Channel complementarity and multiplexity in long-distance friends’ patterns of communication technology use

Abstract

Understanding how people use communication technologies (CTs) in close relationships requires examining interdependencies in or patterns of CT use in those relationships. We combined channel complementarity theory and media multiplexity theory to investigate first-year college students’ (N = 155) use of CTs (video chat, phone calls, and text messaging) in close, long-distance friendships over a 3- to 10-day period. Overall, CTs were used complementarily. However, complementary use of phone calls and text messaging was higher in closer friendships. In contrast, phone calls and video chat were complementary at low but not high levels of closeness. These findings suggest utility in combining channel complementarity theory and media multiplexity theory to understand the “web” of CTs used in daily communication in long-distance friendships and point to similarities in and differences between CTs that might affect complementarity.

Keywords

Channel complementarity theory communication technology friendship long-distance relationships media multiplexity theory relational closeness

Communication technologies (CTs), such as text messaging and video chat, allow friendships to be “flexible” in the sense that people have a wide range of options available for maintaining friendships (Becker et al., 2009; Johnson and Becker, 2010). As Ledbetter (2009) notes, the flexibility of friendships makes them “an intriguing site for examining emergent patterns of media use” (p. 1188). Because friendships can be maintained in a variety of ways (e.g. varying in types of CT used and frequency of contact), they can provide insight into what factors contribute to patterns of communication in relationships. In particular, close, long-distance friendships among students who have moved away for college offer a relevant site for understanding CT use in friendships for two reasons. First, the transition from high school to college and from geographically-close to long-distance friendship often raises concerns about maintaining the friendship, and such friendships can be an important source of emotional and relational continuity during a time of change (Ellison et al., 2007; Johnson et al., 2009). Thus, communication with these friends is likely to be particularly salient. Second, previous research suggests that in less-developed relationships, people restrict themselves to less interactive and more public forms of communication, such as Facebook (Liu and Yang, 2016; Yang et al., 2014). Close friends use a wider range of CTs and communicate more frequently (Liu and Yang, 2016; Yang et al., 2014), facilitating the exploration of CT use patterns within those friendships.

Although previous research finds that CTs can aid in the maintenance of long-distance friendships, most of this research has focused on individual CTs (e.g. Cummings et al., 2006; Johnson et al., 2008; Ledbetter et al., 2011; Piwek and Joinson, 2016; Ranney and Troop-Gordon, 2012); relationships between multiple CTs lack clarity. Integrating aspects of channel complementarity theory (CCT; Dutta-Bergman, 2004) and media multiplexity theory (MMT; Haythornthwaite, 2005) would offer a more comprehensive picture of interdependencies in or patterns of CT use in long-distance friendships. Network tie strength, a key construct in MMT that is often operationalized as relational closeness or interdependence (e.g. Baym and Ledbetter, 2009; Caughlin and Sharabi, 2013; Ledbetter, 2009; Ledbetter et al., 2016), likely serves as a motivating factor for complementary CT use in friendships. Thus, combining CCT and MMT could allow us to not only predict use of a particular CT but also predict and explain when and why use of various CTs is complementary in long-distance friendships. This study uses this integration of CCT and MMT to examine the role of closeness in first-year college students’ daily CT use with close long-distance friends.

CT use in long-distance friendships

Friendships lack the formalization and societal expectations that characterize other relationship types, such as family or romantic relationships. As a result, friendships are vulnerable to dissolution with distance (Oswald and Clark, 2003; Wiseman, 1986). When students move away to college, it is often difficult to maintain regular communication with friends from home. At the same time, the development of new relationships further constrains the time available to maintain long-distance friendships (Cummings et al., 2006). Despite the relational and emotional continuity provided by high school friendships (Ellison et al., 2007; Johnson et al., 2009), the transition to college often leads to the loss of high school friendships (Fehr, 1999; Oswald and Clark, 2003). When first-semester college students can maintain communication in high-quality long-distance friendships, however, they often exhibit better adjustment and lower emotional distress (Ranney and Troop-Gordon, 2012).

The increasing popularity of CTs, such as text messaging and video chat, has made distance less of an issue in long-distance friendships. As a result, friendships are more “flexible”—they can be maintained via a variety of channels (Becker et al., 2009; Johnson and Becker, 2010). Similarly, Stafford (2005) argues that proximity is not a requirement for the maintenance of close friendships. Because they can reduce costs associated with maintaining long-distance friendships (e.g. increased time and energy and reduced ability to give and receive emotional and instrumental support), CTs can make long-distance friendships less vulnerable to dissolution or de-escalation (Johnson and Becker, 2010).

