TY - JOUR
AU1 - Taylor, Samuel, Hardman
AU2 - Bazarova, Natalya, N
AB - Abstract This study investigates how people use multiple media in their romantic relationships, as well as the effects of using multiple media on relational closeness. Revisiting media multiplexity theory, we proposed multimedia frequency, multimedia disclosure, multimedia frequency variability, and multimedia disclosure variability as types of media multiplexity that vary as a function of relational closeness and geographic distance. A six-week longitudinal study tracked 151 college students’ communications across media with their romantic partners. Results suggest that relational closeness is independently associated with each type of media multiplexity. Geographic distance moderated this relationship, such that geographically-close couples engaged in more multimedia frequency than long-distance couples, but the inverse was found for multimedia disclosure. Lagged analysis suggested a curvilinear relationship between multimedia disclosure and relational closeness in the following week. Results advance knowledge about how people integrate multiple media into their relationships and the effects of communication technologies on romantic intimacy. Modern romantic relationships have become “mixed-media relationships,” with dynamic online and offline communication unfolding across multiple media (Parks, 2017). Romantic partners blend face-to-face interactions (FtF), texting, phone calls, social network posts, and other media to communicate, and these multiple media interactions impact their relational quality (Jiang & Hancock, 2013; Toma & Choi, 2016). As a result, understanding how people connect across multiple media has presented itself as a pressing question for understanding social relationships in general and romantic relationships in particular (Caughlin & Sharabi, 2013). Yet, most research focuses on the use of different media in isolation or in competition with one another, rather than in combination (e.g., Taylor & Ledbetter, 2017). A focus on multiple media requires a holistic view of how media operate as a conglomerate rather than as separate entities. One useful perspective for studying multimedia1 communication is media multiplexity theory (MMT; Haythornthwaite, 2005). MMT provides an account of how the use of multiple media in everyday life can simultaneously engage and disengage social connectivity. Despite the potential of this theory to explain mixed-media relationships, MMT provides a limited account of what it means to connect on multimedia (Taylor & Bazarova, 2018), and there is equivocation about how media multiplexity evolves over time (Ledbetter, Taylor, & Mazer, 2016). In this manuscript, we revisit MMT’s propositions with the goal of capturing both patterns of interpersonal multimedia communication and their reciprocal connections with relational closeness. First, we propose new constructs of media multiplexity to address how media work in combination with one another and to elaborate additional propositions of MMT. Second, we test the effects of multimedia communication on relational closeness over six weeks to examine media multiplexity longitudinally. Third, we explore the differences in media multiplexity between geographically-close relationships (GCR) and long-distance relationships (LDR). Media multiplexity theory There is a pervasive and persistent debate about whether mediated communication displaces or enhances interpersonal relationships (Burke & Kraut, 2014). MMT aimed to answer this debate by arguing that media use can simultaneously displace and enhance personal relationships, and shifts questions from a focus on one medium to all media used by partners. MMT advances five propositions about interpersonal relationships and multimedia communication. First, it proposes a positive association between tie strength and the number of media used for communication (Haythornthwaite, 2005). Tie strength characterizes relational closeness, defined by the amount of time, level of intimacy, emotional intensity, and reciprocal services within a relationship (Granovetter, 1973). The second proposition is that communication patterns differ by tie strength, not by medium. In other words, it is the relationship, not the characteristics of the media, that drives use and sharing via multiple media. Close relational partners use many media for similar types of exchanges, because they are interdependent for support, information, goals, tasks, and so forth. Third, MMT proposes a reciprocal, linear, causal relationship between tie strength and media use across time. Despite placing tie strength as MMT’s central causal agent, media use is also expected to shape a dyad’s relationship over time. Fourth, changes in media use affect weak ties more than they do strong ties, as strong ties adapt their communication patterns to new media more easily than weak ties (Haythornthwaite, 2005). Adding or subtracting a new communication medium (e.g., a couple stops using Facebook but continues to talk regularly on the phone) may go relatively unnoticed for strong ties, but is disruptive to weak ties. Fifth, media use is organized into tiers of media based on tie strength, such that all social ties within a group are granted access to the same one or two media (e.g., Facebook and email), while more private media (e.g., text messaging) are introduced as tie strength increases (Haythornthwaite, 2000). MMT has been tested across several different types of social relationships (e.g., friends, extended families) and many different types of media (e.g., social network sites, mobile phones, and online gaming; for a review, see Ledbetter, 2015). There has been consistent support for the first proposition—a positive relationship between tie strength and media use (Baym & Ledbetter, 2009; Ruppel, Burke, & Cherney, 2017)—although the majority of MMT studies have focused on relational closeness, or “a subjective experience of intimacy, emotional affinity, and psychological bonding” (Ledbetter et al., 2011, p. 34), instead of tie strength. A recent meta-analysis found that relational closeness differentiates media use among friends, making relational closeness a cogent mechanism for media multiplexity (Liu & Yang, 2016). The first proposition of MMT was extended beyond the number of media to show an association between relational closeness and frequency of communication on a single medium, although negative attitudes toward a specific medium weaken this association (Ledbetter et al., 2016). The other MMT propositions have received less interrogation, with the exception of the fourth proposition. Consistent with this proposition, increasing or reducing media use on one channel was more disruptive for weak ties than strong ties among extended family (Taylor & Ledbetter, 2017). Despite growing support for MMT, there are several consistent critiques that call for its revisiting. One concern addresses the limited explanation of multimedia communication beyond a simple count of media (Parks, 2017). Another criticism is that research on MMT is reliant on cross-sectional data; therefore, the longitudinal effect of media multiplexity on relational closeness is uncertain. A final critique of the theory is an “undifferentiated articulation of relationship characteristics relevant to media use” (Ledbetter, 2015, p. 371). A more diverse set of relational and contextual characteristics that predict media use, such as geographic distance, would improve the relational focus of MMT. In the following sections, we addressing the critiques in order to continue building MMT. We contextualize multimedia within romantic relationships, because romantic relationships tend to use many media to communicate (Toma & Choi, 2016) and there is variance in relational closeness across the various stages of romantic relationships, such as initial attraction or mutual commitment (Caughlin & Sharabi, 2013). Explicating media multiplexity If the number of media does not fully capture the complexity of how romantic partners connect over multiple media, then the first step in revisiting MMT requires reconceptualizing multimedia communicative processes. Here, we develop types of media multiplexity beyond the number of media: (a) multimedia frequency, (b) multimedia disclosure, (c) multimedia frequency variability, and (d) multimedia disclosure variability. Each type of media multiplexity is integrated into MMT, to offer a more comprehensive account of mixed-media relationships. Number of media The basis of media multiplexity is a positive association between relational closeness and number of media (Haythornthwaite, 2005). Haythornthwaite and Wellman (1998) originally observed that those with closer social ties tended to use more media to communicate. This finding was replicated by others (Baym & Ledbetter, 2009). Hristova, Musolesi, and Mascolo (2014) concluded that the number of media used was a good indicator of relational closeness. We anticipate replication of this foundational proposition of MMT. H1: Relational closeness is positively