Organizational Norm Congruency and Interpersonal Familiarity in E-Mail: Examining Messages From Two Different Status Perspectives
Keri K. Stephens,
University of Texas at Austin
concerning this paper should be addressed to Keri K. Stephens, 512-471-0554, Department of Communication Studies, 1 University Station A 1105, Austin, TX 78712-0115. Electronic mail may be sent to firstname.lastname@example.org.
concerning this paper should be addressed to Keri K. Stephens, 512-471-0554, Department of Communication Studies, 1 University Station A 1105, Austin, TX 78712-0115. Electronic mail may be sent to email@example.com.
Relying on SIDE theory, this 2 × 2 experimental design tested the effects of adherence to organizational norms and interpersonal familiarity in e-mail messages from both superior and subordinate perspectives. Results reveal that using norm-congruent e-mail messages account for over 50% of the variance in both superior and subordinate attitudes toward the message and between 30% and 56% of the variance in perceptions of source credibility. Data from the superiors indicate that norm congruent e-mail messages account for 14.4% of the variance in message compliance. Results from subordinates reveal an interaction effect that is consistent with SIDE-based predictions. Subordinates believe superiors are most likely to comply with e-mail requests from unfamiliar subordinates that are crafted in a manner congruent with organizational norms.
E-mail has quickly become an extensively used method of communication in organizations (Bunz & Campbell, 2004; Kim, Kim, Park, & Rice, 2007; Weber, 2004) including higher education (e.g., Hassini, 2006). As a communication medium it allows users to send and receive messages that reach a wide variety of individuals. Despite the prevalence of e-mail, there are other communication technologies that have recently entered the workplace such as text messaging and instant messaging. These are all text-based communication media, yet there is still conflicting advice concerning how we can best communicate using specific media, especially e-mail (Baron, 2002).
Research has demonstrated that e-mail style—salutations, grammar, and word use—is often different from that found in a formal letter (Burgess, Jackson, & Edwards, 2005; Colley, Todd, Bland, Holmes, Khanom, & Pike, 2004; Weber, 2004). In her essay on the history and predictions concerning e-mail use, Baron (2002) claims that the vast number of books and Websites crafted in the late 1990s offering advice on e-mail style are contradictory. Some books, like Wired Style (Hale & Scanlon, 1999) advise e-mail writers to disregard sentence structure, spelling, and punctuation since “no one reads e-mail with red pen in hand” (p. 3). But others, such as The Elements of E-Mail Style (Angell & Heslop, 1994), claim that e-mail messages should be crafted using rules similar to Standard English grammar. Cunningham and Green (2002) suggest people should “treat e-mail as an opportunity to put your best foot forward” (p. 20). Essentially, credibility perceptions may be cultivated via e-mail messages and are important outcomes of strategically managing impressions (Leathers, 1998).
An alternative approach to making broad sweeping claims and recommendations concerning e-mail use is to examine the organizational context, norms, and expectations surrounding e-mail use. Perhaps the advice books and conflicting research findings are all correct and it is the context that determines appropriateness. In their review of e-mail literature, Ducheneaut and Watts (2005) claim that conflicting research has led us to view e-mail as content-dependent and we should focus on how users appropriate this tool in their local environment. One theoretical perspective that speaks directly to the importance of organizational norms in a computer-mediated context is the social identity model of deindividuation effects (SIDE; Postmes, Spears, & Lea, 1998a; Postmes, Spears, & Lea, 1998b; Spears & Lea, 1992).
While e-mail is commonly used today in organizations, text messaging is increasingly being used on mobile devices, especially by younger generations (Kim et al., 2007; Rideout, Roberts, & Foehr, 2005). Kim et al. (2007) examined the uses of a variety of communication media and found that older people tend to use e-mail, while younger people use mobile phones, SMS, and instant messaging. Text messages tend to contain shortcuts—BTW for by the way or WFM meaning works for me—and can be typically shorter than an e-mail. The difference in media use preferences creates an interesting dilemma at work since quite often members of the younger generation are in subordinate positions.
Organizational communication literature refers to people who have power over others as superiors and those lower in status as subordinates. These status variables are also considered important in instructional communication research (e.g., McCroskey & Richmond, 1983; Turman & Schrodt, 2006) and experimental psychology (e.g., Hall, Rosip, LeBeau, Horgan, & Carter, 2006). In many organizational contexts, e-mail is increasingly being used for strategic and political reasons, which can affect the power and control functions found in organizational structure (Ducheneaunt & Watts, 2005). These power and control considerations are often manifest in superior-subordinate relationships. Increasingly, these superior-subordinate relationships are occurring over a distance such as in virtual work teams and distributed work environments (Kirkman, Rosen, Gibson, Tesluk, & McPherson, 2002). Subordinates often need to impress their superiors, even with their e-mails, if they hope to gain compliance for their requests.
