Introduction

According to Ginsberg et al. (1987) with research in the USA, friendship is a type of interpersonal relationship that serves important functions in human experience throughout the lifespan such as providing companionship and affirmation of self-worth. In the United Kingdom, Baron-Cohen and Wheelwright (2003, p. 509) define those with high friendship quality as individuals that “…enjoy close, empathetic, supportive, caring friendships that are important to them; that they like and are interested in people; and that they enjoy interacting with others for its own sake.” One theory of friendship (based upon USA research) that relates to the possible underlying factors in gender differences in friendship and gossip may be the conceptualization of friendship with two different dimensions: communion and agency (Wright 1988). A recent meta-analysis by Hall (2011), using 76% North American and 24% cross-cultural samples, identified two main gender differences in friendship: communion is higher in females, agency is higher in males. Communion refers to the intimacy or closeness needs that are met through friendship. The agency element of friendship provides individuation and power needs according to Canadian research by Zarbatany et al. (2004). The different emphasis on agency and communion is relative, as both genders value communion in friendships (Zarbatany et al. 2004; Wright 2006, in the USA). However, it is possible that this difference in the balance between communion and agency will produce gender differences in how gossip functions in friendships.

Collectivism versus individualism is a possible basis for cross-cultural differences in the processes of both friendship and gossip. Therefore, the focus of the present investigation is upon more individualistic cultures such as the United States and Canada. Due to the overall cultural similarity of Canada and the United States, the following theoretical and empirical studies reviewed are from these two countries, unless otherwise noted. In addition, research from United Kingdom, Belgium and the Netherlands is mentioned as these countries have also been found to be rated highly on individualism-collectivism (e.g., Suh et al. 1998).

There are several reasons why gender differences in friendship have been observed and these underlying processes may provide some insight into possible predictions with gender differences in friendship and gossip. According to Wright (2006), female friendships are more intimate because women are more likely to be involved in more communal activities such as “…a baby shower..” versus the more agentic, instrumental activities such as “…shingling a roof…” (p. 47). Another aspect of friendship is that women are more likely to have broader more, holistic friendships rather than more circumscribed friendships. For example, having specific work friends, sports friends, etc. (Wright 2006). Another set of factors that tend to produce more agentic friendships in males are dispositional factors such as “…emotional restraint, masculine identity and homophobia…” (Bank and Hansford 2000, p. 64). Lastly, there is also the tendency of males to have a form of friendship that involves being fused within the performance of roles within an organization to a greater degree compared to females (Wright 2006). With the fused friendship, the individuals become friends in the context of a structured role, for example at work, in a club, etc. This reduces the likelihood of the friendship being more communal as the individuals must balance out successfully performing the role, with maintaining the friendship (Wright 2006).

These differences in friendship with males and females may produce gender differences in how gossip functions within friendship. Male friendships may be either facilitated by the use of gossip or the other possibility is that these characteristics of the male friendship are less vulnerable to the negative effects of gossip. For example, according to Foster (2004), gossip can be used to increase the status or power of an individual in a group, therefore this function of gossip may be more conducive to the male emphasis on agentic friendship, compared to the female emphasis on communal friendship.

According to research in the United Kingdom by Emler (1994), 70% of conversations are gossip, which has been defined as “…positive or negative information exchanged about an absent third party…” (Grosser et al. 2010, p. 179). According to Foster’s (2004) theory, gossip is strongly related to friendship as it has an important role in building and enhancing relationships through a friendship or intimacy function. According to Foster (2004), there is little evidence of women gossiping more than men and a general conclusion is that the differences between males and females in gossip are small. However, an earlier study by Levin and Arluke (1985) found a pattern of both similarities and differences, but some evidence that might suggest gender differences in gossip and friendship. While females had a slightly higher level of gossip, the balance of positive and negative gossip was the same for males and females. The differences that were found can be related to the friendship function of gossip, as with the females, the tendency was to talk about people they were closest to in terms of their social networks. Males however, were reluctant to be as intimate in the conversation and tended to discuss more distant individuals (Levin and Arluke 1985).

As there are gender differences in the functions of friendship, there are also likely both general and specific differences in how the functions of friendship are related to the various types of gossip that have been identified. Some of this variation may stem from gender differences in friendship, other dissimilarities may be a result from gender differences in the functions of gossip.

