Understanding Information and Communication Technology Use in Workplace Romance Escalation and De-Escalation

Initiating WRs

The Model of Workplace Romances asserts that propinquity, attitude similarity, attraction, job autonomy, attitude, and desire to engage in a WR, as well as culture, were factors that led to the formation of a WR (Pierce et al., 1996). Research has identified work context has also been found to be a motivating factor in the initiation of WRs. Mano and Gabriel (2006), for example, found that that organizational climate plays a role in understanding the development and perceptions of WRs. Moreover, Salvaggio, Streich, Hopper, and Pierce (2011) found factors such as perceived workplace sexualization (e.g., focus/encouragement on appearance, flirting) and high male-female contact were associated with higher levels of WR. Research recently found employees voiced their motives to engage in WRs as time, similarity, easy of opportunity, and hooking up (Cowan & Horan, 2014b). Although there is some overlap between Pierce et al. and Cowan and Horan, and these studies inform WR formation, they do not specifically speak to ICTs unique role in the process of WR (de)escalation.

Maintaining WRs

Studying how and why WRs are maintained is important because multiple negative consequences have been associated with failed WRs (Mainiero & Jones, 2013b; Pierce et al., 1996). One reason WRs may dissolve, resulting in some negative outcomes, may be due to difficulty with WR maintenance. Understanding challenges associated with maintenance is important because research has uncovered positive outcomes of maintained WRs. These include productivity (Anderson & Hunsaker, 1985; Dillard, 1987), a happier work environment, and increased motivation by those in the relationship (Mainiero, 1989). Organizational culture is also a factor important to maintaining WRs.

An organization’s culture entails the attitudes, values, beliefs, ideologies, and norms particular to the organization (Pierce et al., 1996). Mainiero (1989) found those cultures that are more traditional, slow-paced, and conservative discourage WRs. In contrast, organizations with dynamic, fast-paced, and more liberal cultures can stimulate WRs. It follows that those trying to maintain a WR in cultures that are more open to these kinds of relationships might have an easier time. Informing Mainiero’s nearly 30-year-old findings, Cowan and Horan (2014a) found organizational culture and expectations for behavior affected coworkers’ WR reactions. If culture and work context make it easier to stay in a WR (i.e., WR friendly policies, coworker support), successfully maintaining that WR might be easier.

Research reveals that WR participants send about 4 emails a day to their romantic partner (Hovick, Meyers, & Timmerman, 2003), demonstrating that workplace ICTs are used to help maintain WRs. These emails were used for nonorganizationally driven content, including flirting (as well as asking questions, small talk, talk about work, expressing intimacy, starting a relationship, gaining more information, and managing conflict). Given that this research is 13+ years old, and the rate that technology changes, it is important to further examine ICTs and WRs. Modes have changed and other ICTs, such as texting, have become prevalent in society, therefore presenting new challenges that warrant study (e.g., Mainiero & Jones, 2013a). Texting presents its own set of issues, particularly when the employee uses a personal device. This would not allow organizations to save messages or trace them.

Dissolving WRs

As not all relationships result in life-long partnerships, WRs—like other romantic relationships—can result in dissolution. Surprisingly, little research has examined technology’s role in dissolution, instead focusing largely on the consequences of dissolved WRs. Potentially, and as previously discussed, sexual harassment behavior and/or claims might be a consequence (Mainiero & Jones, 2013b; Pierce et al., 2000, 2004). In two studies, Pierce et al. (2000, 2004) investigated subjects’ judgments of sexual harassment complainants and perpetrators with a WR history. Pierce et al. (2000) found judgments were influenced by perceptions of WR motives and recommended personnel actions were influenced by the status of WR participants, the subject’s sex, and raters’ WR attitudes. Importantly, disciplining the accused perpetrator was seen as less appropriate in peer-peer relationships. They also found those with more favorable attitudes toward WRs saw discipline as less appropriate, and compared with women, men felt it was more appropriate to ignore the complaint. Such findings are particularly important considering that most individuals do have some communication with former romantic partners following termination (e.g., Dailey, Hampel, & Roberts, 2010; Lannutti & Cameron, 2002). Importantly, communication following a breakup with a former partner can be unwanted and also occur via ICT (Langhinrichsen-Rohling Palarea, Cohen, & Rohling, 2000) and this could be a very problematic situation when individuals still work together and interact via ICTs. The previously reviewed research helps explain why Mainiero and Jones (2013a) recommend that organizations employ love contracts. They describe these documents as a contract signed by both parties in the relationship explaining the relationship “is consensual, voluntary, welcome, and unrelated to their professional relationship at work, and each employee is free to end the romance at any time without coercion, prejudice, or any job-related consequences” (p. 197).