associated with the number of media used in romantic relationships. Multimedia frequency Apart from the number of media, previous research also established that individual medium frequency is associated with relational closeness (e.g., Ledbetter et al., 2016). Yet, to consider media multiplexity requires moving the lens beyond a single medium to capture the multimedia ecosystem. Extending medium use frequency research to a multiplexity-level, we propose multimedia frequency as a communication construct capturing the total frequency of communication on all media used, which encompasses both the number of media and frequency of communication across all of them. Building upon MMT, closer partners should not only use a greater number of media overall, but also have a higher frequency of communication across all media compared to less-close partners. Number of media and multimedia frequency represent distinct aspects of multimedia communication, because communicators can interact on various media at different rates (Ledbetter, 2009). As such, our theorizing adds to the first proposition of MMT. Advancing beyond the number of media, relational closeness should also share a positive association with multimedia frequency. H2: Relational closeness is positively associated with multimedia frequency in romantic relationships, controlling for number of media. Multimedia disclosure While most MMT studies focus on communication frequency, Haythornthwaite (2002) argues that social relationships are also influenced by what is communicated in the multimedia landscape. One way to capture communication content is via examination of self-disclosure processes, or the extent to which relational partners share personal information, thoughts, or feelings with each other (Laurenceau, Barrett, & Pietromonaco, 1998). As with medium use frequency, the intimacy of self-disclosure on a single medium is associated with greater relational closeness (Ruppel, 2015); thus, how self-disclosure occurs across media is likely to influence romantic relationships. The totality of intimate self-disclosure across media can represent another type of multimedia communication. We refer to the amount of intimate self-disclosure communicated over multiple media as multimedia disclosure. Close relational partners disclose more with one another than with strangers or acquaintances (Bazarova, 2012). According to MMT (Haythornthwaite, 2002), relational closeness between individuals is the factor determining what is shared, not whether the relationship is being maintained FtF or in a mediated environment. While there are differences in self-disclosure between individual media in romantic relationships (Jiang & Hancock, 2013), when all media used by romantic partners are considered, we anticipate a positive linear relationship between closeness and multimedia disclosure. Therefore, we extend the first proposition of MMT to predict a positive association between relational closeness and multimedia disclosure. H3: Relational closeness is positively associated with multimedia disclosure in romantic relationships, controlling for number of media and multimedia frequency. Multimedia variability Although studying multimedia frequency and disclosure broadens the understanding of media multiplexity, considering multimedia invites questions about the connections between those media. The connections between media impact impression management, emotion regulation, and relational closeness (Caughlin & Sharabi, 2013; Ramirez & Wang, 2008; Scissors & Gergle, 2013). We argue that there is a need to address the consistency of communication across media against the backdrop of multimedia frequency and multimedia disclosure. Whereas the latter constructs emphasize similarity across media, investigating consistency offers a way to understand differences between media. There are several current approaches to understanding consistency between media. According to channel complementarity theory, communication between media is relatively consistent, as people use multiple media to pursue similar goals (Dutta-Bergman, 2004). Frequency of use of one medium to pursue a goal predicts frequency of use of another medium for that same goal (Ruppel et al., 2017). The communication interdependence perspective (CIP, Caughlin & Sharabi, 2013) defines consistency as an integration between FtF and mediated communication. Partners who converse about the same topics across media have a high level of integration. Mode segmentation is a type of media inconsistency described in the CIP. Mode segmentation occurs when couples limit some conversation to either FtF or mediated channels. We refer to the degree of heterogeneity in communication across all media used by partners as multimedia variability, to capture differences in medium-level communication. When variability is high, people use media in diverse ways. In parallel to multimedia frequency and multimedia disclosure, we theorize about both multimedia frequency variability, or how frequency of use differs from medium-to-medium, and multimedia disclosure variability, or the degree of discrepancy in the intimacy of self-disclosure across media. When multimedia frequency variability is low, partners use all media similarly (e.g., with either low or high frequency across all media); when multimedia frequency variability is high, partners use different media with different frequencies (e.g., high frequency of text messaging and low frequency of email). In terms of multimedia disclosure variability, partners may engage in intimate self-disclosure on all media, or they may use some media for intimate disclosure and other media for non-intimate disclosure. Two perspectives can inform how relational closeness may be connected to multimedia variability. First, the second proposition of MMT argues that communication is homogenous across media when analyzed within a single dyad. Because MMT predicts that relationships drive communication, there should be little variability within the dyad across media, because the overarching needs and motivations for connecting with a partner do not change when a partner picks up the telephone versus answers a text message (Haythornthwaite, 2002, 2005). Thus, variability should be quite low in close relationships, because partners’ communication patterns should not be determined by the medium. Although never explicitly tested as multimedia variability, this pattern of consistent frequency of media use has been observed among close friends (Ruppel et al., 2017) and in romantic relationships during conflict communication (Scissors & Gergle, 2013). Second, CIP posits that communication integration across media is a sign of relational closeness, but a recent survey of college students in romantic relationships found that limiting some conversations to FtF communication (mode segmentation) was associated with greater relational closeness (Caughlin & Sharabi, 2013). The association between relational closeness and mode segmentation appears to depend on medium and conversation type: limiting conversations to computer-mediated and not FtF communication predicted less relational closeness (Caughlin & Sharabi, 2013), and integration of conflict had a negative association with relational closeness (Pusateri, Roaché, & Wang, 2015). Because MMT and CIP predict divergent associations between relational closeness and multimedia variability, we pose the following RQ: RQ1: How is relational closeness associated with (a) multimedia frequency variability and (b) multimedia disclosure variability? Media multiplexity across time Thus far, we have focused on broadening MMT’s theoretical scope to explain mixed-media relationships, but another limitation of MMT research is a gap in understanding the connections between relational closeness and multimedia over time. Early longitudinal studies of MMT supported the role of both the number of media and medium use frequency in relationship-building (Haythornthwaite, 2000, 2002), which was consistent with the original MMT proposition of a positive, reciprocal feedback loop between relational closeness and media use. However, subsequent research of MMT has been mostly limited to cross-sectional data examining the association between medium use frequency and relationship characteristics (for a review, see Ledbetter, 2015). In a rare longitudinal study of medium use frequency among adolescents, instant messaging frequency was predictive of relational quality six months later, but the reverse association was not supported (Valkenburg & Peter, 2009). Thus, our next hypothesis predicts a positive effect of multimedia communication on subsequent relational closeness. H4: (a) Number of media, (b) multimedia frequency, and (c) multimedia disclosure are positively associated with relational closeness in the following week. Multimedia variability should also matter for relational closeness because patterns of media use regulate how partners maintain closeness to one another. People demonstrate preferences and expectations for where certain types of information