The present study focuses on both the cognitive and strategic aspects of SIDE in an effort to better understand subordinates' e-mail message behavior—which can resemble the informality of a text message—and how that behavior affects superiors' perceptions of them. In addressing the strategic aspect of SIDE, we are concerned with the presentation of self to powerful others in the computer-mediated communication (CMC) context. Some scholars claim that the strategic aspect of SIDE warrants further investigation particularly in the area of direct assessment of message behavior (Walther & Parks, 2002) because these studies are likely important when there are power and status differences between the communicators (Postmes et al., 1998b). This study contributes an additional explanation of how the strategic and cognitive aspects of SIDE help us better understand organizational e-mail use and the different perceptions between superiors and subordinates.
Review of Literature
The social identity model of deindividuation effects (SIDE, Postmes et al., 1998b; Spears & Lea, 1992) provides an organizational and interpersonal theoretical framework useful for exploring organizational e-mail use. This theory claims that in the absence of physical cues (as in e-mail use) that would identify communication partners, people are deindividuated or not thought of as individuals. Under these conditions, people can relate positively to one another because they assume there is similarity based on shared group norms (Postmes et al., 1998b). Walther (1996) describes this as, “when participants are led to perceive that they are in a group relationship, each tends to hold a ‘social self-categorization’ rather than an ‘individual self-categorization,’ [resulting in] attributions of greater similarity and liking with one's partners” (p. 18). This theory is particularly useful to the present research because it addresses the importance of organizational norms, individuating information, and their influence on the behaviors of organizational members communicating via computer-mediated environments like e-mail (Douglas & McGarty, 2001; Lea & Spears, 1991; Postmes, 1997; Postmes et al., 1998b; Spears & Lea, 1992; Spears, Lea, & Lee, 1990).
SIDE considers the normative influences on a group's behavior and predicts that deindividuation in the CMC environment has two possible effects: cognitive and strategic. Cognitive or self-categorical effects refer to the salience of group identification on the behavior of individuals in the CMC environment. Specifically SIDE predicts that contexts like e-mail (or those low in individuating information) will increase group member adherence to social norms. However, when greater individuating information is present, adherence to group norms will be reduced. Strategic effects suggest that individuals communicating in the CMC context will tend to manage the impressions of important others, in this case organizational superiors, by adhering to the social norms for CMC communication (Douglas & McGarty, 2001). The strategic aspect of SIDE illuminates the link between shared norms and managing impressions.
Strategic Effects: Formality in Superior-Subordinate E-mail Communication
Strategically managing impressions is particularly important in relationships where status and power differentials are present such as the superior-subordinate or the instructor-student. Organizational superiors can have tremendous power over subordinates including holding a great deal of control over the subordinate's success within the organization (Fiske, 1993). Because of this, subordinates have much at stake when communicating with superiors (Harris, Lightner, & Manolis, 1998). Fiske (1993) contends that because the superior has the power to affect the subordinate's outcomes, subordinates will be motivated to please superiors and strategically manage their impression in an effort to ensure their superior views them positively. Instructional communication has also identified power differentials as important because instructors and students communicatively negotiate power use (Turman & Schrodt, 2006).
One way subordinates can manage superiors' impressions is to conform to the organizational norms related to communication behavior in e-mails. While normative e-mail communication behavior varies from organization to organization, research has identified some general norms for formal and informal e-mail messages (Bunz & Campbell, 2004; Crowther & Goldhaber, 2001; Focazio, 1997; Pirie, 2000; Schade, 2004; Weber, 2004). Formal e-mail messages are those that include salutations, openings, and closings that incorporate politeness cues like; “Please” and “Thank you,” as well as those that use correct grammar and spelling (Bunz & Campbell, 2004). In addition, the subject line typically communicates the nature and content of the e-mail message and these messages are signed by the sender (Crowther & Goldhaber, 2001; Focazio, 1997; Pirie, 2000). Informal e-mail messages tend to lack many of the characteristics present in formal e-mail messages and they resemble the informality found in text messaging. Specifically, typical openings and closings are missing or abbreviated, politeness cues are absent, grammar and spelling are less important, and abrupt or terse language use is common (Schade, 2004). In an organization where the norm for e-mail message behavior is to communicate informally, SIDE would predict that superiors and subordinates, in an effort to strategically manage their impressions, would conform to this norm (Postmes et al., 1998b; Lea & Spears, 1991). In the reverse context, where the norm is formal e-mail message behavior for subordinates and superiors, employees will tend to conform to this type of behavior in an effort to strategically manage impressions. Therefore, organizational norms for e-mail message behavior, either formal or informal, should affect how superiors and subordinates use e-mail within their organizational context.
