Agreement between web-based and paper versions of a socio-demographic questionnaire in the NutriNet-Santé study

Web-based studies nowadays raise a major interest as they can improve all steps involved in observational studies. Our objective was to compare the web-based version of the NutriNet-Santé self-administered socio-demographic and economic questionnaire with the traditional paper version.

Both versions of the questionnaire were sent to 170 volunteers and filled in by 147 of them (either paper first, n = 76, or web-based first, n = 71). Agreement between versions was assessed by intraclass correlations (ICC) and kappas.

Agreement between both versions was high, with ICC and kappas ranging between 0.81-1.00 and 0.76-1.00, respectively, similarly across groups of administration order, age, gender and self-estimated web knowledge in general. The web-based version was the one preferred by 93.7% of the subjects and enabled to avoid 553 missing values (2.00% of the total entries), 24 inconsistent data (0.09%), 8 aberrant data (0.03%), 472 data entry errors (0.85%) and to save 2,800 (US $4,072) when sent to 170 subjects.

The web-based socio-demographic and economic questionnaire provided information of similar-to-superior quality compared to the traditional paper version, with substantial logistic and cost advantages.