In this study, the robustness of an online tool for objectively assessing singing ability was examined by: (1) determining the internal consistency and test-retest reliability of the tool; (2) comparing the task performance of web-based participants (n = 285) with a group (n = 52) completing the tool in a controlled laboratory setting, and then determining the convergent validity between settings, and (3) comparing participants’ task performance with previous research using similar singing tasks and populations. Results indicated that the online singing tool exhibited high internal consistency (Cronbach’s alpha = .92), and moderate-to-high test-retest reliabilities (.65–.80) across an average 4.5-year-span. Task performance for web- and laboratory-based participants (n = 82) matched on age, sex, and music training were not significantly different. Moderate-to-large correlations (|r| =.31–.59) were found between self-rated singing ability and the various singing tasks, supporting convergent validity. Finally, task performance of the web-based sample was not significantly different to previously reported findings. Overall the findings support the robustness of the online tool for objectively measuring singing pitch accuracy beyond a controlled laboratory environment and its potential application in large-scale investigations of singing and music ability.

中文翻译：

---

建立基于网络的歌唱研究的可靠性和有效性

在这项研究中，通过以下方式检验了客观评估歌唱能力的在线工具的稳健性：（1）确定该工具的内部一致性和重测信度；(2) 比较基于网络的参与者 (n = 285) 与在受控实验室环境中完成工具的一组 (n = 52) 的任务表现，然后确定设置之间的收敛有效性，以及 (3) 比较参与者的与先前使用类似歌唱任务和人群的研究的任务表现。结果表明，在线歌唱工具在平均 4.5 年的跨度内表现出较高的内部一致性（Cronbach's alpha = .92）和中到高的重测信度（0.65-.80）。基于网络和实验室的参与者 (n = 82) 在年龄、性别和音乐训练方面匹配的任务表现没有显着差异。在自评歌唱能力和各种歌唱任务之间发现了中到大的相关性（|r| =.31-.59），支持收敛效度。最后，基于网络的样本的任务性能与先前报告的发现没有显着差异。总体而言，研究结果支持在线工具的稳健性，该工具可在受控的实验室环境之外客观地测量歌唱音高的准确性，及其在歌唱和音乐能力的大规模调查中的潜在应用。