Cognitive Effects: Familiarity and Superior-Subordinate E-mail Communication
The strategic component of SIDE predicts that when interacting in the CMC environment, superiors and subordinates will use social cues like organizational norms for e-mail message behavior to act strategically and manage the impressions of others. The cognitive aspect of SIDE predicts that because of the deindividuation effects of the CMC environment, interactants will conform to social norms when more group-level information is present. However, when more individuating information is present, the reliance on group norms as a basis of relating should be reduced (Postmes et al., 1998b; Walther & Parks, 2002). Essentially, interpersonal cues such as those that indicate familiarity could reduce reliance on group norms in the e-mail context. In this study, familiarity reflects how well the e-mail correspondents know one another; this could range from having no familiarity with the other person to having a high degree of familiarity. SIDE would predict that organizational norms, like those for e-mail message behavior, and individuating information about communication partners will influence behavior in the CMC environment (Postmes et al., 1998b; Spears & Lea, 1992). For example, if a superior is familiar with a subordinate, their interpersonal way of relating should matter more than adherence to group norms. So over time, as familiarity increases, crafting e-mail messages that adhere to organizational norms should matter less.
Relevant Outcomes of Norm Adherence
Attitudes refer to the affect component of a response or the way we “feel” about some stimulus (McCroskey, 2006). Wilson (2005) found evidence that message recipients develop more negative attitudes about the message and its sender when the message does not adhere to appropriate norms. Such negative perceptions can be detrimental in organizations since they reduce the effects of negotiation and persuasion (Wilson, 2005). As noted above, SIDE predicts that messages adhering to organizational norms will most likely be evaluated more positively than those that do not adhere to these norms (Postmes et al., 1998b). In this study, the stimulus is the e-mail message and prior research suggests the following hypothesis:
H1: Adherence to organizational formality norms in e-mail will positively affect a superior's attitude toward the message.
Recent findings have demonstrated that the quality of CMC (i.e. richness, frequency, and accessibility) affects the superior-subordinate relationship (Huang, 2002; Lau, Wong, Chan, & Law, 2001). Turner, Grube, Tinsely, Lee, and O’Pell (2006) reported employees who adhered to organizational norms for e-mail and instant messaging (IM) use were given higher performance ratings by superiors. In addition, Hamel (2006) found that organizational members saw misuse of communication technology, “as a routine negative side effect of media use in the workplace with dire implications for work relationships” (p. 123). She also found that organizational members have low tolerance for coworkers, superiors, and subordinates who break organizational norms by the misuse of technology. This research suggests adherence to organizational norms for e-mail use should influence source credibility, yet prior research does not guide us to make specific predictions concerning which dimension(s) of credibility are most affected. Therefore, the following hypothesis and research question were created:
H2: Adherence to organizational formality norms in e-mail will positively affect a superior's perception of source credibility.
RQ1: Which dimensions of source credibility (competence, trust, goodwill) are most heavily influenced by adherence to norms?
The research on credibility perceptions in CMC contexts suggests that an unfamiliar source is considered less credible than a familiar one (Dennis, 1996; Petty & Cacioppo, 1986; Rains, 2005). Petty and Cacioppo (1986) found that the source of information was imperative to the acceptance and processing of the message. Specifically, anonymity had a significant negative effect on source credibility. In the CMC context, Dennis (1996) theorized that anonymity resulted in group members viewing information exchanges through the group decision support system (GDSS) as less valid and credible. This leads to the following hypothesis:
H3: Less familiar subordinates will be viewed as less credible regardless of adherence to organizational formality norms.
Willingness to Comply
When choosing whether or not to comply with a message, receivers often consider their assumptions regarding the reason for the message (Smith & Hunt, 1987). In addition, if the message recipient attaches negative attributions to, or develops a negative impression of the source, the recipient tends to be less likely to accept the message as valid or comply with the request (Smith & Hunt, 1987). As previously noted, the strategic aspect of SIDE suggests adhering to organizational norms (either formal or informal e-mail messages) should help subordinates strategically manage the impression of their superiors. Strategically managing a superior's impressions could positively influence their willingness to comply with requests. Furthermore, familiarity is increased as more individuating information is shared, and this might alter the impact organizational norms have on willingness to comply. What this suggests is that unfamiliar communication partners should adhere closely to organizational formality norms, but once they become familiar, norm adherence should matter less. This leads to the following two hypotheses:
H4: Adherence to organizational formality norms in e-mail will increase a superior's willingness to comply with requests.
H5: A superior will be more willing to comply with requests from an unfamiliar subordinate when the request adheres to the organizational formality norms.
To explore organizational norms, we chose to examine a specific organizational context that likely has strong norms relevant to e-mail use, higher education. We verified this belief through measurement of norm congruency. Since our interest was to explore how SIDE theory is interpreted from two different power perspectives, we chose to conduct an experiment on e-mail messages and perceptions within the instructor-student relationship. This relationship, while complex, has been considered as similar to other status differential relationships such as the superior-subordinate (Hall et al., 2006).
