Web Survey Bibliography

Title A online-offline method comparison based on quasi-experimental data from two surveys to family stress
Author Haenggi, Y., Heldner, C.
Source General Online Research Conference (GOR) 2006
Year 2006
Access date 20.09.2006
Abstract In recent years, many studies have been undertaken to detect whether methodology matters or not. Do online surveys lead to same conclusions as traditional paper-pencil surveys do? Many authors underline the comparability of both methods. There measurements of different scales and different data acquisitions methods resulted in comparable values. Other researches claim that there exits differences between methods. For example, data collected by means of internet seam to have wider distribution. Also should the major anonymity given in online studies lead to more self-exposure. Further differences are to be expected, if study design is truly experimental or just quasi-experimental. Experimental studies, in which participants are randomly grouped to conditions, online and offline groups shouldn’t differ in there demographic characteristic. In quasi-experimental studies, however, comparability of group characteristics is often not given. Differences found between groups can be due to differences in such variables. If we are faced to an artificial effect, differences should disappear as soon as demographic variables are balanced out. If not, we found a method effect. In our one study participants were asked about family stress and their coping behaviour. The aim was to determine true methodological effects against artificial effects as a result of differences between groups. We collected online (N=376) and paper-pencil (N=150) data. Data acquisitions of both methods were arranged independently, as if two individual studies were conduced. This procedure guaranteed that self-selection effects could be established, as to be expected. The artificial effects, therefore, could occur. By balancing out demographic differences the self-selection effect were cancelled out. Methodological effects could then be detected. Result show, that there existed both types of effects. Kind of story reported was affected by sample characteristics. But participants in the online condition reported stories with higher personal impact. We conclude that there are differences between methodologies. Anonymity, as one variable, should matter. But there are also many other variables witch aren’t affected, as long as samples are comparable.
Access/Direct link Conference homepage (abstract)
Year of publication2006
Bibliographic typeConferences, workshops, tutorials, presentations
Full text availabilityNon-existant
Print

Web Survey Bibliography - General Online Research Conference (GOR) 2006 (29)