When Employees Speak as They Like: Bad Mouthing in Social Media

Abstract

Employee voice can threaten a company’s reputation when employees speak badly about their employer. This research focuses on identifying and quantifying the importance of antecedents of employees’ company-related bad mouthing in social media. Data for this study was collected through a qualitative interview study (N = 33) and a quantitative survey approach (N = 472). The hypothesized model is tested using structural equation modelling. Drawing on social identity theory and using the job demands-resources model as a theoretical lens, we found that job demands are positively associated with turnover intention, which in turn increases company-related bad mouthing in social media of employees. This study employed a cross-sectional survey design, which may be complemented by mono-organizational studies in the future. This study enriches the understanding of employees’ behavior in social media and provides implications for managers such that the strategy of reducing turnover intention is more successful to limit the amount of employees’ bad mouthing than enhancing employees’ commitment. This study is the first which links job demands and job resources with company-related bad mouthing of employees in social media. It proposes and tests antecedence of corporate reputation harming behavior of employees.

Full text

Vol.:(0123456789)
Corporate Reputation Review
https://doi.org/10.1057/s41299-019-00086-w
ORIGINAL ARTICLE
When Employees Speak asThey Like: Bad Mouthing inSocial Media
StefanIvens1· MarioSchaarschmidt1· RaoulKönsgen1
© Reputation Institute and Springer Nature Limited 2019
Abstract
Employee voice can threaten a company’s reputation when employees speak badly about their employer. This research focuses
on identifying and quantifying the importance of antecedents of employees’ company-related bad mouthing in social media.
Data for this study was collected through a qualitative interview study (N = 33) and a quantitative survey approach (N = 472).
The hypothesized model is tested using structural equation modelling. Drawing on social identity theory and using the job
demands-resources model as a theoretical lens, we found that job demands are positively associated with turnover intention,
which in turn increases company-related bad mouthing in social media of employees. This study employed a cross-sectional
survey design, which may be complemented by mono-organizational studies in the future. This study enriches the under-
standing of employees’ behavior in social media and provides implications for managers such that the strategy of reducing
turnover intention is more successful to limit the amount of employees’ bad mouthing than enhancing employees’ commit-
ment. This study is the first which links job demands and job resources with company-related bad mouthing of employees
in social media. It proposes and tests antecedence of corporate reputation harming behavior of employees.
Keywords Social media· Employee behavior· Company-related bad mouthing· Job demands-resources model
Introduction
Various authors have noted social media’s double-edged
character (e.g., VanMeter etal. 2015; Van Zoonen etal.
2016; Van Zoonen and Treem 2019). With social media, cus-
tomers can support products and services by giving favora-
ble ratings that drive future sales, but simultaneously may
spread their dissatisfaction with the company with the online
world—leading to unwanted outcomes such as damaged rep-
utation or brand equity and reduced profitability (Schaar-
schmidt and Könsgen 2019; Walsh etal. 2018; Wetzer etal.
2007). Consequently, researchers such as Wæraas and Dahle
(2019) state that “reputation management is people manage-
ment”. However, while the negative consequences of social
media for companies have been acknowledged for customer
activities, surprisingly little research addresses negative
outcomes of social media use by own employees (Melián-
González and Bulchand-Gidumal 2017).
Dangerous for firms, in terms of long-term reputa-
tion risks, are negative online ratings (along with detailed
reports) of own employees at company review sites such as
Glassdoor or Kununu (Könsgen etal. 2018). Here, employ-
ees post (often anonymously) about former and current
employers. Company review sites are an important vehicle
to attract new talents as they help to reduce job seekers’
information asymmetry (Dabirian etal. 2016; Kluemper
and Rosen 2009), and an unfavorable online reputation at
these places put the acquisition of qualified personnel under
risk. Moreover, customers have started to take more care
on employees’ working conditions, which is why employee
complaints on company review sites may also affect pur-
chase decisions (Könsgen etal. 2018). Similarly, Opitz etal.
(2018) note that employee voice can cause more harm to
a firm’s reputation than customer voice does. This is evi-
denced by multiple examples where employee postings (on
their private accounts) had negative effects for a compa-
ny’s reputation and even stock market performance (Parker
etal. 2019; Walsh etal. 2016). Thus, from a managerial
perspective, knowing why employees start posting negative
comments or provide low ratings is important because then
managers might find strategies to protect their online reputa-
tion against unwanted employee voice. From an employee
* Mario Schaarschmidt
mario.schaarschmidt@uni-koblenz.de
1 Institute forManagement, University ofKoblenz-Landau,
Universitätsstrasse 1, 56070Koblenz, Germany