There were 152 instructors and 183 students who participated in the experiment. To explore the subordinate-superior perspective in e-mail use, we recruited college-level instructors and asked them to evaluate e-mail messages from students. We used a combination of network and criteria-based sampling where the criteria for inclusion were broad and included any level of college instructor with at least one semester of experience. This resulted in a diverse sample of instructors; 100 females, 48 males, and 2 unidentified; their average age was 38 years (M = 37.96, SD = 11.71); and their average tenure as a college instructor was 9 years (M = 8.8, SD = 7.81). Participants reported being full-time instructors (n = 40), part-time instructors (n = 28), adjunct faculty (n = 16), assistant professors (n = 24), associate professors (n = 17), full professors (n = 12), and 12 participants reported that they held college instructor positions with other titles. The majority reported teaching communication courses (n = 110), however other disciplines were also represented including foreign languages (n = 8), science (n = 7), English (n = 6), business/management (n = 4), business communication (n = 4), technology (n = 4), psychology (n = 3), education (n = 2), public relations (n = 2), government (n = 1), and music (n = 1). In addition, 74% reported receiving 1–25 e-mails per day, 66% reported never or only occasionally using text-messaging on their cell phone, and 74% reported never or only occasionally using instant messaging on the computer.
The student sample was recruited from a single university through a departmental extra credit participation pool. This sampling resulted in a diverse set of students who were 41.2% male (n = 73), 58.8% female (n = 104), and had an average age of 19.7 (SD = 1.64). Their majors varied with the highest percentage being business (40%,n = 71), undeclared (15%, n = 26), and communication (12%, n = 22). Most students (77%, n = 141) indicated using text messaging on their cell phone daily, 93% sent at least 10 e-mails a day, and only 36% reported using instant messaging on their computer either daily or weekly.
Congruency of Understanding of Organizational Norms
A basic assumption in the SIDE Theory is that organizational norms will influence media use such as e-mail (Postmes et al., 1998b); therefore, it is important to establish that a shared understanding exists, especially when including two perspectives. To better understand the organizational norms surrounding written communication, the following procedure was used: First, all subjects received a message that resembled a text message. (see Appendix A for sample). After reading this message, the following statements assessed perceptions of organizational norm violation in e-mail use: This message is written too informally to send to a college instructor, Students need to put more thought into e-mail messages prior to sending them, This message is written appropriately to send to a college instructor. The resulting reliability of these three items measuring the norm violation was .78. A t-test between the instructors (M = 5.44, SD = .90) and the students (M = 5.38, SD = .83) confirmed that the two groups did not view organizational e-mail norms in a different manner t(328) = .62, p = .54. Because this measure appeared after an experimental manipulation of formality, an additional t-test between the norm-incongruent (M = 5.42, SD = .98) and the norm-congruent (M = 5.41, SD = .73) message conditions confirmed that the manipulation did not affect how the two groups viewed organizational e-mail norm violations t(328) = −.11, p = .91.
To test the hypotheses and research question, a 2 (norm-incongruent and norm-congruent e-mails) × 2 (high and low familiarity) experimental design was created. The four e-mail conditions included (a) a norm-incongruent/low-familiarity e-mail message, (b) a norm-incongruent/high familiarity e-mail message, (c) a norm-congruent/low- familiarity e-mail message, and (d) a norm-congruent/high-familiarity e-mail message. Norm congruency was manipulated in the text of the e-mail message. The norm-congruent e-mail message was written in traditional letter format including proper openings and closings, grammar, and punctuation. The norm-incongruent e-mail message was written in an informal manner including shortcuts commonly found in text messaging (i.e. R U for “are you”), incorrect grammar and punctuation, and openings and closings were omitted (Burgess, et al., 2005; Colley et al., 2004). Familiarity was manipulated by an instruction banner at the top of the e-mail message which read, “You receive the following e-mail late in the semester from a student that you know quite well” for the high-familiarity condition. A banner reading, “You receive the following e-mail early in the semester from a student that you do not know at all” was presented for the low-familiarity condition. The content of the e-mail messages was consistent across all four conditions and it concerned a student asking to meet with the professor to discuss class content.
All participants clicked on an HTML link which connected them to the study consent form. After consenting to take part in the study the participants were randomly assigned to view one of the four e-mail messages. After reading the message, the participants were asked questions regarding their perception of the dependent measures. Instructors were asked to evaluate the credibility of the student who sent the message, their attitude toward the message, and their likely compliance with the request. The students were asked to evaluate the credibility of the student who sent the message, their attitude toward the message, and their opinion of the instructor's willingness to comply with the request.
