An examination of sample length and reliability of the Interactional Network Tool, a new measure of group interactions in acquired brain injury
An examination of sample length and reliability of the Interactional Network Tool, a new measure of group interactions in acquired brain injury
复制标题
DOI：
10.1080/02687038.2022.2118517
复制
发表时间：
2022-09
期刊：
影响因子：
2
通讯作者：
S. Howell;S. Beeke;Emma Louise Sinnott;R. Varley;T. Pring
中科院分区：
文献类型：
--
作者：
S. Howell;S. Beeke;Emma Louise Sinnott;R. Varley;T. Pring
摘要背景会话是一种具有挑战性的测量。定量和定性措施需要对对话背景、目的和参与者的可变贡献敏感，以捕捉有意义的变化。独立评分员之间的测量也需要一致。之前已经研究过不同样本长度下全球观察性评级量表的可靠性。使用行为频率测度研究样本长度对评分者间信度的影响是一个新的研究领域。目的本研究报告的评分者间的可靠性的InterestingNetwork工具（INT），一个行为编码系统，用于与组的互动数据。它检查了样本长度对可靠性的影响，使用了一种改进的编码系统，旨在提高临床环境中使用的速度和效率。方法对14个获得性脑损伤患者群体互动的视频样本进行分析。两名评分员使用INT编码系统独立编码电影。使用类内相关性（ICC）计算单个代码的可靠性。代码被组合以形成新的编码结构。使用四个样本长度（5、10、15和20分钟）的类内相关性计算新代码的可靠性。采用单因素方差分析比较4种样本长度的平均值。结果和结果使用改进的INT编码系统实现了可接受的评分者间信度。四种样本长度之间无差异。结论这些研究结果表明，训练有素的临床医生在临床实践中使用INT可以实现一个可靠的措施，参与一个小组的互动，从短样本。现在需要与其他临床组进行验证。
ABSTRACT Background Conversation is challenging to measure. Quantitative and qualitative measures need to be sensitive to the conversation context, the purpose and the variable contributions of participants in order to capture meaningful change. Measurements also need to be consistent across independent raters. The reliability of global observational rating scales across differing sample lengths has previously been investigated. An investigation into the effects of sample length on inter-rater reliability using a behavioural frequency measure is a new field of research. Aims This study reports on the inter-rater reliability of the Interactional Network Tool (INT), a behavioural coding system for use with group interaction data. It examines the effects of sample length on reliability using a refined coding system designed to improve the speed and efficiency of use in clinical settings. Methods Fourteen video samples of group interactions for people with acquired brain injury were prepared for analysis. Two raters independently coded the films using the INT coding system. Individual code reliability was calculated using intra-class correlations (ICCs). Codes were combined to form a new coding structure. Reliability of the new codes was calculated using intra-class correlations across four sample lengths (5,10,15 and 20 minutes). A one-way analysis of variance was used to compare the means of the four sample lengths. Outcomes and Results Acceptable inter-rater reliability was achieved using the refined INT coding system. There was no difference between the four sample lengths. Conclusions These findings indicate that trained clinicians using the INT in clinical practice can achieve a reliable measure of participation in a group interaction from short samples. Validation with other clinical groups is now indicated.