In support of this idea, previous research has found few or no differences in closeness, maintenance, or commitment between long-distance and geographically-close friends. For example, one study of college students found that although geographically-close friends engaged in more maintenance behaviors than did long-distance friends, geographically-close and long-distance friendships did not differ in closeness (Johnson, 2001). A more recent study examining the enactment of maintenance behaviors via email found no differences in the frequency of maintenance behaviors between geographically-close and long-distance friends (Johnson et al., 2008). College students also report high and increasing commitment to long-distance friends (Johnson et al., 2009). These findings suggest that long-distance friendships and maintenance of those friendships are salient in college students’ lives (Ranney and Troop-Gordon, 2012).

CTs play a potentially important role in the maintenance of long-distance friendships. One study of first-year college students’ friendships from high school found that both communication and closeness declined over the 36-month study period, but that more frequent email and instant message communication mitigated these declines in closeness (Cummings et al., 2006). Although the use of CTs is central to maintaining long-distance friendships (Cummings et al., 2006; Johnson and Becker, 2010; Oswald and Clark, 2003), some research suggests that associations between CTs and closeness are not simply attributable to higher overall levels of communication but that various CTs are differentially associated with closeness (e.g. Cummings et al., 2006; Ledbetter et al., 2016; Utz, 2007).

Furthermore, CTs do not function independently of each other in friendships but are interdependent in the sense that use of one CT is tied to use of another CT. For example, research examining long-distance friendships found that closeness was positively associated with the raw amount of email communication in the friendship but negatively associated with the proportion of communication that occurred via email (as opposed to the telephone; Utz, 2007). In their review of research regarding friendship maintenance, Johnson and Becker (2010) argue that CTs must be examined in combination, not isolation. That is, researchers should aim to understand the “web” or patterns of CT use in friendships, rather than examine how friends use individual CTs. CCT is designed to predict and explain these interdependencies, whereas MMT explains the role of relational characteristics in CT use. Taken together, these theories offer a more comprehensive picture of patterns of CT use in long-distance friendships.

Patterns of CT use

Channel complementarity theory

CCT (Dutta-Bergman, 2004) views channel use as an active process in which people use channels to meet specific interpersonal or instrumental needs. Because CCT sees channel use as being driven by needs, as opposed to by the channel itself, it predicts that people who use a given channel to satisfy a particular need will also use other channels that satisfy that need (Dutta-Bergman, 2004). That is, CCT argues that people use channels in complementary ways, such that increased use of one CT is associated with increased use of another CT (Dutta-Bergman, 2004).

Previous research has supported this idea in the domain of interpersonal communication. People who communicate more frequently via instant messaging in a particular relationship also communicate more via email, cell phone, landline phone, and face-to-face (FtF) communication in that relationship (Ramirez and Broneck, 2009). Similarly, a survey of friendships that were initiated on a social network site (SNS) found that uses of several different forms of communication, including phone calls, email, instant messaging, text messaging, chat, and FtF communication, were associated with each other (Baym and Ledbetter, 2009). A study of college students’ daily use of CTs also found that use of phone calls, email, text messaging, and Facebook was positively associated with each other (Ruppel and Burke, 2015). CCT and supporting findings suggest that in long-distance friendships, college students’ use of various CTs should be positively related. For example, the more friends use video chat, the more they would be expected to also use phone calls and text messaging. Our first hypothesis reflects the core prediction of CCT, that channel use is complementary:

Hypothesis 1. Use of CTs will be complementary, such that higher use of a given CT will be associated with higher use of other CTs.

One benefit of CCT is that it allows for the examination of interdependencies in use of various CTs in relationships, as opposed to examining CTs in isolation. Although researchers have begun to understand patterns of CT use in friendships, further refinement in our understanding of these patterns is needed. With few exceptions, research has only recently begun to focus more directly on the factors that are associated with patterns of interpersonal CT use (e.g. Liu and Yang, 2016; Ruppel and Burke, 2015; Utz, 2007; Yang et al., 2014). It is therefore still unclear exactly why these patterns might exist, and researchers are often limited to observing and reporting, as opposed to predicting, those patterns. Because CCT posits that channel use is driven by motivations for particular types of content or connection, complementary use of CTs in friendship is likely to vary as a function of characteristics of the friendship. MMT (Haythornthwaite, 2005) suggests that closeness is a potentially important factor in the extent to which CTs are used complementarily.

Media multiplexity theory

MMT (Haythornthwaite, 2005) predicts that the use of a greater number of channels is associated with higher network tie strength. Although this prediction was originally tested in task-focused groups (for a review, see Haythornthwaite, 2005), research in the contexts of friendships and romantic relationships has been consistent with these original findings. For example, research focused on geographically-close and long-distance friendships found that relational interdependence was positively associated with frequency of communicating via phone calls, instant messenger, SNS, blogs, other online forms of communication, and FtF communication (Ledbetter, 2009). A study of one SNS found that frequency of using FtF, telephone, text messaging, email, instant messenger, postal mail, and other forms of online communication was associated with friendship closeness (Baym and Ledbetter, 2009). Similarly, a study of romantic relationships found that frequencies of CT use and FtF communication were positively associated with closeness and satisfaction (Caughlin and Sharabi, 2013). Although Haythornthwaite (2005) focused on a number of channels, as opposed to how much a particular channel is used, MMT has also been tested using frequency of CT use (e.g. Baym and Ledbetter, 2009; Caughlin and Sharabi, 2013; Ledbetter, 2009; Ledbetter et al., 2016). Our second hypothesis reflects both of these approaches to testing the core prediction of MMT:

Hypothesis 2a. Number of CTs used will be positively associated with closeness.