S.Ivens et al.
perspective, company-related bad mouthing in social media
(CRBM) may be classified as a specific form of coun-
terproductive work behavior (Fox etal. 2001; Lyons and
Bowling 2017). Hence, sets of drivers that are responsible
for other forms of counterproductive work behavior (e.g.,
absenteeism, bullying, incivility), are candidates to predict
CRBM, too, but no research has systematically addressed
this assumption.
Given the lack of research focusing on antecedents of
employees’ posting negative ratings, this research aims to (1)
identify important antecedents for CRBM and (2) quantify
their relative importance. For identifying potential drivers of
CRBM, we draw upon the job demands-job resources model
(Demerouti etal. 2001), in conjunction with social identity
theory (SIT), and conducted a qualitative prestudy. From
a long list of potential predictors of employee behavior,
we found five aspects that seem to be representative of job
demands and job resources in determining CRBM, namely
job dissatisfaction, unfair treatment, perceived external repu-
tation, supervisor support and organizational climate.
The remainder of the article is organized as follows. First,
we review related literature and develop hypotheses that
are based on the job demands/job resources model, SIT, as
well as own qualitative fieldwork. Second, we report on our
main study, which involves a cross-sectional survey of 472
employees. Finally, we provide our results and discuss them
in light of current social media discussions and potential
managerial actions.
Theoretical Background, Qualitative
Prestudy, andHypotheses
Job Demands‑Resources Model
To explain CRBM, we draw upon the job demands-resources
model (JD-R), which already has been used in a variety of
contexts including social media (Walsh etal. 2016) and
willingness to report complaints (Walsh etal. 2015). The
JD-R model categorizes working conditions into two broad
categories: job demands and job resources that both show
specific effects on employees (Bakker and Demerouti 2007).
Job demands are defined as “those physical, social, or organ-
izational aspects of the job that require sustained physical
or mental effort” (Demerouti etal. 2001, p. 501). They are
“associated with certain physiological and psychological
costs (e.g., exhaustion)” (Demerouti etal. 2001, p. 501).
Additionally, job resources are defined as “those physi-
cal, psychological, social, or organizational aspects of the
job that […] (a) be functional in achieving work goals; (b)
reduce job demands […]; (c) stimulate personal growth and
development.” (Demerouti etal. 2001, p. 501). The original
model postulates that a high level of job demands and a
limited availability of job resources can lead to energy deple-
tion and a lower level of work motivation (Demerouti etal.
2001). Consequently, the model has been used to explain
direct outcomes of demands in the absence of necessary
resources (e.g., burnout, emotional exhaustion) as well as
indirect ones (e.g., job satisfaction).
Low levels of work motivation generally may affect
behavioral intentions, such as an increased turnover inten-
tion and less commitment (Schaufeli and Bakker 2004). For
example, low commitment can be a driver of counterpro-
ductive work behavior such as CRBM (Lyons and Bowling
2017), as well as turnover intention can be (Fox etal. 2001).
We adopt these rationales and treat job demands and job
resources as potential drivers of CRBM. While the under-
lying notion of JD-R suggests that demands and resources
interact, we only focus on direct effects in this study. In the
following, we report on a qualitative study aimed at iden-
tifying further bad mouthing-related job demands and job
resources.
Qualitative Insights Into Job Demands andJob
Resources inRelation toBad Mouthing
Literature on employees’ motivations to post-negative com-
ments online is rare (Schaarschmidt and Walsh 2018). To
identify drivers of bad mouthing in social media, we there-
fore, started complementing insights from relevant organiza-
tional behavior literature with qualitative data. For this pur-
pose, we conducted a face-to-face interview series with 33
(employed) interviewees that represent a wide area of jobs
and backgrounds (e.g., teacher, bus driver, manager, service
employee). The aim of the interviews was to identify job
demands and job resources that are particularly relevant for
bad mouthing in social media. To be able to compare results,
we developed an interview guide that targets forms of bad
mouthing in social media as well as potential reasons for
bad mouthing. Interviewees aged between 19 and 52; all of
them were full-time employees. Respondents were asked to
talk freely about their general experiences with social media,
circumstances in which they use or would use social media
to express opinions as well as work conditions. Interviews
lasted between 20 and 30min and were first transcribed
and then content analyzed using MAXQDA software. Two
authors applied Straussian coding to the interviews1 in rela-
tion to social media-related job demands and job resources
(Corbin and Strauss 1990). Straussian coding is one instance
of a grounded theory approach, which is particularly useful
when phenomena are emergent and poorly understood. At
the same time, we did not start from a green field, as the
1 As the authors did not code the interviews independently from each
other, we cannot provide interrater reliability values.

When Employees Speak asThey Like: Bad Mouthing inSocial Media
literature pertaining to counterproductive work behavior is
quite mature (Fox etal. 2001). The authors, therefore, used
open and axial coding, but only for concepts that would
clearly point to job demands and job resources; potential
mediators as well as CRBM itself were not coded. "Appen-
dix 1" provides an overview of emergent topics. Overall,
we found five antecedents of bad mouthing in social media,
three of them might be considered job resources (i.e. per-
ceived external reputation, supervisor support, organiza-
tional climate) while one may be categorized as demand
(i.e. unfair treatment)(Bakker and Demerouti 2007). From
the coding cycle, job dissatisfaction also emerged as a con-
cept, which could not clearly be positioned as a demand as
it is at the same time a reaction to demands. We, therefore,
positioned job dissatisfaction, defined as a reaction to job
demands, as an additional driver of turnover intention.
Theoretical Background andHypotheses
Figure1 depicts our conceptual model, which is based on
the JD-R model. The model helps to disentangle different
forms of CRBM drivers. Based on our qualitative insights,
we focus on unfair treatment as an organizational aspect of
job demands, which might negatively affect employees’ atti-
tudes at work (Essed 1991). Regarding job resources, we
concentrate on perceived external reputation2 of the firm,
supervisor support and organizational climate (we will
introduce and define them when developing hypotheses). In
addition, through our qualitative study we identified job dis-
satisfaction, which we position as a reaction to job demands.
To derive our hypotheses, we draw on SIT, which uses
individual’s self-concept and their perceived membership in
relevant social groups to predict intergroup behaviors. SIT
stats that individuals are intrinsically motivated to achieve
positive distinctiveness and strive for a positive self-concept
(Tajfel and Turner 1979). Their social identity is built out
of the comparison of relevant out-groups and strengthens
their identity if they perceive that their group is more repu-
table and recognized than the comparison group. Otherwise,
individuals tend to leave the group and try to join the more
reputable group. They also have the option to improve their
groups’ reputation, but this strategy is used less often as it
requires excessive effort and resources.
In line with SIT, job demands such as employees’ per-
ception of unfair treatment decrease employees’ self-con-
cepts. In particular, unfair treatment is a state that deviates
from the optimum impression that no other firm has a bet-
ter workplace. Thus, as impressions arise that imply that
out-groups might provide better personal environments with
fair treatment (Cooper-Thomas and Wright 2013), evolving
Fig. 1 Conceptual model
2 We note that the term “perceived external reputation” is close to
“perceived external prestige” (e.g. Smidts etal. 2001). In this study,
we treat these concepts as synonyms.

