If they can do it,so can I: When vicarious experiences of coworker voice create a spiral of voice

Abstract

This study examines how vicarious experiences of voice, i.e., observing or hearing about coworkers’ voice experiences, can affect worker voice, i.e., expressing dissatisfaction about a work-related issue. Using data of 829 Dutch workers, the authors find that vicarious experiences of supervisor support of coworker voice are positively related to voice efficacy and perceived supervisor responsiveness. Conversely, vicarious experiences of supervisor suppression of coworker voice are associated with lower levels of perceived supervisor responsiveness. Furthermore, the authors find that workers’ voice efficacy is positively related to worker voice. These vicarious learning effects show how voice can spill over to other actors within an organization.

Keywords

Coworker voice supervisor response vicarious experience voice efficacy worker voice

Introduction

We study how workers’ voice is affected by vicarious experiences of voice, i.e., observing or hearing about coworker voice. Individual voice is an important way for workers to address unsafe, unfair, and undesirable situations at work. Its importance is growing, and is likely to continue to grow if union power continues its decline. Additionally, the conditions necessary for workers to express their problems individually are also deteriorating: the increasing level of job insecurity associated with labor market flexibilization changes the position of workers vis-a-vis their employers, reducing their ability to raise issues and increasing their vulnerability in terms of retaliation (Sluiter et al., 2022). Various alarming consequences of workers’ inability to speak up for themselves have emerged, including increasing levels of burnout especially among young flex workers (Dorenbosch et al., 2015), exploitation of migrant workers (Dekker et al., 2013), and increasing levels of anti-organizational behavior as an alternative channel for expressing discontent (Akkerman et al., 2020; De Lange, 2015; Dix et al., 2008; Hebdon, 2005; Hebdon and Stern, 1998; Marsden, 2013).

While the effects of workers’ increased vulnerability on voice have become apparent at the societal level, this phenomenon can also foster what Morrison and Milliken (2000) call a ‘climate of silence’ at the organizational level. A climate of silence arises when the absence of voice is perceived by workers as an indication that voice is undesirable behavior. The observation of coworkers voicing issues, on the other hand, provides workers with a signal that voice is welcomed and that it is normal and acceptable. Thus, voice in teams encourages others to voice issues as well, potentially leading to a ‘spiral of voice’ (Bowen and Blackmon, 2003). Empirical research on whether coworker voice stimulates or hampers worker voice, and the conditions under which these phenomena occur, improves our understanding of workers’ voice within the organization. In particular, the response of the supervisor to coworker voice is considered an important indicator of whether voice is acceptable behavior or not (Bashshur and Oc, 2015). Therefore, this article examines the ways in which vicarious experiences of coworker and supervisor behavior affect workers’ voice.

Despite its growing importance, there is a surprising lack of studies focusing on how individual workers speak up for their own interests (Barry and Wilkinson, 2016; Pohler et al., 2020), especially when compared to the extensive amount of research on individual voice that focuses on the interests of organizations (mainly referred to as employee or promotive voice; see Morrison, 2011). It is likely that these two types of voice share certain internal mechanisms, but that they are driven by different motives, e.g., workers’ desire to protect worker rights or to stimulate organizational growth. In this study, we explicitly examine workers’ individual voice related to their own interests (e.g., not being paid enough or experiencing a high level of work stress). We call this type of voice worker voice and define it as individual workers speaking up to the supervisor about an issue at work related to their interests (Sluiter et al., 2022). Thus, we study vicarious experiences of coworker voice and aim to answer the following research question:

To what extent do vicarious voice and supervisor responses affect worker voice?

We adopt an interdisciplinary theoretical approach by integrating theories on efficacy and vicarious learning from political science, social psychology, organizational behavior, and industrial relations (Mowbray et al., 2015). We test our hypotheses by applying structural equation modeling (SEM) to our unique survey data (Akkerman et al., 2017, 2018), which was collected in two waves and contains information about the vicarious experiences of coworker voice, observed supervisor responses and subsequent worker voice of 829 workers in the Netherlands.