Norm congruency independent variable. The norm congruency independent variable included a norm-incongruent and a norm-congruent condition. To confirm an accurate manipulation, the variable was subjected to a manipulation check. A series of questions asked the participants to evaluate the norm congruence in terms of the formality of the e-mail message they just read. Using a 7-point Likert-type response scale, the participants were asked to indicate to what degree they agreed or disagreed with four statements; this message was well written, this message was very poorly written, this message was written in a formal manner, and this message was written in an informal manner. Two items were reverse coded and the four items combined to create a reliable measure achieving Cronbach's α of .89 (n = 329). A t-test between the norm congruent (M = 3.83, SD = 1.33) and norm incongruent (M = 1.46, SD = .67) conditions confirmed a successful manipulation t(327) = −20.13 p < .001.
Familiarity independent variable. The familiarity independent variable included a familiar and unfamiliar condition. To confirm an accurate manipulation, the variable was subjected to a manipulation check. Four items were used to check the participants' understanding of their familiarity with the student. Using a 6-point Likert-type response scale, the participants were asked to indicate to what degree they agreed or disagreed with four statements: The instructor was completely unfamiliar with this student, The instructor knew the student who wrote the e-mail very well, The instructor was quite familiar with the student who wrote the e-mail, and the instructor did not know the student well who wrote the e-mail. Two items were reverse-coded and the four items combined to create a reliable measure achieving Cronbach's α of .95. A t-test between the low-amiliarity (M = 1.92, SD = 1.10) and high-familiarity (M = 4.70, SD = 1.06) conditions confirmed a successful manipulation of formality in the e-mail conditions t(328) = −24.61, p < .001.
Message attitude. This dependent measure was assessed using a modified version of the Generalized Attitude Scale (McCroskey & Richmond, 1989). The scale contained six items (four that were reverse coded) and were measured using a 7-point semantic differential scale. Past research has reported reliability coefficients ranging from .93 to .98 (Cole & McCroskey, 2004; McCroskey, 2006) for this scale. The present study produced a Cronbach's α of .95 with an n = 338 (M = 3.68, SD = 1.76). The higher the value, the more positive attitude people had toward the message.
Credibility. McCroskey and Teven's (1999) 18-item scale (nine of which are reverse coded) was used to measure credibility. The scale uses a 7-point semantic differential scale and measures three dimensions of credibility: competence, goodwill, and trustworthiness. McCroskey and Teven (1999) obtained reliability coefficients ranging from .85 to .92 for the three dimensions and an overall alpha reliability of .94. The present study produced a Cronbach's α of .93 for competence (M = 3.29, SD = 1.29), α of .89 for trustworthiness (M = 4.51, SD = 1.04), and α of .90 for goodwill (M = 3.85, SD = 1.24).
Willingness to comply. Willingness to comply was measured using a modified version of Mottet, Beebe, Raffeld, and Pausel's (2004) seven-item Likert-type scale. Participants were asked to indicate how willing (1 = very unwilling, 7 = very willing) they (or the instructor) were to comply with the student's immediate request for a meeting and the student's future requests (assist them again during regularly scheduled office hours, assist them outside of regularly scheduled office hours, offer them an opportunity to earn extra credit, extend the due date on an assignment, and raise their final grade if it fell just below the cut-off point). A factor analysis of this scale revealed two factors: granting meeting compliance and grade-related compliance. The one containing four items related to granting meeting compliance were most relevant to this study, so that factor was retained and those four items had an α = .79 (M = 4.69, SD = 1.34).
Zero-order correlations among the manipulation checks, dependent measures, and potential control variables were examined (see Table 1). The correlations indicated that the formality in the e-mail had a significant effect on most of the dependent measures for both the superior (range r(144) = .38 to .79) and subordinate sample (range r(159) = .46 to .80), while familiarity seemed to only have minimal relationships (for superior: range r(144) = −.02 to .16; for subordinate: range r(159) = −.10 to .05). The correlation results also indicated that sex, age, and use of text messaging might be important control variables for some of the analyses. These controls were entered into a series of 2 (norm incongruent vs. norm congruent) × 2 (unfamiliar vs. familiar) ANCOVA models. Sex was the only significant control, and only in the superior sample, so it was used as a covariate for goodwill and compliance.
Table 1. Intercorrelations Between Variables and Covariates From Two Perspectives
*p < .05.
**p < .01; (Super) means superior and (Sub) means subordinate.
H1 predicted that adherence to organizational formality norms in e-mail would positively affect a superior's attitude toward the message. This hypothesis was supported from both perspectives. Formality had a main effect on message attitude (F(1,176) = 174.98, p < .001, η2 = .50) for the subordinate perspective and also for the superior perspective (F(1,154) = 226.66, p < .001, η2 = .60). See Table 2 for the means and standard errors.
Table 2. Effect of Norm Adherence on Dependent Measures
Variance Accounted For
*The subordinate and superior data were combined on these analyses due to no significant differences.