Hypothesis 2b. Frequency of CT use will be positively associated with closeness.

Integrating the theories

CCT argues that CT use (and, by extension, complementarity) is driven by motivations for use (Dutta-Bergman, 2004), but it is unknown what factors might influence complementarity in long-distance friends’ CT use. One potential explanation comes from MMT’s prediction that higher use of CTs is associated with greater network tie strength (Haythornthwaite, 2005), often operationalized as relational closeness or interdependence (e.g. Baym and Ledbetter, 2009; Caughlin and Sharabi, 2013; Ledbetter, 2009; Ledbetter et al., 2016). In their study of CT use in romantic relationships, Caughlin and Sharabi (2013) extended MMT to examine the extent to which FtF and CT-based communication were interconnected, or interdependent, in relationships. They argued that understanding how CTs are used in relationships requires examining broader patterns of interaction (as opposed to individual communication channels) and the ways in which various communication channels relate to each other. They found that better integration of CTs and FtF communication (i.e. using CTs to discuss topics already discussed FtF, and vice versa) in romantic relationships was associated with more closeness (Caughlin and Sharabi, 2013). Although integration is not strictly equivalent to complementarity because the former is topic-focused and the latter is not, both address connections or associations between CT use and FtF communication. When combined with what we know about complementary channel use in relationships, MMT suggests that closeness is a potential explanatory factor for complementarity in long-distance friendships.

By definition, FtF communication is infrequent in long-distance relationships (Stafford, 2005). However, friends can use a variety of CTs, such as phone calls, text messaging, social media, email, and video chat, to communicate with each other. CCT argues that channel use is driven by people’s desires to fulfill particular needs or goals (Dutta-Bergman, 2004). In contrast, MMT argues that channel use is driven by relational characteristics—in particular, closeness or tie strength. However, these approaches are not necessarily contradictory. In the context of close, long-distance friendships, needs such as interpersonal connection or goals such as relationship maintenance could drive complementary CT use. Integrating CCT and MMT could allow us to maximize the strengths of each theory—CCT’s focus on patterns of CT use as opposed to use of single CTs and MMT’s inclusion of relational characteristics in predictions of CT use. At the same time, it could minimize key weaknesses of both theories—CCT’s lack of specification of particular factors that drive complementarity (and, by extension, limited ability to predict when and why complementarity will occur) and MMT’s limited focus on the interdependencies between various CTs.

Closeness likely serves as a motivational factor in complementary CT use in close friendships. As a result, CCT would predict that complementarity can be driven by closeness, and we would expect friends with closer relationships to exhibit higher levels of complementarity. That is, use of a given CT should be more strongly associated with use of other CTs when relationships are closer. The third hypothesis reflects this expectation:

Hypothesis 3. Complementary use of CTs will be stronger in friendships that are closer.

Method

Participants

Participants were first-year college students recruited from undergraduate communication courses at a large Midwestern university and a large Southern university. They were asked to complete an initial questionnaire and daily questionnaires over a 10-day period about a close friend with whom they communicate frequently but are unable to see FtF on a frequent basis due to geographical separation. This definition of a long-distance relationship is consistent with previous research and definitions of long-distance relationships (e.g. Jiang and Hancock, 2013; Stafford, 2005) and has been found to be a more valid indicator of geographical separation than distance in miles or contact frequency (Pistole and Roberts, 2011). The questionnaires were completed in October of participants’ first year of college. Participants received extra credit in exchange for completing the initial questionnaire. Participants who completed between five and nine daily questionnaires received a US$5 Amazon gift card as compensation, and those who completed 10 daily questionnaires received a US$10 Amazon gift card as compensation.

A total of 226 participants completed the initial questionnaire; 155 of these completed at least three daily questionnaires and were retained for this study. Independent-samples t-tests comparing participants who did and did not complete at least three daily questionnaires revealed no differences in friendship closeness (t(224) = 1.70, p = .09), age (t(224) = 0.99, p = .32), or sex (χ (1, N = 155) = 0.22, p = .64). The final sample consisted of 37 male and 118 female participants (age M = 18.53, standard deviation [SD] = 1.53) who completed a mean of 6.94 daily questionnaires (SD = 4.24).

Participants reported having known their friends for a mean of 6.72 years (SD = 4.24, range = 0.50–25.00 years). They indicated how often they communicate with their friends overall on a 7-point Likert-type scale with anchors of less than once a month (1) and multiple times per day (7). They were also asked how often they get to see their friend FtF, with response options ranging from never (1) to daily (7). On average, participants reported communicating with their friends between once per day and a few days per week (M = 2.82, SD = 1.75) and seeing them between once a month and less than once a month (M = 2.62, SD = 1.33). Male participants primarily reported on a male friend (n = 28; female friend n = 9), and female participants primarily reported on a female friend (n = 83; male friend n = 28).

