Public Relations Review 44 (2018) 393–406

Contents lists available at ScienceDirect

Public Relations Review

journal homepage: www.elsevier.com/locate/pubrev

Seeking sincerity, finding forgiveness: YouTube apologies as image

T
repair☆
⁎
Jean Kelso Sandlin , Monica L. Gracyalny
California Lutheran University, United States

A R T IC LE I N F O ABS TRA CT

Keywords: Social media have become important communication tools for organizations and public figures,
Apologies particularly in times of crisis. Public figures are frequently advised to use social media platforms
Crisis management to apologize to publics, and their apologies are often posted by news outlets or individual social
Social media media users. However, evidence suggests social media may function on an interpersonal level, yet
Image repair
traditional image repair strategies are based on a mass media model. Using image repair stra-
YouTube
tegies based in theoretical frameworks from mass mediated and interpersonal communication,
this research examined the verbal behaviors and emotions displayed by public figures apol-
ogizing on YouTube and the relationships these had to audience perceptions of sincerity and
forgiveness as expressed via YouTube comments. Two studies analyzed 335 segments of video
from 32 public apologies on YouTube spanning from 2009 to 2014, and 1971 posted responses.
The interpersonal strategies and expression of emotions were largely unrelated to the perceptions
of sincerity and forgiveness; and the image repair strategies were limited in their relatedness.
However, the content of the comments, a majority of which focused on the reputation of the
public figure, was associated with perceptions of sincerity. Reducing offensiveness was associated
with perceptions of insincerity, as was the combination of reducing offensiveness, denial and
evasion. Negative comments regarding the offender’s reputation were also associated with per-
ceptions of insincerity. Audiences were non-forgiving if the apology was perceived as insincere,
but forgiving if they perceived the apology as sincere. Implications of these results in relation to
the practice and scholarship of public relations are discussed.

1. Introduction

As far back as 2006, the report from the Commission on Public Relations recognized the impact of social media on crisis man-
agement, “The contemporary practice of public relations requires practitioners to immediately respond to … crisis situations via Web
sites, blogs and other new media” (VanSlyke Turk, 2006). However, there is no consensus among scholars on the effectiveness of
apology for image repair and crisis management (Brown, Billings, & Devlin, 2016; Coombs & Holladay, 2008; Coombs & Holladay,
2009; Kim, Avery, & Lariscy, 2009; Walsh & McAllister-Spooner, 2011). The varied conditions under which apologies are made (e.g.
guilty vs. not guilty, history of prior offenses, etc.) make it challenging to measure effectiveness, yet apology continues to be a broadly

☆
This study was supported by California Lutheran University’s Culver Mentor and Research Fellows Program. Technical assistance was provided by Culver
undergraduate research fellows Lauren DuCasse, Shaleena Bautista, Aliyah Navarro and Talia Vanwingerden. The funding from this institutional endowment poses no
conflicts of interest. An earlier version of this paper was presented at the International Public Relations Research Conference in Florida in March 2017 and appears in
the proceedings.
⁎
Corresponding author at: California Lutheran University, 60 W Olsen Road, MC 3800, Thousand Oaks, CA, 91360, United States.
E-mail address: jsandlin@callutheran.edu (J.K. Sandlin).

https://doi.org/10.1016/j.pubrev.2018.04.007
Received 25 August 2017; Received in revised form 13 April 2018; Accepted 16 April 2018
Available online 25 April 2018
0363-8111/ © 2018 The Authors. Published by Elsevier Inc. This is an open access article under the CC BY-NC-ND license
(http://creativecommons.org/licenses/BY-NC-ND/4.0/).
J.K. Sandlin, M.L. Gracyalny Public Relations Review 44 (2018) 393–406

accepted practice in public relations (PR) strategy, and an ethical mandate if the offender is truly at fault (Coombs & Holladay, 2008).
Today, public figures are frequently advised to use social media platforms to apologize to audiences, particularly when the offense
originated on social media (Baer & Naslund, 2011; Matejic, 2015; Schultz, Utz, & Göritz, 2011; Ulmer, Sellnow, & Seeger, 2013; Utz,
Schultz, & Glocka, 2013). Even if public figures do not initially post their apologies on social media, the public and news outlets often
do. For example, singer Chris Brown’s videotaped apology, posted on YouTube by On Demand Entertainment (2009), received more
than 1.5 million views and more than 4000 comments. Celebrity chef Paula Deen’s video, posted on YouTube by The Young Turks
(2013), received more than 723,000 views and 5000 comments (as of March 2018). Unlike a one-to-many media environment where
users have limited interactions with producers or other users, social media is interactive and users recognize their potential influence
over other users (Kang, 2014; Muntinga, Moorman, & Smit, 2011).
The use of social media in public relations practice has increased in each of the last 12 years, and PR professionals strongly agree
that new media have changed the way PR is practiced (Wright & Hinson, 2017). A survey of more than 4400 PR professionals
worldwide identified the top three issues affecting their communication strategies and practices as 1) dealing with the speed and
volume of information flow, 2) being prepared to deal effectively with crises and 3) managing the digital revolution and rise of social
media (Meng & Berger, 2017).
As they strive to meet these challenges, PR professionals and scholars must also consider the points of incongruence between
traditional image repair strategies that use a mass media framework and social media’s ability to function on an interpersonal level
(Caplan, 2001; Kelleher, 2009; O’Sullivan & Carr, 2017; Ott & Theunissen, 2015; Schultz et al., 2011; Walther et al., 2010). This is an
important distinction because research on effective interpersonal apologies differs significantly from image repair strategies re-
commended by PR scholars (Gracyalny & Mongeau, 2010; Schlenker & Darby, 1981; Schmitt, Gollwitzer, Förster, & Montada, 2004).
Emotions, typically seen as the purview of interpersonal scholars, impact crisis management, yet PR professionals have just begun to
scratch the surface of their role in crisis communication (Coombs & Holladay, 2005; Jin, Liu, & Austin, 2014; Jin, Pang, & Cameron,
2010; Meer & Verhoeven, 2014; Nabi, 2003; Roschk & Kaiser, 2013; Utz et al., 2013; ten Brinke & Adams, 2015; Zhang, Wang, Wu,
Wang, & Buck, 2017).
Using image repair strategies based in theoretical frameworks from both mass mediated and interpersonal communication re-
search, we conducted two studies to examine the verbal behaviors and nonverbal emotions displayed by public figures apologizing on
YouTube, a video-enabled social media platform, and the relationship these behaviors may have on perceptions of sincerity and
forgiveness of the online commenting audience. Fostering a better understanding of the online environment, including the percep-
tions of the commenting public, will help PR professionals more effectively counsel clients in today’s dynamic media environment
when a public apology may be a strategy for image repair or crisis management. It also offers scholars and educators an additional
lens through which to critique image repair strategies.

2. Literature review

Apology research has interested PR scholars and professionals, particularly as part of image repair strategies and crisis man-
agement. A variety of disciplines and approaches have informed apology scholarship, and a growing number of resources provide a
comprehensive view related to crisis communication (Austin & Jin, 2017; Benoit, 2014; Blaney, 2016; Coombs, 2011; Hearit, 2006).
This literature review is narrowly focused on three main factors that have converged to add a sense of urgency in providing a deeper
understanding of online apologies for the PR profession. One factor is the increased accessibility and use of social media. Another
related, but separate consideration is the ability of social media to enable publics to directly communicate with each other, at times in
very visible ways such as leaving comments. Finally, interdisciplinary research suggests that social media communications may be
perceived as interpersonal, rather than mass media, or some unique combination of the two (Caplan, 2001; Morris & Ogan, 1996;
O’Sullivan & Carr, 2017; Procopio & Procopio, 2007; Walther et al., 2010). Therefore, we consider interpersonal theory, specifically
the components of an effective apology within interpersonal communication. To understand how a shift to a more interpersonal or
masspersonal approach (O’Sullivan & Carr, 2017) might impact the framework of apologies in the practice of public relations, it is
helpful to understand one of PR’s most influential guiding theories related to apologies–Benoit’s image repair discourse (1995, 1997,
2000, 2014).

2.1. Benoit’s image restoration/repair discourse as a starting point

Benoit’s seminal work on the theory of image repair (1995, 1997, 2000, 2014) has shaped the work of PR scholars and profes-
sionals, especially in the area of crisis management response. Benoit described it as a theory to be “used by practitioners to help
design messages during crises and by critics or educators to critically evaluate messages produced during crises” (1997, p. 177).
Although Benoit’s theory has been criticized for its descriptive nature and speculative conclusions, critics also admit that the theory
and research supporting it merit critical commentary because of its depth and breadth (Burns & Bruner, 2000; Coombs, 2007;
Coombs, Frandsen, Holladay, & Johansen, 2010).
Benoit’s theory focuses on message options for organizations and individuals accused of wrongdoing, and identifies five categories
of image repair strategies: denial, evasion of responsibility, reducing the offensiveness of the act, corrective action, and mortification.
Denial includes simple denial and shifting the blame. Evasion of responsibility includes: provocation, defeasibility, accident, and good
intentions. Benoit defines defeasibility as the lack of information or ability. Reducing offensiveness includes: bolstering, minimization,
differentiation, transcendence, attacking accusers and compensation. When bolstering, an organization stresses the good traits over
the offensive act. For example, in the case of an oil spill, the company would stress the quick response to cleanup the spill–bolstering

394
J.K. Sandlin, M.L. Gracyalny Public Relations Review 44 (2018) 393–406

the positive trait of quick response over the act of the spill. When using a differentiation strategy, an organization would stress how
the act is different or distinct from other similar and more offensive acts. For example, if a car dealer was accused of making
unnecessary repairs, the dealership might label these repairs as “preventative maintenance” to differentiate them from the fraudulent
practice of overcharging customers for unnecessary repairs. Corrective action is the plan to solve or prevent the problem. Mortification
refers to admitting guilt and seeking forgiveness. Benoit’s strategies can be combined (Benoit, 1997) and often are (Caldiero, Taylor,
& Ungureanu, 2009). A study that analyzed 18 years of PR crisis strategies based on Benoit’s image restoration theory and Coombs’
situational crisis communication found more than 90% of the cases combined strategies (Kim et al., 2009).
Benoit (2014) admits mortification is a particularly complex image repair strategy, and writes, “No universally agreed conception
of ‘apology’ stipulates exactly what an apology must include” (p. 26). He discusses the potential impact of sincerity and forgiveness
on an audience’s response to apology. He stresses the importance of being sincere to increase credibility and effectiveness, while
recognizing it may be difficult for an audience to judge actual sincerity. “If we believe the apology is sincere, we may choose to
pardon the wrongful act … Hopefully the audience will forgive, but forgiveness is not certain” (Benoit, 2014, p. 26–27). Both in
Benoit’s work and in the critique and experimentation of his work, there is attention given to the key role of the perception of the
audience (Benoit, 1997, 2000, 2014; Burns & Bruner, 2000; Haigh & Brubaker, 2010). Our second study addresses perception in a
social media environment via the commenting audience.