Superior perspective - main effect for norm-congruence
H2 predicted that adherence to the organizational norms of formality will positively affect a superior's perceptions of the source's credibility. This hypothesis was supported from both perspectives. A MANCOVA that included the three dimensions of credibility confirmed that from a subordinate perspective, competence (F(1,155) = 210.50, p < .001, η2 = .58), trust (F(1,155) = 79.85, p < .001, η2 = .34), and goodwill (F(1,155) = 105.68, p < .001, η2 = .41), were all significant. From a superior perspective, competence (F(1,140) = 175.88, p < .001, η2 = .56), trust (F(1,140) = 57.56, p < .001, η2 = .29), and goodwill (F(1,140) = 109.98, p < .001, η2 = .44) were also all significant. Familiarity had no main effects for the superior sample (F(1,140) = 1.81, p > .05) or for the subordinate sample (F(1,155) = 1.53, p > .05), and there were no interaction effects obtained for the superior sample (F(1,140) = .45, p > .05) or for the subordinate sample (F(1,155) = 1.68, p > .05). Both the superior sample and the subordinate sample believed that when the norms of formality are followed, the superior will view the subordinate as more credible.
RQ1 asked which dimension of credibility (competence, trust, or goodwill) was most heavily influenced by adherence to norms. Using the variance accounted for from the MANCOVA as reflected by effect size coefficients, adherence to formality norms impact the perception of a source's competence the most (56% for superiors and 58% for subordinates), followed by goodwill (44% for superiors and 41% for subordinates), and finally trust (29% for superiors and 34% for subordinates).
H3 predicted that less familiar subordinates would be viewed as less credible regardless of adherence to organizational formality norms. An ANOVA demonstrated no main effect for familiarity on any dimension of credibility (F(1,295) = .93, p > .05) thus this hypothesis was not supported.
H4 predicted that adherence to the organizational norms of formality would increase compliance with a request. This hypothesis was supported. Compliance was only measured from the superior's perspective because they are the people who decided whether to grant a meeting or not and from this perspective, the hypothesis was supported. Formality had a main effect on superior's compliance (F(1,155) = 24.93, p < .001, η2 = .14). However, perceived compliance from the subordinate perspective was also measured. Subordinates can respond to whether they think that superiors will comply with their request. The test from the subordinate perspective also supported the hypothesis (F(1,175) = 90.97, p < .001, η2 = .34).
As predicted by H5, the main effect on compliance for the subordinate perspective was qualified by a significant (yet small) interaction effect between compliance and familiarity. H5 predicted that superiors are more willing to comply with requests from an unfamiliar subordinate when the request adheres to the organizational formality norms. This hypothesis was supported, but only from the subordinate perspective. The interaction effect between formality and familiarity was significant (F(1,175) = 4.02, p < .05, η2 = .02), though the effect size was small. The pattern of the means confirmed that the highest compliance condition existed when the subordinate was unfamiliar and adhered to the organizational norms of familiarity. Table 2 and Figure 1 provide the descriptive statistics related to the experimental conditions. When the interaction was decomposed by level of norm adherence, there was no significant difference between the levels of familiarity when the message was written in an informal manner. There was a significant difference (p < .05) between the familiarity levels when the message was written formally. From the superior's perspective, this hypothesis was not supported because there was no interaction effect found with this ANOVA (F(1,152) = .96, p > .05).
The purpose of this study was to determine how organizational e-mail norms and interpersonal familiarity work together to influence superiors' and subordinates' attitude toward the message, source credibility, and willingness to comply with requests. Applying the strategic and cognitive aspects of SIDE theory to e-mail use, we found that organizational formality norms influenced all the outcome variables examined and that familiarity only influenced compliance when it interacted with formality and was measured from the subordinate perspective. A major contribution of this study was that we explored perceptions from two different perspectives that differ in power relations: the superior-subordinate relationship. We verified that there was a norm of formality expected in e-mail and there was no difference between superiors' and subordinates' perspectives on the type of e-mail that violates that norm. Furthermore, the participants in this study did follow the patterns of media use suggested by Kim et al. (2007) because while 77% of the student sample used text messaging daily, 66% of the instructors never or only occasionally used text messaging. Many of our hypotheses were supported; however, the findings clearly demonstrate that superiors and subordinates have different perceptions and those differences might provide us theoretical insight and directions for future research.
Norm-Congruent E-mails Matter
The first hypothesis was supported from both perspectives and claimed that adhering to the formal norms will increase a superior's attitude toward the message. From both perspectives, this variable received the most benefit when a subordinate sent an e-mail message that adhered to the organizational norms of formality. From the subordinate perspective, 50% percent of the variance in message attitude was explained by formality and 60% of the variance in message attitude was explained from the superior's perspective. This finding clearly links a formal e-mail style to positive attitude toward the message.