Procedure

Students who met the study criteria and were interested in participating provided their email address and a 10-digit code that was used to link their initial and daily questionnaires. On the first day of the 10-day study period, participants received an email with a link to the initial questionnaire. That evening, they received an email with a link to the daily questionnaire. The daily questionnaire was open from 7 p.m. to 5 a.m. each day of the study. Each evening, participants were sent reminder emails that included the link to the daily questionnaire. The daily questionnaire asked participants to respond according to the time since they had completed the previous daily questionnaire (e.g. if they had completed the questionnaire the night before at 10.00 p.m., they should think about the time that had elapsed since then).

Measures

Closeness

In the initial questionnaire, participants were asked to indicate the extent to which they agreed with the statement, “I feel that my relationship with this friend is very close” on a 7-point Likert-type scale with anchors of strongly disagree (1) and strongly agree (7) (M = 6.37, SD = 1.15). We chose a single-item measure to minimize response burden and because previous research has successfully employed single-item measures of closeness (e.g. Cummings et al., 2006; DeHart et al., 2004; Jones et al., 1985).

CT use

The daily questionnaire asked participants to indicate how much they had communicated with their friend that day FtF and via five CTs: video chat (e.g. Skype or FaceTime), phone calls, emails, social media (e.g. Facebook or Twitter), or text messaging. An “other” category was also included. Response options were derived from Jiang and Hancock’s (2013) measures of interaction length. For FtF, video chat, and phone calls, response options ranged from less than 15 minutes (1) to more than 1.5 hours (7) in 15-minute increments. For email, social media, text, and other, response options ranged from fewer than 5 messages (1) to more than 30 messages (7) in 5-message increments. Participants who indicated that they had not used a particular CT to communicate with their friend that day were assigned a score of “0” for that CT. Participants reported using an average of 1.05 (SD = 0.77) CTs per day to communicate with their friends. Mean CT use frequency was as follows: video chat: M = 0.25, SD = 1.04; phone calls: M = 0.56, SD = 1.39; email: M = 0.01, SD = 0.12; social media: M = 0.33, SD = 0.92; text messaging: M = 2.23, SD = 2.60; other: M = 0.08, SD = 0.63. Examples of responses to the “other” category were Snapchat and letter. Because of their low use, email and “other” were excluded from the analyses. We also omitted social media because of its breadth as a category, both in variety of potential sites used and types of interactions possible. Thus, with the exception of the analysis of number of CTs used, this study focused on the following CTs: video chat, phone calls, and text messaging.

Analyses

Correlations between study variables and intraclass correlations for use of each CT are presented in Table 1. Daily responses are nested within participants, and the intraclass correlations for use of each CT were significant, indicating nonindependence of responses. Adjustments for nonindependence in daily data varied depending on the analysis and are described separately for each analysis below.

Table 1.

Correlations between study variables.

	M (SD)	2	3	4	5	6	7	8	9	10
1. Age	18.53 (1.53)	.18*	−.15⁺	−.04	.17*	.01	.02	.01	−.01	−.12
2. Relationship length (years)	6.72 (4.24)		−.19*	.09	.37**	.13	−.08	−.10	−.06	−.20*
3. Sex	–			.40**	−.05	.09	−.02	−.07	.03	.09
4. Friend’s sex	–				.50**	−.04	−.30**	−.17*	−.34**	−.27*
5. Sex composition	–					−.06	−.36**	−.31**	−.31**	−.52*
6. Closeness	6.28 (1.16)						.20*	.12	.17*	.15⁺
7. Daily number of CTs used	1.05 (0.77)						.47**	.25**	.45**	.47**
8. Video chat	0.24 (1.04)							.44**	.09*	.12**
9. Phone calls	0.56 (1.39)								.41**	.19**
10. Text messaging	2.23 (2.60)									.59**

SD: standard deviation; CTs: communication technologies.

Variables 1–6 are at the individual level (N = 155). Variables 7–10 are at the daily level (N = 1,075). Daily-level variables were group-mean-centered to remove nonindependence by subtracting each participant’s mean score for that variable from his or her daily score for that variable. Bolded values on the diagonal indicate intraclass correlations within participants. For sex and friend’s sex, −1 = male and 1 = female. For sex composition, −1 = a cross-sex friendship and 1 = a same-sex friendship.

p < .10; *p < .05; **p < .001.

Hypothesis 1 was that daily use of CTs would be complementary, such that higher daily use of a given CT was associated with higher daily use of other CTs. Bivariate correlations between use of each CT were used to examine this hypothesis and are reported in Table 1. Group-mean-centered scores were used to account for statistical nonindependence in scores across days. Each participant’s mean use of a particular CT was subtracted from his or her daily use of that CT to remove nonindependence before correlations were calculated.