Nevo et al. (1993) conducted research in Israel with intent of creating a general measure for gossip. According to this perspective, gossip is a disposition, and therefore, it may be that differences in gender role socialization produce stable differences in how gossip is used in the context of friendship. Nevo et al. (1993) conceptualize gossip in terms of three specific components that may be related to friendship in different ways for males and females. These are social information, achievement, and physical appearance.

The social information component of gossip tendency involves discussion of different social topics, and so is a measure of social involvement according to Nevo et al. (1993). In males, this aspect of gossip is more likely to be related to the agency function of friendship, as social information gossip involves status and control of information (Nevo et al. 1993). Agency oriented friendship involves shared activities and teamwork (e.g., De Vries 1996), and gossip has a role in this agency by regulating and maintaining the norms of the group and preventing social loafing. This form of gossip may be important to the maintenance of the more agentic friendship as it provides a means of exerting control and enforcing norms without resorting to direct physical altercation (Ellickson 1991).

Achievement gossip (e.g., grades, salaries) is another aspect of gossip that may be related to the male emphasis on agency in friendship. This form of gossip may be more associated with male friendships, given the greater interest in individuation through emphasis on individual accomplishments (e.g., Zarbatany et al. 2004). Also male friendships are more concerned with establishing status and reputation (Emler 1994; Salove 2007). As gossip is part of conversation, and friends are more likely to gossip compared to acquaintances (e.g., Blumberg 1972, in the United Kingdom), gender differences in conversation may explain a possible greater emphasis on achievement related gossip in male friendships. With research in the United Kingdom, Dunbar (2010) points to the asymmetry of male and female conversation. Males are more self-focused, females converse more about others. Females emphasize the building and maintaining social networks and males are more concerned about display and status. Dunbar (2010) argues that gender differences in conversation can considered along evolutionary lines, with males using conversation as a form of self-promotion for the purpose of attracting a mate, “…a kind of vocal form of the peacock’s tail” (p.75). Dunbar (2010) refers to male speech as a form of advertizing, particularly if females are present “…more showy, more designed to stimulate laughter as a response…” (p. 76). In addition, the talk becomes more intrusive, more competitive and political in the presence of females.

Physical appearance gossip (e.g., clothes) has been found to be higher in females, (Nevo et al. 1993). In the context of gender differences in friendship, this type of gossip may be more of a threat to the more communal female friendship which values self-disclosure and intimacy more so than in the case of males (Hall 2011). According to research conducted in the Netherlands by Massar et al. (2012), gossip about physical appearance and sexual reputation are the focal elements of the evolutionary value of gossip, as this is a means of intrasexual competition for potential mates. In Belgium, De Backer et al. (2007) found gender differences in the recall of gossip from same-gender rivals. Females recalled more attractiveness information, while males recalled more cues relating to wealth status. Reputation gossip is often used as form of aggression with females, as it is can be a highly effective method of reducing the attractiveness of a same-gender rival (De Backer et al. 2007). Hence in the context of mating, females are vulnerable when it comes to reputation (Hess and Hagen 2002). This finding may be some of the basis for gender differences in friendship and physical appearance related gossip. A moderate amount of gossip may be important for the formation of friendships in females. However, a higher amount of gossip may be perceived as an aggressive threat and therefore lower the level of friendship.

One previous study on friendship and gossip was conducted with a University age population by Jaeger et al. (1994). This study compared high, moderate and low gossipers with a sample of 36 females ranging in age between 18 and 22 years that were members of a sorority. These students were measured in terms of need for social approval, self-esteem, anxiety, and three indicators of popularity, including number of close friends. The results indicated that moderate gossipers had a higher number of close friends compared to the low and high gossip groups. The low gossipers had a higher need for social approval compared to the low or medium level gossipers. In addition, the high gossipers were rated as less likable compared to the low gossipers. While this research does demonstrate the importance of the relationship between gossip and friendship, it is likely that examination of friendship and gossip with both males and females would further clarify this relationship.

While gender differences in friendship and differences in gossip by gender have been examined separately in the literature, there is no research on how gossip might function differently in the context of friendship for men and women. Therefore, the present study will build upon this previous work as it will examine how gossip processes might be different within friendships in males and females. For these reasons, outlined above, it is likely that the relationship between gossip and friendship will be different for males and females. Therefore, the following hypotheses are proposed.

  • Hypothesis 1. The correlation between gossip and friendship quality will be stronger in males compared to females. This is prediction is consistent with the greater emphasis on agency in male friendships and communality in females.