are shared (Bazarova, 2012). When those expectations are violated, there are negative relational consequences (Taylor & Ledbetter, 2017). According to the communication interdependence perspective, mode segmentation of limiting some conversations to only FtF may be advantageous for relational closeness (Caughlin & Sharabi, 2013). Whereas Caughlin and Sharabi compared FtF to all types of mediated communication combined, the use of mediated channels can be further differentiated. For example, partners maintain psychological closeness with brief, lightweight mobile text messages, in addition to regular, intimate discussions on other channels (Ling, 2008). One explanation for this finding is that the selection of media for some conversations signals their relational importance. Multimedia variability could foster closeness by communicating meta-messages to a partner about the relationship, which go beyond the other types of messages exchanged (Taylor & Ledbetter, 2017). Thus, variability in communication media may be a communication strategy that facilitates closeness. H5: (a) Multimedia frequency variability and (b) multimedia disclosure variability are positively associated with relational closeness in the following week. Although we expect a positive relationship between media use and relational closeness, there is mounting evidence that excessive multimedia communication can increase stress and relational strain, due to the demands of being constantly available. The mutual expectation of availability from the multimedia environment draws social relationships together, while simultaneously coercing partners to sacrifice autonomy (Ling, 2016). In college friendships, partners’ high expectations of constant connectedness over mobile media both increased and decreased relational satisfaction (Hall & Baym, 2012). Together, these findings suggest a curvilinear relationship between media multiplexity and relational closeness across time that is akin to the Goldilocks effect, showing that while a moderate amount of media is beneficial, too little and too much media use can be detrimental for one’s well-being (Przybylski & Weinstein, 2017). The constant communication afforded by the multimedia environment is good for intimacy, but high amounts of multimedia communication may be detrimental to the health of the relationship. H6: (a) Number of media, (b) multimedia frequency, and (c) multimedia disclosure have a curvilinear relationship with relational closeness in the following week. Media multiplexity and geographic distance Although media multiplexity places relational closeness as the primary mechanism in predicting media use, relational closeness reflects only a small portion of the known relational and contextual characteristics that determine how romantic partners communicate (Taylor & Ledbetter, 2017). A couple’s geographic distance likely impacts their multimedia communication patterns, as LDRs rely exclusively on mediated communication to maintain intimacy. MMT anticipates that similar patterns of multimedia use will hold for both collocated and geographically-dispersed partners (Haythornthwaite, 2005), but does not make predictions about whether geographic distance will amplify or reduce these patterns. On average, LDRs tend to use more electronically-mediated forms of communication than GCRs (Jiang & Hancock, 2013). However, mediated communication also matters for GCRs, even though they have the option of frequent FtF interaction (Toma & Choi, 2016). In fact, the majority of phone calls and text messages are sent between ties who are geographically close (Ling, 2016). A consistent, yet surprising, finding of geographic distance in romantic relationships is that LDRs tend to report greater levels of intimacy and satisfaction than GCRs (Stafford, 2010). While there is conflicting evidence on the role of media in determining this effect (see Jiang & Hancock, 2013; Stafford & Merolla, 2007), what is clear is that geographic distance influences how romantic partners use media, because partners’ needs and motivations for media use vary depending on their geographic distance (Stafford & Merolla, 2007). Therefore, we will investigate how partners’ geographic distance moderates the association between relational closeness and media multiplexity. RQ2: How does geographic distance moderate the association between relational closeness and (a) number of media, (b) multimedia frequency, and (c) multimedia disclosure? RQ3: How does geographic distance moderate the association between relational closeness and (a) multimedia frequency variability and (b) multimedia disclosure variability? Method Participants 151 students from a large, northeastern university in the United States were recruited from communication and information science courses to participate, in exchange for either partial course credit or $20. Ages ranged from 18 to 42 (M = 20.45, SD = 2.51), with the following education levels: 15.9% freshmen, 25.2% sophomores, 27.2% juniors, 24.5% seniors, 6.6% graduate students, and .7% unspecified. The majority of the participants were female (78.8%); 60.9% were Caucasian, 23.2% Asian, 8.6% Hispanic/Latino, 2.6% African American, and 4.6% other. Participants were eligible for participation if they were able to identify someone who they had a romantic relationship with or a romantic interest in (Theiss & Solomon, 2006). On a weekly basis, we requested they report on their most established romantic relationship, which resulted in a range of relationships, from romantic interest to committed romantic relationships. During the first week of data collection, 60.3% of respondents were seriously dating their romantic partner, 22.5% were casually dating, 14.6% were friends, 1.3% were married or partnered, 0.7% were engaged, and 0.7% described their relationship as “complicated.” The median length of romantic involvement with the other individual was 1 year (M = 1.53 years, SD = 1.84 years). There were 145 participants that reported on heterosexual relationships (97.3%). Those with serious dating relationships had the highest relational closeness (M = 6.68, SD = .51), followed by casual relationships (M = 5.66, SD = .98), and then friends (M = 5.10, SD = 1.28; F[2, 142] = 44.84, p < .001 based on the analysis of relational closeness at week 1, with all the means being significantly different from one another at p < .05 or greater. The relationship types of engaged (n = 1), married/partnered (n = 2), and other (n = 1) were excluded from this comparison because of their small sample sizes. Participants were asked to self-identify their relationship as GCR or LDR (Jiang & Hancock, 2013): 53.6% of participants reported their relationship as GCR and 46.4% as LDR. GCR and LDRs did not differ in relational closeness (MGCR = 6.16, SD = 1.14; MLDR = 6.13, SD = 1.12; p > .05). Procedures Participants were contacted once a week for six continuous weeks to report on their relational closeness and media use. The once-a-week interval was selected to obtain longitudinal data and to avoid being too demanding on the participants. Knobloch and Theiss (2010) suggested collecting data on relationships for six weeks, because relationships can change significantly within a few weeks. Individuals, rather than dyads, were utilized for the study because each member of a dyad experiences their own level of relational closeness. Individuals consenting to participate completed a pre-study survey of demographic information and were emailed a link to the survey each Wednesday, and they had until Sunday evening to complete the survey. There were 40 students (21%) who dropped out over the course of the six weeks, which is a comparable rate to other longitudinal studies (e.g., Knobloch & Theiss, 2010; Theiss & Solomon, 2006). Independent sample t-tests found that participants who dropped out (M = 5.76, SD = 1.34) reported less relational closeness than those who completed the study (M = 6.20, SD = .99; p = .001). Measures Number of media Participants indicated which media they had used with their romantic partner in the previous week from a list of commonly-used media: (a) face-to-face; (b) voice telephone; (c) mobile text messaging; (d) email; (e) desktop instant messaging; (f) public social networking sites (e.g., Facebook, Twitter, Instagram); (g) video chat; (h) Snapchat; (i) other forms of online communication (e.g., online gaming). The number of media participants used each week was determined by summing the number of media marked from this list (M = 3.27, SD = 1.47), similar to Haythornthwaite’s (2001) study. Means and standard deviations from week-to-week for all variables are available in the online Supplementary Appendix. Multimedia frequency Adapting Haythornthwaite’s (2001) procedure from monthly to weekly media reports, we designed an original measure of multimedia frequency that accounts for both the number of media and total communication across media at the same time. For each week, participants indicated their frequency of use for each of the nine media listed above, using a 7-point