Hypothesis 2 was that number of CTs used (H2a) and amount of CT use (H2b) would be positively associated with closeness. H2a was tested by calculating the partial correlation between closeness and the average daily number of CTs each participant used, controlling for age, sex, friend’s sex, and sex composition of the friendship. H2b was tested by calculating the partial correlation between closeness and average daily frequency of using each CT, using the same control variables.

Hypothesis 3 was that complementarity of CT use would be associated with friendship closeness, such that complementarity would be higher in closer friendships. These associations were tested using multilevel modeling. Predictor variables were grand-mean-centered to aid interpretation of the results. Age, sex, friend’s sex, and sex composition of the friendship (same-sex or cross-sex) were significant control variables in at least one model, so they were included as controls in all models. For each model, participants’ daily use of a given CT (video chat, phone calls, or text messaging) served as the outcome variable. Level 1 predictors were participants’ daily use of each other CT (e.g. phone calls and text messaging when video chat was the outcome variable). Level 2 predictor variables included age, sex, friend’s sex, sex composition, and closeness. Cross-level interactions were the two-way interactions between closeness and use of each CT. Results of the multilevel models are shown in Table 2.

Table 2.

Multilevel models predicting communication technology use as a joint function of use of other communication technologies and closeness.

	Outcome variables
	Video chat	df	Phone calls	df	Text messaging	df
Age	.06⁺ (.04)	133	.01 (.05)	132	−.07 (.09)	137
Sex	−.09 (.08)	141	.12 (.10)	139	.18 (.19)	142
Friend’s sex	.06 (.08)	142	−.27* (.10)	140	−.10 (.19)	144
Sex composition	−.20* (.08)	150	.03 (.10)	150	−.92** (.19)	145
Closeness	.03 (.05)	153	.05 (.06)	153	.19 (.12)	149
Video chat	–	–	.16** (.04)	1056	.27** (.07)	1015
Phone calls	.07*(.02)	1055	–	–	.26** (.05)	1028
Text messaging	.07* (.01)	1032	.12** (.02)	1022	–	–
Video chat × closeness	–	–	−.14* (.05)	1054	.03 (.10)	997
Phone calls × closeness	−.03 (.02)	1055	–	–	.17* (.05)	986
Text messaging × closeness	.02 (.01)	823	.05** (.02)	815	–	–

Each column reports the multilevel model for the outcome variable listed on the top of that column. Variables were grand-mean-centered to aid interpretation. Coefficients are unstandardized, and standard errors are in parentheses. Degrees of freedom were rounded to the nearest whole number. For sex and friend’s sex, −1 = male and 1 = female. For sex composition, −1 = a cross-sex friendship and 1 = a same-sex friendship.

p < .10; *p < .05; **p < .001.

Results

Complementary CT use

Positive associations (i.e. complementarity) emerged for participants’ daily use of video chat and phone calls (r = .09, p = .002), daily use of video chat and text messaging (r = .12, p < .001), and daily use of phone calls and text messaging (r = .19, p < .001). On days participants used a particular CT more, they also used other CTs more. Hypothesis 1 was supported.

Closeness and CT use

The partial correlation between closeness and average daily number of CTs used, controlling for age, sex, friend’s sex, and sex composition of the friendship, was significant, r = .19, p = .02. Regarding the frequency of CT use, closeness was not significantly associated with average daily use of video chat (r = .12, p = .16), phone calls (r = .15, p = .07), or text messaging (r = .13, p = .12) after accounting for the control variables. Hypothesis 2a was supported, and Hypothesis 2b was not supported.

Closeness and complementarity

Video chat

Contrary to Hypothesis 3, closeness did not moderate the associations between daily use of video chat and daily use of phone calls or text messaging. Although video chat was complementary with phone calls and text messaging, this complementarity was not associated with closeness.

Phone calls

Closeness moderated the associations between daily phone calls and both daily video chat (b = −.14, standard error [SE] = 0.05, p = .01) and daily text messaging (b = .05, SE = 0.02, p < .001). To explore these interactions, the associations between use of phone calls and use of video chat and text messaging were examined at low (1 SD below the mean) and high (the scale maximum of 7) levels of closeness. The association between phone calls and video chat was positive (i.e. complementary) at low levels of friendship closeness (b = .32, SE = 0.09, p < .001) but not significant at high levels of friendship closeness (b = .07, SE = 0.04, p = .11). The association between phone calls and text messaging was stronger at higher levels of friendship closeness (low closeness: b = .06, SE = 0.02, p = .01; high closeness: b = .16, SE = 0.02, p < .001). Thus, phone calls were complementary with video chat at low, but not high, levels of friendship closeness, whereas phone calls and text messaging were more complementary in closer friendships. Hypothesis 3 was supported for complementarity of phone calls and text messaging but not for complementarity of phone calls and video chat.