  • Hypothesis 2a: The information gossip scale will have a stronger correlation with friendship quality in the males compared to females. This prediction is made due to the importance of status and control of information in male friendship.

  • Hypothesis 2b: Achievement gossip will be more strongly correlated with friendship quality in males compared to females. This prediction is consistent with the greater interest with individuation in male friendships.

  • Hypothesis 2c: Physical appearance gossip will be more strongly correlated with friendship quality in males compared to females. This result is likely as reputation gossip about physical appearance may possibly be less of a threat to male friendships compared to that of females.

Method

Participants

The participants were 167 female and 69 male undergraduates from a Western Canadian University. This University has a student composition of 80% non-minority and 20% visible minority students, including 6% aboriginal students. The students participated as part of their introductory psychology research experience. Participants were given informed consent that the research consisted of a set of questionnaires about conversational topics. The students were fully debriefed as to the purpose of the study upon completion of the questionnaires, and free to withdraw from the study at any time without consequence. The mean age of the participants was 19.98 and the range was 17–29.

Instruments

The Tendency to Gossip Questionnaire (TGQ) Nevo et al. (1993, 1994) was developed in Israel and was designed for use with a college student population. The original normative sample was 120 of students with an average age of 23.4. The test was originally administered in Hebrew and has been translated into English for use with Western samples. The underlying theoretical background of the measure draws upon past work conducted in North America (e.g., Rosnow 1977; Suls 1977). In addition, Nevo et al. (1993, 1994) describe the development of a test for more general use, rather than specifically in an Israeli context. Therefore, this test was deemed suitable for a study with a North American population. The TGQ has 20 items using a 1 ‘never’ to 7 ‘always Likert scale with a possible range of 20–140. Higher scores indicate a greater tendency to gossip. Nevo et al. (1993, p.975) define gossip as a disposition “…a social activity requiring two or more individuals who discuss other persons”. The test also measures four different content areas of gossip. The scales were developed through a principal components analysis which obtained four factors which are labeled: (a) physical appearance, e.g. “Talk with friends about other people’s clothes”; (b) achievement-related gossip, e.g. “Talk with friends about other people’s salaries”; (c) social information, e.g. “Know what is going on, who is dating, etc.; and (d) sublimated gossip e.g., “analyze with friends other people’s motives”, (Nevo et al. 1993, p. 978). Nevo et al. (1993) reported a full scale Cronbach’s of α = .87 for the measure. With external validation of the scale, the TGQ was found to correlate with vocational interests in terms of people oriented professions (Nevo et al. 1993). The sublimated gossip scale was omitted from the present study as the Cronbach’s alpha was less than α = .60, at α = .59.

The Friendship Questionnaire (FQ) is a 35-item self-report questionnaire by Baron-Cohen and Wheelwright (2003) that was designed to measure friendship. The instrument was developed with a sample of British adults. Baron-Cohen and Wheelwright (2003) emphasize that this instrument was designed with items that are more neutral in terms of different friendship styles that may be found in males and females. Therefore, confiding in others (females) was not given more or less merit compared to shared activities (males). Baron-Cohen and Wheelwright (2003) used 27 of the items, with a range of 0–135 and reported a Cronbach’s alpha of α = .75. With this particular study, 33 of the items were scored (range 33–165) with an obtained alpha of α = .66. The scale uses a combination of selection between one of several statements, e.g. a. “I have one or two particular best friends”. b. “I have several friends who I would call best friends” c. “I don’t have anybody who I would call a best friends” and Likert scales, e.g. 1 ‘easy’ to 5 ‘difficult’. Items 34 and 35 were not used as these are more qualitative questions concerning specific topics that people converse about with friends (item # 34) or at social occasions (item # 35). For example, item 34: “When talking with friends, what proportion of your time do you spend talking about the following? …politics and current affairs, hobbies interests, personal matters, work, family and friends, the weather…” (Baron-Cohen and Wheelwright 2003, p. 516–517).

