Compulsive Internet Use and Working: a self-reinforcing cycle?

Quinones, Cristina (2015). Compulsive Internet Use and Working: a self-reinforcing cycle? In: Work, Stress, and Health: Sustainable Work, Sustainable Health, Sustainable Organizations, 6-9 May 2015, Atlanta, Georgia.


Statement of the problem: Workaholism refers to the incontrollable need to work excessively, thus it is constituted by two dimensions: working excessively and working compulsively (Oates, 1971; Schaufeli et al., 2009ab). The negative consequences for employee well-being have been extensively documented (e.g. Andreassen et al., 2007; Burke, 2001; 2008). A maladaptive behaviour with apparent communalities in the underlying phenomenology of workaholism is Compulsive Internet Use (CIU). CIU denotes the maladaptive use of the internet characterised by excessive engagement to the point of experiencing loss of control over the use, inter and intra-personal conflict (Meerkerk et al., 2010; Davis, 2001; Griffiths, 2000; Young & Rogers, 1998). Since the internet has become an essential tool in many jobs, it has been argued that a self-reinforcing pattern of workaholism and CIU could develop from those who are inclined to one or the other (Porter & Kakabadse, 2006). In this study we tested the extent to which reciprocal relationships between CIU and compulsive working actually exist. The latter was tested whilst controlling for neuroticism, extraversion and hours of usage, as these variables have been associated with CIU in the past. Furthermore, since the literature on CIU relies largely on cross-sectional data and little is known about the temporal relationships of commonly discussed correlates, we also explored the temporal relationships between neuroticism, extraversion, hours of use and CIU.
Procedures: A total of 244 British employees with ages ranging 18-65 participated in this study. Participants were UK based and were recruited through a market research panel. Data was collected twice within a 6 month interval, and the same variables were included at the two data collection points. The survey contained previously validated scales for the core variables of study: Compulsive Internet Use (Meerkerk et al., 2010) Workaholism (Schaufeli et al., 2009b), Neuroticism and Extraversion (Donnellan et al., 2006). We also included demographic questions and questions about time spent online. Analyses: We used Cross-Lagged longitudinal analysis with Structural Equation Modelling and AMOS 20. First we conducted confirmatory factor analysis to confirm the appropriateness of measuring our latent variables with the indicators. We then examined whether the measurement model was stable over time. Finally we tested the structural model with bidirectional relationships between CIU and workaholism; and between CIU and hours of usage, neuroticism and extraversion.
Results: A reciprocal relationship between CIU and workaholism was not found. Instead, CIU at time 1 predicted working compulsively and higher neuroticism at time 2 but the opposite was not the case. As expected, CIU was not significantly related to the working excessively dimension of workaholism. Finally, we found a reciprocal relationship between hours of usage and CIU.
Practical Implications: In light of the risks that excessive engagement with the internet has for well-being and healthy working practices, we urge organisations to evaluate the extent to which their working practices are somehow encouraging an “always connected culture” even when this is done with a view to enhance employee flexibility. Thus, we suggest that companies should evaluate their work and communication practices to identify potential messages that encourage individuals to remain connected if they do not want to miss out. Although the latter can be an advantage in the short term as it may allow us to meet specific project deadlines, the risks associated with developing a habit of being always accessible through technology need to be understood and prevented. In this sense, our results confirmed that the risk of developing a maladaptive use of the internet seems to increase with high usage and this in turn, increases the likelihood of developing maladaptive relationships with one’s work. We believe this illustrates the importance of adopting a sustainable perspective in the management of organisations and employee’s well-being, as both are intrinsically related.
Conclusions: The two-wave design allowed us to capture a dynamic process which had only been tested with qualitative data before. The mutually reinforcing model of workaholism and CIU should be revisited as causation seems to be unidirectional. These findings also contribute to improve the current body of knowledge in CIU as most studies focused on teenage samples and are cross-sectional in nature. It is nonetheless important to note that problems of omitted variables bias cannot be ruled out, hence claims of causality need to be taken with caution and more waves would have provided stronger support for these potential causal relationships.

Viewing alternatives

Item Actions