Text messaging

The interaction between closeness and daily video chat was not significant, but there was a significant interaction between closeness and daily phone calls (b = .17, SE = 0.05, p = .001). Text messaging and phone calls were complementary at high levels of closeness (b = .37, SE = 0.05, p < .001) but not at low levels of closeness (b = .07, SE = 0.08, p = .40). In summary, use of text messaging was complementary with video chat regardless of closeness level and was complementary with phone calls at high, but not low, levels of closeness. Hypothesis 3 was supported for complementarity of text messaging and phone calls but not for complementarity of text messaging and video chat.

Discussion

This study used CCT and MMT to examine closeness and complementary CT use in first-year college students’ long-distance friendships. Consistent with CCT (Dutta-Bergman, 2004), video chat, phone calls, and text messaging were used complementarily. MMT was also partially supported (Haythornthwaite, 2005); participants who reported higher closeness with their friends also reported using a higher average daily number of CTs, although use of individual CTs was not associated with closeness. Regarding the integration of CCT and MMT, closeness was useful in predicting complementarity, but the nature of this association varied. The results suggest that in close, long-distance friendships, patterns of daily CT use are complex and dependent upon the nature of the friendship. The results also suggest utility in combining CCT and MMT to understand these patterns. Both of these implications will be explored below.

Although CCT predicts that use of various channels will be positively related (Dutta-Bergman, 2004), little research has investigated the nature of these patterns and factors that are associated with them. MMT predicts that closeness is associated with patterns of channel use (Haythornthwaite, 2005), and extensions of MMT predict that closeness is associated not only with use of various channels but also with integration of those channels (Caughlin and Sharabi, 2013). Building on this previous research, several of the complementary associations we observed varied as a function of closeness. However, the role of closeness in complementarity varied depending on the particular CTs examined. As expected, complementarity of phone calls and text messaging was stronger in closer relationships. However, phone calls and video chat were complementary at low levels of closeness but not at high levels of closeness. These findings suggest a pattern in which higher closeness is associated with a stronger link between use of phone calls and text messaging but a weaker link between use of phone calls and video chat.

It might be the case that in closer friendships, video chat and phone calls serve similar purposes—to engage in synchronous, cue-rich (relative to text messaging) conversations. Phone calls and text messaging, however, might allow friends to fulfill motivations for connection but in different ways. Phone calls are often seen as more personal than text-based forms of communication (Stafford et al., 1999; Yang et al., 2014) and are used by long-distance friends for intimate conversations, particularly in closer friendships (Utz, 2007). This interpretation is consistent with niche theory, which posits that the extent of competition between media is a function of the degree of overlap in the gratifications that those media serve (Dimmick et al., 2000). It is also consistent with previous research findings that complementarity can vary as a function of whether or not pairs of CTs offer similar cues (Ruppel and Burke, 2015). It might be the case that CCT is more applicable to certain channels or gratifications, whereas others are better explained by niche theory (e.g. Dimmick et al., 2000; Liu and Yang, 2016) or the displacement hypothesis (Bryant and Fondren, 2009). In particular, channel complementarity might be more likely among channels with dissimilar cues or that serve different interpersonal functions (e.g. mundane everyday conversation vs more personal or intimate conversation). The current findings suggest that video chat and phone calls might serve similar functions but text messaging and phone calls serve dissimilar functions (though still within the overarching function of communicating with that particular friend).

It is also possible that text messaging is more habitual, whereas video chat and phone calls tend to be intentional or planned ahead of time. This finding is consistent with the higher intraclass correlation for text messaging than for video chat or phone calls, which suggests that text messaging is used more consistently from day to day. More intentional or planned use could have led closer participants to choose video chat over phone calls, or vice versa, whereas the same decision process would not occur for text messaging. In contrast, participants whose friendships were less close used fewer channels overall, so they might have engaged in different decision-making processes. Future research should further examine the roles of closeness, channel characteristics, and habit in channel use.

Theoretical implications

Combining the perspectives of CCT and MMT allowed us to examine the “web” of CTs used in long-distance friendships as it relates to characteristics of the relationship, specifically closeness. The finding that the role of closeness in complementarity varied for different pairs of CTs is consistent with previous research that has found differential associations between closeness and various CTs (Cummings et al., 2006; Utz, 2007) and suggests that identifying how different CTs fit into the landscape of communication in friendships is important for understanding communication as a whole in those friendships. Integrating CCT and MMT can allow for a more comprehensive understanding of CT use in relationships that is informed by channel characteristics (via CCT), relational characteristics (via MMT), and their interaction. The role of closeness in complementarity is also noteworthy because the average level of closeness participants reported was relatively high. This suggests that even within the context of close friendships, variations in closeness are important for understanding patterns of CT use.

