Tailored motivation of business respondents (Improving data collection by soft computing)

(A) Presenter

Miroslav Hudec, University of Economics in Bratislava

(B) Category

Early-stage idea

(C) Partners needed

Partners working in fields of mathematics and especially fuzzy logic, statistical methodology, motivation of respondents, statistical data analysis, ICT (especially graph databases and web applications) are especially welcome.


Data collections rely on businesses and their willingness to respond timely and accurately. However, businesses are increasingly reluctant to cooperate in surveys. Recent studies revealed that as much as 30 % of the total survey cost is spent on data editing (imputation) and for businesses it concerns more irritating than time consuming tasks (0.5% of total administrative burden). Currently statistical institutes are working on reusing administrative and other data sources for statistical purposes. However, not all surveys could be replaced with these sources. On the other hand, data users (especially businesses) are becoming ever more demanding for timely and relevant data, but are less willing to provide their own data to statistical institutes. Therefore, data collection is influenced with data dissemination and vice versa.

Adaptive survey designs bring advantage to the field of designing surveys by classifying respondents into classes and adapt the survey and communication design to the businesses’ characteristics to enhance response through motivation. However, sharp classes could cause that similar respondents might end up in different classes. Fuzzy logic is based on flexible sets, logic and relations. Therefore, similar entities are always similarly treated by less complex models. Moreover, rich relations between respondents and designs could be managed by graph databases.                                                  
A new solution supported by fuzzy logic and graph databases could solve these problems, i.e. integrating fuzzy logic, ASDs and graph databases is able to create a robust, low cost and efficient way for managing respondents’ motivation. ASDs’ fuzzy logic and graph databases could create foundation for supporting further activities like tailored data dissemination and even data imputation.