Participants and Sampling

Participants had to be at least 18 years old and currently or previously engaged in a WR. Following institutional review board approval, both authors recruited participants using convenience and snowball sampling techniques (Lindlof & Taylor, 2011). Both authors contacted possible people in their networks inviting them to participate in the study if they met the aforementioned requirements. The recruitment message included information about the study, who to contact for participation, and anticipated participation time. Email, social networks, and posted flyers were all used for recruitment. These efforts helped us garner 14 participants who met the criteria. Following the snowball sampling technique (Tracy, 2013), we interviewed two additional participants. Students enrolled in the authors’ classes were also given the opportunity to participate in the study if they met the qualifications and received minimal course credit for participation. Using these sampling techniques, we were able to conduct in-depth interviews with a total of 23 participants who met the criteria. The majority of the participants were female (n = 18) and held nonmanagement positions (n = 19). All but two of the relationships were cross-sex. We were able to talk to two couples who eventually married (in separate interviews). The participants’ relationships lasted between 9 months and over 10 years with the majority being between 1 and 2 years (n = 15). All participants were assured confidentiality and were given information sheets with relevant participation information.

Data Collection

The goal of this inquiry was to gain an in-depth understanding of how and why ICTs are used in WRs; because of this, the first author conducted one-on-one semistructured interviews with all participants. The interviews took place outside of participants’ organizations and at a location they felt comfortable. The interviews lasted between 30 and 75 minutes. Because of the distance, 10 of the interviews were conducted via telephone. The interviews were considered “narrative” and “respondent” in nature (Lindlof & Taylor, 2011), because after reading a definition of what is meant by a workplace romance (Horan & Chory, 2011), we asked the participant to tell us their story and then asked questions relating to ICT use in the (de)escalation of the relationship. If the relationship had terminated, we asked questions about this dissolution including if, how, and why ICT was used. Although the interviews were recorded with participant consent, the first author also took notes throughout the interview noting important stories and information. Immediately after each interview, the first author would write down her thoughts and ideas generated during the interview as a form of memoing, noting similarities and differences with other interviewees’ experiences as well as interesting examples and overall questions (Charmaz, 2000). Interviews were conducted until we started seeing high reoccurrence of developing themes (around the 20th interview). We then interviewed the remaining participants and left the field to concentrate on data analysis. Interviews were transcribed by a professional transcriptionist resulting in 308 single spaced pages of text; the first author checked the transcriptions for accuracy. The appropriate institutional review board approved all procedures.

Data Analysis

We took a constructivist grounded theory approach to data analysis because we were interested in building theory from the ground up on the initiation, maintenance, and termination stages of WR and ICT use. Constructivist grounded theory (Charmaz, 2000) is a technique used to inductively create theory from textual data. Using the constant comparative method (or uncovering categories, comparing these to new data, adjusting and refining the emerging categories based on new data, comparing again, and so on), we used a two-step coding process (often discussed as open, axial, coding procedures; Charmaz, 2000; Strauss & Corbin, 1990). Analysis began in the data collection phase as the first author began to generate ideas and codes as she took notes and reflected on the interviews about what was being learned from participants in relation to our research question. More focused coding began as both authors conducted more focused readings of the interview transcripts and generating initial codes. During this first stage of coding, data is broken down and emerging concepts are labeled (Charmaz, 2000). We separately open coded the data by asking ourselves, “What is this?” and “what is going on here?” in terms of what types of ICTs were used, how, and why in the three WR stages. Some examples of these initial codes included “hiding,” “privacy,” “fun and secret,” “sneaky use,” “awkward communication,” and “emotional.” We then met to discuss these categories and discuss possible larger themes. During the second stage of coding, the initial categories identified during initial coding were further developed by identifying larger themes in the emerging categories. We asked ourselves how are these initial codes similar or different? How do they fit together and how do they answer our research question? The larger themes suggested a clear process of ICT use in WR development and termination (see Table 1). To better understand whether our interpretation resonated with our participants, we checked our interpretation with four participants who agreed to serve as member checks (Lincoln & Guba, 1985). After reading our findings and interpretation all four related they felt our interpretation reflected their experience.