S.Ivens et al.
dissatisfaction might enhance employees quitting inten-
sions. Nowadays, this perception of a comparably unsatis-
fying workplace is fortified by company review sites such
as Glassdoor or Kununu, where ratings of the own employer
are directly comparable to other company ratings (Könsgen
etal. 2018). As only few employees are able to increase
their self-concept despite feelings of dissatisfaction, again,
increased turnover intention is the consequence. Therefore,
it seems to be unlikely that the employee chooses the group-
improving strategy mentioned above. This coherence is true
for dissatisfied employees, too. Therefore, we postulate that
job demands (e.g. unfair treatment) as well as resulting
states of feelings (e.g. dissatisfaction) are positively related
to employees’ turnover intention (e.g., Biron and Boon 2013;
Schaufeli and Bakker 2004).
H1 Employees’ job dissatisfaction is positively associated
with their turnover intention.
H2 Employees’ perception of an unfair treatment is posi-
tively associated with their turnover intention.
Of the identified job resources, perceived external reputa-
tion is the impression of how outsiders evaluate the group
and accrues through a recursive process of impressions of
meta-stereotypes (Schaarschmidt 2016). It is a stereotype
that members of the in-group (e.g. employees) believe to be
perceived by members of relevant out-groups such as indi-
viduals who are external to the company (Schaarschmidt and
Könsgen 2019; Vorauer etal. 1998). Other literatures have
referred to perceived external reputation as construed image
as a “mental associations that organization members believe
others outside the organization hold about the organization”
(Brown etal. 2006, p. 101). In accordance with SIT, employ-
ees who think that their in-group is perceived as reputable by
externals might create a positive self-concept. In turn, these
employees tend to stay within the group and increase their
organizational commitment. In addition, supervisor support,
understood as an important factor in employees’ definition
of their contextual environment (Cole etal. 2006), might be
an indicator for employees to compare with out-groups. If
employees find that their supervisor support is better than
the supervisor support of relevant out-groups (e.g., again
through comparison in company review sites), this might
lead to an increased organizational commitment. In accord-
ance with this argumentation, finally, organizational climate
might become an indicator of employees’ comparison pro-
cess, too. Organizational climate is related to, but different
from organizational culture (Moran and Volkwein 1992).
Organizational climate, identified as a job resource through
our qualitative research, has also been suggested to influence
organizational commitment (Schwepker Jr 2001). There-
fore, we hypothesize a positive effect between perceived
external reputation, the perceptions of supervisor support,
and organizational culture with individuals’ organizational
commitment.
H3 Employees’ perceived external reputation of the firm is
positively related to employees’ organizational commitment.
H4 Employees’ perceived supervisor support is positively
associated with their organizational commitment.
H5 Employees perceived organizational climate is positively
related with their organizational commitment.
While we have detailed how commitment and turnover
intentions are shaped by CRBM-related job demands and
job resources, the focus of this study is how commitment
and turnover intentions shape CRBM itself. CRBM might
be influenced both by individuals’ turnover intention (posi-
tive relationship) as well as by individuals’ organizational
commitment (negative relationship), as evidenced for other
forms of counterproductive work behavior (Ones 2018). In
line with SIT, organizational commitment might encourage
employees to defend their company in social media, result-
ing in less CRBM. In turn, turnover intention positively
affects negative bad mouthing behavior in social media.
Thus, we hypothesize a positive relation of turnover inten-
tion on CRBM and a negative association between organi-
zational commitment and CRBM.
H6 Turnover intention is positively associated with employ-
ees’ company-related bad mouthing in social media.
H7 Organizational commitment is positively associated with
employees’ company-related bad mouthing in social media.
Methodology
Sample andData Collection
We used a survey-based approach to test the hypotheses. To
ensure external validity, that is, varying levels of organiza-
tional climate and perceived external reputation, this study
uses employees form various organizations as respondents
(Helm 2013). We used a snowball technique to distribute
the questionnaire via social media such as Facebook and
Twitter. Additionally, we asked master students of a research
lecture to share the questionnaire within their personal net-
works of employed friends and colleagues. Our invitation
referred to an academic study regarding “social media usage
of employees,” thus obscuring the real intention such that
no potential self-selection bias would arise. In addition, we
ensured that the questionnaire could only be completed once

When Employees Speak asThey Like: Bad Mouthing inSocial Media
from the same computer. The questionnaire was accessible
for approximately 3weeks. A total of 650 employees par-
ticipated in our study of which 4993 (76.8%) have completed
the questionnaire.
Because we did not include a completeness check, we
face data sets with missing values. Thirteen cases were
dropped because of systematically missing values, revealed
by analyzing missing values in SPSS 25. The same analysis
yielded six cases that are affected by unsystematic missing
values (i.e., not more than two missing values per subject).
We handled these cases by imputing them with the mean
replacement procedure, which leads to 501 cases with-
out missing values. Our pre-test with 28 subjects pointed
out that 11min is the average time required to answer the
questionnaire carefully. Thus, 21 employees were excluded
because they answered the questionnaire within < 5min.
Finally, we had to drop eight subjects because they did not
pass the attention checks in the middle and at the end of
the questionnaire (e.g., Answering a question about rode
traffic such as which color must be shown on a traffic light
to grant a pedestrian to cross the road.). We were thus able
to use the responses of 472 participants for our analysis:
36% of them were female4; on average, the employees were
31years old; 70% of the respondents are full time employees
and 33% have a university-entrance diploma. Additionally,
24% stated that they use social media up to 12h/week. 55%
of the employees responded that their company does not
use social media guidelines. "Appendix 1" provides a more
detailed overview about the sample. We also investigated
the potential for non-response bias by comparing the earli-
est 25% and the latest 25% of the respondents with regard
to our conceptual variables (Armstrong and Overton 1977).
The t test of the group means for aggregated measures of
main constructs revealed no significant difference. For this
reason, non-response bias seem to be unlikely in this study.
Measures
The questionnaire contained multi-item measures of the
model variables. The items we used were translated from a
bi-lingual speaker into German for the final questionnaire.
The authors relied on well-established conceptualizations
and measurements and developed variables for this research.
We measured most of the variables on seven-point Likert
scales ranging from 1 = ‘strongly disagree’, to 7 = ‘strongly
agree’. Our literature review revealed that no adequate
items for employees’ company-related bad mouthing fit to
our social media context. Therefore, we relied on our initial
interview data to develop an appropriate measure. Respond-
ents predominantly named (1) general posting in social
media, (2) rating on company review sites, and (3) com-
ments and likes as options to voice their work experiences.
In addition, the majority of the respondents (30 employees)
indicated that anonymity must be ensured. In combination of
these information, we developed four items. Those items are
“I use social media to report anonymously on bad experience
that I have made in my working life,” “I rate my employer
anonymously as “poor” on a company rating platform,”
“I use social media to comment on grievances within my
company anonymously,” “I publicly express my displeasure
about my employer when I am anonymous.” Finally, we vali-
dated the items by performing an explorative factor analy-
sis, while using a maximum likelihood extraction method in
combination with a Promax rotation. We were able to extract
distinct factors and found no cross-loadings > .2.
Further, we adopted the three-item conceptualization
of job satisfaction from Cammann etal. (1983) to meas-
ure dissatisfaction and enriched it while adding the item “I
am dissatisfied with my job.” To assess unfair treatment,
the authors adapted the three-item operationalization of
Essed (1991) while adding the following items based on
the interview prestudy. “I get a disproportionate number of
tasks,” “I am getting less privileges than colleagues,” “I have
a poorer working environment (e.g., office) or equipment
(e.g., computer) than colleagues,” and “I was exposed in
front of colleagues.” To assess employees’ perceived reputa-
tion, we asked the subjects “How do you feel people outside
your company would rate it on the following attributes?”,
as suggested by Helm (2013), and provided a list of eight
aspects (e.g., innovative company, good employer, etc.). This
allowed us to assess employees’ perceptions of their com-
panies’ reputation and reducing the probability of a social
desirability bias. For supervisor support, we relied on the
five-item operationalization of Eisenberger etal. (1986).
Organizational climate is known as a multi-dimensional con-
ceptualization, which is why we used the dimension friendly
team spirit, which fit best to our context (Johannesson 1973).
To assess employees’ turnover intention, we adapted three
items from Helm (2013) and two items from Lum etal.
(1998). For organizational commitment, we used Allen and
Meyer’s (1990) eight-item operationalization including the
reverse-coded items. All descriptive statistics for multi-item
measures are listed in Table1. We also used the follow-
ing controls: Gender, age, education, income (in ranges; see
"Appendix 1"), employees’ social media usage in h/week,
organizational tenure, industry, job type, hierarchical levels,
the presence of social media guidelines within the company
(yes/no), and number of employees.
3 The vast majority (136 subjects) of those who did not complete
the survey aborted the questionnaire on the welcome page where the
intention of the study was explained.
4 We measured gender on a 7-point Likert scale where 1 was female
and 7 was male to account for the increasing sensitivity in the LGBT
community against a clear positioning as either male or female.

