Claremont Graduate University, USA.
Claremont Graduate University, USA.
Soc Sci Med. 2015 Oct;142:256-60. doi: 10.1016/j.socscimed.2015.08.020. Epub 2015 Aug 15.
Investigations conducted through Amazon's Mechanical Turk (MTurk) sometimes explicitly note eligibility requirements when recruiting participants; however, the impact of this practice on data integrity is relatively unexplored within the MTurk context.
Contextualized in the organ donor registration domain, the current study assessed whether overtly listing eligibility requirements impairs the accuracy of data collected on MTurk.
On day 1, the first and third round of data collection did not list eligibility requirements; the second and fourth round overtly listed a qualification requirement: status as a non-registered organ donor. On day 2, the approach was identical, except the order was reversed-the first and third round overtly listed the study qualifications, while the second and fourth did not. These procedures provided eight different waves of data. In addition, all participants were randomly assigned to read an elevating (i.e., morally inspiring) story or a story not intended to elicit any emotion. Regardless of recruitment approach, only participants who were not registered as donors were included in the analysis.
Results indicated that the recruitment script that explicitly requested non-registered donors resulted in the collection of participants with higher mean intentions scores than the script that did not overtly list the eligibility requirements. Further, even though the elevation induction increased intentions to register as a donor, there was not a significant interaction between recruitment approach and the influence of the elevation manipulation on registration intentions.
Explicitly listing eligibility requirements can influence the accuracy of estimates derived from data collected through MTurk.
通过亚马逊的 Mechanical Turk（MTurk）进行的调查在招募参与者时有时会明确注明资格要求；然而，在 MTurk 背景下，这种做法对数据完整性的影响相对尚未得到探索。
在器官捐献登记领域，本研究评估了明确列出资格要求是否会损害在 MTurk 上收集的数据的准确性。
在第 1 天，第一轮和第三轮数据收集没有列出资格要求；第二轮和第四轮数据收集则明确列出了一个资格要求：非注册器官捐献者的身份。在第 2 天，方法相同，只是顺序颠倒了——第一轮和第三轮数据收集明确列出了研究资格，而第二轮和第四轮数据收集则没有。这些程序提供了八波不同的数据。此外，所有参与者都被随机分配阅读一个提升（即道德上令人鼓舞）的故事或一个不打算引起任何情感的故事。无论采用哪种招募方式，只有未注册为捐赠者的参与者才被纳入分析。
结果表明，明确要求非注册捐赠者的招募脚本导致参与者的平均意图得分高于没有明确列出资格要求的脚本。此外，尽管提升诱导增加了注册为捐赠者的意愿，但招募方式和提升操作对注册意愿的影响之间没有显著的交互作用。
明确列出资格要求会影响从 MTurk 收集的数据得出的估计的准确性。
