Instructor Response to Uncivil Behaviors in the Classroom: An Application of Politeness Theory

We investigated student perceptions of instructor responses to classroom incivility with a 2 (passive or active student incivility) x 2 (instructor avoidance or bald-on-record response) experimental design. Undergraduate students (n = 281) were randomly assigned to view one of four videos of an incivility incident. They then evaluated the instructor’s behavior with respect to her credibility (competence, caring, and trustworthiness) and effectiveness, as well as how motivating the instructor was. Results indicated that when students in the video engaged in active incivility, bald-on-record responses in comparison to avoidance were considered to be more motivating and effective, and resulted in higher perceptions of instructor caring and trustworthiness. In the passive incivility condition, instructor response did not predict any outcome variable except trustworthiness.

Students do not always perform incivility behaviors to be rude and may not have a strategic goal in mind. Sometimes students simply act with their own self interests in mind, despite being in a classroom environment. For example, Wei and Wang (2010) found text messaging in class was most strongly predicted not by classroom factors, but by cell phone habits outside of class. Most uncivil behaviors are of this sort, what Berger (2002) has labeled passive incivility behaviors-mild disruptions like reading during class time or talking quietly to another student. In contrast, active incivility behaviors are directly insulting and disrespectful and may even involve direct challenges to the instructor (Berger, 2002).
Instructors can to some degree prevent incivility from occurring if they are credible and nonverbally immediate (Boice, 1996;Miller, Katt, Brown, & Sivo, 2014). Even the most warm and approachable instructor, however, may occasionally have to deal with unanticipated uncivil behavior during class. When this occurs, instructors may feel pressure to respond correctly and quickly. Indeed, some scholars suggest students want instructors to take immediate effective action to end poor classroom behaviors (Alkandam, 2011;Boysen, 2012;Myers, 2003). In such cases, a variety of options are available to instructors. Some take a "soft approach" such as staring at the student, making a joke, making eye contact, moving toward that part of the classroom, or speaking to the student privately. This type of response has been recommended for dealing with passive incivility behaviors (Ali, Papakie, & MsDevitt, 2012;Nilson & Jackson, 2004;Nordstrom, Bartels, & Bucy, 2009). For behaviors that demonstrably affect the flow of the class, a more direct approach has been recommended such as reassuring the class, being honest if something is not working, speaking directly with the student about his/her behavior (Boysen, 2012) and, if necessary, making use of the campus chain of command (Berger, 2002).
Despite wide-ranging suggestions, little empirical research has compared the effectiveness of various approaches. A notable exception is Boysen's (2012) use of text vignettes to test the perceived effectiveness of several instructor response strategies (e.g., discussing the behavior with the class, telling the student privately the behavior was inappropriate). Boysen found instructors directly or privately addressing inappropriate behaviors with students were seen as the most effective by other students. However, he did not investigate how the type of uncivil behavior affected student preferences for instructor action.

Facework and Politeness Theory
Research on facework and politeness theory provides a theoretical framework for identifying effective instructor responses to different magnitudes of classroom incivility behaviors. Face has been defined as "the positive social value a person effectively claims for himself by the line others assume he has taken during a particular contact" (Goffman, 1955, p. 5). For example, instructors expect that they will be respected as the authority in a classroom, if students do not respond to the instructor in a manner that conveys the expected respect, they threaten the face of the instructor in front of the class. Students, too, expect to be treated as valued and intelligent members of the classroom environment. If an instructor corrects or criticizes them, they may feel they have lost face because their competence has been called into question. When people interact, they automatically activate a combination of face concerns about their own face and that of others. Individuals look bad when they maintain their own face without showing any regard for others with whom they are speaking. Facework is the communication used to maintain the image of individuals in the exchange (Holtgraves, 1992;Metts, 2000). Brown and Levinson's (1978) politeness theory proposed that the main mechanism by which individuals conduct facework is by selecting from among five levels of politeness, or indirectness, to mitigate a potentially face threatening remark. The least face threatening strategy is not performing Yrisarry, Neuberger, and  the act at all, also called avoidance. Avoiding making a face threatening act altogether, however, negates the possibility that potential target may agree to a request. The most face threatening strategy, baldon-record politeness, completely disregards hearers' face concerns by making an unqualified request or remark. Three intermediate levels of politeness were also proposed, though they have rarely been operationalized in research. To determine which level of indirectness to employ in a potentially face threatening action speakers are posited to consider three factors: 1) the power difference between the two individuals, 2) the social distance between the two individuals, 3) and the level of imposition from the specific act (Brown & Levinson, 1978). Individuals are usually more polite when another person maintains a higher power position or is not socially close to them, and when the act is highly face threatening. They can be less polite when the situation is reversed.

