Abstract
Variables were analyzed from 13 mail surveys addressing tourism issues, each of which collected data from three waves of respondents. A conceptual model was developed based on previous literature suggesting that differences in wave responses may be attributable to type of population (special interest groups or general populations) and/or to type of variable (demographic/profile or interest/behavioral), but the data did not support this contention. On 82% of the variables tested across the data sets, the additions of Waves 2 and 3 led to no change in the results, but a second wave typically reduced the variance on responses to the first wave by between 28% and 49%. It was recommended that instead of using resources to send Waves 2 and 3, those resources be allocated to intensive follow-up sampling of nonrespondents in a more representative way.
Get full access to this article
View all access options for this article.
