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

Web Survey Bibliography

Title Grids and Online Surveys: Do More Complex Grids Induce Survey Satisficing? Evidence from the Gallup Panel
Year 2016
Access date 09.06.2016
High satisficing behaviors and item non-response can be greatly detrimental to survey quality; therefore online surveys must be designed with great care in order to minimize satisficing, increase responses and in turn to maximize quality. One derivative design is grid-style questions. However, this question type represents a double-edged sword to online survey market. On the one hand, the benefits that grid questions bring to the survey industry is to reduce time-consuming formats and shorten the length of questionnaire. On the other hand, there are several challenges that grid formats pose to measurement quality, including higher item non-response and dropout rates as well as an increase in measurement error. Operationally, grid questions can have very different levels of complexity which presents many challenges when using these types of questions in surveys. This research focuses on an experiment designed to explore data quality difference when responding to grid questions depending on: 1) the objective complexity; 2) the position of the grid; and 3) the interaction between grid's position and complexity. This study will allow us to evaluate the paradata, in combination with substantive data, from a fully randomized experiment conducted on the web component of the Gallup Panel. To increase the power of this analysis from previous studies, each respondent was assigned a questionnaire including a fully randomized ordering of three grid questions of varying complexity between the beginning, middle and end of the survey. Evidence from preliminary analyses indicates that grids with higher complexity are more likely to induce data quality problems, which demonstrated prominent measurement error among panel respondents. Moreover, as the survey progresses, the effects of grid complexity on data quality carried over to later questions, indicating respondents may be fatigued and thus become morelikely to be satisficing on more complex grids at the end of the survey.
Year of publication2016
Bibliographic typeConferences, workshops, tutorials, presentations

Web survey bibliography (1211)