Facework, Politeness, and the Classroom Environment
The concept of facework and face threat mitigation has been applied to the university classroom context. In general, "competent instruction must include the ability to mitigate face threats and negotiate mutually acceptable identities during key instructional interactions" (Kerssen-Griep, Trees & Hess, 2008, p. 314). Students have been found to respond positively to instructors who make face respecting comments (Kerssen-Griep, 2001;Sabee & Wilson, 2005), and who mitigate face threat when they provide negative feedback on student performance (Kerssen-Griep, Hess, & Trees, 2008;Trad, Katt, & Miller, 2014;Witt & Kerssen-Griep, 2011). Appealing to a student's responsibility to the class has been identified as threatening to face. Students observing such interactions want to feel solidarity with their classmates and when this cohesion was threatened they also feel threatened (Rudick & Martin, 2011).
When instructors pay attention to students' face needs they also motivate students. Instructors who use high levels of politeness when making requests prompt positive emotional responses from their students that in turn influence intentions to comply (Zhang, 2011). Analyzing faculty-student disagreements, Rees-Miller (2000) observed politeness strategies like using inclusive pronouns, positive comments, and humor, as well as downplaying disagreements, can "enhance the face of the addressees and thus encourages students to participate actively" (p. 1107), especially when they felt like a valued member of the class. Rees-Miller concluded this tactic can be especially beneficial when the students want something from the instructor and the instructor wants to maintain positive relationships with students.
Although none of these studies specifically addressed student incivility behaviors, they imply that instructors who experience student incivility will experience a dilemma. According to politeness theory, damaging or protecting the face of a student will also damage or protect the face of the instructor (Holtgraves, 1992). Bald-on-record politeness strategies in which the instructor directly rebukes a student for incivility could be highly face threatening to the student, which in turn could make the instructor appear less credible. The way they respond to one uncivil student can negatively affect the entire class (Martin, 2011). At the same time, previous research has demonstrated that instructors are not seen as effective when they avoid responding to student incivility behaviors (Boice, 1996;Boysen, 2012). If instructors allow their own face to be attacked via active student incivility, a real loss of the social value they claim for themselves in terms of authority in the classroom may ensue.
Put another way, instructors dealing with student incivility may find themselves debating between loss of different aspects of credibility. Since the time of Aristotle, source credibility has been assumed to be comprised of three dimensions: intelligence, character, and goodwill-now more commonly called competence, trustworthiness, and caring Teven & McCroskey, 1997). Instructors who do not address incivility in the classroom strongly enough may risk loss of students' perception that they can be trusted to keep control over the classroom. However, Yrisarry, Neuberger, and  if they address incivility in a way that threatens the face of students, they may lose ground in students' perception of the caring dimension of their credibility.
Politeness theory suggests that an additional factor in instructors' decision about how to address uncivil student behavior is the magnitude of the imposition the incivility presents to the instructor's face. Classroom incivility includes a large span of behaviors, from more subtle actions, such as texting while in class, to more intentionally rude behaviors such as making an inappropriate comment directly to an instructor (Berger, 2002;Boysen, 2012). Thus, the two main categories of incivility, passive versus active, may require different types of response on the part of the instructor because they represent different levels of imposition. Based on literature about classroom face and politeness, we anticipated active uncivil behaviors would be viewed by students as highly face threatening to the instructor, not to mention damaging to the overall classroom environment, and students would approve a more direct, less polite approach by the teacher (Holtgraves, 1992). Literature indicates that in such cases a direct approach should be motivating for students and give students the impression that their instructor cares about their welfare and can be trusted to protect the overall classroom environment. In the case of passive uncivil behaviors, in which little imposition was made on the face of the instructor, we predicted the opposite, that students would not endorse a direct, face-threatening response. Instructors who over-responded, as it were, would do so at the cost of their own image in the minds of students, and would end out demotivating them. Formally stated, then, we hypothesized the following: H1: There will be an interaction between type of student incivility and type of instructor politeness response such that in the passive student incivility condition, avoidance will be seen as a more effective instructor response, but in the active student incivility condition, bald-onrecord politeness will be seen as a more effective instructor response. H2: There will be an interaction between type of student incivility and type of instructor politeness response such that in the passive student incivility condition, avoidance will elicit higher perceived instructor caring, but in the active student incivility condition, bald-on-record politeness will elicit higher perceived instructor caring. H3: There will be an interaction between type of student incivility and type of instructor politeness response such that in the passive student incivility condition, avoidance will elicit higher perceived instructor trustworthiness, but in the active student incivility condition, baldon-record politeness will elicit higher perceived instructor trustworthiness. H4: There will be an interaction between type of student incivility and type of instructor politeness response such that in the passive student incivility condition, avoidance will elicit higher student motivation, but in the active student incivility condition, bald-on-record politeness will elicit higher student motivation.
Research is not clear on what effects instructor decisions about politeness strategies in the face of either active or passive incivility would have on students' perceptions of the competence dimension of instructor credibility. Therefore, we posed a research question: RQ1: What effect will active vs. passive student rudeness and avoidance vs. bald-on-record politeness strategies have on perception of instructor competence? Yrisarry, Neuberger, and

