Notice: the WebSM website has not been updated since the beginning of 2018.

Web Survey Bibliography

Title The quality of ego-centered social network data in web surveys: experiments with a visual elicitation method
Year 2014
Access date 11.06.2014
Abstract

Relevance & Research Question: Traditionally, the measurement of ego-centered social networks is done with the help of an interviewer who is available for assistance and who can motivate the respondent to continue with the answering procedure. The most often used method to collect ego-centered network data was proposed by Burt (1984): first a contact list with name generators, then name-interpreters, and finally the inter-alter response matrix. Nowadays, network data are often collected via web surveys using the same three kinds of questions. However, research has shown that this procedure leads to a reduction of data quality, probably because respondents tend to answer items mechanically (Matzat & Snijders 2010). We present the results of an online experiment that tests the usefulness of a visual way of eliciting participants’ responses to network items. We test whether the ‘quality of the data’ (see below) collected in this way improves, when compared to the standard procedure implemented in web surveys. We hypothesize that participants find the experience more enjoyable, leading to an improved data quality.
Methods & Data: The implemented web survey tool presents the name interpreter and the inter-alter-response matrix in such a way that the participants immediately observe their own social network emerging. We test our hypotheses on a student sample and a sample of members of a commercial opt-in internet panel (total n=725). The randomized experiment uses a between subject design with the visual elicitation method as treatment and the standard data collection method as control condition. We test hypotheses about effects of the visualization method on, among others, the drop-out rate, the number of missing values, and the tendency to answer questions mechanically, using multivariate linear and logistic regression analyses.
Results: An important finding is that respondents answered less often mechanically. At the same time, more respondents had difficulties in understanding how to answer, leading to more missing values. Drop-out was negligible under both conditions.
Added Value: The findings indicate that the tool should be simplified more for respondents. Nevertheless, they demonstrate that a visual way of eliciting participants’ answers about ego networks in web surveys is in principal useful.

Year of publication2014
Bibliographic typeConferences, workshops, tutorials, presentations
Print

Web survey bibliography - General Online Research Conference (GOR) 2014 (29)