In the past decade, various scales have been developed for measuring engagement with social network sites (SNS), but validity concerns have recently been raised about some of them. The present study thus provides a systematic review of the psychometric properties of these scales. This review included articles that aimed at either developing an SNS engagement scale or providing a systematic test of the psychometric properties of the scale. We conducted keyword-based searches of several broad multidisciplinary databases, along with reference list searches and article citation searches. These search strategies yielded a total of 14 reports, revealing validation evidence for 12 SNS engagement scales among 13,861 participants from 11 countries. There was mixed evidence for the various types of validity tests, with some scales having been validated more rigorously with multiple studies and samples while others having not yet been systematically validated. Sampling and acquiescence biases were also present for some scales. The present review provides recommendations for researchers intending to study SNS engagement. Although the literature search was multi-faceted, it may conceivably have missed studies that provided less rigorous validity evidence. Overall, this study contributes to evaluating and strengthening the methodological foundations of SNS research.
Measuring SNS engagement
Though SNS are becoming an integral part of many people's daily lives (Pew Research Center, 2011; Statista, 2018), multiple studies have shown that there are important individual differences in how SNS are used (C. Y. Lai & Yang, 2014; N. Park, Lee, & Kim, 2012). To capture these differences, researchers have used a variety of measurement techniques. Some techniques focus on objective behaviors (e.g., time spent on SNS, frequency of SNS behaviors), and data mining-based methods provide promising new opportunities to collect this type of data (Burke & Kraut, 2013; February). These methods may be particularly useful in situations where selfreport methods are likely to be inaccurate, such as in estimating amount of time spent on Facebook or the Internet (Junco, 2013; Scharkow, 2016). These techniques, however, may be less adequate in measuring the psychological aspects of SNS use, such as emotional connection to the SNS (Jenkins-Guarnieri, Wright, & Johnson, 2013; Orosz, Toth-Kir aly, & Bothe, 2016 } ) or the motivations underlying SNS behaviors (Lomborg & Bechmann, 2014; Mahrt & Scharkow, 2013). Thus, misinterpretations of SNS use and behavior are more likely for studies that rely solely on objective measures (Ellison & Boyd, 2013). For example, an objective behavior such as re-posting someone else's status can indicate an affirmation of their relationship, or simply a superficial passing along of information (Boyd, Golder, & Lotan, 2010). To capture psychological aspects of SNS use, several scales have been developed that assess the level of an individual's engagement with the SNS. Engagement refers to “a quality of user experience with technology that is characterized by challenge, aesthetic and sensory appeal, feedback, novelty, interactivity, perceived control and time, awareness, motivation, interest, and affect” (O'Brien & Toms, 2008, p. 23).