Method
A 2 (passive or active incivility) x 2 (avoidance or bald-on-record instructor response) experimental design exposed participants to videos depicting a classroom-based scenario in which type of student incivility and level of instructor response politeness were manipulated.

Participants
Participants (n = 281) were students at a large southeastern university recruited from several undergraduate communication courses. The self-reported race of the participants was reflective of the campus composition with 11.4% African American/Black, 6.0% Asian/Pacific Islander, 61.6% Caucasian/White, 17.4% Hispanic/Latino and 1.8 % mixed race participants (1.8% chose not to answer). Participants were 43.1% freshmen, 21.0% sophomores, 21.7% juniors, 13.2% seniors, and 1.1% fifth year or higher and average age was 20.08 (SD=3.34). Females made up 60.1% of the sample, and males 39.9%. Some instructors offered course credit or extra credit to students who participated in the online survey.

Procedure
Students were recruited via an email sent by their instructors with a link to the online study. Once they opened the link, they were informed about the study structure and answered several demographic items before being randomly assigned to one of four video conditions. After viewing the video, they answered questions about their perceptions of the instructor's behavior.

Stimulus Materials
Each of the videos produced for this study depicted an excerpt of a college lecture including a student interruption and instructor response. All videos were approximately 60 seconds long and identical except for sections containing the experimental manipulations. Passive student incivility behavior was presented as one student saying in an aside to another, "This is so boring. I don't know why these lectures take so long." Active student incivility behavior was presented by a student saying directly to the instructor, in front of the entire class, "This is so boring. I don't know why these lectures take so long." The instructor responded to the student in one of two ways based on the levels of politeness developed by Brown and Levinson (1987). The first condition was avoidance in which the instructor ignored the incivility being committed and said nothing. The second condition was bald on record with the instructor saying, "Hey Alexis...it is really inappropriate for you to interrupt lecture like this. If you have any comments to make about the lecture you need to find me in my office. Now is not the time." A pilot test confirmed the manipulations were viewed by participants in the intended way.

Measures
Unless otherwise noted, items were measured using seven-point semantic differential scales.

Demographics
Participants indicated their age, gender, race/ethnicity, and year in school. Students were asked to rate instructor effectiveness by indicating options from very ineffective to very effective, as modeled after Boysen (2012). The scale consisted of seven items, was developed for this study, and demonstrated good reliability (α=.93).