OPEN IN VIEWER

Table 1.

ICT Use in Workplace Romantic Relationships.

Stage	Category	Category description
Stage 1:  Initiation	The introductory phase	Exchange of personal contact information due to benign and work-focused motives.
	Uncertainty reduction phase	Covert gathering of information about the other person through social networking sites such as Facebook and Twitter as a way to get to know the other person and reduce uncertainty.
	Private exploration phase	Increase in personal communication, attempts to hide the budding relationship, and a shift to using several different kinds of technology to talk to and learn more about the other person both at work and during personal time.
	Privacy concerns	Participants take steps to hide or downplay the relationship from coworkers and bosses and use technology to hide the relationship from others.
Stage 2:  Maintenance	Questioning whether to disclose	Conversations about if the couple should disclose the relationship or keep it secret (organizational culture, rumors and gossip were determining factors). Disclosing is often done through technology.
	Maintaining professionalism	Participants wanted to maintain professionalism while at work and this included not discussing personal topics at work, not being physical and touching, saving arguments and emotional conversations until after they are away from the workplace, and generally keeping work and personal life separated. Technology such as texting allowed the employees to maintain professionalism and at the same time engage in maintaining behaviors while at work.
	Technology as maintenance	Technology was used to maintain these WRs by keeping partners in contact with one another even though they had busy lives and often could not physically be together. Technology was also used to erect barriers between the couple and others and as a tool to keep the relationship strong. Some examples include posting videos on their partners SNS sites, texting to coordinate schedules as well as having conversations about work and the people at work.
Stage 3:  Termination	Technology as indicator of impending dissolution	Changes in technology use, including communicating via these technologies less and less was seen as an indicator that the WR was dissolving.
	Technology as break-up tool	Technology, including SNS and texting, were used as predominant tools in the termination of the WR.

Stage 1: Technology Use in the Initiation of WRs

Several technologies are used in the initiation of WRs including personal cell phones (particularly, the text feature), SNS sites such as Facebook, Instagram, Twitter, Foursquare, Skype, work email, work landlines, and internal intranets. A wide variety of technologies are used as partners begin to “figure out” one another’s communication preferences and style. We found that there were clear patterns in how and why these technologies are used at this stage of the relationship. We labeled these steps: (a) the introductory phase, (b) the uncertainty reduction, and (c) the private exploration phase. Privacy was an important driver in Phases two and three.

Stage 2: Technology Use in the Maintenance of WRs

In terms of WR maintenance, our participants discussed three main factors associated with how and why certain technologies are used in WR the maintenance. We labeled these themes: (a) questioning whether to disclose, (b) maintaining professionalism, and (c) technology as maintenance.

Stage 3: Technology Use in the Dissolution of WRs

Seven of the participants we spoke to were still in a relationship, the rest (n = 16) had been terminated due to several different factors including infidelity, drifting apart, not enough time spent together, and loss of interest. It was clear that technology played an important role in the termination process. In answering how and why technology was used in the dissolution process, we found two main themes, technology use as indicator of impeding dissolution and technology as an essential dissolution tool.

Theoretical Implications

Strauss and Corbin (1990) explain that a good grounded theory will be understood by those studied, be broad enough to apply to a variety of contexts related to the area of investigation, and provide control in the form of conditions in which the theory applies. Our member check responses demonstrate our model does resonate with those who were studied. We also believe the model is broad enough to apply to various types of WRs including peer-to-peer and asymmetrical relationships. However, the model should be used with caution and in consideration of the particular organization’s culture. Regarding control, we believe organizational culture, policy, and norms play a significant role in how ICT use actually plays out in WRs. As the findings demonstrate, our model tends to reflect a culture that is more open to WRs and personal communication and involves experienced technology users, using their own devices and in predominantly long-term relationships (the majority being between 1 and 2 years as previously reported). That said, findings demonstrate support for ideas previously presented, particularly the idea that today’s WRs are largely love-motived and long-term, serious endeavors (Mainiero & Jones, 2013b). Our findings have implications for multiple theories, reviewed below.