Likert scale (0 = never to 6 = daily). All frequencies were summed together to create a multimedia frequency variable, indicating a total amount of communication across all the media they had used in the previous week with their partner (M = 16.13, SD = 5.33). Multimedia disclosure We extended Valkenburg and Peter’s (2009) measure of instant messaging self-disclosure to operationalize the frequency of intimate self-disclosure across media. Respondents were asked to answer six items for each of the media they had used in the previous week with their partner (e.g., “When you were using text messaging to communicate with your partner in the past week, how much did you tell about … your personal feelings?”). The 5-point scale ranged from “I tell nothing about this” (1) to “I tell everything about this” (5; alpha = .87). To determine the intimacy of self-disclosure across all the media, we added together reports for each media into one total score of multimedia disclosure (M = 9.53, SD = 5.33). Multimedia variability Multimedia variability captured consistency in how different media were used throughout the week. In accordance with Roberson, Sturman, and Simons’ (2007) recommendation for calculating within-level variability for cross-level interaction effects, we used a measure of standard deviation to operationalize multimedia variability. Multimedia frequency variability was represented by the standard deviation around the mean of multimedia frequency, and multimedia disclosure variability was represented by the standard deviation around the mean of multimedia disclosure within each week. Higher variability indicates more variation in communication on different media. Lower variability indicates less variation in frequency of use and disclosure behaviors across different media, regardless of whether frequency of use or disclosure intimacy was low or high. Each variability score is relative to the number of media used that week, and media that were not used in that week were not included in the assessment of variability. The 14% of participants who reported only using either one or zero media in a week were excluded from the calculations of variability. Multimedia frequency variability had a mean of 1.33 (SD = .77), and the mean of multimedia disclosure variability was 0.90 (SD = .60). Relational closeness Relational closeness was measured using 7 items from Vangelisti and Caughlin’s (1997) relational closeness measure (e.g., “How close is this person to you?”). This measure uses a 7-point Likert scale (with 1 = not at all to 7 = very much), with alphas ranging from 0.93 to 0.97 throughout the six-week study (M = 6.15, SD = 1.13). Results Relational closeness predicting media multiplexity Analytical strategy We employed multilevel models to account for the nesting in the data from repeated measurements and to analyze how participant responses varied during the six weeks. The data had a two-level structure; six repeated observations (e.g., multimedia frequency, relational closeness) were nested within each individual. Data were analyzed using the lme4 package in R v3.3.2. All the models included a random intercept for each individual, with a random slope for week. Variables were grand-mean centered to estimate the overall effect of relational closeness on media multiplexity (Snijders & Bosker, 2012). Geographic distance was dummy coded. Cross-level interaction effects for geographic distance and relational closeness were included in all the models. Covariates in each model were week of data collection (weeks one to six), participant age, partner age, participant gender, partner gender, and length of relationship. In multimedia variability analyses, the corresponding multimedia communication variables (e.g., multimedia frequency for multimedia frequency variability) were controlled. Bivariate correlations and intraclass correlations are reported in the online Supplementary Appendix. Number of media H1 predicted a positive relationship between relational closeness and number of media, with a potential moderation from geographic distance (RQ2a). Table 1 contains the results for the multilevel model, with relational closeness and geographic closeness predicting the number of media. Consistent with H1, there was a significant, positive association between relational closeness and number of media. There was no main effect for geographic distance, nor was the interaction of relational closeness and geographic distance significant. Thus, as relational closeness increased, the number of media also increased, regardless of geographic distance, in line with the foundational proposition of MMT. Table 1 Relational Closeness and Geographic Distance Prediction of Media Multiplexity Media Multiplexity Number of Media Multimedia Frequency Multimedia Disclosure Multimedia Frequency Variability Multimedia Disclosure Variability b SE b SE b SE b SE b SE Intercept 4.22*** .53 18.05*** 1.82 7.41** 1.15 1.18 .33 .85 .20  Distance −.32 .17 −.59 .59 1.16** .37 .05 .07 −.08 .07 Slopes  RC .34*** .08 1.53*** .27 .41* .16 .19*** .05 .23*** .04  Week −.27*** .02 −.65*** .09 .46*** .06 −.03 .02 .00 .01  No. of media 3.23*** .13 2.15*** .10 .31*** .03 .17*** .02  MF .14*** .02 −.05*** .01  MD −.02*** .01 Interactions  Distance × RC .03 .11 −1.00* .40 .67** .24 −.04 .07 −.06 .05 Residuals .67 10.06 3.08 .36 .14 Random effects  Participant 1.25 14.31 7.33 .08 .10  Week .05 .41 .20 .00 .01 Media Multiplexity Number of Media Multimedia Frequency Multimedia Disclosure Multimedia Frequency Variability Multimedia Disclosure Variability b SE b SE b SE b SE b SE Intercept 4.22*** .53 18.05*** 1.82 7.41** 1.15 1.18 .33 .85 .20  Distance −.32 .17 −.59 .59 1.16** .37 .05 .07 −.08 .07 Slopes  RC .34*** .08 1.53*** .27 .41* .16 .19*** .05 .23*** .04  Week −.27*** .02 −.65*** .09 .46*** .06 −.03 .02 .00 .01  No. of media 3.23*** .13 2.15*** .10 .31*** .03 .17*** .02  MF .14*** .02 −.05*** .01  MD −.02*** .01 Interactions  Distance × RC .03 .11 −1.00* .40 .67** .24 −.04 .07 −.06 .05 Residuals .67 10.06 3.08 .36 .14 Random effects  Participant 1.25 14.31 7.33 .08 .10  Week .05 .41 .20 .00 .01 Note: Distance reference category is geographically-close relationships. Coefficients in Table 1 are from the full model, including interaction effects. MD = multimedia disclosure; MF = multimedia frequency; RC = relational closeness. *p < .05, **p < .01, ***p < .001. Table 1 Relational Closeness and Geographic Distance Prediction of Media Multiplexity Media Multiplexity Number of Media Multimedia Frequency Multimedia Disclosure Multimedia Frequency Variability Multimedia Disclosure Variability b SE b SE b SE b SE b SE Intercept 4.22*** .53 18.05*** 1.82 7.41** 1.15 1.18 .33 .85 .20  Distance −.32 .17 −.59 .59 1.16** .37 .05 .07 −.08 .07 Slopes  RC .34*** .08 1.53*** .27 .41* .16 .19*** .05 .23*** .04  Week −.27*** .02 −.65*** .09 .46*** .06 −.03 .02 .00 .01  No. of media 3.23*** .13 2.15*** .10 .31*** .03 .17*** .02  MF .14*** .02 −.05*** .01  MD −.02*** .01 Interactions  Distance × RC .03 .11 −1.00* .40 .67** .24 −.04 .07 −.06 .05 Residuals .67 10.06 3.08 .36 .14 Random effects  Participant 1.25 14.31 7.33 .08 .10  Week .05 .41 .20 .00 .01 Media Multiplexity Number of Media Multimedia Frequency Multimedia Disclosure Multimedia Frequency Variability Multimedia Disclosure Variability b SE b SE b SE b SE b SE Intercept 4.22*** .53 18.05*** 1.82 7.41** 1.15 1.18 .33 .85 .20  Distance −.32 .17 −.59 .59 1.16** .37 .05 .07 −.08 .07 Slopes  RC .34*** .08 1.53*** .27 .41* .16 .19*** .05 .23*** .04  Week −.27*** .02 −.65*** .09 .46*** .06 −.03 .02 .00 .01  No. of media 3.23*** .13 2.15*** .10 .31*** .03 .17*** .02  MF .14*** .02 −.05*** .01  MD −.02*** .01 Interactions  Distance × RC .03 .11 −1.00* .40 .67** .24 −.04 .07 −.06 .05 Residuals .67 10.06 3.08 .36 .14 Random effects  Participant 1.25 14.31 7.33 .08 .10  Week .05 .41 .20 .00 .01 Note: Distance reference category is geographically-close relationships. Coefficients in Table 1 are from the full model, including interaction effects. MD = multimedia disclosure; MF = multimedia frequency; RC = relational closeness. *p < .05, **p < .01, ***p < .001. Multimedia frequency H2 predicted that relational closeness was positively associated with multimedia frequency: the frequency of communication across all media. As predicted by H2, relational closeness was positively associated with multimedia frequency (see Table 1). Addressing RQ2b, about whether geographic distance moderates media multiplexity, geographic distance interacted with relational closeness to predict multimedia frequency. The positive association between relational closeness and multimedia frequency was stronger for GCRs than LDRs. Multimedia disclosure Our third hypothesis concerned multimedia disclosure, or the overall level of disclosure intimacy across media. We predicted that relational closeness would share a positive relationship with multimedia disclosure (H3), but RQ2c noted that geographic distance might moderate this association. The number of media and multimedia frequency were positively associated