Instructor Credibility
Instructor credibility was measured using McCroskey and  18-item scale that measured three dimensions of credibility (i.e., competence, caring, trustworthiness) with six items per subscale. Reliabilities in the current study were: α=.91 for competence, α=.85 for caring, and α=.86 for trustworthiness with an overall reliability rating of α=.91 for the complete scale.

Manipulation Check
Participants were asked to assess the politeness of both student and instructor responses in the video on five item scales to determine the fidelity of the intended manipulations. Reliabilities for both the student (α=.92) and instructor scales (α=.94) were good.

Data Preparation
Survey responses with more than 20% of the information left incomplete were removed. Results in which the respondent did not watch a full minute of the stimulus video were also removed as these participants would not have witnessed the full manipulation.

Manipulation Checks
An independent samples t-test was run to verify that participants perceived the active and passive incivility conditions differently. Perceptions of student politeness in the passive incivility condition (M=2.48, SD=1.13) were significantly higher than perceptions of student politeness in the active condition (M=1.60, SD=0.91; t (417) = 7.17, p < .001). A t-test also demonstrated levels of instructor politeness were perceived differently in the videos displaying avoidance (M=3.72, SD=1.34) and baldon-record responses (M=2.92, SD=1.82) techniques [t (278) = 16.05, p < .001]. The bald on record technique was seen as significantly less polite than the avoidance response, as intended.

Hypothesis and Research Question Testing
The two hypotheses and one research question were tested with a series of two-way ANOVAs. Results are presented in Table 1. Yrisarry, Neuberger, and  Hypothesis one predicted an interaction effect between type of student incivility and type of instructor politeness response such that in the passive student incivility condition, avoidance would be seen as a more effective instructor response, but in the active student incivility condition, bald-onrecord politeness would be seen as a more effective instructor response. An interaction effect did emerge, such that the bald-on-record strategy (M = 5.13, SD = 1.68) was perceived to be more effective than the avoidance strategy (M = 3.93, SD = 1.71) in the active incivility condition. In the passive incivility condition no significant difference was present between the bald-on-record strategy (M = 3.84, SD = 1.68) and the avoidance strategy (M = 3.88, SD = 1.43). Thus, the predicted interaction effect was observed, but not in the expected direction in the passive incivility condition. Hypothesis 1 was partially supported.
Hypothesis two predicted an interaction effect between type of student incivility and type of instructor politeness response such that in the passive student incivility condition, an instructor avoidance response would be associated with higher perceived instructor caring, but in the active student incivility condition, an instructor bald-on-record response would be associated with higher perceived instructor caring. An interaction effect did emerge, such that the bald-on-record strategy (M = 4.16, SD = 1.27) was perceived to be more effective than the avoidance strategy (M = 3.70, SD = 1.01) in the active incivility condition. In the passive incivility condition, although the effect was in the predicted direction, the difference between the bald-on-record strategy (M = 3.48, SD = 1.22) and the avoidance strategy (M = 3.84, SD = 1.14) was not statistically significant. Thus, hypothesis two was partially supported. Yrisarry, Neuberger, and  Hypothesis three predicted an interaction effect between type of student incivility and type of instructor politeness response such that in the passive student incivility condition, an instructor avoidance response would be associated with higher perceived instructor trustworthiness, but in the active student incivility condition, an instructor bald-on-record response would be associated with higher perceived instructor trustworthiness. No interaction effect emerged. A main effect for instructor response type was evident, with students expressing more trust for the instructor in using the bald-on-record strategy (M = 4.78, SD = 1.05) than the avoidance strategy (M = 4.47, SD = 1.06) condition, regardless of the type of student incivility. Hypothesis 3, therefore, was not supported.
Hypothesis four predicted an interaction effect between type of student incivility and type of instructor politeness response such that in the passive student incivility condition, an instructor avoidance response would be associated with higher student motivation, but in the active student incivility condition, an instructor bald-on-record response would be associated with higher student motivation. An interaction effect did emerge, such that the bald-on-record strategy (M = 3.40, SD = 1.32) was associated with higher student motivation than the avoidance strategy (M = 2.76, SD = 1.03) in the active incivility condition. In the passive incivility condition, although the effect was in the predicted direction, the difference between the bald-on-record strategy (M = 3.77, SD = 1.26) and the avoidance strategy (M = 3.12, SD = 1.37) was not statistically significant. Thus, hypothesis 4 was partially supported.
Research question one asked what effect active versus passive student rudeness, and avoidance versus bald-on-record politeness strategies, would have on perception of instructor competence. No main or interaction effects emerged with respect to the competence dimension of instructor credibility.