Results

Table 1 illustrates the descriptive statistics for the gossip and friendship measures. The overall MANOVA for gender was F (8,227) = 12.57, p < .001. Wilk’s λ = .693. The largest gender difference was a lower score for the males with the Friendship Questionnaire: males, M = 82.99, SD = 15.36 versus the females, M = 94.57, SD = 13.28. The difference was significant with an F (1,234) = 33.76, p < .001 and an effect size of δ = .81. This result is similar to Baron-Cohen and Wheelwright (2003) that obtained an effect size of d = 1.24. With gossip, the TGQ had higher scores for the females F (1,234) = 10.47, p < .001 as well as the physical appearance, F (1,234) = 30.97, p < .001 and social information, F (1,234) = 10.53, p < .01 gossip subscales. In summary, the overall gender differences were that males had lower friendship quality scores, and females had higher overall gossip scores and higher physical appearance and information subscale scores.

Table 1 Descriptive statistics and gender comparisons
Full size table

Sublimated gossip scale was omitted as the obtained alpha was α = .59.

The relationship between gossip and friendship is examined through analysis of correlations between friendship quality and gossip as shown in Table 2. With males, a relationship between gossip and friendship quality was demonstrated as all of the correlations were significant. In the case of the females, there were no significant correlations. In order to test the differences between these correlations, a series of Z tests for two independent correlations were performed, see Table 3.

Table 2 Intercorrelations for males and females
Full size table
Table 3 Correlations between gossip measures and friendship quality
Full size table

Hypothesis 1 was confirmed with a stronger correlation between the TGQ and FQ with the males at r (69) = .420, p < .001, and the females at r (167) = .128, p = .099. The difference was significant at Z = 2.21, p < .05.

Hypothesis 2a was demonstrated as there was a larger correlation for males between social information and friendship quality. Males had a correlation of r (69) = .458, p < .001 and females had a correlation of r (167) = .110, p = .156. The difference was significant at Z = 2.66, p < .01.

Hypothesis 2b was that males would have a stronger correlation with achievement and friendship quality. With males, the correlation was r (69) = .291, p = .015. Females had a correlation of r = .082, p = .292. The difference was a suggestive trend at Z = 1.50, p = .066.

Hypothesis 2c was that physical appearance gossip and friendship quality would have a stronger correlation for the males with r (69) = .364, p < .01 and for females r (167) = .160, p < .05. The difference was a suggestive trend at Z = 1.52, p < .064.

Overall, the results are in support of Hypothesis 1 of a stronger relationship between friendship quality and gossip in males. There was also support for Hypothesis 2a of a greater relationship between social information gossip and friendship quality in males. Both achievement gossip, Hypothesis 2b and physical appearance gossip, Hypothesis 2c were related to male friendship quality, but not related to friendship quality in females. The obtained differences were suggestive trends in the expected direction.

Discussion

Core Friendship Expectations and Gossip

The results of this investigation support Hypothesis 1 of gender differences in the relationship between friendship and gossip. With the females, there was very little relationship between gossip and friendship quality. With the males however, gossip tendency was strongly related to friendship quality. One possible interpretation of these obtained differences in gossip and friendship may be that gossip is potentially more damaging to the female friendship compared to the male friendship. Turner et al. (2003, p. 129) found that gossip can be either “…relational ruin or social glue…” depending upon whether the gossip was positive or negative. When the gossip was negative, liking was decreased especially if the receiver of the information was a friend rather than a stranger. One possibility suggested by Turner et al. (2003) is that negative gossip is a violation of the expectancies for friendship. Gender differences in friendship expectancies may be an explanation of why gossip appears to function differently in males and females. According to Hall (2011, p. 742) the core expectations of friendship “…trust, commitment, locality, and genuineness…” are similar in men and women. However, women tend to place more emphasis on the importance of friendship than do males (Lansford et al. 2006; Macoby 1998). Hall (2011) has argued that the core expectations of friendship are more related to communion, which is more valued in females and that peripheral aspects of friendship are related to agency, which is more valued by males. Higher friendship expectations and a greater emphasis on communion make the female friendship more vulnerable to damage compared to males. If agency expectations are violated in the male friendship, this is not as potentially damaging to the friendship as agency is a more a more peripheral aspect of friendship (Hall 2011).

