The omnipresence of workplace gossip makes understanding gossip processes imperative to understand social life in organizations. Although gossip research has recently increased across the social sciences, gossip is conceptualized in disparate ways in the scientific literature. This conceptual confusion impedes theoretical integration and providing practical advice. To resolve this, we systematically reviewed 6114 scientific articles on gossip and identified 324 articles that define gossip. From these definitions, we extracted two essential characteristics of gossip on which there seems to be agreement within the literature, namely, (1) that gossip is communication between humans involving a sender, a receiver, and a target, and (2) that the target is absent or unaware of the communicated content. These two characteristics formed the basis of a broad, integrative definition of gossip: a sender communicating to a receiver about a target who is absent or unaware of the content. Furthermore, some definitions include characteristics on which there is less agreement: gossip valence (from negative to neutral to positive) and formality (from informal to intermediate to formal). We incorporate these characteristics in a dimensional scaling framework that can guide future research. Our broad, integrative definition of gossip and the dimensional scaling framework provide the building blocks for a systematic, integrated knowledge base on the role of gossip in human social life in general as well as in organizations. This can foster future theory development and hypothesis testing, ultimately helping organizations to manage gossip.

The omnipresence of workplace gossip makes understanding gossip processes imperative to grasp social life in organizations. Although gossip research has recently experienced an upsurge across the social sciences, findings regarding the consequences of gossip are conflicting. A potential reason is that gossip is conceptualized in myriad different manners in the scientific literature, causing conceptual confusion and rendering theoretical integration impossible. In order to resolve this, we systematically reviewed 6114 scientific articles on gossip and identified 324 papers that define gossip. From the definitions we extracted two essential characteristics of gossip on which there seems to be good agreement within the literature, namely (1) that gossip is communication between humans involving a sender, receiver, and target, and (2) that the target is absent or unaware of the communicated content. These formed the basis of a broad, integrative definition of gossip: a sender communicating to a receiver about a target who is absent or unaware of the content. Furthermore, our review revealed that some definitions discuss characteristics on which there is less agreement: gossip valence (from negative to neutral to positive) and formality (from informal to intermediate to formal). We propose incorporating these characteristics in a multidimensional scaling framework that can guide future research. Our broad, integrative definition of gossip and the multidimensional scaling framework provide the building blocks for a systematic, integrated knowledge base on the role of gossip in human social life, which can foster future theory development and hypothesis testing, and thereby ultimately help organizations to manage gossip.

Gossip, or sharing information about absent others, has been identified as an effective solution to free rider problems in situations with conflicting interests. Yet, the information transmitted via gossip can be biased, because gossipers may send dishonest information about others for personal gains. Such dishonest gossip makes reputation-based cooperation more difficult to evolve. But when are people likely to share honest or dishonest gossip? We build formal models to provide the theoretical foundation for individuals' gossip strategies, taking into account the gossiper's fitness interdependence with the receiver and the target. Our models across four different games suggest a very simple rule: when there is a perfect match (mismatch) between fitness interdependence and the effect of honest gossip, the gossiper should always be honest (dishonest); however, in the case of a partial match, the gossiper should make a choice based on their fitness interdependence with the receiver and the target and the marginal cost/benefit in terms of pay-off differences caused by possible choices of the receiver and the target in the game. Moreover, gossipers can use this simple rule to make optimal decisions even under noise. We discuss empirical examples that support the predictions of our model and potential extensions. This article is part of the theme issue ‘The language of cooperation: reputation and honest signalling’.

No abstract

Introduction Analysis of covariance (ANCOVA) remains a widely misunderstood approach for dealing with group differences on potential covariates (Miller & Chapman, 2001). This misunderstanding of the ANCOVA has a long history and its discussion is dispersed across fields and journals, making it difficult to obtain a systematic overview. Here we present a network method to organize the results of a literature search conducted by 44 Master's students as part of the 2016 University of Amsterdam course "Good Research Practices". The ANCOVA Pitfall Dora wants to assess whether, in her own university, men earn more than women. She has access to the salaries of a subset of researchers, and, as expected, men earn significantly more than women (p < .005). But wait! The men in her sample are also older than the women, and this confounds the results: perhaps the salary difference is due to age rather than gender. To address this confound and "control for" age, Dora includes age as a covariate in an ANCOVA. This procedure is tempting but statistically problematic. The ANCOVA is easier to interpret correctly when age influences salary but does not differ across the groups. As explained in Miller and Chapman (2001; but see chapter 10 in Judd, McClelland, & Ryan, 2011, and Field, 2013, pp. 484-486), when groups differ on a covariate (e.g., age), removing the variance associated with the covariate also removes the shared variance associated with the group (e.g., gender). As a result, the grouping variable loses some of its representativeness. This occurs mostly when groups are pre-existing and are not obtained by random assignment (Jamieson, 2004). As an example, assume one has access to the height of several mountain peaks in the Himalayas and the Pyrenees (Cohen & Cohen, 1983). One may test whether the mountain ranges differ in height and it may be tempting to include air pressure as a covariate; after all, air pressure differs across the