Researchers have used Work-Family border theory to better understand WRs, arguing that both blending and spillover occur (e.g., Horan & Chory, 2011). WR participants are border crossers (Clark, 2000), and researchers argued that, for those in WRs: “Work issues are discussed in the life domain (e.g., a date, dinner, etc.) and life issues are discussed in the work domain (e.g., discussing relationship plans at work, scheduling of a personal event, etc.)” (p. 565). The previous description supports the recent description that WRs “involve boundary permeability” (Methot & LePine, 2016, p. 118). Thus, as previously described, these blended domains (Clark, 2000) with permeable borders (Methot & LePine, 2016) allow for spillover (Horan & Chory, 2011). Findings here inform those claims, as ICT use likely facilitates both blending and spillover. Consider results regarding social media as a means to reduce uncertainty. Situations wherein participants used social media to reduce uncertainty while in the life domain about a coworker from the work domain would represent blending. Likewise, feeling frustration with a romantic partner about an organizational situation while at home would represent spillover. Thus, our results add explanation (e.g., Chaffee & Berger, 1985) to this theory.

Work–family border theory’s (Clark, 2000) first proposition argues that “when domains are similar, a weaker border will facilitate work/family balance” (p. 765). Our results regarding technology’s role in WR appear to support this assertion, as those working in organizational cultures perceived to be more “relaxed” also seemed more supportive of WRs. Additionally, this perspective is consistent with Greenhaus and Powell’s (2006) theory of Work-Family Enrichment. They specifically framed instances where “work and family roles are ‘allies’” (p. 72). These situations result in enrichment, which they described as “experiences in one role improve the quality of life in the other role” (p. 73). Based on their theoretical arguments, experiences in WR may result in enrichment for WR participants, as their theory argues, and concurrently, balance as Clark argued. Golden’s (2013) findings illustrate how technology is used at work to communicate with members of the home/life domain and, equally, how ICTs are used in the home domain to communicate about work. As both theories add to our limited understanding of WR, additional research and theoretical framing is necessary to further understand how WR operates in organizations, both for organizations and for members of the WR.

Study participants repeatedly cited the role of privacy concerns in WRs, pointing to the importance of privacy management (Petronio, 2002). These concerns were largely seen during the initiation stage, and also in the maintenance stage when individuals struggled with the decision of whether to disclose their WR. This finding is consistent with Petronio’s theory (2002) as we found those in WRs discuss and negotiate how and when to reveal the relationship to others in the organization. Some participants pointed to the idea that WR information is indeed shared by the relationship partners and revealing this information should include input from both people and entail privacy rules. The unique context of WR as well as fear of disclosure reactions are likely key factors in understanding how individuals form privacy rules (Petronio, 2002).

Applied Implications

Our findings also point to several practical implications that should be considered by employees, managers, and organizations. First, we found ICT use in the early stages of a relationship, and in the maintenance phase, may result in positive spillover for WR participants. Still, during times of conflict and/or dissolution, ICT use may lead to negative spillover. Problematically, situations of unwanted communication following romantic termination can cause negative emotions and may even constitute obsessive relational intrusion (e.g., Spitzberg & Cupach, 2003; Spitzberg, Nicasto, & Cousins, 1998). These situations are linked to a number of negative outcomes for victims of unwanted pursuit and/or stalking (Spitzberg & Cupach, 2003). When unwanted pursuit (Langhinrichsen-Rohling et al., 2000) happens within organizations, a challenging situation emerges. This issue seems particularly important for organizations to be aware of as often employees use privately owned devices at work to facilitate the WR. The “Bring Your Own Device” phenomenon in organizations occurs when employees “use their personal mobile devices to access the resources of the company for both work as well as personal use” (Gajar, Ghosh, & Rai, 2013, p. 62). With this increased use of personal technology comes, of course, an increased risk for the organization and the employee. This increased risk stems from less organizational control of a personal device than a company owned and monitored device. For instance, an employee may use their own device to access company applications/websites that allows the employee to contact his/her former relational partner. In these situations, employees blend personal/professional boundaries to communicate with someone who may not want such communication; this could be perceived, then, as harassment. Using personal devices increases the potential for the organization to not even know the communication is taking place, monitor it, and take different steps to protect employees. Future research would be wise to explore how organizations are managing the “Bring Your Own Device” phenomenon and the increased use of personal devices in WRs.