S.Ivens et al.
Results
Measurement Model Evaluation
The measurement model was assessed by a confirmatory
factor analysis (CFA) using AMOS 25 and a maximum like-
lihood estimator. The fit indexes for assessing the model
fit include Chi square (χ2), degrees of freedom (df), root
mean square error of approximation (RMSEA), standardized
root mean square residual (SRMR), comparative fit index
(CFI) and parsimony-adjusted normed fit index (PNFI), as
recommended by Kline (2005). The analysis of the meas-
ured variables job dissatisfaction, unfair treatment, turnover
intention, perceived external reputation, supervisor support,
organizational climate, organizational commitment and
employees’ company-related bad mouthing in social media
revealed a good model fit with χ2/df = 1.73 (χ2 = 1.302,
df = 751, p < .001). Accordingly, the requested threshold for
χ2/df of ≤ 2 is not exceeded (Hair etal. 2010) We discovered
a close fit regarding RMSEA of 0.039 with a 90% confi-
dence interval ranging from 0.036 to 0.043, and a PCLOSE
of 1.000. Our RMSEA-value fit perfectly within the range
from 0.00 to 0.05 recommended by Browne and Cudeck
(1993). With 0.04, the model did not exceed the suggested
threshold for SRMR of 0.05. Additionally, we established
good CFI of .95 and a tolerable PNFI of 0.83 (Mulaik etal.
1989). Finally, the CFA revealed that all of the standardized
factor loadings were greater than 0.60 except one item of
organizational commit that has a loading of 0.54.
To ensure discriminant and convergent validity of the
measurement model, additional quality criteria must be
satisfied. Those criteria are indicator reliability expressed
by factor loadings > 0.5, construct reliability measured by
composite reliability (CR) which should exceed 0.7, con-
vergent validity which is represented by average variance
extracted (AVE) which should be higher than 0.5. Finally,
discriminant validity is an important quality criteria which
is calculated by the square root of AVE. This square root
must be greater than inter-construct correlations (Fornell and
Larcker 1981). The presented measurement model passes all
the quality criteria and thus the model revealed good reli-
ability and validity (see Table1). Table1 shows comparably
high correlations between some variables of the measure-
ment model. Especially, the high correlation values above .6
draw our attention, because those might be an indicator for
multicollinearity issues. Therefore, we calculated the vari-
ance inflation factor recommended by Hair etal. (2010). The
highest variance inflation factor revealed by our test was 2.2,
which is below the recommended threshold. Thus, multi-
collinearity is no problem for our hypotheses tests.
We analyzed the measurement model for common method
bias because we used the same source and method for our
independent and dependent variables (Podsakoff etal. 2003).
Therefore, three methods were used: (1) Harman’s single
factor test in SPSS 24, (2) a common unmeasured latent
factor analysis, and (3) a common measured latent factor
analysis in AMOS 25 (Lindell and Whitney 2001). The first
test showed that a single factor with the highest explanatory
Table 1 Convergent validity,
discriminant validity and
correlations
The diagonal (bold) displays the square root of AVE
JDS job dissatisfaction, UFT unfair treatment, PER Perceived external reputation, SS supervisor support,
OC organizational climate, TI turnover intention, COM organizational commitment, CRBM company-
related bad mouthing in social media
Mean SD CR AVE JDS UFT PER SS OC TI COM CRBM
JDS 2.65 1.43 0.83 0.55 0.74
UFT 2.22 1.33 0.91 0.60 0.71 0.77
PER 4.78 1.09 0.89 0.51 − 0.43 − 0.45 0.72
SS 5.10 1.44 0.91 0.66 − 0.60 − 0.55 0.55 0.81
OC 5.10 1.19 0.86 0.55 − 0.66 − 0.61 0.60 0.70 0.74
TI 3.06 1.78 0.91 0.66 0.74 0.67 − 0.38 − 0.46 − 0.52 0.81
COM 4.26 1.41 0.90 0.53 − 0.65 − 0.51 0.53 0.54 0.64 − 0.63 0.73
CRBM 1.87 1.39 0.94 0.80 0.60 0.66 − 0.32 − 0.42 − 0.46 0.67 − 0.44 0.90
Table 2 Results of structural equation modelling
JDS job dissatisfaction, UFT unfair treatment, PER perceived exter-
nal reputation, SS supervisor support, OC organizational climate, TI
turnover intention, COM organizational commitment, BM company-
related bad mouthing, n.s. not significant
***p < 0.001; *p < 0.05
Path Finding Hypothesis
JDS → TI 0.45*** H1Supported
UFT → TI 0.28*** H2Supported
PER → COM 0.22*** H3Supported
SS → COM 0.15* H4Supported
OC → COM 0.40*** H5Supported
TI → BM 0.66*** H6Supported
COM → BM 0.00n.s. H7Not supported