2.2. Increased accessibility and use of social media

The advent and increased use of social media is one of the most significant changes in the image repair landscape since Benoit first
undertook his research. As social media emerge, researchers have used Benoit’s theory to analyze image repair efforts on social media
by celebrities (Moody, 2011), corporations (Caldiero et al., 2009; Muralidharan, Dillistone, & Shin, 2011) and sports figures (Glantz &
Benoit, 2017; Hambrick, Frederick, & Sanderson, 2013; Sheckels, 2013). Other studies analyzing the use of social media in crises have
used other theories such as Coombs’ (2007) situational crisis communication theory (Brown & Billings, 2013), Jin and Liu’s social-
mediated crisis communication model (Jin et al., 2014) and the networked crisis communication model (Utz et al., 2013).
Some scholars argue that social media should be viewed as new media (Procopio & Procopio, 2007), while others view it as a
convergence of media (Jenkins, 2006; Walther et al., 2010), and still others view it as a new phenomenon (Caplan, 2001; Cathcart &
Gumpert, 1983; O’Sullivan & Carr, 2017). Regardless of the view, more exploration is warranted into how audiences perceive
apologies in today’s online environment. Benoit (2014) is explicit in his urging, “As we adopt new media (e.g., the Internet, Facebook,
Twitter) we need to investigate image repair in those media” (p. 132), and others scholars have also advocated this position (Fawkes
and Gregory, 2001; Moody, 2011; Morris and Ogan, 1996; Muralidharan et al., 2011; Taylor and Kent, 2010). Therefore, the advent
and widespread adoption of social media serves as an opportunity to reconsider previous theoretical frameworks related to image
repair strategies.
Alfonso and de Valbuena (2006) observed “The very power of the internet also magnifies the weaknesses in an organization’s
public relations practice” (p. 274). That statement is more true today as the influence and reach of the internet continues to increase,
and PR practices and scholarship struggle to keep pace. Today, 73% of Americans go online every day (Perrin, 2015), and 87% of
American adults have access to smartphones (Nielson Social Media Report, 2016). In emerging and developing countries, smartphone
ownership rose to 37% in 2015 (Poushter, 2016). The increased use of social media via smartphones is of particular note when
undertaking image repair strategies and crisis management because audiences seek out social media during crises for unfiltered and
up-to-date communication (Austin, Fisher Liu, & Jin, 2012; Procopio & Procopio, 2007).

2.3. The convergence of mass and interpersonal communication

In 2015, when NBC’s The Today Show YouTube Channel posted a video of Chipotle’s founder and co-chief executive officer Steve
Ells apologizing for the company’s food-borne illness crisis, the video received 30,000 views (as of March 2018), and elicited 75
comments. One YouTube user commented that Ells “…seems like an honest guy, and I believe he's actually sorry for what happened.”
Another added, “he seems genuine. a likeable character that dude” (Today Show YouTube Channel, 2015). Unlike television, social
media platforms such as YouTube, have the ability to allow for audience feedback and interaction. The comments of the respondents
demonstrate the ability of social media to function on both mass and interpersonal levels, and heightens the need to understand how
this convergence might impact strategies for effective apologies on social media.
Scholars disagree whether it is the messaging (Coombs, Claeys, & Holladay, 2017; Coombs & Holladay, 2009) or the medium
(Schultz et al., 2011; Utz et al., 2013) that most impacts reactions to crisis and to post-crisis reputational repair, so it is important to
consider both the message and the interactivity of social media, including the ability to shift power from the organization or public
figure to the public (Ki & Nekmat, 2014; Sheth & Solomon, 2014) and empower audiences (Kang, 2014; Muntinga et al., 2011). If Ells’
appearance had been limited to traditional TV broadcast, the members of the public who commented on YouTube could not have
shared their opinions with each other, and the viewers could not have shared the video 187 times through their own social media
networks.
Although public comments posted on social media sites may not be representative of social media audiences, PR professionals are
urged to monitor and engage active publics. Comments on social media may indicate a segment of active publics (Bronner & De Hoog,
2011; Chewning, 2016; Cho, Schweickart, & Haase, 2014; Coombs & Holladay, 2012; Dozier, Shen, Sweetser, & Barker, 2016; Kang,
2014; Muntinga et al., 2011). Sentiment expressed via social media comments can be early warning signs of crises or a measure of
audience response and engagement, and can shape the discourse surrounding the issue (Jin et al., 2014; Matejic, 2015; Valentini,

395
J.K. Sandlin, M.L. Gracyalny Public Relations Review 44 (2018) 393–406

Romenti, & Kruckenber, 2017).

Scholars have grappled with how to categorize and study social media communication behaviors, particularly how to accom-
modate the fluidity of the boundaries between mass and interpersonal communication forged by technology (Caplan, 2001; Cathcart
& Gumpert, 1983; O’Sullivan & Carr, 2017; Procopio & Procopio, 2007; Reardon & Rogers, 1988; Sheth & Solomon, 2014; Walther
et al., 2010). Even in the early years of the Internet, scholars recognized the duality of the technology. Morris and Ogan (1996) wrote,
“The Internet is a multifaceted mass medium… Its varied forms show the connection between interpersonal and mass communication
that has been an object of study since the two-step flow associated the two (Lazarsfeld, Berelson, & Gaudet, 1944)” (p. 42). More
recently, terms such as masspersonal (O’Sullivan & Carr, 2017) or hyperpersonal (Caplan, 2001) have been proposed to help bridge the
dichotomy between the two functions. It is an evolving conversation, and an important one for those studying and practicing image
repair and crisis management strategies.
For PR professionals and scholars, this conversation relating the convergence of mass and interpersonal communication is a call to
re-examine guiding theories in all practices (Duhé, 2015). This paper serves as an entry into the conversation by considering two
dominant image repair theories–one from mass communication and one from interpersonal communication–to better understand
how this convergence might impact strategies for effective apologies on social media. Benoit’s theory is familiar to many PR pro-
fessionals and scholars, and is discussed earlier in this paper. Interpersonal theory, drawn from the work of Goffman (1967),
Schlenker and Darby (1981), and Schmitt et al. (2004), differs in that it includes an emphasis on relational content as well as on the
emotions of victims and offenders.

2.4. Interpersonal apologies

According to Schmitt et al. (2004), five verbal components are necessary for an effective interpersonal apology: admit fault, admit
damage, express remorse, ask for forgiveness, and offer compensation. Apologies that contain these components are more likely to be
perceived as sincere and lead to forgiveness (Darby & Schlenker, 1982; Schlenker & Darby, 1981; Schmitt et al., 2004). Furthermore,
apologies containing a greater number of the components are perceived as more sincere and effective (in reducing offender blame and
punishment) than those with fewer components, particularly when the offense is severe (Lewicki, Polin, & Lount, 2016; Scher &
Darley, 1997). Research also suggests that stronger or more intense apologies (e.g., saying “I’m sorry” four times as compared to
once) are more effective in repairing the damage caused by a transgression (Roschk & Kaiser, 2013).
In addition to a verbal apology, offenders are also expected to display an appropriate level of remorse for their actions (Feeney,
2004; Scher & Darley, 1997; ten Brinke & Adams, 2015). A perceived lack of remorse following an interpersonal transgression both
increases the estimation that the offender will repeat the behavior in the future and decreases evaluations of the offender’s moral
character (Gold & Weiner, 2000). In contrast, remorseful offenders are often viewed as less blameworthy (Darby & Schlenker, 1982),
typically receive lighter punishment (Gold & Weiner, 2000), and are more likely to be forgiven for their behaviors (Fisher & Exline,
2006; Gold & Weiner, 2000; McCullough, Worthington, & Rachal, 1997; Waldron & Kelley, 2008; Younger, Piferi, Jobe, & Lawler,
2004). It is important to note, however, that people rely on a variety of cues (e.g., verbal, vocal, and facial) to infer others’ emotions
(Planalp, 1998) and remorse is likely no exception. In fact, Schmitt et al. (2004) argued that “a verbal statement of remorse is neither
necessary nor very effective for making the victim feel that the harm-doer is remorseful” (p. 482), as it is likely that perceived remorse
stems from assessments of what people say as well as what they do. For these reasons, it is important to include offenders’ nonverbal
expressions of emotion in an analysis of effective apologies. Thus, two studies investigated the following questions (study I in-
vestigated RQ1 and RQ2 and study II investigated RQ3):
RQ1: What emotions do public figures display when apologizing to audiences on video-enabled social media platforms?
RQ2: What verbal behaviors do public figures use to apologize to audiences on video-enabled social media platforms?
RQ3: How do the displayed emotions and verbal behaviors used by public figures to apologize on video-enabled social media
relate to receivers’ perceptions of sincerity and forgiveness as expressed via YouTube comments?