Based on our findings highlighting the use of technology such as cell phones and the text feature during relationship termination, it seems clear organizations need to be proactive and manage communication following termination between former partners, possibly through pertinent organizational policy. Such a policy would be dependent on norms, culture, and legalities as well as being specific to how organizations use technology and ICTs. Potentially, this policy could be explicitly spelled out in new employee orientation as well as in any potential love contracts. HR practitioners and managers would be wise to consult other studies about workplace romance policies and concerns regarding approaches to management (e.g., Pierce & Aguinis, 2009).

Second, our findings demonstrate that personal communication during the initiation, maintenance, and even termination stages of relationships does occur during work hours and ICT use aids in this type of communication. We also found those in WRs feel a tension with regard to revealing and discussing the relationship with other organizational members and not being so open. Similarly, struggles with openness have been identified in workplace friendship studies: it is a tension in workplace friendships (Bridge & Baxter, 1992), yet also a maintenance behavior (Sias et al., 2012). Though openness has its benefits, maintaining secrets can lead to sources’ rumination about the secret (Afifi & Caughlin, 2006) and avoidance is a negative maintenance behavior (Dainton & Gross, 2008). When secrets are shared about WRs with other organizational members, yet asked to remain private with trusted actors, privacy rule violations could result in gossip about the romance (e.g., Cowan & Horan, 2014a; Mano & Gabriel, 2006) and/or workplace distractions. As discussed previously, our participants mentioned gossip and distraction from work were issues they were aware of and tried to mitigate. Although we know these participants were trying to mitigate possible negative effects of the WR on the work environment, we do not know whether these strategies are effective or whether the people in WRs tend to do this well. The answer to this question is likely complicated, but it seems clear how others find out about the relationship and the organizational culture are important factors. Future research should explore WRs and employee distraction more fully.

Last, findings have implications for the positive functioning of work groups and coworker relationships. We found those in WRs often erect barriers with coworkers as a way to maintain the relationship and keep it healthy and secure. This could have negative implications for work groups and teams especially those that have a high level of interdependence. Cohen and Bailey (1997) reported that when there is conflict between group members (which can happen when WRs are present and barriers are erected between members), the interdependence of coworkers increased the negative impact of conflict. The kinds of barriers discussed by our participants could negatively affect group cohesion which is related to group performance. Future research should explore these issues and determine when these barriers could be positive or detrimental to work groups, individuals, and group functioning.

Conclusion, Limitations, and Future Research

The model of ICT use in WR discussed here adds to our limited understanding of WR (de)escalation by specifically focusing on how and why certain channels are used. Results of this study should be interpreted with limitations—many of which are avenues for future research. First, our sample was largely cross-sex and therefore does not speak to same sex WRs (e.g., Horan & Chory, 2013). Second, the WRs studied were largely noncontroversial; additional studies should explore WRs with large age difference and WRs where one or both parties are married to other people. Other fruitful areas might look at how age, job position, time in the relationship, and other demographic variables might affect WR technology use. Though we used qualitative data to generate a model, future studies should quantitatively test the ideas presented. Naturally, qualitative data can lead to scale development which would allow for additional exploration of the impact of WRs on both organizational outcomes (e.g., satisfaction, motivation, productivity, turnover intentions) and interpersonal communication processes necessary for a productive work environment (e.g., antisocial organizational communication behaviors—see Chory & Hubbell, 2008; Horan & Chory, 2013). Findings of such studies will offer additional understanding for both researchers and practitioners.

As described previously, WRs are common and likely will continue to grow in popularity. Equally, organizations and individuals rely heavily on ICT use in their personal and professional lives. Consequently, our model should be of value to organizations and individuals as they help describe the how and why of ICT use in WRs.