Agency-Communion Balance

Another interpretation of these findings is that the obtained overall differences may be due to differences in agency-communion balance. According to Zarbatany et al. (2004), male friendships function differently in terms of the relationship between agency and communion. While both genders value communion more than agency in friendship, males will use communion in a different way compared to females. Males will have close communal relationships and use these close friends to meet the agency needs of status and social prominence. Therefore, males are better able to balance these two aspects of friendship. With males, the relative weight is on agency, as males desire the increase in status that friendship can provide, but will not sacrifice their communion, close friendships, in order to increase status (Zarbatany et al. 2004). Gossip may be more conducive to the more agentic quality of male friendships, being more external, involving a shared activity and more emotionally detached compared to females (Hall 2011). Therefore, gossip enhances the ability of the group members to communicate about the shared activity and perhaps enforce norms about the collective endeavor without direct physical confrontation as Brison (1992) has found in Papua New Guinea and Ellickson (1991) in the USA. Acheson (1988) has illustrated this in a study of predominately male lobster fisherman (91% males) and the role of gossip in establishing friendship within this community and enforcement of the normative standard of this particular group. Ellickson (1991) also demonstrated similar functions of gossip in mainly male, cattle ranchers.

Female friendships may have the ability to be communal without extensive use of gossip. These results were somewhat different from Jaeger et al. (1994) with a sample of females, differences were found between high, medium and low level gossipers, with the moderate level gossipers having a higher number of close friends. This previous study was in the context of a sorority group with perhaps more densely interconnected social networks compared to the larger student sample in the present study. Jaeger et al. (1994) found that the high gossip group were more likely to be the influential, leader-type members of the group, and perhaps were more distant from the others in terms of the depth of the friendships that were formed. This compares to the medium level gossipers, who in this context were not the influential members and therefore used gossip more for forming friendships rather than influence which may have been the more predominant motivation for the gossip in the high gossip, more socially dominant group. Therefore, some gossip is necessary to bond with the group, but if the level is too high, then the person is viewed more negatively.

In the current study, the relationship between gossip and friendship quality was found only with the males. This result may be due to the lack of a densely interconnected context as was the case in the Jaeger et al. (1994) sorority study. With males, it is likely that the relationship between gossip and friendship quality may be stronger, because in order for gossip to occur in the first place, it is more likely to occur within the context of a close relationship. Males are more reticent to enter into these close relationships (e.g. De Vries 1996), so when you do find gossip within male friendships, it is more likely to be related to a closer, higher quality friendship.

Gender Differences in the Functions of Gossip

Other than overall gossip tendency scores, the strongest gender difference in the relationship between gossip and friendship is with social information, which provides support for Hypothesis 2a. While females had higher scores on the information scale, the relationship between social information and friendship was stronger with the males. Access to information and control of information can be valuable commodities in a social group. Social information gossip has been described as a way of obtaining a quick assessment of the norms and the hierarchy of the group, as well as a method of getting and spreading information about a social group (Foster 2004). The greater male concern with status and reputation in friendship could be related to the notion of gossip as an “…information management technique…” (Paine 1970, p.186); whereby the individual is in possession of the knowledge has a type of currency, private information about a person’s life. Gossip is said to be “…mutually empowering to its participants…” (Ayim 1994, p. 99). This is closely associated with friendship, as part of the foundation of friendship is access to private information about another, and trust between the two individuals (Bergmann 1993). Therefore, in the context of friendship, the individual with this type of knowledge is in possession of information that can be used to attain the status and power more characteristic of the male friendship.

Achievement had no relationship to friendship quality in females and was related to friendship quality in males in support of Hypothesis 2b. A stronger relationship between achievement gossip and friendship in males could be related to the greater emphasis on teamwork, shared activities and status in male friendships. The relationship between achievement gossip and friendship quality in males may also be due a greater emphasis on individuation in males. Baumeister and Sommer (1997) argue that individuation is a social process that has the effect of enhancing one’s prominence within a social group. This process is not something that involves separation or distancing males, but can instead enhance friendship, as the individuation process happens with the context of a social group. Zarbatany et al. (2004) emphasize that agentic friendship can provide the mastery or power needs more typical in males. Power, according to Baumeister and Sommer (1997) is not something that separates individuals, but something that connects individuals together as the powerful person is someone that is sought after by others.

Physical appearance gossip scores were found to be higher in females, but the trend with the males was to have a higher correlation with friendship in support Hypothesis 2c. The difference may be that with females, this type of gossip is more common, but leads to a high level of social comparison and therefore, lower friendship quality. With males, discussing physical appearance may be less common, but when the topic emerges; it is more related to establishing status. Males are better able to integrate and balance competition with communality into friendship (Zarbatany et al. 2004), and therefore when physical appearance does enter the conversation, it is more related to friendship quality.