When Employees Speak asThey Like: Bad Mouthing inSocial Media
power explains only 37.3% of the variance, which is less
than the suggested threshold of 50%. Second, we compared
standardized regression weights for models with an unmeas-
ured common latent factor, a comparison between regres-
sion weights for models with and without common latent
factors revealed deltas much < 0.2, which is a commonly
used threshold (Chin etal. 2012). Finally, we performed the
third test while using a measured common latent factor (i.e.,
honesty, measured with three items, see Katz etal. 1994).
This test supported the results of the Harman’s single fac-
tor test and the unmeasured common latent factor analysis
and found that the variables are not affected by common
method bias.
Hypotheses Testing
To assess the hypothesized direct effects, the authors used
structural equation modeling in AMOS 25 and a maxi-
mum likelihood estimator. The model reveals an adequate
fit with the data (χ2 = 1809.75, df = 804, χ2/df = 2.251,
p < .001, RMSEA = 0.052, CFI = 0.93, PNFI = 0.82). The
variances of the endogenous variables are all explained
to a substantial amount; all R2 values are of consider-
able size (R2(JDS) = 0.42, R2(TI) = 0.56, R2(COM) = 0.44,
R2(CRBM) = 0.43). In support of H1, the effect of job dis-
satisfaction on turnover intention is positive (β = 0.45,
p < 0.001). Additionally, we found support for H2 in that
unfair treatment is positively associated with turnover inten-
tion (β = 0.28, p < 0.001). The positive effect of perceived
external reputation on organizational commitment hypoth-
esized in H3 can be supported, too (β = 0.22, p < 0.001).
Supervisor support is positively associated with organiza-
tional commitment (β = 0.15, p < 0.05), in support of H4.
We were also able to find support for H5; organizational
climate is positively related to organizational commitment
(β = 0.40, p < 0.001). The driving variable turnover intention
is positively associated with the CRBM, as hypothesized
in H6 (β = 0.66, p < 0.001). Furthermore, we claimed that
organizational commitment is negatively related to CRBM,
but we did not find support for H7 in our sample (β = 0.00,
n.s.) (see Table2).
We included control variables to test their effect on our
dependent variable CRBM. All nominal variables (e.g., job,
education, etc.) were dummy-coded. Of the control varia-
bles, age was significantly related with CRBM (β = − 0.11,
p < 0.01). In addition, having a bachelor degree decreases
the likelihood of bad mouthing (β = − 0.18, p < 0.05). The
time employees spend per week in social media is also con-
nected to CRBM (β = 0.10, p < 0.01). Surpr isingly, organi-
zational tenure slightly increases the likelihood of displaying
CRBM (β = 0.09, p < 0.05). Finally, out of the number of
employees (grouped into < 10, 10–49, 50–249, 250–499,
500 and above), employee of small and medium-sized com-
pany tend to show less CRBM (β10–49 = − 0.10, p < 0.01,
β50–249 = − 0.10, p < 0.01) while employees of very small
firms show more CRBM (β<10 = 0.11, p < 0.01). Occupa-
tion group, gender, job type (full/part time), hierarchical
level, and presence of social media guidelines did not affect
CRBM.5 Table3 summarizes the findings of this research.
Additional Analysis
Although no mediation effects were postulated in this study,
the conceptual model suggest that these effects could exist.
For this purpose, we again used AMOS and 2000 bootstrap
Table 3 Additional analysis
Model included direct paths and is different from conceptual model in Fig.1
IV independent variable, M mediator, DV dependent variable, CRBM company-related bad mouthing in
social media, JDS job dissatisfaction, UFT unfair treatment, PER perceived external reputation, SS supervi-
sor support, OC organizational climate, TI turnover intention, COM organizational commitment
Variable Direct effect Indirect effect Result
IV M DV IV → DV IV → M → DV
JDS TI CRBM β = 0.08, p = 0.338 β = 0.19, p < 0.001 Full mediation
UFT TI CRBM β = 0.30, p < 0.001 β = 0.28, p < 0.001 Indirect effect
PER COM CRBM β = 0.01, p = 0.912 β = − 0.02, p = 0.138 No mediation
SS COM CRBM β = − 0.05, p = 0.446 β = − 0.02, p = 0.104 No mediation
OC COM CRBM β = − 0.03, p = 0.771 β = − 0.04, p = 0.173 No mediation
COM TI CRBM β = 0.04, p = 0.573 β = − 0.14, p < 0.001 Full mediation
5 We provide full results of dummy coded control variables upon
request.

S.Ivens et al.
samples with a 95% bias-corrected bootstrap confidence
interval, as suggested by Preacher and Hayes (2008).
The bootstrap samples were based on the original sam-
ple with 472 cases and also included the direct effects of
job resources and job demands on CRBM (Table3). First,
for paths through commitment, no mediation effect was
found. For paths through turnover intention, full mediation
could be established. In addition, for theoretical reasons,
(1) unfair treatment could affect turnover intention through
job dissatisfaction, and (2) commitment could indirectly
influence CRBM through turnover intention (see Fig.1).
Unfair treatment has a significant indirect effect on turnover
intention (β = 0.29, p < 0.001, LLCI = 0.20, ULCI = 0.39).
Similarly, commitment has a significant indirect effect on
CRBM through turnover intention (β = − 0.22, p < 0.001,
LLCI = − 0.28, ULCI = − 0.15).
Finally, we tested for alternative effects within the JD-R
part of the model. Resources could potentially buffer the
negative effects of demands on turnover intention (Demerout
etal. 2001)—either directly or in a moderating role. None
of the direct effects of job demands (i.e. perceived external
reputation, supervisor support, and organizational climate
and none of the possible moderations yielded significance.
Discussion
Summary ofResults
The purpose of this study was to investigate the links
between job demands and job resources and turnover inten-
tion, organizational commitment, and CRBM. Drawing on
previous research and SIT, this research found support for
the association of job demands and job resources on the
antecedences of CRBM. In particular, job dissatisfaction,
positioned as a reaction to job demands, and unfair treatment
are associated positively with turnover intention, which, in
turn, is positively associated with CRBM. Furthermore, sup-
port is found for the positive relation of job resources on
organizational commitment. However, this research did not
find support that employees’ commitment has a dampening
effect on CRBM. However, commitment reveals an indirect
effect on CRBM through turnover intention. Among the non-
hypothesized results are those that relate to control variables;
CRBM seem to decrease with age and social media guide-
line seem to not affect CRBM.
Theoretical Implications
First, our research indicates the applicability of the JD-R
model in the context of social media and employee voice
(Walsh etal. 2016). However, while we found support for an
effect of turnover intention on bad mouthing, we found no
(dampening) effect for organizational commitment –at least
no direct one. An explanation for this finding might be that
organizational commitment is not able to reduce employees’
CRBM directly because of situational aspects. Commitment
is a state that holds over time while quitting intentions are
more short-term focused. Further research therefore might
find support for the dampening effect of organizational
commitment when shedding light on CRBM in relation to a
specific incident. This incident could be related to personal
experiences such as a job termination or a missed promo-
tion. Nevertheless, commitment influences turnover inten-
tion negatively. The mediating role of turnover intention
also explains a large proportion of how commitment affects
CRBM.
One strength of the JD-R model is its ability to predict
an interacting influence of job resources on the effect of job
demands on particular outcomes (Demerouti etal. 2001).
However, in our study in a social media context, no such
interaction was observable, indicating that the identified
resources do not have the power to limit the effect of unfair
treatment and dissatisfaction. In addition, we identified
unexpected findings, for example, that companies’ social
media guidelines seem not to be associated with CRBM.
This is surprising, because those guidelines were explic-
itly built to prevent reputational incidents. An explanation
might be that employees are not aware of the social media
guidelines of their company, or that existing guidelines do
not address CRBM. Furthermore, we found that employ-
ees with a higher level of tenure are more willing to report
badly in social media. This finding is surprising because
we expected that employees with a higher-level von ten-
ure feel more commitment for the company, which would
decrease the company-related bad mouthing in social media.
The well-established informal networks of employees with
a higher level of tenure might be an explanation. Employees
who just joined a company may fear more consequences
in comparison, to those with a higher level of tenure. We
encourage other researchers to investigate these issues.
Furthermore, our research provides core implications for
the understanding of employees’ behavior in social media.

