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

Web Survey Bibliography

Title Mode Effects on Subjective Well-being Research: Do they Affect Regression Coefficients?
Year 2016
Access date 04.06.2016
Abstract
Indicators of subjective well-being have become essential to measure prosperity and societal progress across the world, with much of the produced research being based on survey data at a time in which traditional survey modes have trouble maintaining data quality and costs. One of the most widespread alternatives in overcoming data collection challenges is to use various modes of data collection but as a consequence, there is great concern that data collected through different modes may not be comparable. Using data from a mode comparison experiment implemented in Switzerland, we compared telephone, mail and web modes and found both selection and measurement effects for some widely used variables such as happiness and job satisfaction. In this paper, we examine the impact that mode effects have on the statistical relationship between variables. To do this, we replicate OLS multiple regression analyses that are widely used by researchers to predict happiness, satisfaction at work and social trust for each mode of data collection, with and without controlling for selection effects. Theregressions are implemented for three different scenarios depending on whether mode effects were found a) on the dependent variable but not on independent variables; b) on independent variables, but not on the dependent variable; and c) on both dependent and independent variables. Results show differences in regression coefficients and the relationship between the variables, indicating that mode effects play an important role on the conclusions obtained in cross-sectional research into well-being.
 
 
Year of publication2016
Bibliographic typeConferences, workshops, tutorials, presentations
Print