Industrial/Organizational Psychology, The Pennsylvania State University, University Park , Pennsylvania.
Cyberpsychol Behav Soc Netw. 2015 Mar;18(3):181-7. doi: 10.1089/cyber.2014.0418.
Cyberpsychology is a recently emergent field that examines the impact of technology upon human cognition and behavior. Given its infancy, authors have rapidly created new measures to gauge their constructs of interest. Unfortunately, few of these authors have had the opportunity to test their scales' psychometric properties and validity. This is concerning, as many theoretical assumptions may be founded upon scales with inadequate attributes. If this were found to be true, then previous findings in cyberpsychology studies would need to be retested, and future research would need to shift its focus to creating psychometrically sound and valid measures. To provide inferences on this concern, the current study examines the article reporting, scale creation, and scale reliabilities of every article published in Cyberpsychology, Behavior, and Social Networking from its inception to July 2014. The final data set encompassed the coding of 1,478 individual articles, including 921 scales, and spanning 17 years. The results demonstrate that the simple survey methodology has become more popular over time. Authors are gradually applying empirically tested scales. However, self-created measures are still the most popular, leading to concerns about the measures' validity. Also, the use of multi-item measures has increased over time, but many articles still fail to report adequate information to assess the reliability of the applied scales. Lastly, the average scale reliability is 0.81, which barely meets standard cutoffs. Overall, these results are not overly concerning, but suggestions are given on methods to improve the reporting of measures, the creation of scales, and the state of cyberpsychology.
网络心理学是一个新兴的领域，研究技术对人类认知和行为的影响。由于其处于起步阶段，作者迅速创建了新的措施来衡量他们感兴趣的结构。不幸的是，很少有作者有机会测试他们的量表的心理测量特性和有效性。这令人担忧，因为许多理论假设可能建立在属性不足的量表之上。如果这是事实，那么网络心理学研究中的先前发现将需要重新测试，未来的研究将需要将其重点转移到创建具有心理测量学意义和有效性的措施上。为了对这一关注提供推断，本研究检查了从开始到 2014 年 7 月在 Cyberpsychology，Behavior 和 Social Networking 上发表的每一篇文章的文章报告，量表创建和量表可靠性。最终数据集包括对 1478 篇文章的编码，包括 921 个量表，跨越 17 年。结果表明，随着时间的推移，简单的调查方法变得越来越流行。作者逐渐应用经过实证检验的量表。然而，自创的措施仍然最受欢迎，这导致人们对措施的有效性产生担忧。此外，多项目措施的使用随着时间的推移而增加，但许多文章仍未报告足够的信息来评估所应用量表的可靠性。最后，平均量表可靠性为 0.81，勉强达到标准截止值。总体而言，这些结果并不令人过于担忧，但给出了一些建议，以改进措施报告、量表创建和网络心理学的状况。