When Employees Speak asThey Like: Bad Mouthing inSocial Media
Finally, this research extends our knowledge of the driving
forces of employees’ organizational harming behavior as
well as counterproductive work behavior.
Managerial Implications
From our findings, some notable implications for man-
agement could be derived. First, this research also has
highlighted that of the two possible strategies, namely
enhancing commitment versus reducing turnover inten-
tion, only one is suitable to affect CRBM. As commit-
ment is not correlated with the outcome variable, only
the strategy of reducing turnover intention is potentially
successful to limit the amount of employees’ bad mouth-
ing. Second, this research identified drivers of turnover
intention and organizational commitment in relation to
counterproductive work behavior in social media. Among
the drivers of turnover intention are unfair treatment, and
a consequence to it, job dissatisfaction. Given the effect
of bad mouthing in social media, which might harm a
firm’s reputation (Könsgen etal. 2018; Schaarschmidt
and Walsh 2018), it is important to devote more attention
to employees job dissatisfaction and their perceptions of
unfair treatment. Managers have to find ways to identify
reasons for employees’ dissatisfaction and find ways to
remove or dampen the effect of job dissatisfaction as sug-
gested by Helm (2013) and Schaarschmidt etal. (2015).
Additionally, we recommend reducing employees’ percep-
tions of unfair treatments by explaining their tasks and
actions in greater detail. This often helps to balance per-
ceptions against expectations. This might not come as a
surprise but in turbulent environments, managers might be
agnostic to employee signals of unfair treatment. All these
approaches tie employees to the organization and enable
them to comprehend management decisions better. These
recommendations might lead to a decrease in turnover
intention, which in turn help to reduce bad evaluations in
social media. Thus, managers should be aware of the fact
that the effects of states of dissatisfaction are no longer
narrowed to organizational boundaries; they also affect
the company’s image in the public domain.
Finally, we derive managerial implications out of the
effects provided by the control variables. For example, our
results indicate that the presence of social media guidelines
(i.e. verbal or textual descriptions of appropriate social
media usage) are ineffective. We postulate that these guide-
lines have to be enriched by social media trainings to unfold
their power.
Limitations andFurther Research
As with any research, this one is not free of limitations.
First, we employed a cross-sectional approach to capture a
wider set of perceived external reputation and organizational
climate—data from one firm might have low variance in
these two aspects as employees share similar perceptions
of their employer (Helm 2013). On the other hand, cross-
sectional survey data need a larger amount of control vari-
ables. While we have accounted for many of them, other
important control variables were not included (e.g., posi-
tion in organization). Another limitation pertains to the
use of self-reported data. Future research could find other
measurement instruments for CRBM, for example, using
observations from real Twitter or Glassdoor accounts (e.g.,
Schaarschmidt and Walsh 2018). We noted that several ways
exist to damage the employer’s reputation, but our measure
of bad mouthing grouped many of them into one construct.
For example, there might be a difference in drivers when
the posting in anonymous versus non-anonymous. Future
research could use a more fine-grained version of our meas-
urement approach.
Regarding the theoretical framework of JD-R, we did
not test for associations between different job demands or
between job dissatisfaction and unfair treatment because
our focus was on CRBM. However, from an organizational
behavior perspective, such analyses can also reveal useful
results. Finally, researchers should investigate other kinds
of job demands and job resources and their effect on ante-
cedence of employees’ company-related bad mouthing in
social media.
Appendix1
See Table4.

S.Ivens et al.
Appendix2
See Table5.
Table 4 Qualitative evidence on BM-related job demands and job resources
a This column displays the number of instances in each of the JD-R categories and the percentage of the total number of JD-R categories. About
377 quotes were identified in 33 face-to-face in-depth interviews and related to the JD-R categories
BM-related job demands and resources Number of
codes (%a)
Key quotes
Job dissatisfaction 84 (22.3) “If I am dissatisfied with my job or my employer I mention it on Kununu.com but only if
anonymity is guaranteed otherwise I would not talk about my employer on such sites.”
(Nadine, 27years, junior sales agent in media industry, 2years of job experience)
“The lack of flexible scheduling raises my dissatisfaction about my employer. I mean if
companies fail to recognize the need for their employees to maintain a healthy life and
work balance this is definitely a thing which I would bring on a company valuation plat-
form.” (Martin, 32years, member of purchasing department in the automotive industry,
8years of job experience)
Unfair treatment 63 (16.7) “If I am treated unfair in comparison to colleges I will more likely evaluate my employer
worse on company valuation platforms” (Oliver, 24years, mechatronics specialist in the
mechanical engineering industry, 2years of job experience)
“Unfair treatment at work is a reason for me to talk about my experiences in social media.”
(Marie, 22years, shop assistant in the fashion industry, 4years of job experience)
Perceived external reputation 76 (20.2) “A misfit of external and internal reputation perception would be a reason for me to talk
about in social media, because I feel the need to defend my company.” (Georg, 54years,
head of marketing in the chemical industry, 21years of job experience)
“Working for a reputable company makes it easier for me to do my job, because I am part
of a successful team. Additionally, this encourages me to overcome the monotonous part
of work. (Andrei, 24years, postman, 3years of job experience)
Supervisor support 79 (21) “I think a good manager should provide inspiration, motivation and support its employees.
If a manager does not do this, it hinders me to achieving my work goals and might be a
reason I would talk about on company valuation platforms.” (Hakim, 19years, ware-
houseman in the food industry, 1year of job experience)
“If I don’t get support from my boss for a long time or if I am dependent on her decisions,
but she has no time for me, it annoys me. Of course, this can happen, but it is decisive if
it happens or if it is a permanent condition. If it is a permanent condition, I would report
that on Kununu.com.” (Eva, 26years, assistant of the managing director in the healthcare
industry, 1year of job experience)
Organizational climate 54 (14.3) “An ongoing unfriendly atmosphere is a reason for me to report on a valuation platform
like Kununu.com, but anonymity must be ensured otherwise I would not talk about such
internal information.” (Daniel, 31years, clerk in the civil service industry, 4years of job
experience)
“If I notice that there are not much personal loyalty from employees towards the company
this is a thing I would report.” (Tim, 23years, computer specialist in the logistic industry,
3years of job experience)