3. Study I: Methods

3.1. Sample

In order to be included in the study, apologies had to meet the following criteria: a) made by a highly visible public figure – as
determined by coverage of the apology on either the top Nielson-rated broadcast news channel (NBC) or cable news channel
(FOXNEWS) for that year; b) made between 2009 and 2014, and c) posted on YouTube as a video. Although collected in 2016,
apologies only spanned to 2014 because that was the last year the video archives for the top rated broadcast and cable stations were
both available on Waybackmachine.org. Because political communication is influenced by partisanship, politicians were excluded
from the study. To obtain the sample of public figure apologies, two search engines were used: Google and Waybackmachine.org.
First, Google was searched using the terms “celebrity,” “sports,” “corporate,” “entertainment,” “artists,” “company,” or “public
figure;” followed by the word “apology,” and then the year (ex. 2009, 2010, 2011, 2012, 2013, or 2014). From the initial list of
available apologies, Waybackmachine.org was used to determine whether the apology appeared on either of the top-rated broadcast
news or cable news networks. Of these apologies, only the videos that were posted on YouTube were selected for inclusion. This
resulted in a final sample of 32 video apologies.

396
J.K. Sandlin, M.L. Gracyalny Public Relations Review 44 (2018) 393–406

3.2. Measures

Apologies were coded for both verbal behaviors and nonverbal facial expressions of emotion. Coding for nonverbal display of
emotion was based on Ekman and Friesen’s (1978) and Ekman et al. (2002) Facial Action Coding System (FACS) to identify the
following emotions: sadness, anger, contempt, disgust, fear, and surprise. Happiness/joy was not deemed relevant to the study’s aims
and was coded as “no negative emotion.” Verbal content of the apologies was also coded according to both mass and interpersonal
communication frameworks. To examine verbal apology behaviors from a public relations standpoint, Benoit’s (1995, 1997, 2014
typology of image restoration/repair strategies was used, including: denial, evasion, reducing offensiveness, corrective action, and
mortification. Schmitt et al.’s (2004) five components of an effective interpersonal apology (admit fault, admit damage, express
remorse, ask for forgiveness, and offer compensation) were used to code the interpersonal apology strategies in each video. Finally, in
addition to verbal and nonverbal behaviors, the sex and type of public figure were also coded.

3.3. Procedure

Apology videos were coded by two trained independent raters; nonverbal behaviors were examined separately from the verbal
content. Raters were trained to identify facial expressions related to emotion based on Ekman and Friesen’s (1978) and Ekman et al.
(2002) Facial Action Coding System. This system classifies emotions according to movement of specific muscle groups in the face. For
instance, sadness is displayed by activating the inner brow raiser, brow lowerer, and lip corner depressor (action units 1, 4, and 15).
To code nonverbal content, each video was unitized into 15-s segments and analyzed without audio. Each segment was coded
according to the dominant emotion identified during that segment (e.g., sadness, anger, fear, disgust). Because the apologies varied
significantly in duration (M = 177.17 s; SD = 145.73 s), amount of expression was converted to a percentage of the apology by
dividing the number of segments in which the emotion was expressed by the total number of segments in the video. In coding the
verbal content, all verbal strategies present during each segment were identified and recorded. For example, if one 15-s segment
contained a statement of “I’m sorry, please forgive me,” it would be coded as both an expression of remorse and asking for forgiveness
(according to Schmitt et al.’s classification) or two instances of mortification (according to Benoit). To account for the intensity or
strength of the apology, the verbal strategies were also converted to percentages by dividing the number of times each strategy was
used by the total number of all strategies in the apology.
The 32 apology videos yielded a total of 335 segments for analysis; coding for approximately 20% of the segments was compared.
Intercoder reliability using pooled Cohen’s kappa was 0.71 for nonverbal content (Sadness = 0.76; Contempt = 0.70; Anger = 0.72;
Fear = 0.64; No negative emotion = 0.69; Disgust not observed in data). The pooled Cohen’s kappa score for verbal content was
0.86. Schmitt et al.’s (2004) apology components consisted of: Admit fault =0.80; Admit damage = 0.93; Express remorse = 0.87;
Offer compensation = 0.74; Request forgiveness = 1.00. Benoit’s strategies included: Denial = 0.73; Evasion of responsibility,
comprised of Provocation = 0.74; Defeasibility = 0.85; and Accident = 1.00 (no instances of good intentions were observed); Re-
ducing offensiveness, which included Bolstering = 0.78; Minimization = 1.00; Differentiation = 1.00; and Offer compensa-
tion = 0.74 (no instances of transcendence or attacking accusers were observed); Corrective action = 0.73; and Mortification, in-
cluding Admit fault, Admit damage, Express remorse, and Request forgiveness (see above for reliability). All reliability coefficients
indicated good to very good agreement, with the exception of Fear and No negative emotion which were determined to be “mod-
erate” (Landis & Koch, 1977).

3.4. Results

Of the 32 YouTube apologies examined, 11 (34%) were made by entertainment figures, 10 (29%) made by for-profit CEOs, and 11
(34%) by sports figures. Twenty-nine (91%) of the figures were male; 3 (9%) were female.
Regarding research question one, the vast majority of apologies included some form of nonverbal expression of negative emotion;
only two apologies (6%) did not. Nonverbal sadness was seen in the greatest number of apologies (21 videos, 66%), followed by fear
(15 videos, 47%), then anger (11 videos, 34%), contempt (8 videos, 25%), and surprise (4 videos, 13%). None of the apologies
included an expression of disgust. (See Table 1 for average percentages of emotion; Table 2 displays the frequencies of emotion
expression.)

Table 1
Percentages of Nonverbal Emotions Expressed in YouTube Apologies.
Min% Max% M SD

Sadness 0.00 100.00 34.05 36.22

Anger 0.00 50.00 11.30 17.51
Contempt 0.00 64.71 11.58 16.21
Disgust 0.00 0.00 0.00 0.00
Fear 0.00 100.00 11.79 20.52
Surprise 0.00 60.00 4.28 13.64
No emotion 0.00 100.00 26.98 30.43

Note: N = 32.

397
J.K. Sandlin, M.L. Gracyalny Public Relations Review 44 (2018) 393–406

Table 2
Nonverbal and Verbal Behaviors in Public Figure Apologies.
Behaviors n %

Nonverbal Expression
Sadness 114 34
Anger 47 14
Contempt 43 13
Disgust 0 0
Fear 27 8
Surprise 13 4
No negative emotion 91 27
Total Nonverbal Codes 335
Benoit’s Image Repair Strategies
Denial 16 4
Evasion 29 8
Reduce Offensiveness 77 21
Corrective Action 54 15
Mortification 189 52
Total Image Repair Codes 365
Schmitt et al.’s Interpersonal Apology
Admit fault 90 36
Admit damage 33 13
Express remorse 59 24
Ask for forgiveness 7 3
Offer compensation 60 24
Total Interpersonal Apology Codes 249

The second research question asked about the verbal behaviors used by public figures to apologize on social media. Results
indicate that public figures use a variety of behaviors reflecting both image repair and interpersonal apology strategies.

3.4.1. Image repair strategies

Of the 32 apologies, 5 (15.6%) included statements of denial, 12 (37.5%) included a form of evasion, 21 (65.5%) had statements
reducing offensiveness, 21 (65.5%) used corrective action, and nearly all of the apologies (30, 93.8%) included attempts at morti-
fication. The frequencies of each image repair strategy as well as examples are displayed in Tables 2 and 3.

3.4.2. Interpersonal apology components

The apologies also contained the elements of an effective interpersonal apology. Twenty-nine apologies (90.6%) included

Table 3
Verbal Components: Image Repair and Apology Examples.
Benoit’s Image Repair Example

Denial “I am not a racist. …I was baited.”

Evasion
Defeasibility “When I entered the pros I was a young kid …I was 18 years old, right out of high school. I thought I knew everything, and I clearly
didn’t. …My cousin starting telling me about a substance you could purchase over-the-counter in the DR. My cousin and I, one more
ignorant than the other, decided it was a good idea to start taking it. It was his understanding it would give me a dramatic energy
boost, and otherwise harmless. My cousin would administer it to me, but neither of us knew how to use it properly, providing just
how ignorant we both were.”
Provocation “There was a paparazzi guy and he was antagonizing me – attacking me personally and my family personally, and I was genuinely
hurt by this and made angry by this.”
Accident I didn’t mean for that to happen
Reduce Offensiveness
Bolstering “I pride myself in 20 years – 18 years – of doing standup of using it to heal people, and not hurt; that was my whole thing.”
Minimization “No client, customer, or taxpayer money was impacted by this event.”
Differentiation “We were just trying to do something fresh that no one had ever done, but you know, it turns out some people don’t believe in
Father Christmas.”
Corrective Action “We’re-examining all of our hiring practices to make sure that people like this don’t make it into our stores.”
Mortification See below – Admit fault, Admit damage, Express remorse, Request forgiveness
Schmitt et al.’s Apology
Admit fault “We made a mistake, and I am absolutely responsible.”
Admit damage It’s been extremely difficult knowing that I hurt someone
Express remorse “I am truly, truly sorry.”
Request forgiveness “I can only ask and pray that you forgive me.”
Offer compensation “We’d like to make it up to you by giving you your money back, another return journey for free, and 150 pounds a head as a gesture
of apology.”

