The impact of survey design on research outcomes: A case study of seven pilots measuring cash usage in the Netherlands

We employ a unique dataset of transaction records to analyse whether results of consumer surveys are influenced by the survey setup. We have asked more than 5,000 consumers to report their payments using seven different data collection methods. The results of the seven pilot studies were validated against actual payments data from retailers and the owner of the Dutch debit card scheme. The results of both the validation exercise and econometric analyses reveal that both the data collection mode and the length of the registration period significantly influence consumers' registration of payments. Measurement errors are minimised when consumers are asked to use a self-reported transaction diary for one single day.