Understanding of how channel complementarity varies as a function of relational characteristics can help predict patterns of channel use in long-distance friendships and can potentially be extended to identify individual or relational outcomes of these patterns and how these patterns can be altered. For example, if the complementarity of phone calls and text messaging gets stronger as friends get closer, but complementarity of phone calls and video chat is tempered (as found in this study), these changing interdependencies might also alter what friends talk about and how they talk about it, ultimately affecting the quality and trajectory of the friendship (and, reciprocally, patterns of CT use in that friendship). Furthermore, the extent to which conversation topics are carried from one topic to another has been implicated in relationship quality (Caughlin and Sharabi, 2013) and could potentially be affected by changing patterns of complementarity.

Limitations and future directions

Although this study contributes to existing knowledge about complementarity CT use and closeness in long-distance friendships, some limitations and directions for future research warrant discussion. We examined patterns of CT use in students’ close long-distance friendships at a time when maintenance of those friendships is particularly salient to them: in the transition to college. Furthermore, women were overrepresented in the sample, and we did not ask participants about the nature of their friendship in terms of issues such as romantic attraction (e.g. Guerrero and Chavez, 2005; Reeder, 2000). Future research is needed to determine how generalizable the results in this study are to other populations and relationship types.

Our focus on daily CT use in close friendships provides a relatively conservative test of CCT, MMT, and their integration for two reasons. First, the observed patterns occurred on a daily basis, as opposed to being general patterns over time that would be more likely to exhibit complementarity (Ruppel and Burke, 2015). That is, it is more likely that high use of one CT would be associated with high use of another CT over a period of weeks or months than over the period of a single day. Second, although we focused on close friendships because we expected them to be more strongly motivated to maintain contact (a key factor of CCT) and to use a wider range of CTs more frequently, the range of closeness observed was restricted by the focus on close friendships. That this study suggests utility in integrating the two theories to understand interdependencies in or patterns of CT use despite these restrictions implies that future research examining these patterns over longer periods of time or in a wider range of relationships will likely also benefit from this theoretical integration.

Long-term studies should be also used to explore the causal direction of the associations between complementary CT use and closeness and the potential for cumulative effects over time. CCT posits that motivations drive channel use (Dutta-Bergman, 2004), and MMT expects tie strength to affect channel use (Haythornthwaite, 2005). Thus, both theories would predict that closeness causes complementarity. Longitudinal research should confirm this prediction. Long-term studies could also explore the possibility that the relatively small daily effects sizes observed in this study might accumulate and become magnified over time.

Finally, this study focused on the frequency of CT use. However, some research finds that the integration of communication content across various channels is also relevant in close relationships (Caughlin and Sharabi, 2013). Future research might extend our proposed integration of CCT and MMT to examine the extent to which relational closeness is associated not only with complementary channel use but also with complementary communication content across different channels. This extension could add another dimension of understanding to research on the “web” of CT use in close relationships.

Conclusion

The transition to college is a stressful time during which the maintenance of pre-existing friendships is a salient concern for students. Understanding interrelationships between CTs is important for elucidating communication in long-distance friendships. The results of this study support a combination of CCT and MMT and contribute to a more holistic understanding of CT use in long-distance friendships. Future research should continue to explore these interrelationships to demonstrate causal relationships between complementarity and closeness and patterns of transition between particular CTs.

Footnotes

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

Author biographies

Erin K Ruppel (PhD, University of Arizona) is an assistant professor at the University of Wisconsin-Milwaukee. Her research focuses on how people integrate different modes of communication in interpersonal relationships.

Tricia J Burke (PhD, University of Arizona) is an assistant professor at Texas State University. Her research focuses on the intersection of interpersonal communication processes in the contexts of health and technology.

Maura R Cherney is a PhD student at the University of Wisconsin- Milwaukee. Her research focuses on communication technology use and its effects in interpersonal and educational contexts.

References

Baym

Ledbetter

(2009) Tunes that bind? Information, Communication & Society 12: 408–427.

Becker

JAH

Johnson

Craig

, et al. (2009) Friendships are flexible, not fragile: turning points in geographically-close and long-distance friendships. Journal of Social and Personal Relationships 26: 347–369.

Bryant

Fondren

(2009) Displacement effects. In: Nabi

Oliver

(eds) The SAGE Handbook of Media Processes and Effects. Thousand Oaks, CA: SAGE, pp. 505–516.

Caughlin

Sharabi

(2013) A communicative interdependence perspective of close relationships: the connections between mediated and unmediated interactions matter. Journal of Communication 63: 873–893.

Cummings

Lee

Kraut

(2006) Communication technology and friendship during the transition from high school to college. In: Kraut

Brynin

Kiesler

(eds) Computers, Phones, and the Internet: Domesticating Information Technology. New York: Oxford University Press, pp. 265–278.

DeHart

Pelham

Murray

(2004) Implicit dependency regulation: self-esteem, relationship closeness, and implicit evaluations of close others. Social Cognition 22: 126–146.

Dimmick

Kline

Stafford

(2000) The gratification niches of personal e-mail and the telephone. Communication Research 27: 227–248.