398
J.K. Sandlin, M.L. Gracyalny Public Relations Review 44 (2018) 393–406

statements admitting fault, 18 (56.3%) admitted damage, 21 (65.5%) included expressions of remorse, 6 asked for forgiveness
(18.8%), and 18 (56.3%) offered compensation. Tables 2 and 3 display frequencies and examples of each interpersonal apology
component.

3.4.3. Relationships between emotion displays and verbal behaviors

Several behaviors classified as image repair strategies by Benoit, and components of effective interpersonal apologies by Schmitt
et al. (2004), related to nonverbal expressions of emotion. Regarding image repair, results of correlation analyses indicated that
denial (r = 0.525, p = .002) and reducing offensiveness (r = 0.441, p = .012) associated positively with the public figure’s anger
expression, whereas mortification related negatively to expressing anger (r = −0.465, p = .007), and positively to expressing sad-
ness (r = 0.502, p = .003). Similarly, statements of remorse positively related to expressing nonverbal sadness (r = 0.762, p < .001)
and negatively related to expressing no negative emotion (r = −0.485, p = .005). Making an offer of compensation associated
positively with expressing no negative emotion (r = 0.464, p = .007). Public figures’ facial expressions of emotion appropriately
reflected their verbal statements, consistent with normative apology or defensive behavior (Ekman, 1992; ten Brinke & Adams,
2015).

4. Study II: methods

4.1. Sample

The sample for study II consisted of public comments posted to 28 of the original 32 YouTube videos analyzed in the first study.
Three of the videos coded in study I had public comments disabled, thus, those videos were not included in study II. To obtain the
sample, we selected the first 100 relevant user comments posted to each YouTube video. Comments were deemed relevant if they
related specifically to the apology or to the public figure (excluding advertisements or spam). If there were fewer than 100 comments
posted, the entire set of relevant comments was retained. The final sample consisted of 1971 comments posted to 28 YouTube apology
videos.

4.2. Procedure

Three coders analyzed the public comments posted to the apologies in several steps. In the first step, coders independently sorted
responses into conceptual units (words or phrases representing a comment topic). In the second step, coders generated categories and
examples following Strauss and Corbin’s (1990) open coding system. Coders grouped similar comments into the same category and
created new categories for different comments. Coding for approximately 20% of the responses was compared, with the average
intercoder agreement of 0.85 using the pooled Kappa score which accounts for agreement by chance across a set of codes for three or
more raters (De Vries, Elliott, Kanouse, & Teleki, 2008). Individual reliability coefficients were acceptable: Apology sincere = 0.74;
Apology insincere = 0.74; Positive verbal content of apology = 1.00; Negative verbal content of apology = 0.79; Positive nonverbal
content of apology = 0.84; Negative nonverbal content of apology = 0.88; Positive character = 0.80; Negative character = 0.84;
Forgiveness = 0.94; Withholding forgiveness = 1.00; Intended audience public = 0.89; Intended audience offender = 0.96; Intended
audience unclear = 0.75. These categories are described in the following section.

4.3. Results

Content analysis of the YouTube comments identified 3542 total conceptual units grouped into 7 main categories. The categories
included comments about: a) the nonverbal content of the apology (e.g., “He doesn’t look sad at all!”), b) verbal content of the
apology (e.g., “Good for him for admitting he’s wrong and taking responsibility”, c) public figure’s reputation (e.g., “this guy is a
hero…and has helped a lot of people”), d) perceptions of apology sincerity (e.g., “I believed her, it seemed sincere.”), e) perceptions
of apology insincerity (e.g., “He’s not sorry, he obviously got paid to say this”, f) decisions about forgiveness (e.g., “I forgive him, he’s
soo cute!”) including implied forgiveness (e.g., “He seemed contrite, everybody makes mistakes, he gets a second chance”), and g)
decisions about withholding forgiveness (e.g., “I’m sorry to say I don’t forgive!”) as well as implied non-forgiveness (e.g., “Apology…
not accepted!”). In addition, comments were coded according to whether the intended audience for the comment was the public
figure who apologized or if it was directed to other commenters on the site. Table 4 displays the frequencies of comments for each
category.
To answer RQ3, the facial expressions and verbal apology behaviors from the YouTube videos in study I were examined in relation
to the percentages of public comments regarding sincerity/insincerity and forgiveness/no forgiveness posted in response to the
apologies. These relationships were tested using a series of multiple regression analyses.

4.3.1. Nonverbal facial expression of emotion

To test for the effects of facial expression of emotion on audience perceptions of sincerity expressed via YouTube comments, the
five emotion expressions (anger, sadness, fear, contempt and surprise) were entered as predictor variables, with perceived sincerity as
the continuous criterion variable. The results of the analysis were not significant, F(5,22) = 0.41, p = .981, R = 0.176, adj
R2 = −0.19. In addition, the regression analyses showed no significant effects of facial expression of emotion on perceived insin-
cerity (F(5,22) = 1.01, p = .436, R = 0.432, adj R2 =0.002), audience forgiveness (F(5,22) = 0.221, p = 0.95, R = 0.219, adj

399
J.K. Sandlin, M.L. Gracyalny Public Relations Review 44 (2018) 393–406

Table 4
Frequencies of YouTube Comments Relating to Verbal/Nonverbal Apology Content, Public
Figures’ Reputation, Perceived Sincerity, and Forgiveness.
Comment Topic n %

Verbal content of apology

Positive 59 3
Negative 154 8
Total Verbal Content Codes 213
Nonverbal content of apology
Positive 46 2
Negative 65 3
Total Nonverbal Content Codes 111
Public figure’s character
Positive 493 25
Negative 725 37
Total Character Codes 1218
Perceived sincerity
Sincere 120 39
Insincere 191 61
Total Sincerity Codes 311
Audience forgiveness
Forgive 100 63
Not forgive 60 37
Total Forgiveness Codes 160
Intended audience
Public 1045 67
Offender 333 21
Unclear 160 11
Total Audience Codes 1496

R2 = −0.169), or audience withholding forgiveness (F(5,22) = 0.424, p = 0.83, R = 0.296, adj R2 = −0.12).

4.3.2. Benoit’s image repair strategies

With respect to Benoit’s typology, multiple regression analyses were conducted to test the effects of Benoit’s image repair stra-
tegies on audience perceptions of sincerity and forgiveness as expressed via social media comments. To test of the effects of Benoit’s
typology on perceptions of sincerity, the image repair strategies were entered as predictor variables with sincerity entered as the
criterion. The results of the regression were not significant (F(3,23) = 0.73, p = .607, R = 0.39, adj R2 = −0.06). To examine the
effects of image repair strategies on audience perceptions of insincerity, the strategies of denial, evasion, and reducing offensiveness
were entered as predictor variables with perceived insincerity as the criterion. Results of the analysis were significant, (F
(5,25) = 3.40, p = .036, R = 0.56, adj R2 =0.22), indicating that nearly one quarter of the variance in perceived insincerity was
predicted by the public figure’s use of denial, evasion, and reducing offensiveness. There was a significant effect of reducing of-
fensiveness (β = 0.52, p = .009), such that reducing offensiveness increased audience levels of perceived insincerity. The individual
effects of denial (β = −0.19, p = .313) and evasion (β = −0.21, p = 0.256) were not significant in the final model. Finally, with
respect to forgiveness, the results of the analyses were not significant for either audience forgiveness (F(5.20) = 0.52, p = .728,
R = 0.35, adj R2 = −0.096) or withholding forgiveness (F(5.20) = 0.70, p = .628, R = 0.39, adj R2 = −0.06).

4.3.3. Interpersonal apology components

Multiple regression analyses found no significant relationships between Schmitt et al.’s (2004) components of an effective in-
terpersonal apology (i.e., admit fault, admit damage, express remorse, offer compensation, request forgiveness) and audience per-
ceptions of sincerity (F(5,22) = 0.31, p = .900, R = 0.28, adj R2 = −0.15) or insincerity (F(5,22) = 0.83, p = .543, R = 0.39, adj
R2 = −0.03) as expressed in comments. There were also no significant relationships between the interpersonal apology components
and audience forgiveness (F(5,22) = 0.41, p = .840, R = 0.29, adj R2 = −0.12) or withholding forgiveness (F(5,22) = 1.42,
p = .257, R = 0.49, adj R2 = −0.07).

4.3.4. Public figure’s reputation and perceptions of sincerity and forgiveness

In addition to verbal apology strategies and facial expressions of emotion, we also examined comments relating to reputation. A
majority of the comments posted in response to the videos related to people’s thoughts about the public figure’s reputation, both
positive (e.g., “I’ve always loved her style”) and negative (e.g., “he’s a complete jerk”). In fact, 62% of all conceptual units referred to
commenters’ beliefs about the public figure’s reputation (See Table 4). Not surprisingly, these beliefs also related to commenters’
perceptions of sincerity/insincerity (See Table 5). There was a significant correlation between making positive comments about the
public figure’s reputation and perceptions of apology sincerity (r = 0.53, p = .004). Moreover, making negative comments about the
public figure’s reputation was related to perceptions of apology insincerity (r = 0.375, p = .050). These associations are important to
note, considering that perceptions of apology sincerity were related strongly to forgiveness (r = 0.761, p < .001), whereas

400
J.K. Sandlin, M.L. Gracyalny Public Relations Review 44 (2018) 393–406

Table 5
Percentages of YouTube Comments Regarding Sincerity of Public Figure and Forgiveness.
Min% Max% M SD

Apology Insincere 0.00 31.0 10.20 9.0

Apology Sincere 0.00 33.0 5.10 8.5
No Forgiveness 0.00 38.0 3.68 9.3
Forgiveness 0.00 30.0 4.40 8.3

perceptions of insincerity related to commenters withholding forgiveness (i.e., “no forgiveness”, r = 0.383, p = .040).