Empirically, we contribute to the research on voice in three important ways. First, we contribute to the understanding of how voice can spill over within organizations by measuring vicarious experiences with voice. Although the idea of vicarious experiences with voice was previously mentioned by Detert and Treviño (2010) in their qualitative interview study, we expand on this notion theoretically and use survey data to further examine such experiences. Second, we expand the understanding of the influence of supervisor behavior on voice by measuring supervisor responses to coworker voice. Previous research has studied the influence of supervisors through factors such as: transformational leadership (Detert and Burris, 2007; Liu et al., 2010) and the relationship between supervisor and subordinate (Botero and Van Dyne, 2009; Burris et al., 2008). Few studies have measured supervisors’ actual responses to voice. We follow recent studies that focus on the direct effects of supervisor responses to worker voice (Sluiter et al., 2022; Stanojevic et al., 2020a, 2020b), and study the effects of such behavior on workers’ peers. Third, in contrast to the large amount of research that focuses exclusively on supportive supervisor behavior, we test the effect of both supportive and suppressive supervisor responses.

Theoretical framework

In what follows, we elaborate our theoretical argument about the role of vicarious learning in determining the likelihood that workers will engage in worker voice. Our argument is built as follows: First, we explain the role of voice efficacy and perceived supervisor responsiveness as important antecedents to worker voice. Thereafter, we explain why vicarious learning should be considered as an antecedent of both voice efficacy and perceived supervisor responsiveness, and we elaborate this argument specifically for coworker voice and supervisor response to coworker voice. Connecting these two elements results in the vicarious voice model, according to which vicarious experiences with (supervisor responses to) coworker voice affect a worker’s likelihood to speak up at work.

Voice efficacy, perceived supervisor responsiveness, and worker voice

To understand why workers do or do not voice, we focus on two relevant concepts from the voice literature: voice efficacy and perceived supervisor responsiveness. First, efficacy, as a general concept, refers to a person’s belief that they have the necessary capabilities, as well as the ability to use those capabilities, to take control over their actions (Bandura, 1997). In addition to general beliefs about one’s capabilities across different aspects of their life, efficacy can also be related to beliefs about specific tasks or activities (Bandura, 1986). Research in the context of work has for example shown that higher levels of efficacy can lead to increased work performance (Stajkovic and Luthans, 1998), lower levels of work-related stress and burnout (Schwarzer and Hallum, 2008), and increased job satisfaction (Judge and Bono, 2001).

More recently, voice efficacy was coined by Kish-Gephart et al. (2009) to refer to a person’s belief that they are capable of speaking up at the workplace. Political scientists use a similar concept of efficacy, namely, political efficacy, which is defined as a person’s belief that they are capable of undertaking political action (Balch, 1974; Campbell et al., 1954; Morrell, 2003). Research has shown a positive relationship between political efficacy and political participation (e.g., vote in elections) (Finkel, 1985; Gastil and Xenos, 2010; Valentino et al., 2009). Similar to these insights from political science, and in line with previous literature which suggests that workers with higher levels of efficacy are more likely to speak up to their supervisor (Aryee et al., 2017; Duan et al., 2014), we expect that the decision to express voice at work is dependent on a person’s beliefs of being capable to speak up at work. As such, we hypothesize:

Hypothesis 1 (H1): Voice efficacy is positively related to worker voice.

Second, we study the role of perceived supervisor responsiveness for the likelihood to engage in worker voice. We argue that the decision to express voice at work is dependent not only on a person’s capability to voice but also on the responsiveness of those to whom the voice is directed (Holland et al., 2017; Withey and Cooper, 1989). This is in line with previous research that suggests a link between perceptions regarding supervisors’ levels of responsiveness to voice and voice: workers who perceived their supervisor as approachable and responsive to new ideas were more likely to speak up to their supervisors. Likewise, uncertainty regarding supervisors’ responsiveness decreased the likelihood that workers would speak up (Detert and Burris, 2007; Saunders et al., 1992; Tangirala and Ramanujam, 2012; Van Dyne et al., 2008). We expect that this effect will be similar for worker voice and therefore hypothesize the following:

Hypothesis 2 (H2): Perceived supervisor responsiveness has a positive effect on worker voice.