Limitations and Future Directions

The use of self-report methods to assess gossip and friendship is a possible limitation. Therefore, future research could be conducted that would examine gossip and friendships using more naturalistic observational methods. The age range of 17–29 years of age is also a limitation. Older and younger populations of both genders should be examined to investigate possible differences in the gossip and friendship relationship.

For example, in a study using an evolutionary perspective, Massar et al. (2012) found that older women were less likely to gossip about a rival, but this effect was mediated by mate value. It would be interesting to see if this effect is still present in males and the possible effects on friendships in men and women.

Cross-cultural investigation of these findings would also be an important area for future research, as these gender differences in friendship obtained in a more individualistic culture may not apply to more collectivistic cultures. Friendship expectations can vary cross-culturally and gossip may have patterns of cross-cultural similarity and differences. With friendship, individualism-collectivism is a possible basis for cross-cultural differences for example, Koh et al. (2003) compared Korean and Canadian University students and found many similarities in friendship when individualism-collectivism was statistically factored out e.g., level of intimacy. In another study, Gonzalez et al. (2004) compared friendship expectations in collectivistic Cuba with more individualistic Canada. While both countries emphasized loyalty and acceptance in friendship, the Cuban participants emphasized reciprocity more than the Canadians. A study by Adams and Plaut (2003) in Ghana, West Africa and the USA demonstrated that collectivism can produce seemingly paradoxical affects, as the collectivistic culture in Ghana emphasizes the possible negative aspects of friendships. In this study, the Ghanaian participants considered friendship as something to be approached with caution, as friends can be envious, jealous and hide behind a social façade. Adams and Plaut (2003) argue that this apparent paradox is due to the more interdependent view of the self found in a more densely interconnected culture. Therefore, the participants in Ghana were found to be relatively more agentic, rather being than communal in their friendships, as practical or material assistance was emphasized in the relationship. The North American participants were more likely to stress the communal aspects of friendship, as participants mentioned companionship and self-disclosure as the key defining elements of friendship.

With gossip, there are no systematic comparisons of the cross-cultural aspects of gossip. However, Foster (2004) cites several field studies in both North American and cross-cultural contexts, whereby the researchers observed common functions of gossip.

Gossip was found to be a means of control through the functions of social information and influence. For example, in Zanacatan, Mexico, Haviland (1977) emphasized that gossip can be used as a means of social control “…gossip is one sort of behavior by which people manage their social faces: keeping an eye out while limiting other people’s view of oneself” (p. 101). Besnier (1989, 2009) mentions the use of gossip in Nukulaelae society as a means of control though withholding of information. In a study in Papua New Guinea, Brison (1992) found that gossip is a method of enforcing norms without direct physical confrontation.

While these studies point to cross-cultural similarities, there are also possible differences, therefore the results of the present study cannot be assumed to be universal. The present research was conducted in North America and so the assumption is that these findings will be applicable to North American or other Western individualistic countries such as the United States, Canada, United Kingdom, Belgium and the Netherlands which have been found be more individualistic (e.g., Suh et al. 1998). Further studies will be required to address the issue of the cross-cultural implications of these hypotheses.

Another aspect that would merit further investigation is the communal versus agentic dimension as described by Zarbatany et al. (2004). Particularly in the context of male friendships, these results point to the need for additional research into the characteristics and dynamics of male gossip. The role of social information, physical appearance, and achievement gossip in male friendships needs further examination.

The issue of same-gender versus cross-gender friendships is another avenue for further research in the context of gossip as male–female friendships have their own dynamics which may have a different relationship with gossip.

The gate-keeping aspect of gossip or notion of betweenness is also another characteristic that could be further researched, particularly in the context of male friendships. Social network analysis has demonstrated that two important factors with gossip are density (more interpersonal connections) in the social network and norm coherence (Foster and Rosnow 2006). Social groups are enhanced by dense social networks and when the gossip network is also dense, there is enhanced access to information and the ability of a single individual to influence the group is increased (Foster and Rosnow 2006). Also with this perspective, is the notion that one person can act as a gatekeeper for information to others, the notion of “…betweenness…” (Foster and Rosnow 2006, p. 170). This analysis illustrates the importance of friendship and gossip as those with higher network density (more interpersonal connections) and more betweenness (control of information through a gate keeper) have more adherence to group norms. According to this perspective, gossip can be useful in terms of establishing status in terms of more influence, enforcement of norms and greater control of information within the group. Therefore, another possible area for further research is to examine gender differences in social network density and how these relate to friendship quality.