When Employees Speak asThey Like: Bad Mouthing inSocial Media
References
Allen, N.J., and J.P. Meyer. 1990. The measurement and antecedents of
affective, continuance and normative commitment to the organiza-
tion. Journal of Occupational Psychology 63 (1): 1–18.
Armstrong, J.S., and T.S. Overton. 1977. Estimating non-response bias
in mail surveys. Journal of Marketing Research 14: 396–402.
Bakker, A.B., and E. Demerouti. 2007. The job demands-resources
model: State of the art. Journal of Managerial Psychology 22
(3): 309–328.
Biron, M., and C. Boon. 2013. Performance and turnover intentions: A
social exchange perspective“. Journal of Managerial Psychology
28 (5): 511–531.
Brown, T.J., P.A. Dacin, M.G. Pratt, and D.A. Whetten. 2006. Identity,
intended image, construed image, and reputation: An interdis-
ciplinary framework and suggested terminology. Journal of the
Academy of Marketing Science 34 (2): 99–106.
Browne, M.W., and R. Cudeck. 1993. Alternative ways of assessing
model fit. In Testing structural equation models, ed. K.A. Bollen
and J.S. Long, 136–162. Beverly Hills: Sage.
Cammann, C., M. Fichman, D. Jenkins, and J. Klesh. 1983. Assessing
the attitudes and perceptions of organization members. In Assess-
ing organizational change: A guide to methods, measures and
practices, ed. S. Seashore, E. Lawler, P. Mirvis, and C. Cammann,
71–138. New York: Wiley.
Chin, W., J.B. Thatcher, and R.T. Wright. 2012. Assessing common
method bias: Problems with the ULMC technique. MIS Quarterly
36 (3): 1003–1019.
Table 5 Sample description
Sample characteristic (n =472) Number Percentage
Gendera
Female 169 36
Male 257 54
Other 46 10
Age
≤ 20 14 3
21–25 186 39
26–30 131 28
31–35 45 10
36–40 19 4
41–45 9 2
≥ 46 68 14
Education
Lower secondary education 22 5
Middle school degree 76 16
Technical high school degree 71 15
High school degree 134 28
College graduate 155 33
Other 14 3
Income
< 1000€ ($1176) 102 22
1000–2000 € 182 39
2001–3000 € 100 21
3001–4000 € 33 7
> 4000€ 20 4
n/a 35 7
Internet usage (h/week)
< 1 36 8
1–5 215 46
6–10 108 23
11–15 40 8
16–20 31 6
> 20 42 9
Tenure (year)
< 1 6 1
1–5 320 68
6–10 61 13
11–15 37 8
16–20 15 3
> 20 33 7
Employee groups
Temporary staff 36 8
Trainees 22 5
Blue-collar worker 42 9
White-collar worker 372 79
Employees with customer contact 114 24
IT and Media staff 37 8
Marketing professionals 9 2
Manager 89 19
Other employees 223 47
a Gender was measured on a 7-point Likert scale where 1 was female
und 7 was male
b The participants were asked to indicate how many hierarchical levels
were between them and the CEO according to the organization chart
Table 5 (continued)
Sample characteristic (n =472) Number Percentage
Hierarchical levelb
0 9 2
1 59 13
2 85 18
3 117 25
4 87 18
5 58 12
6 27 6
7 12 3
≥ 8 18 4
Social media guidelines
Existing 259 55
Not existing 213 45
Number of employees
1–9 (very small) 37 8
10–49 (small) 96 20
50–249 (medium small) 118 25
250–499 (medium large) 54 11
500–10,000 (large) 125 27
> 10,000 (very large) 42 9