with multimedia disclosure (Table 1). Consistent with H3, more relational closeness predicted more multimedia disclosure. This relationship was modified by a significant interaction between geographical distance and relational closeness (RQ2c), with relational closeness being a stronger predictor of overall disclosure intimacy in LDRs than GCRs. Multimedia frequency variability RQ1a was concerned with the association between relational closeness and multimedia frequency variability. As seen in Table 1, more relational closeness was associated with more variation in frequency of use across media. Romantic partners reporting a higher degree of closeness were more likely to use some media very frequently and other media less frequently. On the other hand, people reporting low relational closeness to their romantic partner were more likely to have consistent reports of frequency across media. Furthermore, multimedia frequency variability was positively associated with the number of media, but negatively associated with multimedia frequency. There was no significant association between geographic location and multimedia frequency variability, nor an interaction between geographic distance and relational closeness (RQ3a). Multimedia disclosure variability Similar to the results on multimedia frequency variability, individuals with more relational closeness reported more multimedia disclosure variability (RQ1b). People in close romantic relationships used some media for intimate disclosure and other media for non-intimate conversations (see Table 1). Geographic distance was not a significant predictor, and it did not interact with relational closeness (RQ3b). As with the previous analysis, multimedia disclosure variability was positively associated with number of media, but negatively associated with multimedia disclosure. Predicting relational closeness in the following week Analytical strategy A lagged variable analysis was run to determine if multimedia variables predicted relational closeness in the subsequent week (Theiss & Solomon, 2006). Relational closeness at wave t for weeks two through six were the outcomes of interest; multimedia variables from wave t-1 (number of media, multimedia frequency, multimedia disclosure, multimedia frequency variability, and multimedia disclosure variability) were predictors. Number of media, multimedia frequency, and multimedia disclosure were evaluated in separate models to reduce multicollinearity; the models with multimedia frequency and disclosure also included the corresponding multimedia variability variable as a predictor. The dependent variable at wave t-1 was entered as a covariate to control for relational closeness from the previous week. All predictor variables were grand-mean centered. Age, partner age, gender, partner gender, and relationship length were used as covariates, and a random effect for individual was included in the models. An interaction effect for each media multiplexity variable, with geographic distance, was also included in the models (see Table 2). Table 2 Media Multiplexity Prediction of Relational Closeness in the Following Week Relational Closeness at Wave t Number of Media Multimedia Frequency Multimedia Disclosure b SE b SE b SE Intercept 6.04 .23 6.05 .14 6.08 .26  Distance .02 .04 .00 .02 .02 .04 Slopes  t − 1 multimedia use .02 .02 .01 .00 .01 .01  t − 1 multimedia variability .02 .04 .03 .05  t − 1 relational closeness .92*** .02 .93*** .02 .92*** .02 Geographic distance interaction  t − 1 multimedia use −.02 .03 −.00 .01 −.01 .01  t − 1 multimedia variability .00 .06 .06 .07 Residuals .28 .26 .26 Random effect  Participant .00 .00 .00 Relational Closeness at Wave t Number of Media Multimedia Frequency Multimedia Disclosure b SE b SE b SE Intercept 6.04 .23 6.05 .14 6.08 .26  Distance .02 .04 .00 .02 .02 .04 Slopes  t − 1 multimedia use .02 .02 .01 .00 .01 .01  t − 1 multimedia variability .02 .04 .03 .05  t − 1 relational closeness .92*** .02 .93*** .02 .92*** .02 Geographic distance interaction  t − 1 multimedia use −.02 .03 −.00 .01 −.01 .01  t − 1 multimedia variability .00 .06 .06 .07 Residuals .28 .26 .26 Random effect  Participant .00 .00 .00 Note: Distance is dummy coded (0 = GCR, 1 = LDR). Independent variables were grand-mean centered. Coefficients in Table 2 are from the full model, including interaction effects. GCR = geographically-close relationships; LDR = long-distance relationships. ***p < .001. Table 2 Media Multiplexity Prediction of Relational Closeness in the Following Week Relational Closeness at Wave t Number of Media Multimedia Frequency Multimedia Disclosure b SE b SE b SE Intercept 6.04 .23 6.05 .14 6.08 .26  Distance .02 .04 .00 .02 .02 .04 Slopes  t − 1 multimedia use .02 .02 .01 .00 .01 .01  t − 1 multimedia variability .02 .04 .03 .05  t − 1 relational closeness .92*** .02 .93*** .02 .92*** .02 Geographic distance interaction  t − 1 multimedia use −.02 .03 −.00 .01 −.01 .01  t − 1 multimedia variability .00 .06 .06 .07 Residuals .28 .26 .26 Random effect  Participant .00 .00 .00 Relational Closeness at Wave t Number of Media Multimedia Frequency Multimedia Disclosure b SE b SE b SE Intercept 6.04 .23 6.05 .14 6.08 .26  Distance .02 .04 .00 .02 .02 .04 Slopes  t − 1 multimedia use .02 .02 .01 .00 .01 .01  t − 1 multimedia variability .02 .04 .03 .05  t − 1 relational closeness .92*** .02 .93*** .02 .92*** .02 Geographic distance interaction  t − 1 multimedia use −.02 .03 −.00 .01 −.01 .01  t − 1 multimedia variability .00 .06 .06 .07 Residuals .28 .26 .26 Random effect  Participant .00 .00 .00 Note: Distance is dummy coded (0 = GCR, 1 = LDR). Independent variables were grand-mean centered. Coefficients in Table 2 are from the full model, including interaction effects. GCR = geographically-close relationships; LDR = long-distance relationships. ***p < .001. Relational closeness in the following week H4 and H5 predicted a positive association between relational closeness and the five variables of media multiplexity: number of media, multimedia frequency, multimedia disclosure, multimedia frequency variability, and multimedia disclosure frequency. In each model, relational closeness had a positive association with relational closeness in the prior week (Table 2). However, no type of media multiplexity at wave t-1 shared a linear relationship with relational closeness at wave t, and there was no interaction effect for geographic distance for either multimedia use (RQ2) or multimedia variability (RQ3). Next, we tested for a curvilinear relationship between media multiplexity (i.e., number of media, multimedia frequency, and multimedia disclosure) and relational closeness in the subsequent week (H6). There was no significant curvilinear effect for number of media (b = .000, SE = .008, p > .05) or multimedia frequency (b = −.000, SE = .000, p > .05) on relational closeness in the following week. Multimedia disclosure had a significant, curvilinear relationship with relational closeness in the subsequent week (b = −.002, SE = .001, p < .01). In the curvilinear model, the linear effect of multimedia disclosure was positive (b = .017, SE = .008, p < .05). Multimedia disclosure predicted increases in relational closeness in the following week, but high amounts of multimedia disclosure predicted slight decreases in relational closeness in the next week (Figure 1). H6a and H6b were not supported, and H6c was supported.2 Figure 1 View largeDownload slide Relational closeness in the following week by multimedia disclosure. Linear estimate, b = .017, SE = .008; curvilinear estimate, b = −.002, SE = .001. Figure 1 View largeDownload slide Relational closeness in the following week by multimedia disclosure. Linear estimate, b = .017, SE = .008; curvilinear estimate, b = −.002, SE = .001. Discussion The goal of this study was to revisit MMT to extend the understanding of how people integrate multiple media for daily communication in today’s diverse multimedia landscape. First, the construct of media multiplexity was expanded to include new patterns of multimedia communication, in addition to the traditional measurement of number of media. Relational closeness shared a positive association with multimedia frequency, multimedia disclosure, multimedia frequency variability, and multimedia disclosure variability, after controlling for the number of media. Second, the longitudinal effects of MMT were tested over six weeks. The only significant relationship was a curvilinear association between multimedia disclosure and relational closeness in the subsequent week. Third, geographic distance was included as a contextual moderator: GCRs tended to use multimedia more frequently than LDRs, but LDRs engaged in more disclosure across media than GCRs. Below, we discuss implications of these findings for understanding romantic relationships and mixed-media relationships in general. Revisiting media multiplexity MMT expanded a discussion of how communication technologies influence social relationships beyond “use of a single medium to examination of all media available” (Haythornthwaite, 