4.3.5. Comment audience: public figure or other commenters

The final analysis examined the intended audience of the YouTube comments. Of the total 1496 audience units coded, the large
majority of statements (n = 1045, 70%) were directed to the viewing public or other commenters, while significantly fewer (n = 333;
22%) were directed to the public figure who apologized in the video. The intended audience was unclear in approximately 11% of the
statements (n = 160).

5. Discussion

Although there are some differences in the theoretical traditions and practical strategies of interpersonal apology and image
repair, both traditions agree apologies play an important role in repairing the social, relational and reputation damage caused by a
transgression. These studies sought to extend the research in three key ways: by determining what strategies public figures used when
apologizing on YouTube, assessing the commenting audience’s response to those strategies, and determining what strategies were
more likely to lead to perceptions of sincerity and forgiveness as expressed via YouTube comments.
Results indicated that public figures use both image repair and interpersonal apology strategies when apologizing on social media
and their facial expressions are consistent with normative expectations of apology behaviors. However, as we attempted to determine
the strategies that led to perceptions of sincerity and forgiveness, there was a striking lack of evidence that online audience per-
ceptions, as reflected in the comments, were related to the theories of either interpersonal apology or image repair. For example,
there were no significant results for apologies that adhered to Schmitt et al.’s (2004) components of an effective interpersonal
apology. Nor were there significant effects of facial expression of emotion on perceived sincerity/insincerity and forgiveness, as had
been previously indicated in the work of ten Brinke and Adams (2015) based on Ekman and Friesen’s (1978) and Ekman et al. (2002)
Facial Action Coding System. In addition, the majority of Benoit’s image repair strategies did not relate to perceptions of forgiveness
or sincerity/insincerity. However, an explanation may be in the comprehensive consideration of the following results:

• A majority of comments (70%) were directed to the public or other commenters.

• A majority of comments (62%) referred to beliefs about the public figures’ reputations.
• Negative comments about the public figure’s reputation related to perceptions of apology insincerity.
• Positive comments about the public figure’s reputation related to perceptions of apology sincerity.
• Perceptions of apology sincerity were related to forgiveness.
• Perceptions of apology insincerity related to withholding forgiveness.
YouTube users motivated to comment may differ from non-commenting users, yet PR professionals–especially those responding to
crises–are encouraged to monitor the online conversions “with the full understanding that the online environment has its own
rules…” (Valentini et al., 2017, p. 64). We must not assume, for example, that the results of study II are an indication that long-held
interpersonal or image repair strategies are not useful or effective, rather, we must consider why these long-held strategies do not
reliably relate to perceptions of sincerity and forgiveness among YouTube commenters.
A possible explanation for the study’s results is to consider the motivations of the commenters, and the way in which YouTube’s
interactive, online environment facilitates the expression and influence of commenters’ opinions. Research has already determined
that commenters engage more with negative posts than positive (Thelwall, Pardeep & Farida, 2012); and individuals use media that
fulfill their needs (Katz, Blumler, & Gurevitch, 1973), with the expression of opinions being identified as one of 10 uses of social
media (Whiting & Williams, 2013). In our study, prior to viewing the apology, commenters who were drawn to view the apology may
have already been familiar with the public figure and formed an opinion about him or her and the incident for which they were
apologizing. Commenters may be using the YouTube platform as a way to confirm and express their biases about public figures. For
example, one commenter wrote, “IT'S SIMPLE [sic]. If you're a true John Mayer fan you'll stick with him no matter what, ‘cause you
know that he's got a good heart. … I will always love him no matter what. Whatever stupid thing he might do, he's still a brilliant
musician and a wonderful person. Nothing will ever change that fact:).” The comment clearly indicates that the commenter was a fan
of John Mayer prior to the incident he apologized for, and that her pre-existing attitude influenced her perception of his apology. In
another example, after comedian Jimmy Kimmel made a racial joke, one commenter responded to his YouTube apology with, “This
was a poor joke. I am very embarrassed by this stupidity. Kimmel should leave the network. This was not a very sincere apology. Only
trying to save his own ass and to save his ad money. Kimmel has always been a trashy comedian. Not cool Jimmy.”

401
J.K. Sandlin, M.L. Gracyalny Public Relations Review 44 (2018) 393–406

Although we know online users are more likely to select information that aligns with their attitudes (Knobloch-Westerwick,
Johnson, & Westerwick, 2015; Winter, Metzger, & Flanagin, 2016; Yeo, Xenos, Brossard, & Scheufele, 2015; Yin, Mitra, & Zhang,
2016), the comments in the above examples reveal that commenters may have sought out and interpreted the apologies based on
previously held opinions and then were motivated to comment by their desire to express their opinions (eg., “he's still a brilliant
musician and a wonderful person” and “[he] has always been a trashy comedian”). In this way, the apology response of the com-
menters appears to be consistent with confirmation bias–a preference for information seekers to choose and interpret information
consistent with their views and beliefs (Johnston, 1996; Jonas, Schulz-Hardt, Frey, & Thelen, 2001; Snyder, 1984). These results
indicate a need to design a study that can assess viewers’ pre-existing opinions about the public figures and the incidents for which
they are apologizing, and this is discussed in greater detail in the future research section below. However, the results also provide new
insights and concerns in relation to the practice and scholarship of public relations.
Although YouTube commenters are not typical and their responses cannot be generalized to the YouTube viewing population, that
the commenters’ responses did not reliably relate to long-held image repair and interpersonal strategies is a concern, particularly
when considered in relation to recent studies. First, YouTube comments impact how viewers evaluate associated videos (Walther
et al., 2010) and can shape community linguistic norms that may impact how topics are framed (Kwon & Gruzd, 2017). In addition,
network analysis research challenges the concept that a few “influentials” can shape public opinion in online networks and, instead,
finds that influence can be driven by a critical mass of easily influenced individuals (Watts & Dodds, 2007). Therefore, this atypical
group of YouTube viewers (the commenters) may be influencing non-commenting viewers with regard to perceptions of sincerity and
forgiveness. In study II, 70% of the comments were directed at the public or other commenters. And since YouTube commenters
engage more with negative posts than positive (Thelwall et al., 2012), they may be influencing other viewers to perceive the apology
more negatively.
Social media algorithms that recommend content for individual users (Bennett & Iyengar, 2008) exacerbate this concern for
negative responses to YouTube apologies. YouTube’s algorithm is undisclosed and understudied, and acts as a “filter bubble” to
further narrow exposure to content (Arthurs, Drakopoulou, & Gandini, 2018, p. 7). According to a Wall Street Journal investigation
(Nicas, 2018), of the more than a billion YouTube hours watched daily world-wide, YouTube’s recommendations drive more than
70% of its viewing time, making the algorithm among the single biggest deciders of what people watch. The investigation demon-
strated how, for example, when a user shows a political bias in what they choose to view, YouTube typically recommends videos that
echo those biases, often with more-extreme viewpoints and “isolate users in ‘filter bubbles’ where they hear largely likeminded
perspectives” (p. 1). Thus, users congregating around content form echo chambers that affirm existing beliefs of the micro-audience
(Cacciatore, Scheufele, & Iyengar, 2016). Recent research in computational social science found evidence of the echo chamber effect
specifically with online videos (Bessi et al., 2016). The researchers studied how the same videos were consumed on Facebook and
YouTube in a sample of 12,000 users. Their result found that “content drives the emergence of echo chambers on both platforms” (p.
1). In other words, on the social media platforms of Facebook and YouTube, users were selecting videos consistent with their pre-
existing beliefs and making comments affirming those beliefs. The researchers were able to predict the formation of echo chambers by
users’ commenting patterns.
In study II, the finding that positive comments about the public figure’s reputation were related to perceptions of apology sincerity
is consistent with other studies that concluded: 1) a positive attitude towards the CEO increased the persuasiveness of the CEO’s
YouTube apology (Manika, Papagiannidis, & Bourlakis, 2015); 2) that stakeholders’ conversations via comments on social media have
a role in image repair strategies (Chewning, 2016); and 3) monitoring online comments can help assess stakeholder reactions
(Coombs & Holladay, 2012). However, questions remain. For the non-commenting YouTube viewers, after watching a public figure
apologize via a YouTube video: 1) Do the interpersonal and image repair strategies used by the public figure affect their perception of
sincerity and forgiveness? 2) How do the online comments affect their perceptions of sincerity and forgiveness? 3) Does their pre-
existing opinion of the public figure, or lack there of, affect their perceptions of sincerity and forgiveness? Perhaps the apology video
on social media only serves to confirm users’ pre-existing attitudes, as was found in previous studies related to online searches
(Knobloch-Westerwick et al., 2015; Winter et al., 2016), online reviews (Yin et al., 2016), and online information seeking (Yeo et al.,
2015). According to Edgerly, Vraga, Dalrymple, Macafee, and Fung (2013), comments, under certain conditions, can resemble de-
liberation, serve as a metric for public opinion, ignite mobilization, foster incivility, or be idle chatter. However, scholars have yet to
definitively identify the conditions that create those outcomes. As Bennett and Iyengar (2008) warned, these new “undertheorized
sociotechnological conditions,” (p. 708) present a new set of challenges for researchers.
The results of study II indicate that positive comments about reputation can lead to perceptions of sincerity, and perceptions of
sincerity can lead to forgiveness as expressed via YouTube comments. Although more research is needed, prudent public relations
professionals should continue to heed earlier recommendations regarding the importance of fostering a positive pre-crisis reputation
(Aula, 2011; Shim & Yang, 2016; VanSlyke Turk, Jin, Stewart, Kim & Hipple, 2012).