Both the superiors' and subordinates' perspectives of source credibility were positively influenced by the norm-congruent e-mail message and this message enhanced all three dimensions of source credibility. The variance explained by using a formal message varied slightly by perspective, but overall the superior and subordinate viewed these effects in similar ways. Competence benefited the most from a formal message because it accounted for 56% of the variance from the superiors' perspectives and 58% from the subordinates' perspectives. Goodwill also benefited heavily from a formal message accounting for 44% of the variance from the superiors' perspectives and 41% from the subordinates' perspectives. Finally, trust also benefited from formality, accounting for 29% of the variance from the superiors' perspectives and 34% of the subordinates' perspectives.
Applying what we know about the higher education context and the norms associated with proper writing in a formal manner, it makes sense that when people write norm-congruent e-mails, perceptions of competence is the credibility dimension most strongly affected. However, it is also intriguing that writing norm-congruent e-mails also affects goodwill perceptions so highly. Goodwill is associated with caring for others, being sensitive, and understanding. It appears that by writing a formal e-mail message, people convey their focus on others. This illuminates a major concern that the MIS Quarterly Editor, Ron Weber (2004) raised concerning e-mail: “Perhaps the most problematical behavior we exhibit as senders of e-mail messages is that we do not carefully consider the costs we impose on the recipients of our messages” (p. vi). The goodwill findings might speak to the appreciation we have when others take the time to craft messages that are easy to read and impose few costs on us.
The final outcome examined in this study was that compliance and adherence to organizational formality norms in e-mail did increase a superior's willingness to comply with requests. In this investigation, 14.1% of the variance in compliance was explained by formality of the e-mail message. Superiors in this organizational environment of higher education were more willing to grant a standard meeting with their student when the message was written formally. It is important to note that the request was nothing out of the ordinary because the student had kept up with coursework, attended class, read the book, and still had questions for the instructor. This effect might be even higher for a more intrusive request. Still, a formal message mattered when indicating a willingness to comply with this simple request.
Subordinates' Perspectives Differ From Superiors
The findings from the subordinates' perspectives, however, tell a different story. By exploring the relationship between the cognitive and strategic aspects of SIDE, this study found that subordinates believed superiors were most willing to comply with requests that were formal (met the organizational norms) and were received from an unfamiliar subordinate. This finding is supportive of the strategic aspect of SIDE because it shows subordinates believe it is important to conform to norms especially when there is no familiarity in the relationship. Yet familiarity did not actually influence how the superiors said they would comply with requests.
This finding raises an important consideration for organizational research as it demonstrates that groups with differing amounts of organizational power can perceive the effects of their e-mail messages differently. Quite often our organizational communication and workplace technology use research only examines a single perspective, even though most communication scholars acknowledge the importance of both message senders and receivers. This interaction finding (that unfamiliar subordinates who send norm-congruent e-mail messages are thought to be more likely to gain superiors' compliance) would have never emerged if this study had only included the superiors' perspectives. Yet it could also be argued that only the superior's perspective matters here because this is the person who will ultimately choose to comply or not. While this is a good assumption, knowing that subordinates have different beliefs about compliance from their superiors, raises important questions: Do subordinates know how to increase the likelihood of compliance with their requests? If subordinates believe that formal messages matter most when the superior does not know them, subordinates might also assume that as they become familiar with their superior, norm adherence matters less. This study suggests that there could be some misunderstandings that develop between superiors and subordinates when they operate under different assumptions in their e-mail use.
Familiarity Matters When it Interacts With Formality
The prediction of a main effect for familiarity on source credibility seemed plausible, yet was not supported. Essentially throughout this study, interpersonal familiarity does not appear to have any effects on the outcomes examined except when it is combined with adherence to the norms of formality and when it is measured from the subordinates' perspective. This finding is important to explore further in the future as the effect size in this study was quite small.
One explanation for this finding is found in the nature of the superior-subordinate relationship and the specific type of student-instructor relationship examined in this study. In this case, the instructor does not need to develop trust in the student because the relationship is more hierarchical than typically found in the virtual teams that tend to be studied (e.g., Jarvenpaa et al., 1998; Jarvenpaa & Leidner, 1999; Knoll & Jarvenpaa, 1995; Meyerson, Weick, & Kramer, 1996). If we extend these trust findings into other dimensions of credibility, the same applies to competence and goodwill. There is no need for the instructor to depend on the student; therefore, even when the student is familiar to the instructor, just knowing the student does not impact credibility perceptions. This finding highlights the importance of including organizational variables, like hierarchy, into theoretical models like SIDE.