2005, p. 126). A primary challenge of MMT in explaining the use of all media available was reducing the complexity of multimedia communication to a single account of the number of media (Parks, 2017). This paper addresses this limitation of MMT by introducing new dimensions of multimedia communication and building them into MMT. The totality of multimedia communication was represented by multimedia frequency (i.e., the total use of all media available) and multimedia disclosure (i.e., aggregated intimacy of self-disclosure over all media). Controlling for the number of media, relational closeness was positively associated with both multimedia frequency and multimedia disclosure, showing that close romantic partners have both a higher communication frequency and more intimate disclosure across all media they use. The positive association between relational closeness and all three measures of media multiplexity (number of media, multimedia frequency, and multimedia disclosure) suggests that relational closeness is not only signaled by a more diverse set of media, but also by how much partners communicate and disclose across these media. Whereas previous extensions of MMT examined the frequency of communication on each medium separately (e.g., Ledbetter et al., 2016), multimedia frequency considers a totality of communication, thus accounting for both the number of media and total communication across media together. Similarly, multimedia disclosure reflects both the number of media and the combined frequency of disclosure on all media used by a couple. This makes communicating a great deal on one or two media not identical to communicating a great deal on three or four media, suggesting that accounts of romantic relationships that neglect their use of multimedia communication fail to capture the experience of these mixed-media relationships. One future direction would be to explore the underlying reasons for the association between relational closeness and greater multimedia engagement. For example, partners may be using multimedia to satisfy the need to maintain availability of each other (Ling, 2008) and to create a sense of connected presence (Licoppe, 2004) in cases where digital media complement or compensate for the lack of (in case of long-distance couples) FtF communication. Understanding availability via multimedia could move MMT toward predicting well-being, as the availability of a close partner regulates affective well-being (Taylor & Bazarova, 2018). Beyond the summation of multimedia communication, two types of multimedia variability were introduced to offer a holistic view of how couples weave together media in their daily interactions. Multimedia frequency variability refers to the degree of discrepancy between how often media are used for communication within a dyad, with more variability indicating less consistency. Similarly, multimedia disclosure variability captures the degree of discrepancy in disclosure intimacy across all media, with more variability indicating more differentiation between media. Greater relational closeness predicted more frequency variability and disclosure variability. Close romantic partners tended to use some media more often and other media less often, as compared to those with a low degree of closeness, and this tendency extended to their self-disclosure patterns. These results held when controlling for other types of media multiplexity, indicating that multimedia variability captures a distinct facet of media multiplexity. The variability among close partners did not support the second proposition of MMT, which suggests that communication content differs by relational closeness, and not by medium (Haythornthwaite, 2005). Results were more consistent with the CIP, according to which close partners may limit some of their interactions to specific media (Caughlin & Sharabi, 2013). Based on these results and CIP findings, we propose revising MMT’s second proposition to state that relational closeness predicts greater heterogeneity in how media are used. One explanation for the variability across media in close relationships is that partners tailor communication to the affordances of a medium (Bazarova, 2012). While the original formulation of MMT was affordance-agnostic, downplaying the affordances of each medium, the positive association between relational closeness and multimedia variability suggests that close partners differentiate between media for different communication and relational functions. Future MMT research can explore how people understand mediums in relation to each another or select between media by unpacking communication goals and routines over these media in relation to different affordances they offer (Bazarova & Choi, 2014). As emphasized in the recent integration of the multiple goals framework and CIP (Caughlin, Basinger, & Sharabi, 2017), social goals shape how people use their communication resources, creating interdependencies of conversations across different media. The use of media, in this sense, carries a relational meaning, operating as a proxy for different communication routines. Not only the frequency and intimacy of conversations, but the transfer of topics across media or, vice versa, segmentations of certain topics to certain media, can be laden with relational meanings and implications. This suggests a complex structure of how media are interwoven to satisfy communication goals and routines, contingent on conversational topics, media affordances, and relational experiences. The new multimedia constructs can be extended to study how multimedia use patterns reflect and drive interpersonal processes in family, friendship, and other interpersonal contexts. Another research direction would be to explore how a dyad’s patterns of multimedia use interplay with social factors at different scales (e.g., network-level trends capturing peer effects; Hage & Noseleit, 2018). Whereas MMT examines the use of media through a relational lens that is conditional upon a communication partner, it would be interesting to examine how a dyad’s media ecology—all use of media and their interrelated patterns within a dyad—exists in relation to a broader social media ecology of each of the partners, including all forms of mobile and social media used by each partner separately. This will help not only to delineate the boundaries of MMT, but also understand how MMT, as a dyad-focused theory, interfaces with frameworks focused on an individual’s use of multimedia (e.g., channel complementarity theory; Ruppel et al., 2017). The effect of media multiplexity on relational closeness Whereas previous research using cross-sectional data found a positive association between communication on a single medium and relational closeness (e.g., Ledbetter et al., 2011), this six-week study found no evidence for a linear relationship between media multiplexity and relational closeness from week-to-week. Our analysis did find a curvilinear relationship between relational closeness and multimedia disclosure in the previous week. Intimate disclosure across all media was associated with increased intimacy in the next week, up to moderate levels. At high amounts of multimedia disclosure, the positive association turned into a minor negative association, which suggests diminishing returns on relational closeness. The curvilinear effect challenges the third proposition of MMT, which hypothesizes a reciprocal, linear effect of media use on relational closeness over time. Our results favored the Goldilocks hypothesis explanation of how media impact social relationships (Przybylski & Weinstein, 2017). That is, a moderate amount of media use, not too little or too much, has the most positive effect on a relationship. There are several implications emerging from the analysis of media multiplexity on relational closeness over time. First, these results refocus questions away from the mere use of media, towards the intimacy of conversations happening across media, because self-disclosure on all media was the only significant predictor of relational closeness in the next week. This resembles findings about how self-disclosure on a single medium mediates the effect of medium use frequency on relational quality (Valkenburg & Peter, 2009). Future work about the effects of media multiplexity should continue examining what is communicated across media, in addition to the frequency of multimedia use. Second, the curvilinear effect counters narratives that mobile and social media are holistically good or bad for intimacy. Media can be beneficial or harmful for relational closeness, depending on the amount of intimate self-disclosure over all media. Third, because relational closeness was high for romantic partners in our study, this may have resulted in a ceiling effect, where the other effects of media use on relational closeness in the next week were no longer observable, similar to results in Burke and Kraut’s study (2014). Since weak ties are more affected by changes in media than close ties (Taylor & Ledbetter, 2017), partners