5.1. Limitations

In study II, we coded YouTube comments in an attempt to assess audience perceptions of sincerity and forgiveness. For an online
audience, public comments are typically posted by a certain segment of the audience (e.g., those with knowledge of the public figure/
organization/offense who are motivated to post online). YouTube comments may not be representative of the total YouTube viewing
audience, most of whom do not take action via sharing, liking or commenting (Bullas, n.d.).
This research was also limited because of the small number of apologies that met the criteria. Since more people are getting their
news from online sources, the criteria that the apology be reported on both the top broadcast and top cable stations may have unfairly

402
J.K. Sandlin, M.L. Gracyalny Public Relations Review 44 (2018) 393–406

limited the number of apologies. Being able to code the nonverbal behaviors required video and, until recently, YouTube was one of
the only social media platforms to support video. Today, there are other social media platforms that support video, including
Facebook and Twitter. Another limitation was very few women public figures in the sample of apologies, which could speak to the
larger societal issues of fewer women in positions of power. Moreover, there is some research to suggest that audiences might respond
differently to a woman’s apology than to a man’s. Carli (2001) argues that current gender role stereotypes affect how people perceive
women’s social influence attempts, such that women need to be warm, likeable, and agreeable to be persuasive whereas men do not.
Also, we had no way to assess the demographics of the commenters who responded to the videos. Finally, there was no attempt to
assess the guilt of the offenders, which could play a role in both how they apologized and the audience response (Coombs, Holladay &
Claeys, 2016).

5.2. Future research

Although YouTube comments can be useful in assessing stakeholders’ reactions (Chewning, 2016; Coombs & Holladay, 2012),
they may not be representative of the perceptions of the non-commenting viewing audience. PR professionals and researchers must be
cautioned against assuming that content analysis of comments will give a complete picture of stakeholder perceptions – since
commenters’ motivations may differ from non-commenting viewers. An experimental design could more accurately measure the
effects of interpersonal and image repair strategies on viewers’ perceptions of sincerity and forgiveness. In addition, future research
could investigate how audience demographics, motivation to comment, and prior knowledge and opinions of the public figure
influence viewers’ perceptions of sincerity and forgiveness. This approach would also offer the opportunity to test the effects of
confirmation bias on viewers’ perceptions of sincerity and forgiveness, and the effects of positive or negative comments on viewers’
perceptions of sincerity and forgiveness.

6. Conclusion

The results bring us back to a sentiment expressed by earlier researchers, that one must have “full understanding that the online
environment has its own rules…” (Valentini et al., 2017, p. 64). The interpersonal strategies and expression of emotions were largely
unrelated to the perceptions of sincerity and forgiveness; and the image repair strategies were limited in their relatedness. However,
the content of the comments, a majority of which focused on the reputation of the public figure, was associated with perceptions of
sincerity. Reducing offensiveness was associated with perceptions of insincerity, as was the combination of reducing offensiveness,
denial and evasion. Negative comments regarding the offender’s reputation were also associated with perceptions of insincerity.
Audiences were non-forgiving if the apology was perceived as insincere, but forgiving if they perceived the apology as sincere.
This study underscores how online communication behaviors are even more complex than interpersonal or one-way mass media
communication because of the additional layers of complexity from the online, interactive platforms and the potential differences
between commenting and non-commenting viewers. Social media is relatively new yet widely used, therefore far less research exists
from which public relations professionals can develop “working theories” typically used to design PR strategy. PR professionals must
be cautioned against assuming that working theories based on interpersonal or mass media image repair strategies are reliable when
using social media platforms, such as YouTube.
We agree with Edgerly et al. (2013) who suggest using existing communication frameworks to “uncover how well these existing
frameworks, born from different media environments and technologies, fit the YouTube experience” (p. 287). Our study contributed
to that conversation–as it uncovered that image repair strategies from interpersonal and mass media theories are inadequate to fully
explain the perceptions of sincerity and forgiveness as expressed via YouTube comments. It also informs approaches to future re-
search when working within an online environment.

References

Alfonso, G. H., & de Valbuena, M. R. (2006). Trends in online media relations: Web-based corporate press rooms in leading international companies. Public Relations
Review, 32(3), 267–275.
Arthurs, J., Drakopoulou, S., & Gandini, A. (2018). Researching YouTube. Convergence, 24(1), 3–15. http://dx.doi.org/10.1177/1354856517737222.
Aula, P. (2011). Meshworked reputation: publicists’ views on the reputational impacts of online communication. Public Relations Review, 37(1), 28–36. http://dx.doi.
org/10.1016/j.pubrev.2010.09.008.
Austin, L., & Jin, Y. (Eds.). (2017). Social media and crisis communicationTaylor & Francis.
Austin, L., Fisher Liu, B., & Jin, Y. (2012). How audiences seek out crisis information: Exploring the social-mediated crisis communication model. Journal of Applied
Communication Research, 40(2), 188–207. http://dx.doi.org/10.1080/00909882.2012.654498.
Baer, J., & Naslund, A. (2011). The now revolution: 7 Shifts to make your business faster, smarter and more social. New Jersey: Wiley.
Bennett, W. L., & Iyengar, S. (2008). A new era of minimal effects? The changing foundations of political communication. Journal of Communication, 58(4), 707–731.
http://dx.doi.org/10.1111/j.1460-2466.2008.00410.x.
Benoit, W. L. (1995). Accounts, excuses and apologies: A theory of image repair strategies. Albany: State University of New York Press.
Benoit, W. L. (1997). Image repair discourse and crisis communication. Public Relations Review, 23(2), 177–186. http://dx.doi.org/10.1016/S0363-8111(97)90023-0.
Benoit, W. L. (2000). Another visit to the theory of image restoration strategies. Communication Quarterly, 48(1), 40–43.
Benoit, W. L. (2014). Accounts, excuses, and apologies: image repair theory and research. Albany: State University of New York Press Retrieved from Proquest Ebook
Central.
Bessi, A., Zollo, F., Del Vicario, M., Puliga, M., Scala, A., Caldarelli, G., ... Quattrociocchi, W. (2016). Users polarization on facebook and YouTube. PLoS One, 11(8),
e0159641. http://dx.doi.org/10.1371/journal.pone.0159641.
Blaney, J. R. (2016). Putting image repair to the test: Quantitative applications of image restoration theory. Lanham Maryland: Lexington Books.
Bronner, F., & De Hoog, R. (2011). Vacationers and eWOM: Who posts, and why, where, and what? Journal of Travel Research, 50(1), 15–26.