In addition to the findings directly testing the hypotheses, there were some additional differences between the superior perspective and subordinate perspective that are worth noting. First, superiors—instructors in this study—were very likely to comply with the request for a meeting inside or outside their regular office hours, yet the subordinates—who were students in this study—predicted an overall lower level of anticipated compliance. This is an interesting finding that might explain why students think formal e-mail messages from unfamiliar students have the highest likelihood of compliance. Perhaps they feel that as they become more familiar with an instructor, they need to incorporate more face-to-face communication to achieve positive results. They also could believe as they get to know an instructor, formal e-mail messages matter less.
Implications for SIDE Theory
If we were to pit organizational norms against interpersonal familiarity in this e-mail context, organizational norms would win. More individuating information, does not impact the two variables that appear to have a strong interpersonal link: source credibility and message attitude. This finding demonstrates additional support for the strategic aspect of SIDE and extends its impact into the realm of organizational e-mail use. Other scholars have also found that norms in a CMC context can help define communication patterns and that the influence of those norms is likely bound by the group (Postmes, Spears, & Lea, 2000). It is encouraging that both power perspectives (student and instructor) can acknowledge shared norms and in most cases they agree on how the contents of an e-mail message shape message attitude and perceptions of source credibility.
The cognitive aspect of SIDE has been explored quite a bit, yet the strategic aspect has not (Walther & Parks, 2002). This study suggests that the strategic aspect matters and is worthy of additional research. Specifically, we found that adherence to organizational norms—likely driven by strategic factors in a superior-subordinate relationship—affected most of the relationships explored in this study. Familiarity, a type of increased individuating information, did not directly affect the outcomes in this study.
Limitations and Opportunities for Future Research
While the findings from this study clearly indicate the importance of conforming to organizational e-mail norms and carefully considering multiple perspectives in communication research, there are several limitations that we must acknowledge. One limitation is the use of a single organizational environment, higher education, which likely has a strong organizational norm for formal written work. It appears that this expectation extends to e-mail and this norm is well understood by both instructors and students. This superior-subordinate relationship might be very distinct due to the age and experience-level of the students. It is important to test these findings in an organizational context where there are both formal and informal organizational norms for e-mail use to see if the hypotheses are supported in organizations other than higher education. In addition, while the findings revealed no main effects for familiarity, with the significant manipulation check, it is possible that this manipulation was not reflective of reality and thus people may not be considering their actual feelings of familiarity.
It is also important to examine these findings in other relationships and contexts. While there is a strong organizational norm for formal e-mail messages exchanged between superior and subordinate, this is likely not the case when subordinates and supervisors e-mail their peers. There is also a possibility that informally constructed messages are more valuable than formally constructed messages in other contexts. These findings might be organizationally, relationally and context bound and are worthy of future study.
There are several outcomes, not examined in this study, which should be considered in the future. Communication overload is increasingly a concern in organizations (Farhoomand & Drury, 2002) and future studies should examine how norm violations impact perceived communication overload. It is also worth determining the specific aspects of e-mail messages that resemble a text message (or a norm-incongruent message) that are disliked most by superiors and subordinates. Perhaps by combining many specific instances of informally crafted e-mail elements we missed the one that matters the most. All of these areas should be addressed in future studies.
E-mail will most likely continue to grow as a viable medium for organizational communication. The findings from this study demonstrate additional support for the strategic component of SIDE and extend the theory's impact into the realm of organizational e-mail use. Using a formal e-mail in an organization that values written formality has considerable impact on the desirable outcomes of message compliance, source credibility, and message attitude. Even as a subordinate becomes familiar to the superior, this does not overcome the strong organizational norms that dictate formality, even in e-mail.
The authors would like to thank Russell Corrigan for his initial work on testing the e-mail messages used in this study and Jorge Pena for his helpful comments on this manuscript. A previous version of this paper was presented at the International Communication Association Conference, May, 2008.
About the Authors
Keri K. Stephens is an Assistant Professor in the Department of Communication Studies at the University of Texas at Austin. Her research focuses on workplace information and communication technology (ICT) use, especially complex uses such as sequential, multitasking, and combinatorial ICT use.
Address: University of Texas at Austin, Department of Communication Studies, 2504A Whitis, CMA 7.114, A1105, Austin, TX 78712.
Renee L. Cowan is an Assistant Professor in the Department of Communication at The University of Texas at San Antonio. Her research focuses on dysfunctional organizational communication such as workplace bullying, the intersection between work and life, and communication technologies including e-mail and blogs.
Address: University of Texas at San Antonio, Department of Communication, One UTSA Circle, MB 2.248E, San Antonio, Texas 78249.
Marian L. Houser is an Associate Professor in the Department of Communication Studies at Texas State University. Her research focuses on instructional and interpersona l communication, especially how communication impacts the student-teacher relationship both in and outside of the classroom.
Address: Texas State University-San Marcos, Department of Communication Studies, 601 University Drive San Marcos, TX 78666.
Directions: Please read the student e-mail listed below and answer the following questions.