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

Web Survey Bibliography

Title Using preferred, understood or effective scales? How scale presentations effect online survey data collection
Author Derham, P.
Source Market & Social Research, 19, 2, p. 13
Year 2011
Access date 14.07.2015

This paper reviews the impact of different scale presentations in online surveys on closed ended question completion levels, respondents' preferences for scale presentation types and the influence those preferences have on questions answered and on respondents' future survey intentions. The scale presentations assessed were word scales, number scales, and emoticon scales. The rationale for the project was that marketing research seeks to obtain full and accurate answers from people invited to complete online self-completion surveys, and that survey design influences response and data quality. Self-completion surveys often use Likert scales (expressed in words, numbers or images) to collect data about attitude, intention or feeling. Online surveys can use emoticons (moving imagery) to present scale answers in place of word or number tick-a-box scale presentation. As emoticons have been seen, intuitively, as attractive and useful variations to long lists of word or number scale presentations, it was hypothesised that the use of emoticons in place of word or number scales would strengthen online survey data collection. Safe research practice requires confirmation of intuition, so five online surveys, each showing different presentations of scale formats, were undertaken in 2009 and in 2010. The findings indicate that the most effective scale presentation was number scale presentations if more answers in the questions are sought but that number scales were least preferred and may be better avoided if the survey design aims include being attractive and interesting to respondents. More respondents reported that they preferred word format scale presentations than emoticon or number scale presentations and respondents who completed word scale presentation format surveys were more likely to say they would complete similar such surveys in the future than those who completed number scale presentation format surveys. The least effective of the scale presentations tested were emoticons (images and slider scales) and their use is not recommended.

Year of publication2011
Bibliographic typeMagazine article

Web survey bibliography (8390)