Automatic versus manual forwarding in web surveys

Several authors and software vendors advocate the benefits of auto forwarding in web surveys, but there is little empirical research on this approach. We experimentally tested automatic versus manual forwarding (MF) under different levels of cognitive effort. We manipulated information accessibility (IA; low vs. high) and consistency requirements (CRs; yes vs. no), along with auto forwarding (AF) versus MF in two studies conducted among students in Finland. We find that an AF survey takes less time to complete, but only for those completing a survey on personal computers or tablets; no time advantage is found for smartphone users. We also find that respondents in both AF and MF conditions return more often to items with higher cognitive burden (low IA or a CR). MF respondents change answers more often than AF respondents. AF appears to reduce straightlining slightly. We find no difference in response consistency between two behavioral items between AF and MF, but a slight advantage for AF for two attitude items. Finally, respondents reported more positive experiences with the AF version. Auto forwarding appears to be somewhat more efficient and easy to use but may decrease the quality of responses to cognitively demanding questions.