Speeding in Web Surveys: The tendency to answer very fast and its association with straightlining
Web surveys can be programmed to capture a variety of paradata regarding how respondents answer questions. These paradata provide great opportunities for researchers to assess response quality, specifically whether respondents engage in satisficing – not spending enough effort to provide accurate re...
Saved in:
Published in | Survey research methods Vol. 8; no. 2 |
---|---|
Main Authors | , |
Format | Journal Article |
Language | English |
Published |
European Survey Research Association
01.01.2014
|
Subjects | |
Online Access | Get full text |
Cover
Loading…
Summary: | Web surveys can be programmed to capture a variety of paradata regarding how respondents answer questions. These paradata provide great opportunities for researchers to assess response quality, specifically whether respondents engage in satisficing – not spending enough effort to provide accurate responses. In particular, speeding (i.e., giving answers very quickly) has increasingly been used as an indicator for satisficing and low response quality. However, few studies have examined whether speeding actually leads to compromised response quality. To address this gap in the literature, the current study investigates speeding behaviors among Web respondents from a probability-based panel. We first identify and characterize respondents who speed more frequently than others over the entire questionnaire. To explore the impact of speeding on response quality, we then examine whether respondents who speed more frequently also straightline in more grid questions. The results show that the tendency to speed is related to several respondent characteristics, particularly age (younger respondents are more likely to speed). This study also reveals that more speeding seems to be universally related to more straightlining, and this relationship is particularly strong among the less educated respondents. |
---|---|
ISSN: | 1864-3361 |
DOI: | 10.18148/srm/2014.v8i2.5453 |