Discussion
Uncivil behavior in college classrooms negatively impacts both instructors and students. When students behave uncivilly, instructors must make decisions on the spot about the best way to address the situation. Knowing their responses may impact not only their relationship with the offending student, but also classroom management over the long term, instructors must wrestle with the dilemma of whether to address student incivility directly, or employ some less face threatening response. Although advice for instructors in such situations is readily available, little empirical research has investigated how various instructor strategies are perceived by students, and even less experimental research has been conducted. This study aimed to address that gap.
Our findings indicate that when the incivility is passive, it may not matter so much what the instructor does to address it. The critical issue is how an instructor addresses active, disruptive behaviors. When incivility was active, an instructor who met active classroom incivility with a direct bald-on-record response was perceived as more motivating, caring, and trustworthy than an instructor who used the strategy of avoidance. This result supports assertions in previous scholarship (Boysen, 2012) that not responding when students interrupt the instructor leads to lower perceptions of instructor effectiveness and perhaps even demoralization among students (via attenuation of motivation). Indeed, Boice's (1996) early work on incivility indicates that efficient, authoritative handling of classroom dynamics can prevent incivility. More research is needed to determine what types of disruption would lead class members to prioritize concerns of classroom management over face concerns of a fellow classmate.
The context of uncivil student behavior highlights several limitations to politeness theory as it has traditionally been studied in the social sciences. Politeness theory (Brown & Levinson, 1978) predicts low power interlocutors will be more polite/less direct than their high power counterparts. Incivility on the part of students flips such a scenario on its head, with the lower power party initiating Yrisarry, Neuberger, and  direct, impolite behavior (albeit sometimes unknowingly), with the higher power party as a result left to contemplate whether to meet directness with directness, or to formulate a more polite message. It is also important to note politeness theory originally examined requests as opposed to responses. Although an instructor's response to an uncivil student is in one sense a request for the student to stop, a request and a response do differ. A request may begin an interaction and a response may finish it. This change in sequence raises issues the original presentation of the theory did not address. Finally, politeness theory addresses interaction in which both parties are working to maintain both positive and negative face of the other party in order to facilitate smooth social intercourse. In the case of uncivil classroom behavior, the offending student typically is unaware of face concerns at all. The onus of addressing them falls on the instructor, who is likely to be simultaneously aware of the enacted threat to his/her own face, the threat to the face of observers, and the tension inherent in any decision that threatens the face of the student.

Limitations
The current study was constrained by the fact that participants were recruited from communication courses and the sample was primarily composed of freshmen. Regarding the stimulus materials, all videos featured an encounter between a white female student and white female instructor. Although this design allowed us to avoid extraneous variables, it also limits the applicability of findings. The videos also depicted a small classroom environment (N<20), and results may vary in a large lecture format. Future research in this area should include more diverse participants and actors in video scenarios, as well as varying the context such as large lecture or lab classes. Additionally, the student participants were not actually enrolled in the mock course depicted in the video, so their identification with the student in the video was inevitably lower than it would be in a real classroom situation, and may have resulted in harsher evaluations of the student incivility and less harsh evaluation of instructor responses.

Future Research
Although effects of instructor strategies in this research were small, it does provide an important early step towards empirical investigation of classroom exchanges that occur every day. Most, if not all, instructors have had the experience of quickly responding to student incivility behaviors and then wondering if they should have handled the situation differently. From a practical standpoint, results indicate instructors should evaluate the type of student incivility when determining their response to an uncivil student. For cases in which students are actively disruptive, responding firmly and directly may be beneficial for a classroom by motivating students and giving them the sense that appropriate action has been taken. When students are engaged in passively uncivil behaviors like side conversations or texting, instructors need not be overly concerned to show a firm hand, but may, perhaps, feel free to choose to ignore the behavior, hint, or speak directly as they feel comfortable.