with less relational closeness, such as adolescent friends or extended family members, may undergo more relational changes because of different multimedia use patterns. Moving beyond relational closeness Finally, we contribute to the development of MMT’s relational mechanism by exploring how geographic distance moderates the association between relational closeness and media multiplexity. While multimedia research often focuses on LDRs, GCRs in our study used media more frequently than LDRs. One explanation for this finding is that GCRs use mobile communication to check in throughout the day, though they are likely to see each other FtF too (Ling, 2008). These findings point to the need for more attention to how GCRs maintain relationships via multimedia (see also, Toma & Choi, 2016). The association between relational closeness and multimedia disclosure also depended on distance. LDRs tended to disclose more across media as their relationships increased in closeness, compared to GCRs. This is consistent with research showing that couples in long-distance relationships engage in behavioral adaptation to compensate for the lack of FtF communication: LDRs tend to share more intimate disclosures than GCRs across media, maintaining intimacy despite the lack of everyday, mundane interactions (Jiang & Hancock, 2013; Stafford, 2010). The number of media, multimedia frequency variability, and multimedia disclosure variability did not differ by geographic distance. In addition, a lagged analysis found that the effect of media multiplexity on relational closeness in the following week was not dependent upon geographic distance. The moderating effect of geographic distance signals that contextual factors qualify the positive association between relational closeness and some facets of media multiplexity. Thus, geographic distance contributes to the growing body of research on conditions that attenuate the strength of this association (Ledbetter et al., 2011). The moderating effect of various contextual factors represent one way to find the boundary conditions of MMT’s propositions. Theory development is also needed on distinct relational antecedents of media multiplexity, independent of relational closeness. Future extensions of MMT toward other relational characteristics that are known to influence communication behavior, such as relational turbulence (Theiss & Solomon, 2006) or attachment (Luo, 2014), will continue to build its explanatory power. Where does MMT stand, given this study? Overall, these results propose revisions to several original propositions of MMT (Haythornthwaite, 2005). Adding to the first proposition, not only is there a positive relationship between relational closeness and the number of media, but relational closeness is also associated with more multimedia frequency and multimedia disclosure. Furthermore, this association is qualified by other factors, such as geographic distance. Revising the second proposition, instead of consistency within a dyad’s media ecology use, relational closeness predicts inconsistency across a couple’s media ecology. Finally, rethinking the linear prediction of the third position, we posit an inverted, U-shaped, curvilinear effect of media multiplexity on relational closeness over time, and this effect is a result of the content of communication across media, rather than the frequency. Limitations This research has some limitations, many of which relate to the nature of our sample and self-report data. The majority of participants in our sample were White females, limiting the generalizability of our findings. Next, this analysis only followed participants for six weeks; while this is a long enough time to capture changes in relationships (Knobloch & Theiss, 2010), the limited time span may have affected our ability to observe the association between media multiplexity and relational closeness, especially since closeness was relatively stable across the six weeks. Testing media multiplexity across a larger time span, such as six months, may be required to witness this effect (Valkenburg & Peter, 2009). The stability of relational closeness may explain the differences between our results and previous studies (e.g., Baym & Ledbetter, 2009). Third, although multimedia frequency was independent of number of media, there may be some overlap between multimedia frequency and general communication frequency. Controlling for both the general communication frequency and number of media may be necessary to untangle the multimedia effect from the underlying process happening in media contexts. Conclusion As more and more media are weaved into relationships, communication research is challenged to explain the mix of media people use for daily conversations (Parks, 2017). By revisiting MMT, this study addresses this challenge by explicating multiple ways that media are interrelated for interpersonal communication and examining the longitudinal effects of multimedia communication on closeness in romantic relationships. Relational closeness was uniquely tied to each type of media multiplexity introduced here, which showcases the complex web of media close partners use to fulfill their goals in modern romantic relationships. The curvilinear association between multimedia disclosure and closeness suggests moderation, using the Goldilocks principle for reaping optimal effects from this web of media on relational closeness. Supplementary material Supplementary material are available at Journal of Communication online. Footnotes 1 Multimedia was selected as our term of choice rather than multimodal, for a variety of reasons. Using face-to-face interactions as a medium is consistent with Haythornthwaite’s (2005) discussion of the term media. “Media” has become synonymous with “channel” in research on mixed-media relationships (Parks, 2017). Mode tends to refer to how a message was encoded, such as text, speech, and so on. MMT is explicitly about media, not modes of communication (Haythornthwaite, 2005). 2 Geographic distance did not interact with the curvilinear relationship between multimedia disclosure and relational closeness in the following week. References Baym , N. K. , & Ledbetter , A. M. ( 2009 ). Tunes that bind?: Predicting friendship strength in a music-based social network . Information, Communication & Society , 12 , 408 – 427 . doi:10.1080/13691180802635430 Google Scholar Crossref Search ADS Bazarova , N. N. ( 2012 ). Public intimacy: Disclosure interpretation and social judgments on Facebook . Journal of Communication , 62 , 815 – 832 . doi:10.1111/j.1460-2466.2012.01664.x Google Scholar Crossref Search ADS Bazarova , N. N. , & Choi , Y. H. ( 2014 ). Self-disclosure in social media: Extending the functional approach to disclosure motivations and characteristics on social network sites . Journal of Communication , 64 , 635 – 657 . doi:10.1111/jcom.12106 Google Scholar Crossref Search ADS Burke , M. , & Kraut , R. ( 2014 , April). Growing closer on Facebook: Changes in tie strength through social network site use. Paper presented at the SIGCHI Conference on Human Factors in Computing Systems (CHI ‘14), Toronto, Ontario, Canada. doi: 10.1145/2556288.2557094 Caughlin , J. P. , Basinger , E. D. , & Sharabi , L. L. ( 2017 ). The connections between communication technologies and relational conflict: A multiple goals and communication interdependence perspective. In J. A. Samp (Ed.), Communicating interpersonal conflict in close relationships: Contexts, challenges, and opportunities (pp. 57 – 72 ). New York, NY : Routledge . Caughlin , J. P. , & Sharabi , L. L. ( 2013 ). A communicative interdependence perspective of close relationships: The connections between mediated and unmediated interactions matter . Journal of Communication , 63 , 873 – 893 . doi:10.1111/jcom.12046 Dutta-Bergman , M. J. ( 2004 ). Complementarity in consumption of news types across traditional and new media . Journal of Broadcasting & Electronic Media , 48 , 41 – 60 . doi:10.1207/s15506878jobem4801_3 Google Scholar Crossref Search ADS Granovetter , M. S. ( 1973 ). The strength of weak ties . American Journal of Sociology , 78 , 1360 – 1380 . doi:10.1086/225469 Google Scholar Crossref Search ADS Hage , E. , & Noseleit , F. ( 2018 ). It takes at least two to tango: A population-level perspective on interrelated patterns of media use . Journal of Computer-Mediated Communication , 23 , 222 – 242 . doi:10.1093/jcmc/zmy011 Google Scholar Crossref Search ADS Hall , J. A. , & Baym , N. K. ( 2012 ). Calling and texting (too much): Mobile maintenance expectations, (over)dependence, entrapment, and friendship satisfaction . New Media & Society , 14 , 316 – 331 . doi.org/10.1177/1461444811415047 Google Scholar Crossref Search ADS Haythornthwaite , C. ( 2000 ). Online personal networks: Size, composition, and media use among distance learners . New Media & Society , 2 , 195 – 226 . doi:10.1177/14614440022225779 Google Scholar Crossref Search ADS Haythornthwaite , C. ( 2001 ). Exploring multiplexity: Social network structures in a computer supported distance