Addressing response-shift bias: Retrospective pretests in recreation research and evaluation
Addressing response-shift bias: Retrospective pretests in recreation research and evaluation
复制标题
DOI：
10.1080/00222216.2007.11950109
复制
发表时间：
2007-01-01
影响因子：
3.2
通讯作者：
Ward, Peter
中科院分区：
文献类型：
--
作者：
Sibthorp, Jim;Paisley, Karen;Ward, Peter
自我报告的前测/后测通常被用来评估娱乐研究和评估工作的变化。比较干预前后措施的可行性依赖于这样一个假设，即干预前后的衡量尺度是相同的。对于自我报告测量，该指标驻留在研究参与者中，因此可以直接受到干预的影响。如果参与者的自我认识水平因娱乐计划的结果而发生变化，那么这一指标也可能发生变化，使计划前后的测量结果之间的比较出现问题。本文旨在综合围绕这一问题的理论和文献，并提供一个基于数据的混合方法的例子，说明这个问题在娱乐背景下的问题，并假设报告的课前属性水平因报告时间不同而存在差异的可能原因。结果普遍支持将回溯性预测测试作为一种通过自我报告措施应对不断变化的指标的方式。本文进一步讨论了在娱乐研究和评估中何时以及如何使用回溯性预测测试是合适的。
The self-reported pretest/posttest has been commonly used to assess change in recreation research and evaluation efforts. The viability of comparing pre and post measures relies on the assumption that the scale of measurement, or metric, is the same before and after an intervention. With self-report measures, the metric resides within the study participants and, thus, can be directly affected by the intervention. If participants' levels of self-knowledge change as the result of a recreation program, then this metric may also shift, making comparisons between measures from before and after the program problematic. This article aims to both synthesize the theory and literature surrounding this problem and to offer a mixed-methods, data-based example, which illustrates the problem in a recreation context and posits possible reasons for differences in reported pre-course attribute levels by reporting time. Results generally supported rising a retrospective pretest as a way to address changing metrics with self-report measures. This article further discusses when and how it is appropriate to use retrospective pretests in recreation research and evaluation.