403
J.K. Sandlin, M.L. Gracyalny Public Relations Review 44 (2018) 393–406

Brown, N. A., & Billings, A. C. (2013). Sports fans as crisis communicators on social media websites. Public Relations Review, 39(1), 74–81.
Brown, K. A., Billings, A., & Devlin, M. (2016). Image repair across the racial spectrum: Experimentally exploring athlete transgression responses. Communication
Research Reports, 33(1), 47–53. http://dx.doi.org/10.1080/08824096.2015.1117442.
Bullas, J. (n.d.), 35 Mind numbing YouTube facts, figures and statistics. [web log]. Retrieved from http://www.jeffbullas.com/35-mind-numbing-youtube-facts-
figures-and-statistics-infographic/.
Burns, J. P., & Bruner, M. S. (2000). Revisiting the theory of image restoration strategies. Communication Quarterly, 48(1), 27–39.
Cacciatore, M. A., Scheufele, D. A., & Iyengar, S. (2016). The end of framing as we know it… and the future of media effects. Mass Communication and Society, 19(1),
7–23.
Caldiero, C. T., Taylor, M., & Ungureanu, L. (2009). Image repair tactics and information subsidies during fraud crises. Journal of Public Relations Research, 21(2),
218–228.
Caplan, S. E. (2001). Challenging the mass-interpersonal communication dichotomy: Are we witnessing the emergence of an entirely new communication system.
Electronic Journal of Communication, 11(1).
Carli, L. (2001). Gender and social influence. Journal of Social Issues, 57(4), 725–741.
Cathcart, R., & Gumpert, G. (1983). Mediated interpersonal communication: Toward a new typology. Quarterly Journal of Speech, 69(3), 267–277.
Chewning, L. V. (2016). Measuring the enactment of IRT via social media: What are organizations and stakeholders saying during crisis? In J. R. Blaney (Ed.). Putting
image repair to the test: Quantitative applications to image restoration strategies (pp. 133–155). Lanham, MD: Lexington Books.
Cho, M., Schweickart, T., & Haase, A. (2014). Public engagement with nonprofit organizations on Facebook. Public Relations Review, 40(3), 565–567.
Coombs, W. T., & Holladay, S. J. (2005). An exploratory study of stakeholder emotions: affect and crises. In Ashkanasy, Zerbe, & Härte (Vol. Eds.), The effect of affect in
organizational settings (Research on emotion in organizations): Vol. 1, (pp. 263–280). pp: Emerald Group Publishing Limited. http://dx.doi.org/10.1016/S1746-
9791(05)01111-9.
Coombs, W. T., & Holladay, S. J. (2008). Comparing apology to equivalent crisis response strategies: Clarifying apology's role and value in crisis communication. Public
Relations Review, 34(3), 252–257. http://dx.doi.org/10.1016/j.pubrev.2008.04.001.
Coombs, W. T., & Holladay, S. J. (2009). Further explorations of post-crisis communication: Effects of media and response strategies on perceptions and intentions.
Public Relations Review, 35(1), 1–6.
Coombs, W. T., & Holladay, S. J. (2012). Amazon.com's Orwellian nightmare: Exploring apology in an online environment. Journal of Communication Management,
16(3), 280–295. http://dx.doi.org/10.1108/13632541211245758.
Coombs, W. T., Frandsen, F., Holladay, S. J., & Johansen, W. (2010). Why a concern for apologia and crisis communication? Corporate Communications: An International
Journal, 15(4), 337–349.
Coombs, W. T., Holladay, S. J., & Claeys, A. S. (2016). Debunking the myth of denial’s effectiveness in crisis communication: Context matters. Journal of Communication
Management, 20(4), 381–395. http://dx.doi.org/10.1108/jcom-06-2016-0042.
Coombs, W. T., Claeys, A., & Holladay, S. (2017). Social media’s value in a crisis: Channel effect or stealing thunder? In L. L. Austin, & Y. Jin (Eds.). Social media and
crisis communication (pp. 159–167). New York: Taylor & Francis Retrieved from ProQuest Ebook Central.
Coombs, W. T. (2007). Protecting organization reputations during a crisis: The development and application of situational crisis communication theory. Corporate
Reputation Review, 10(3), 163–176.
Coombs, W. T. (2011). Ongoing crisis communication: Planning, managing, and responding. Thousand Oaks, CA: Sage Publications.
Darby, B. W., & Schlenker, B. R. (1982). Children’s reactions to apologies. Journal of Personality and Social Psychology, 4, 742–753.
De Vries, H., Elliott, M. N., Kanouse, D. E., & Teleki, S. S. (2008). Using pooled kappa to summarize interrater agreement across many items. Field Methods, 20(3),
272–282.
Dozier, D. M., Shen, H., Sweetser, K. D., & Barker, V. (2016). Demographics and Internet behaviors as predictors of active publics. Public Relations Review, 42(1), 82–90.
Duhé, S. (2015). An overview of new media research in public relations journals from 1981 to 2014. Public Relations Review, 41(2), 153–169. http://dx.doi.org/10.
1016/j.pubrev.2014.11.002.
Edgerly, S., Vraga, E. K., Dalrymple, K. E., Macafee, T., & Fung, T. K. F. (2013). Directing the dialogue: The relationship between YouTube videos and the comments
they spur. Journal of Information Technology & Politics, 10(3), 276–292. http://dx.doi.org/10.1080/19331681.2013.794120.
Ekman, P., & Friesen, W. V. (1978). Manual for the facial action coding system. Consulting Psychologists Press.
Ekman, P., Friesen, W. V., & Hager, J. C. (2002). Facial action coding system. Manual and investigator’s guide. Salt Lake City, UT: Research Nexus.
Ekman, P. (1992). Telling lies: Clues to deceit in the marketplace, politics, and marriage. New York: WW Norton & Company.
Fawkes, J., & Gregory, A. (2001). Applying communication theories to the internet. Journal of Communication Management, 5(2), 109–124.
Feeney, J. A. (2004). Hurt feelings in couple relationships: Toward integrative models of the negative effects of hurtful events. Journal of Social and Personal
Relationships, 21, 487–508. http://dx.doi.org/10.1177/0265407504044844.
Fisher, M. L., & Exline, J. J. (2006). Self-forgiveness versus excusing: The roles of remorse, effort, and acceptance of responsibility. Self and Identity, 5, 127–146. http://
dx.doi.org/10.1080/15298860600586123.
Glantz, M., & Benoit, W. L. (2017). The world’s all atwitter: Image repair discourse on social media. In L. L. Austin, & Y. Jin (Eds.). Social media and crisis communication
(pp. 168–179). New York: Taylor & Francis Retrieved from ProQuest Ebook Central.
Goffman, E. (1967). Interaction ritual: Essays on face-to-face interaction. Oxford, England: Aldine.
Gold, G. J., & Weiner, B. (2000). Remorse, confession: Group identity and expectancies about repeating a transgression. Basic and Applied Social Psychology, 22,
291–300.
Gracyalny, M. L., & Mongeau, P. M. (2010). Expressing remorse in close relationships: Verbal and nonverbal components of remorse. Paper presented to the National
Communication Association Convention Top Paper in Nonverbal Communication Division.
Haigh, M. M., & Brubaker, P. (2010). Examining how image restoration strategy impacts perceptions of corporate social responsibility, organization-public re-
lationships, and source credibility. Corporate Communications: An International Journal, 15(4), 453–468. http://dx.doi.org/10.1108/13563281011085538.
Hambrick, M. E., Frederick, E. L., & Sanderson, J. (2013). From yellow to blue: Exploring Lance Armstrong’s image repair strategies across traditional and social media.
Communication & Sport, 3(2), 196–218. http://dx.doi.org/10.1177/2167479513506982.
Hearit, K. M. (2006). Crisis management by apology: Corporate response to allegations of wrongdoing. Mahwah, New Jersey: Routledge.
Jenkins, H. (2006). Convergence culture: Where old and new media collide. New York: NYU press.
Jin, Y., Pang, A., & Cameron, G. T. (2010). The role of emotions in crisis responses. Corporate Communications: An International Journal, 15(4), 428–452. http://dx.doi.
org/10.1108/13563281011085529.
Jin, Y., Liu, B. F., & Austin, L. L. (2014). Examining the role of social media in effective crisis management: The effects of crisis origin, information form, and source on
publics’ crisis responses. Communication Research, 41(1), 74–94. http://dx.doi.org/10.1177/0093650211423918.
Johnston, L. (1996). Resisting change: Information-seeking and stereotype change. European Journal of Social Psychology, 26(5), 799–825.
Jonas, E., Schulz-Hardt, S., Frey, D., & Thelen, N. (2001). Confirmation bias in sequential information search after preliminary decisions: an expansion of dissonance
theoretical research on selective exposure to information. Journal of Personality and Social Psychology, 80(4), 557.
Kang, M. (2014). Understanding public engagement: Conceptualizing and measuring its influence on supportive behavioral intentions. Journal of Public Relations
Research, 26(5), 399–416. http://dx.doi.org/10.1080/1062726X.2014.956107.
Katz, E., Blumler, J. G., & Gurevitch, M. (1973). Uses and gratifications research. The Public Opinion Quarterly, 37(4), 509–523.
Kelleher, T. (2009). Conversational voice, communicated commitment, and public relations outcomes in interactive online communication. Journal of Communication,
59(1), 172–188. http://dx.doi.org/10.1111/j.1460-2466.2008.01410.x.
Ki, E. J., & Nekmat, E. (2014). Situational crisis communication and interactivity: Usage and effectiveness of Facebook for crisis management by Fortune 500
companies. Computers in Human Behavior, 35, 140–147. http://dx.doi.org/10.1016/j.chb.2014.02.039.
Kim, S., Avery, E. J., & Lariscy, R. W. (2009). Are crisis communicators practicing what we preach?: An evaluation of crisis response strategy analyzed in public

