A comparison of the reliability of the take-it-or-leave-it and the bidding game approaches to estimating willingness-to-pay in a rural population in West Africa.

The test-retest reliability of the bidding game and the take-it-or-leave-it (TIOLI) approaches to eliciting willingness-to-pay (WTP) are compared. A random sample of households in the Nouna area of Burkina Faso were interviewed twice with an interval of around 4-5 weeks. One thousand one hundred and eight individuals were asked their individual WTP for community-based health insurance. Three hundred and forty eight of these individuals were household heads who were in addition asked about their WTP for health insurance for the whole household. Median and the mean WTP were higher in the test than in the retest. Despite these differences both methods displayed moderate to good reliability (kappa values ranged from 0.467 to 0.621, Spearman correlations ranged from 0.653 to 0.701 and Pearson correlations ranged from 0.593 to 0.675). There was some evidence that the bidding game was more reliable than the TIOLI method. This study is based on larger sample size than previous studies and also is one of the first studies of the reliability of WTP in a developing country.