learning class . The Information Society , 17 , 211 – 226 . https://doi.org/10.1080/01972240152493065 Google Scholar Crossref Search ADS Haythornthwaite , C. ( 2002 ). Strong, weak, and latent ties and the impact of new media . The Information Society , 18 , 385 – 401 . doi:10.1080/01972240290108195 Google Scholar Crossref Search ADS Haythornthwaite , C. ( 2005 ). Social networks and Internet connectivity effects . Information, Communication & Society , 8 , 125 – 147 . doi:10.1080/13691180500146185 Google Scholar Crossref Search ADS Haythornthwaite , C. , & Wellman , B. ( 1998 ). Work, friendship, and media use for information exchange in a networked organization . Journal of the American Society for Information Science , 49 , 1101 – 1114 . Google Scholar Crossref Search ADS Hristova , D. , Musolesi , M. , & Mascolo , C. ( 2014 , June). Keep your friends close and your Facebook friends closer: A multiplex network approach to the analysis of offline and online social ties. Paper presented at the 2014 AAAI International Conference on Weblogs and Social Media (ICWSM ‘14), Ann Arbor, MI. Jiang , L. C. , & Hancock , J. T. ( 2013 ). Absence makes the communication grow fonder: Geographic separation, interpersonal media, and intimacy in dating relationships . Journal of Communication , 63 , 556 – 577 . doi:10.1111/jcom.12029 Google Scholar Crossref Search ADS Knobloch , L. K. , & Theiss , J. A. ( 2010 ). An actor-partner interdependence model of relational turbulence: Cognitions and emotions . Journal of Social and Personal Relationships , 27 , 595 – 619 . doi:10.1177/0265407510368967 Google Scholar Crossref Search ADS Laurenceau , J. P. , Barrett , L. F. , & Pietromonaco , P. R. ( 1998 ). Intimacy as an interpersonal process: The importance of self-disclosure, partner disclosure, and perceived partner responsiveness in interpersonal exchanges . Journal of Personality and Social Psychology , 74 , 1238 – 1251 . doi:10.1037/0022-3514.74.5.1238 Google Scholar Crossref Search ADS PubMed Ledbetter , A. M. ( 2009 ). Patterns of media use and multiplexity: Associations with sex, geographic distance and friendship interdependence . New Media & Society , 11 , 1187 – 1208 . doi:10.1177/1461444809342057 Google Scholar Crossref Search ADS Ledbetter , A. M. ( 2015 ). Media multiplexity theory: Technology use and interpersonal tie strength. In D. O. Braithwaite & P. Schrodt (Eds.), Engaging theories within interpersonal communication: Multiple perspectives ( 2nd ed. , pp. 363 – 376 ). Thousand Oaks, CA : Sage . Ledbetter , A. M. , Mazer , J. P. , DeGroot , J. M. , Meyer , K. R. , Mao , Y. , & Swafford , B. ( 2011 ). Attitudes toward online social connection and self-disclosure as predictors of Facebook communication and relational closeness . Communication Research , 38 , 27 – 53 . doi:10.1177/0093650210365537 Google Scholar Crossref Search ADS Ledbetter , A. M. , Taylor , S. H. , & Mazer , J. P. ( 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 . doi:10.1016/j.chb.2015.07.053 Google Scholar Crossref Search ADS Licoppe , C. ( 2004 ). ‘Connected’ presence: the emergence of a new repertoire for managing social relationships in a changing communication technoscape . Environment and Planning: Society and Space , 22 , 135 – 156 . https://doi.org/10.1068/d323t Google Scholar Crossref Search ADS Ling , R. ( 2008 ). New tech, new ties: How mobile communication is reshaping social cohesion . Cambridge, MA : MIT Press . Ling , R. ( 2016 ). Soft coercion: Reciprocal expectations of availability in the use of mobile communication . First Monday , 21 . doi:10.5210/fm.v21i9.6814 Liu , D. , & Yang , C. C. ( 2016 ). Media niche of electronic communication channels in friendship: A meta-analysis . Journal of Computer-Mediated Communication , 21 , 451 – 466 . Google Scholar Crossref Search ADS Luo , S. ( 2014 ). Effects of texting on satisfaction in romantic relationships: The role of attachment . Computers in Human Behavior , 33 , 145 – 152 . doi:10.1016/j.chb.2014.01.014 Google Scholar Crossref Search ADS Parks , M. R. ( 2017 ). Embracing the challenges and opportunities of mixed-media relationships . Human Communication Research , 43 , 505 – 517 . doi:10.1111/hcre.12125 Google Scholar Crossref Search ADS Przybylski , A. K. , & Weinstein , N. ( 2017 ). A large-scale test of the Goldilocks hypothesis: Quantifying the relations between digital-screen use and the mental well-being of adolescents . Psychological Science , 28 , 204 – 215 . doi:10.1177/0956797616678438 Google Scholar Crossref Search ADS PubMed Pusateri , K. B. , Roaché , D. J. , & Wang , N. ( 2015 ). The role of communication technologies in serial arguments: A communicative interdependence perspective . Argumentation and Advocacy , 52 , 44 – 60 . doi:10.1080/00028533.2015.11821860 Google Scholar Crossref Search ADS Ramirez , A. , & Wang , Z. ( 2008 ). When online meets offline: An expectancy violations theory perspective on modality switching . Journal of Communication , 58 , 20 – 39 . doi:10.1111/j.1460-2466.2007.00372.x Google Scholar Crossref Search ADS Roberson , Q. M. , Sturman , M. C. , & Simons , T. L. ( 2007 ). Does the measure of dispersion matter in multilevel research?: A comparison of the relative performance of dispersion indexes . Organizational Research Methods , 10 , 564 – 588 . doi:10.1177/1094428106294746 Google Scholar Crossref Search ADS Ruppel , E. K. ( 2015 ). Use of communication technologies in romantic relationships: Self-disclosure and the role of relationship development . Journal of Social and Personal Relationships , 32 , 667 – 686 . doi:10.1177/0265407514541075 Google Scholar Crossref Search ADS Ruppel , E. K. , Burke , T. J. , & Cherney , M. R. ( 2017 ). Channel complementarity and multiplexity in long-distance friends’ patterns of communication technology use . New Media & Society , 20 , 1564 – 1579 . Advance online publication doi:10.1177/1461444817699995 Google Scholar Crossref Search ADS Scissors , L. E. , & Gergle , D. ( 2013 , Februrary). Back and forth, back and forth: Channel switching in romantic couple conflict. Paper presented at the ACM Conference on Computer Supported Cooperative Work (CSCW ‘13), San Antonio, TX. doi: 10.1145/2441776.2441804 Snijders , T. A. B. , & Bosker , R. J. ( 2012 ). Multilevel analysis: An introduction to basic and advanced multilevel modeling ( 2nd ed. ). Thousand Oaks, CA : Sage . Stafford , L. ( 2010 ). Geographic distance and communication during courtship . Communication Research , 37 , 275 – 297 . doi:10.1177/0093650209356390 Google Scholar Crossref Search ADS Stafford , L. , & Merolla , A. J. ( 2007 ). Idealization, reunions, and stability in long-distance dating relationships . Journal of Social and Personal Relationships , 24 , 37 – 54 . doi:10.1177/0265407507072578 Google Scholar Crossref Search ADS Taylor , S. H. , & Bazarova , N. N. ( 2018 ). Social media and subjective well-being: A relational perspective. In Z. Papacharassi (Ed.), A networked self and love (pp. 86 – 112 ). New York, NY : Routledge Taylor , S. H. , & Ledbetter , A. M. ( 2017 ). Extending media multiplexity theory to the extended family: Communication satisfaction and tie strength as moderators of violations of media use expectations . New Media & Society , 19 , 1369 – 1387 . doi:10.1177/1461444816638458 Google Scholar Crossref Search ADS Theiss , J. A. , & Solomon , D. H. ( 2006 ). Coupling longitudinal data and multilevel modeling to examine the antecedents and consequences of jealousy experiences in romantic relationships: A test of the relational turbulence model . Human Communication Research , 32 , 469 – 503 . doi:10.1111/j.1468-2958.2006.00284.x Google Scholar Crossref Search ADS Toma , C. L. , & Choi , M. ( 2016 , February). Mobile media matters: Media use and relationship satisfaction among geographically close dating couples. Paper presented at the 19th ACM Conference on Computer-Supported Cooperative Work & Social Computing (CSCW’16). San Francisco, CA doi: 10.1145/2818048.2835204 Valkenburg , P. M. , & Peter , J. ( 2009 ). The effects of instant messaging on the quality of adolescents’ existing friendships: A longitudinal study . Journal of Communication , 59 , 79 – 97 . doi:10.1111/j.1460-2466.2008.01405.x Google Scholar Crossref Search ADS Vangelisti , A. L. , & Caughlin , J. P. ( 1997 ). Revealing family secrets: The influence of topic, function, and relationships . Journal of Social and Personal Relationships , 14 , 679 – 705 . doi:10.1177/0265407597145006 Google Scholar Crossref Search ADS © The Author(s) 2018. Published by Oxford University Press on behalf of International Communication Association. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/open_access/funder_policies/chorus/standard_publication_model)
TI - Revisiting Media Multiplexity: A Longitudinal Analysis of Media Use in Romantic Relationships
JF - Journal of Communication
DO - 10.1093/joc/jqy055
DA - 2018-12-01
UR - https://www.deepdyve.com/lp/oxford-university-press/revisiting-media-multiplexity-a-longitudinal-analysis-of-media-use-in-DD05n3V15Z
SP - 1104
VL - 68
IS - 6
DP - DeepDyve
ER -