404
J.K. Sandlin, M.L. Gracyalny Public Relations Review 44 (2018) 393–406

relations research from 1991 to 2009. Public Relations Review, 35(4), 446–448.
Knobloch-Westerwick, S., Johnson, B. K., & Westerwick, A. (2015). Confirmation bias in online searches: impacts of selective exposure before an election on political
attitude strength and shifts. Journal of Computer-Mediated Communication, 20(2), 171–187. http://dx.doi.org/10.1111/jcc4.12105.
Kwon, K. H., & Gruzd, A. (2017). Is offensive commenting contagious online? Examining public vs interpersonal swearing in response to Donald Trump’s YouTube
campaign videos. Internet Research, 27(4), 991–1010.
Landis, J., & Koch, G. (1977). The measurement of observer agreement for categorical data. Biometrics, 33(1), 159–174. http://dx.doi.org/10.2307/2529310.
Lewicki, R. J., Polin, B., & Lount, R. B. (2016). An exploration of the structure of effective apologies. Negotiation and Conflict Management Research, 9, 177–196. http://
dx.doi.org/10.1111/ncmr.12073.
Manika, D., Papagiannidis, S., & Bourlakis, M. (2015). Can a CEO's YouTube apology following a service failure win customers' hearts? Technological Forecasting and
Social Change, 95, 87–95. http://dx.doi.org/10.1016/j.techfore.2013.12.021.
Matejic, N. (2015). Social media rules of engagement: Why your online narrative is the best weapon during a crisis. Hoboken, NJ: John Wiley & Sons.
McCullough, M. E., Worthington, E. L., & Rachal, K. C. (1997). Interpersonal forgiving in close relationships. Journal of Personality and Social Psychology, 73, 321–336.
Meer, T. G. L. A., & Verhoeven, J. W. M. (2014). Emotional crisis communication. Public Relations Review, 40(3), 526–536. http://dx.doi.org/10.1016/j.pubrev.2014.
03.004.
Meng, J., & Berger, B. K. (2017). The role of public relations leadership in effective social media and crisis communication management. In L. L. Austin, & M. Yan Jin
(Eds.). Social media and crisis communication (pp. 85–98). New York: Taylor & Francis Retrieved from ProQuest Ebook Central.
Moody, M. (2011). Jon and Kate Plus 8: A case study of social media and image repair tactics. Public Relations Review, 37(4), 405–414. http://dx.doi.org/10.1016/j.
pubrev.2011.06.004.
Morris, M., & Ogan, C. (1996). The Internet as mass medium. Journal of Computer-Mediated Communication, 1(4), http://dx.doi.org/10.1111/j.1083-6101.1996.
tb00174.x.
Muntinga, D. G., Moorman, M., & Smit, E. G. (2011). Introducing COBRAs. International Journal of Advertising, 30(1), 13–46. http://dx.doi.org/10.2501/IJA-30-1-013-
046.
Muralidharan, S., Dillistone, K., & Shin, J. H. (2011). The Gulf Coast oil spill: Extending the theory of image restoration discourse to the realm of social media and
beyond petroleum. Public Relations Review, 37(3), 226–232. http://dx.doi.org/10.1016/j.pubrev.2011.04.006.
Nabi, R. L. (2003). Exploring the framing effects of emotion: do discrete emotions differentially influence information accessibility, information seeking, and policy
preference? Communication Research, 30(2), 224–247.
Nicas, J. (2018). How YouTube drives people to the Internet’s darkest corners. The Wall Street Journal Retrieved from https://www.wsj.com.
Nielson Social Media Report. (2016). Retrieved from http://www.nielsen.com/us/en/insights/reports/2017/2016-nielsen-social-media-report.html.
O’Sullivan, P. B., & Carr, C. T. (2017). Masspersonal communication: A model bridging the mass-interpersonal divide. New Media & Society, 20(3), http://dx.doi.org/
10.1177/1461444816686104.
On Demand Entertainment (2009). Chris Brown: I'm sorry for beating up Rihanna. (July 20). Retrieved from https://www.youtube.com/watch?v=vTj3-2Fdvj0.
Ott, L., & Theunissen, P. (2015). Reputations at risk: Engagement during social media crises. Public Relations Review, 41(1), 97–102. http://dx.doi.org/10.1016/j.
pubrev.2014.10.015.
Perrin, A. (2015). One-fifth of Americans report going online ‘almost constantly’Pew Research Center (December 8). Retrieved from http://www.pewresearch.org/fact-
tank/2015/12/08/one-fifth-of-americans-report-going-online-almost-constantly/.
Planalp, S. (1998). Communicating emotion in everyday life: Cues, processes, and channels. In P. A. Andersen, & L. K. Guerrero (Eds.). Handbook of communication and
emotion (pp. 30–48). San Diego, CA: Academic Press.
Poushter, J. (2016). Smartphone ownership and Internet usage continues to climb in emerging economiesPew Research Center (February 22). Retrieved from http://www.
pewglobal.org/2016/02/22/smartphone-ownership-and-internet-usage-continues-to-climb-in-emerging-economies/technology-report-01-03/.
Procopio, C. H., & Procopio, S. T. (2007). Do you know what it means to miss New Orleans? Internet communication, geographic community, and social capital in
crisis. Journal of Applied Communication Research, 35(1), 67–87.
Reardon, K. K., & Rogers, E. M. (1988). Interpersonal versus mass media communication a false dichotomy. Human Communication Research, 15(2), 284–303.
Roschk, H., & Kaiser, S. (2013). The nature of an apology: An experimental study on how to apologize after a service failure. Marketing Letters, 24(3), 293–309. http://
dx.doi.org/10.1007/s11002-012-9218-x.
Scher, S. J., & Darley, J. M. (1997). How effective are the things people say to apologize? Effects of the realization of the apology speech act. Journal of Psycholinguistic
Research, 26, 127–140. http://dx.doi.org/10.1023/A:1025068306386.
Schlenker, B. R., & Darby, B. W. (1981). The use of apologies in social predicaments. Social Psychology Quarterly, 271–278.
Schmitt, M., Gollwitzer, M., Förster, N., & Montada, L. (2004). Effects of objective and subjective account components on forgiving. The Journal of Social Psychology,
144(5), 465–486.
Schultz, F., Utz, S., & Göritz, A. (2011). Is the medium the message? Perceptions of and reactions to crisis communication via twitter, blogs and traditional media.
Public Relations Review, 37(1), 20–27.
Sheckels, T. F. (2013). The failed comedy of NBA’s Gilbert Arenas: Image restoration in context. In J. R. Blaney, L. R. Lippert, & J. S. Smith (Eds.). Repairing the athlete’s
image: Studies in sports image repair (pp. 169–185). Lanham, MD: Lexington.
Sheth, J. N., & Solomon, M. R. (2014). Extending the extended self in a digital world. Journal of Marketing Theory and Practice, 22(2), 123–132.
Shim, K., & Yang, S. U. (2016). The effect of bad reputation: The occurrence of crisis, corporate social responsibility, and perceptions of hypocrisy and attitudes toward
a company. Public Relations Review, 42(1), 68–78. http://dx.doi.org/10.1016/j.pubrev.2015.11.009.
Snyder, M. (1984). When belief creates reality. Advances in experimental social psychology, 18, Academic Press247–305.
Strauss, A., & Corbin, J. (1990). Basics of qualitative research, Vol. 15. Newbury Park, CA: Sage.
Taylor, M., & Kent, M. L. (2010). Anticipatory socialization in the use of social media in public relations: A content analysis of PRSA's public relations tactics. Public
Relations Review, 36(3), 207–214. http://dx.doi.org/10.1016/j.pubrev.2010.04.012.
ten Brinke, L., & Adams, G. S. (2015). Saving face? When emotion displays during public apologies mitigate damage to organizational performance. Organizational
Behavior and Human Decision Processes, 130, 1–12.
The Young Turks (2013). Paula deen's sloppy N-word apology. (June 22) Retrieved from https://www.youtube.com/watch?v=HKWCJUxXJ6E.
Thelwall, M., Sud, P., & Vis, F. (2012). Commenting on YouTube videos: From Guatemalan rock to El Big Bang. Journal of the American Society for Information Science
and Technology, 63(3), 616–629. http://dx.doi.org/10.1002/asi.21679.
Today Show YouTube Channel (2015). Chipotle founder Steve Ells addresses series of outbreaks. (December 10) Retrieved from https://www.youtube.com/watch?v=
Oeapkmn-xIQ.
Ulmer, R. R., Sellnow, T. L., & Seeger, M. W. (2013). Effective crisis communication: Moving from crisis to opportunity. Thousand Oaks: Sage Publications.
Utz, S., Schultz, F., & Glocka, S. (2013). Crisis communication online: How medium, crisis type and emotions affected public reactions in the Fukushima Daiichi
nuclear disaster. Public Relations Review, 39(1), 40–46. http://dx.doi.org/10.1016/j.pubrev.2012.09.010.
Valentini, C., Romenti, S., & Kruckeberg, D. (2017). Handling crises in social media: From stakeholder crisis awareness and sense-making to organizational crisis
preparedness. In L. L. Austin, & Y. Jin (Eds.). Social media and crisis communication (pp. 57–67). New York: Taylor & Francis Retrieved from ProQuest Ebook
Central.
The professional bond: public relations education, practice: The report of the commission on public relations education. In J. VanSlyke Turk (Ed.). Commission on public
relations education Retrieved from http://www.commpred.org/theprofessionalbond/index.php.
VanSlyke Turk, J., Jin, Y., Stewart, S., Kim, J., & Hipple, J. R. (2012). Examining the interplay of an organization's prior reputation, CEO's visibility, and immediate
response to a crisis. Public Relations Review, 38(4), 574–583. http://dx.doi.org/10.1016/j.pubrev.2012.06.012.
Waldron, V. R., & Kelley, D. L. (2008). Communicating forgiveness. Thousand Oaks, CA: Sage Publications.
Walsh, J., & McAllister-Spooner, S. M. (2011). Analysis of the image repair discourse in the Michael Phelps controversy. Public Relations Review, 37(2), 157–162.

405
J.K. Sandlin, M.L. Gracyalny Public Relations Review 44 (2018) 393–406

http://dx.doi.org/10.1016/j.pubrev.2011.01.001.
Walther, J. B., Carr, C. T., Choi, S. S. W., DeAndrea, D. C., Kim, J., Tong, S. T., & Van Der Heide, B. (2010). Interaction of interpersonal, peer, and media influence
sources online. In Z. Papacharissi (Ed.). A networked self: Identity, community, and culture on social network sites (pp. 17–38). Routledge.
Watts, D. J., & Dodds, P. S. (2007). Influentials, networks, and public opinion formation. Journal of Consumer Research, 34(4), 441–458. http://dx.doi.org/10.1086/
518527.
Whiting, A., & Williams, D. (2013). Why people use social media: A uses and gratifications approach. Qualitative Market Research: An International Journal, 16(4),
362–369.
Winter, S., Metzger, M. J., & Flanagin, A. J. (2016). Selective use of news cues: A multiple-motive perspective on information selection in social media environments.
Journal of Communication, 66(4), 669–693.
Wright, D., & Hinson, M. (2017). Tracking how social and other digital media are being used in public relations practice: A twelve-year study. Public Relations Journal,
11(1) Retrieved from http://prjournal.instituteforpr.org/wp-content/uploads/PRJ-2017-Wright-Hinson-2-1.pdf.
Yeo, S. K., Xenos, M. A., Brossard, D., & Scheufele, D. A. (2015). Selecting our own science: How communication contexts and individual traits shape information
seeking. The ANNALS of the American Academy of Political and Social Science, 658(1), 172–191. http://dx.doi.org/10.1177/0002716214557782.
Yin, D., Mitra, S., & Zhang, H. (2016). Research note—When do consumers value positive vs. negative reviews? An empirical investigation of confirmation bias in
online word of mouth. Information Systems Research, 27(1), 131–144.
Younger, J. W., Piferi, R. L., Jobe, R. L., & Lawler, K. A. (2004). Dimensions of forgiveness: The views of laypersons. Journal of Social and Personal Relationships, 21,
837–855. http://dx.doi.org/10.1177/0265407504047843.
Zhang, J., Wang, Y., Wu, Y., Wang, X., & Buck, R. (2017). Study of Malaysia Airlines’ missing flight crisis. In L. L. Y. Austin Jin, & Y. Jin (Eds.). Social media and crisis
communication (pp. 209–223). New York: Taylor & Francis Retrieved from ProQuest Ebook Central.

406