Vicarious learning from (supervisor responses to) coworker voice

Both voice efficacy and perceived supervisor responsiveness are influenced by the experiences that people have at work. This seems rather intuitive for perceived supervisor responsiveness. After all, while people may have general ideas about how supervisors tend to respond and how they think they should respond, their perceptions of the responsiveness of their own supervisor will be influenced by the behavior of this supervisor in the work context. Following the literature on efficacy, experiences are also likely to influence voice efficacy. Successful experiences will increase efficacy, while unsuccessful experiences or failure will have a negative effect on efficacy (Bandura, 1997). As such, both voice efficacy and perceived supervisor responsiveness will be influenced by experiences with voice at work.

However, workers may have little experience with expressing worker voice, due to the lack of issues in the past or because of the risks associated with speaking up, for instance. This can lead to workers being uncertain about their own capabilities to voice, and about the level of responsiveness they may expect from the supervisor. When people have little or no past experiences on which to base their beliefs regarding their own capabilities, observations of other people’s behaviors, i.e., vicarious experiences, become an important source of information (Bandura, 1997; Gist and Mitchell, 1992; Myers, 2018; Roberts, 2010). As such, vicarious learning can therefore be especially relevant for voice efficacy and perceived supervisor responsiveness.

The most important source of such voice-related vicarious learning are coworker experiences with worker voice. This is because people are likely to compare themselves to peers who they perceive as similar to themselves. Vicarious experiences involving the behavior of such peers will affect their beliefs regarding their own capabilities to engage in similar behaviors (Bandura, 1997; Festinger, 1954). Furthermore, observations of peers’ behavior provide people indications of the social norms to which they should adhere (Cialdini and Trost, 1998; Cialdini et al., 1991). Moreover, supervisor responses to peers’ voice provide workers with valuable information about how their supervisor responds to voice. Observing coworker behavior can increase workers’ beliefs in their capabilities to engage in similar behaviors, leading to the thought, ‘If they can do it, so can I.’ Furthermore, workers can learn vicariously by hearing about their coworkers’ experiences. This can help them to know what to do when they are faced with similar situations at work (Salancik and Pfeffer, 1978; Spouse, 2002).

The behavior of the supervisor influences whether the vicarious experiences are viewed as positive or negative. The supervisors’ response to coworker voice provides information about the potential consequences of worker voice; and, as such, is an important part of the vicarious learning process (Bandura, 1986). The act of speaking up can have negative consequences for workers; it makes them stand out and places them at risk of reprisal from management or their supervisors (Burris, 2012; Milliken et al., 2003). Vicarious experiences of supervisor responses to coworker voice can therefore lead workers to expect a positive or negative outcome when speaking up. We consider two possible supervisor responses to coworker voice, a supportive and a suppressive response, and elaborate on their effects on voice efficacy and perceived supervisor responsiveness respectively.

Supervisor response to coworker voice and voice efficacy

Workers who observe or hear about a supportive supervisor response to coworker voice, i.e., the supervisor responded positively to the coworker, will be likely to view this response as a positive outcome of voice. Regardless of whether the issue was resolved by the supervisor, the receipt of support indicates that the coworker was at least somewhat successful in raising the issue. Such a vicarious experience may condition workers to expect a positive outcome when they engage in worker voice (Bandura, 2008). This experience may give them the confidence to express worker voice themselves, thereby increasing their voice efficacy.

Conversely, if a worker observes a suppressive supervisor response to coworker voice, i.e., the supervisor threatened or punished a coworker in response to their voice, the worker is likely to perceive this observation negatively. Such a vicarious experience leads workers to expect a negative outcome when engaging in worker voice (Bandura, 2008), likely perceiving that worker voice carries a higher risk as a result. This increased emphasis on the risks of speaking up can increase workers’ doubt about their own capabilities, thereby decreasing their voice efficacy.

