa et al , 2012. These analyses are intended to shed greater light on the question of whether response time is an appropriate proxy for cognitive effort in stated preference surveys.…”

Section: Research Questionsmentioning

confidence: 99%

“…The two econometric models employed in this study are the equality-constrained latent class (ECLC) (Scarpa et al 2009(Scarpa et al , 2012 and the generalised multinomial logit (GMNL) models (Fiebig et al 2010;Gu et al 2013). In all models, utility of respondent n from choosing option i in choice occasion t is assumed to consist of representative utility explained by the vector x nit consisting of alternative-and respondent-specific attributes and its coefficient vector β n and an unobserved error term ε nit according to…”

Section: Equality-constrained Latent Class (Eclc) and Generalised Mulmentioning

confidence: 99%

“…The vector z n contains individual-specific variables (such as response time) influencing class membership. What distinguishes the ECLC from a normal latent class model are the restrictions of the parameters in different classes (Scarpa et al 2009(Scarpa et al , 2012. These are threefold: (i) if an attribute is not attended to its coefficient is zero as this attribute does not affect choice probabilities; (ii) if not zero, estimated parameters are equal across classes except for (iii) the alternative-specific constant indicating the change options which is not constrained to be equal across classes (Glenk et al 2014).…”

Section: Equality-constrained Latent Class (Eclc) and Generalised Mulmentioning

confidence: 99%

See 1 more Smart Citation

Are Fast Responses More Random? Testing the Effect of Response Time on Scale in an Online Choice Experiment

Bӧrger

2015

Environ Resource Econ

View full text Add to dashboard Cite

Scepticism over stated preference surveys conducted online revolves around the concerns over "professional respondents" who might rush through the questionnaire without sufficiently considering the information provided. To gain insight on the validity of this phenomenon and test the effect of response time on choice randomness, this study makes use of a recently conducted choice experiment survey on ecological and amenity effects of an offshore windfarm in the UK. The positive relationship between self-rated and inferred attribute attendance and response time is taken as evidence for a link between response time and cognitive effort. Subsequently, the generalised multinomial logit model is employed to test the effect of response time on scale, which indicates the weight of the deterministic relative to the error component in the random utility model. Results show that longer response time increases scale, i.e. decreases choice randomness. This positive scale effect of response time is further found to be non-linear and wear off at some point beyond which extreme response time decreases scale. While response time does not systematically affect welfare estimates, higher response time increases the precision of such estimates. These effects persist when self-reported choice certainty is controlled for. Implications of the results for online stated preference surveys and further research are discussed.

show abstract

Informing water harvesting technology contract design using choice experiments

Tarfasa

Brouwer

Sheremet

et al. 2017

Water Resources Research

View full text Add to dashboard Cite

Introducing water harvesting technology is expected to be more effective and last longer if farm households are involved in their design. The main objective of this study is to inform policymakers in Ethiopia about the most important terms and conditions to incentivize farmers to enter into a contractual agreement to invest in water harvesting on their land. In order to test the influence of the way the specific contractual terms and conditions are communicated to farm households, many of whom are illiterate, a split sample approach is applied with and without visual aids for technical, institutional, and economic contract characteristics. Both samples generate significantly different results, highlighting the importance of how information is conveyed to farm households. This pattern is confirmed when examining the self‐reported importance attached to the various contract characteristics. Equality Constrained Latent Class models show that contract characteristics for which visual aids were developed are considered more attentively, emphasizing the importance of adequate communication tools in a developing country context where literacy rates are limited to increase water technology innovation uptake and reduce farm household vulnerability to droughts.

show abstract

Modelling attribute non-attendance in choice experiments for rural landscape valuation

Cited by 310 publications

References 32 publications

Discrete Choice Experiments in Developing Countries: Willingness to Pay Versus Willingness to Work

Discrete Choice Experiments in Developing Countries: Willingness to Pay Versus Willingness to Work

Are Fast Responses More Random? Testing the Effect of Response Time on Scale in an Online Choice Experiment

Informing water harvesting technology contract design using choice experiments

Contact Info

Product

Resources

About