Dutta-Bergman

(2004) Interpersonal communication after 9/11 via telephone and internet: a theory of channel complementarity. New Media & Society 6: 659–673.

Ellison

Steinfield

Lampe

(2007) The benefits of Facebook “friends”: Social capital and college students’ use of online social network sites. Journal of Computer-Mediated Communication 12: 1143–1168.

10.

Fehr

. (1999) Stability and commitment in friendships. In: Adams

Jones

(eds) Handbook of Interpersonal Commitment and Relationship Stability. New York: Kluwer Academic/Plenum Publishers, pp. 259–280.

11.

Guerrero

Chavez

(2005) Relational maintenance in cross-sex friendships characterized by different types of romantic intent: an exploratory study. Western Journal of Communication 69: 339–358.

12.

Haythornthwaite

(2005) Social networks and Internet connectivity effects. Information, Communication & Society 8: 125–147.

13.

Jiang

Hancock

(2013) Absence makes the communication grow fonder: geographic separation, interpersonal media, and intimacy in dating relationships. Journal of Communication 63: 556–577.

14.

Johnson

(2001) Examining the maintenance of friendships: are there differences between geographically close and long-distance friends? Communication Quarterly 49: 424–435.

15.

Johnson

Becker

JAH

(2010) CMC and the contextualization of “friendship”: how friendships have changed with the advent of new methods of interpersonal communication. In: Wright

Webb

(eds) Computer-Mediated Communication in Personal Relationships. New York: Peter Lang Publishing, pp. 225–243.

16.

Johnson

Becker

JAH

Craig

, et al. (2009) Changes in friendship commitment: comparing geographically close and long-distance young-adult friendships. Communication Quarterly 57: 395–415.

17.

Johnson

Haigh

Becker

JAH

, et al. (2008) College students’ use of relational management strategies in email in long-distance and geographically close relationships. Journal of Computer-Mediated Communication 13: 381–404.

18.

Johnson

Haigh

Craig

, et al. (2009) Relational closeness: comparing undergraduate college students’ geographically close and long-distance friendships. Personal Relationships 16: 631–646.

19.

Jones

Carpenter

Quintana

(1985) Personality and interpersonal predictors of loneliness in two cultures. Journal of Personality and Social Psychology 48: 1503–1511.

20.

Ledbetter

(2009) Patterns of media use and multiplexity: associations with sex, geographic distance and friendship interdependence. New Media & Society 11: 1187–1208.

21.

Ledbetter

Broeckelman-Post

Krawsczyn

(2011) Modeling everyday talk: differences across communication media and sex composition of friendship dyads. Journal of Social and Personal Relationships 28: 223–241.

22.

Ledbetter

Taylor

Mazer

(2016) Enjoyment fosters media use frequency and determines its relational outcomes: toward a synthesis of uses and gratifications theory and media multiplexity theory. Computers in Human Behavior 54: 149–157.

23.

Liu

Yang

(2016) Media niche of electronic communication channels in friendship: a meta-analysis. Journal of Computer-Mediated Communication 21: 451–466.

24.

Oswald

Clark

(2003) Best friends forever?: high school best friendships and the transition to college. Personal Relationships 10: 187–196.

25.

Pistole

Roberts

(2011) Measuring long-distance romantic relationships: a validity study. Measurement and Evaluation in Counseling and Development 44: 63–76.

26.

Piwek

Joinson

(2016) “What do they Snapchat about?” Patterns of use in time-limited instant messaging service. Computers in Human Behavior 54: 358–367.

27.

Ramirez

Broneck

(2009) “IM me”: instant messaging as relational maintenance and everyday communication. Journal of Social and Personal Relationships 26: 291–314.

28.

Ranney

Troop-Gordon

(2012) Computer-mediated communication with distant friends: relations with adjustment during students’ first semester in college. Journal of Educational Psychology 104: 848–861.

29.

Reeder

(2000) “I like you. as a friend”: the role of attraction in cross-sex friendship. Journal of Social and Personal Relationships 17: 329–348.

30.

Ruppel

Burke

(2015) Complementary channel use and the role of social competence. Journal of Computer-Mediated Communication 20: 37–51.

31.

Stafford

(2005) Maintaining Long-Distance and Cross-Residential Relationships. Mahwah, NJ: Lawrence Erlbaum Associates.

32.

Stafford

Kline

Dimmick

(1999) Home e-mail: relational maintenance and gratification opportunities. Journal of Broadcasting & Electronic Media 43: 659–669.

33.

Utz

(2007) Media use in long-distance friendships. Information, Communication & Society 10: 694–713.

34.

Wiseman

(1986) Friendship: bonds and binds in a voluntary relationship. Journal of Social and Personal Relationships 3: 191–211.

35.

Yang

Brown

Braun

(2014) From Facebook to cell calls: layers of electronic intimacy in college students’ interpersonal relationships. New Media & Society 16: 5–23.