Telephone and Web Survey: a Study of the Marginal Mode Effect

Telephone surveys have a long history in social and economic researches and their correlated problems in terms of survey quality have been studied from different points of view in many analyses. On the other hand, in the last years web surveys had a remarkable development due to their advantages in terms of speed in comparison with mail surveys and cost in comparison with telephone or face-to-face interviews. In order to validate the use of Internet it becomes very important to measure the marginal mode effect of the web survey above all during the process of data collection. To this aim, a panel survey is necessary in order to assume that all other attributes of the design are the same as in the survey to which the web survey is compared. In fact, with a panel survey the respondents can be compared and the differences, most likely, can be attributed to the communication mode itself, even though other possible reasons for differences have to be opportunely evaluated. Essentially, this paper aims to compare telephone versus web survey using data on the transition from university-to-work of a panel of graduates in Economics at the University of Naples “Parthenope”. They are interviewed in two times with different modes: by telephone and by web questionnaire.