S.Ivens et al.
Cole, M.S., H. Bruch, and B. Vogel. 2006. Emotion as mediators of the
relations between perceived supervisor support and psychologi-
cal hardiness on employee cynicism. Journal of Organizational
Behavior 27 (4): 463–484.
Cooper-Thomas, H.D., and S. Wright. 2013. Person-environment mis-
fit: The neglected role of social context. Journal of Managerial
Psychology 28 (1): 21–37.
Corbin, J.M., and A. Strauss. 1990. Grounded theory research: Pro-
cedures, canons, and evaluative criteria. Qualitative Sociology
13 (1): 3–21.
Dabirian, A., J. Kietzmann, and H. Diba. 2016. A great place to work!?
Understanding crowdsourced employer branding. Business Hori-
zons 60 (2): 197–205.
Demerouti, E., A.B. Bakker, F. Nachreiner, and W.B. Schaufeli. 2001.
The job demands-resources model of burnout. Journal of Applied
Psychology 86 (3): 499–512.
Eisenberger, R., R. Huntington, S. Hutchison, and D. Sowa. 1986. Per-
ceived organizational support. Journal of Applied Psychology 71
(3): 500–507.
Essed, P. 1991. Understanding everyday racism: An interdisciplinary
theory. Newbury Park: Sage.
Fornell, C., and D.F. Larcker. 1981. Evaluating structural equation
models with unobservable variables and measurement error. Jour-
nal of Marketing Research 18 (1): 39–50.
Fox, S., P.E. Spector, and D. Miles. 2001. Counterproductive work
behavior (CWB) in response to job stressors and organizational
justice: Some mediator and moderator tests for autonomy and
emotions. Journal of Vocational Behavior 59 (3): 291–309.
Hair, J., W. Black, B. Babin, and R. Anderson. 2010. Multivariate data
analysis, 7th ed. New York: Prentice-Hall.
Helm, S. 2013. “A matter of reputation and pride: Associations between
perceived external reputation, pride in membership, job satisfac-
tion and turnover intentions. British Journal of Management 24
(4): 542–556.
Johannesson, R.E. 1973. Some problems in the measurement of organi-
zational climate. Organizational Behavior and Human Perfor-
mance 10 (1): 118–144.
Katz, R.C., J. Santman, and P. Lonero. 1994. Findings on the revised
morally debatable behaviors scale. The Journal of Psychology
128 (1): 15–21.
Kline, R.B. 2005. Principles and practice of structural equation mod-
eling, 2nd ed. New York: The Guilford Press.
Kluemper, D.H., and P.A. Rosen. 2009. Future employment selec-
tion methods: Evaluating social networking web sites. Journal of
Managerial Psychology 24 (6): 567–580.
Könsgen, R., M. Schaarschmidt, S. Ivens, and A. Munzel. 2018. Find-
ing meaning in contradiction on employee review sites—effects
of discrepant online reviews on job application intentions. Journal
of Interactive Marketing 45: 165–177.
Lindell, M.K., and D.J. Whitney. 2001. Accounting for common
method variance in cross-sectional research designs. Journal of
Applied Psychology 86 (1): 114–121.
Lum, L., J. Kervin, K. Clark, F. Reid, and W. Sirola. 1998. Explain-
ing nursing turnover intent: Job satisfaction, pay satisfaction, or
organizational commitment? Journal of Organizational Behavior
19 (3): 305–320.
Lyons, B.D., and N.A. Bowling. 2017. On the effectiveness of peer
reporting policies. Journal of Managerial Psychology 32 (8):
547–560.
Moran, E.T., and J.F. Volkwein. 1992. The cultural approach to the for-
mation of organizational climate. Human Relations 45 (1): 19–47.
Melián-González, S., and J. Bulchand-Gidumal. 2017. Extending
the scope of hotel client reactions to employee injustice: Hotel
employer reviews on the internet. International Journal of Hos-
pitality Management 63: 93–100.
Mulaik, S.A., L.R. James, J. Van Alstine, N. Bennet, S. Lind, and C.D.
Stilwell. 1989. Evaluation of goodness-of-fit indices for structural
equation models. Psychological Bulletin 105 (3): 430–445.
Ones, D.S. 2018. Counterproductive work behaviors 2.0: Assessment
or consequences. International Journal of Selection and Assess-
ment 26 (1): 1–4.
Opitz, M., V. Chaudhri, and Y. Wang. 2018. Employee social-mediated
crisis communication as opportunity or threat? Corporate Com-
munications: An International Journal 23 (1): 66–83.
Parker, J.M., S. Marasi, K.W. James, and A. Wall. 2019. Should
employees be “dooced” for a social media post? The role of social
media marketing governance. Journal of Business Research 103:
1–9.
Podsakoff, P.M., S.B. MacKenzie, J.-Y. Lee, and N. Podsakoff. 2003.
Common method bias in behavioral research: A critical review
of the literature and recommended remedies. Journal of Applied
Psychology 88 (5): 879–903.
Preacher, K.J., and A.F. Hayes. 2008. Asymptotic and resampling
strategies for assessing and comparing indirect effects in multiple
mediator models. Behavior Research Methods 40 (3): 879–891.
Schaarschmidt, M. 2016. Frontline employees’ participation in service
innovation implementation: The role of perceived external reputa-
tion. European Management Journal 34 (5): 540–549.
Schaarschmidt, M., and R. Könsgen. 2019. Good citizen, good ambas-
sador? Linking employees’ reputation perceptions with supportive
behavior on Twitter. Journal of Business Research. https ://doi.
org/10.1016/j.jbusr es.2019.04.004.
Schaarschmidt, M., and G. Walsh. 2018. Social media-driven anteced-
ents and consequences of employees’ awareness of their impact
on corporate reputation. Journal of Business Research. https ://doi.
org/10.1016/j.jbusr es.2018.11.027.
Schaarschmidt, M., G. Walsh, and S. Ivens. 2015. Perceived exter-
nal reputation as a driver of organizational citizenship behavior:
Replication and extension. Corporate Reputation Review 18 (4):
314–336.
Schaufeli, W.B., and A.B. Bakker. 2004. Job demands, job resources,
and their relationship with burnout and engagement: A multi-sam-
ple study. Journal of Organizational Behavior 25 (3): 293–315.
Schwepker Jr., C.H. 2001. Ethical climate’s relationship to job satisfac-
tion, organizational commitment, and turnover intention in the
salesforce”. Journal of Business Research 54 (1): 39–52.
Smidts, A., A.T.H. Pruyn, and C.B. Van Riel. 2001. The impact of
employee communication and perceived external prestige on
organizational identification. Academy of Management Journal
44 (5): 1051–1062.
Tajfel, H., and J.C. Turner. 1979. An integrative theory of intergroup
conflict. In The social psychology of intergroup relations, ed.
W.G. Austin and S. Worchel, 33–47. Monterey: Brooks/Cole.
VanMeter, R.A., D.B. Grisaffe, and L.B. Chonko. 2015. Of “likes”
and “pins”: The effects of consumers’ attachment to social media.
Journal of Interactive Marketing 32: 70–88.
van Zoonen, W., J.W. Verhoeven, and R. Vliegenthart. 2016. Social
media’s dark side: Inducing boundary conflicts. Journal of Mana-
gerial Psychology 31 (8): 1297–1311.
van Zoonen, W., and J.W. Treem. 2019. The role of organizational
identification and the desire to succeed in employees’ use of per-
sonal twitter accounts for work. Computers in Human Behavior
100: 34–46.
Vorauer, J.D., K.J. Main, and G.B. O’Connell. 1998. How do individu-
als expect to be viewed by members of lower status groups? Con-
tent and implications of meta-stereotypes. Journal of Personality
and Social Psychology 75 (4): 917–937.
Walsh, G., Z. Yang, D. Dose, and P. Hille. 2015. The effect of job-
related demands and resources on service employees’ willingness
to report complaints:Germany versus China. Journal of Service
Research 18 (2): 193–209.

When Employees Speak asThey Like: Bad Mouthing inSocial Media
Walsh, G., M. Schaarschmidt, and H.F.O. von Kortzfleisch. 2016.
Employees’ company reputation-related social media competence:
Scale development and validation. Journal of Interactive Market-
ing 36: 46–59.
Walsh, G., M. Schaarschmidt, and S. Ivens. 2018. Assessing the effects
of multichannel service provider corporate reputation on customer
new product adoption and RFM value. Journal of Service Man-
agement 29 (4): 680–702.
Wæraas, A., and D.Y. Dahle. 2019. When reputation management
is people management: implications for employee voice. Euro-
pean Management Journal 1: 1. https ://doi.org/10.1016/j.
emj.2019.08.010.
Wetzer, I.M., M. Zeelenberg, and R. Pieters. 2007. Never eat in that
restaurant, I did!”: Exploring why people engage in negative
word-of-mouth communication. Psychology & Marketing 24 (8):
661–680.
Publisher’s Note Springer Nature remains neutral with regard to
jurisdictional claims in published maps and institutional affiliations.
Stefan Ivens received his PhD from University of Koblenz-Landau
(Germany). He now works as a chief digital officer in a northern Ger-
man city. His research interests include reputation issues (both online
and offline) in the digital age. His work appeared in journals such as
Corporate Reputation Review, Journal of Interactive Marketing, and
Journal of Product & Brand Management.
Mario Schaarschmidt is assistant professor of Technology and Innova-
tion Management at University of Koblenz-Landau (Germany). His
research centers on customer integration, organizational identification
and reputation in social media. His work has been published in, among
others, European Management Journal, Information and Organization,
Journal of Interactive Marketing, Journal of Service Management, and
Journal of Service Research.
Raoul Könsgen received his PhD from University of Koblenz-Landau
(Germany). His research interests include social media marketing and
negative effects of online reviews. His work appeared in journals such
as Journal of Business Research and Journal of Interactive Marketing.