Hypothesis 3 (H3): Vicarious experiences of coworker voice that are met with supportive supervisor responses have a positive effect on voice efficacy.

Hypothesis 4 (H4): Vicarious experiences of coworker voice that are met with suppressive supervisor responses have a negative effect on voice efficacy.

Supervisor response to coworker voice and perceived supervisor responsiveness

In addition to its effect on voice efficacy of workers, observing or hearing about supervisor responses to coworker voice demonstrates how responsive supervisors are to voice and will therefore also impact workers’ expectations about the responsiveness of the supervisor to voice. Detert and Treviño (2010) provided an example of this in their interview study on promotive voice: ‘an R&D informant reported an incident of vicarious learning in which a coworker got “dressed down” by a skip-level leader during a presentation, leaving a “flare in the air that this guy was a failure” and cementing the impression that this leader “destroys those meetings when he’s there – no one will speak up”’ (p. 258). This example illustrates how workers adjust their expectations about the supervisor’s openness to voice based on perceived negative responses to their coworker. Conversely, it is likely that observing or hearing about supervisor support of coworker voice increases the expectation that their supervisor is responsive to voice. This leads to the following two hypotheses:

Hypothesis 5 (H5): Vicarious experiences of coworker voice that are met with supportive supervisor responses have a positive effect on perceived supervisor responsiveness.

Hypothesis 6 (H6): Vicarious experiences of coworker voice that are met with suppressive supervisor responses have a negative effect on perceived supervisor responsiveness.

Figure 1 shows our conceptual model, the vicarious voice model, which depicts the hypothesized relationships involving vicarious experiences of supervisor support or suppression of coworker voice.

Figure 1.

The vicarious voice model: the hypothesized effects of vicarious experience of supervisor support or suppression of coworker voice.

Methods

Data

To test our hypotheses, we used data from two waves obtained via the Work and Politics Panel Survey (Akkerman et al., 2017, 2018). Our sample is drawn from the TNS NIPObase, which consists of 235,000 people in the Netherlands. Respondents provided written consent to participate in the survey. The data used in this study have been anonymized. A sample of 11,942 respondents who were part of the labor force in the Netherlands, meaning that they were currently employed or looking for employment, and between the ages of 15 and 67, were contacted to participate in our survey. They were representative in terms of gender, age, education, and employment situation. Ultimately, 7,599 respondents participated in the first wave of the survey, yielding a response rate of 64%. The data of the first wave were collected between July and September 2017. The data of the second wave were collected between October and November 2018. Due to panel drop-out, 7,244 of the original respondents were still available for the second wave. The second wave had a response rate of 83% and consisted of 6,008 respondents.

Respondent selection

Questions about previous vicarious experiences of coworker voice and supervisor responses to coworker voice were presented only to respondents who had not expressed worker voice themselves in the three years prior to the first wave¹ (N=3,177). From this subset, we selected the respondents who, in the second wave of the survey, were asked whether they had expressed worker voice or not. These were the respondents who indicated that they had experienced an issue related to their interests during the time period (one year) between the first and second wave. These respondents (N=897) were asked whether or not they had addressed this issue. To isolate the effects of vicarious experiences involving supportive or suppressive supervisor responses to coworker voice, we excluded the respondents who, in the first wave, (1) did not indicate that their supervisor had responded to their coworker’s voice (17 respondents), (2) indicated that their supervisor had responded in a way that was neither supportive nor suppressive (36 respondents), or (3) observed a supportive supervisor response to their coworker’s voice in addition to a suppressive response (15 respondents).² These observations involve situations that our theoretical model is unable to predict. Ultimately, this process resulted in a final sample of 829 respondents.

Measures

Worker voice

We measured the dependent variable, worker voice, in the second wave. The respondents were first asked if they had experienced an issue related to their working conditions, employment conditions, career development, or interpersonal conflicts. They could select issues from a list, which can be found in Appendix B. If they selected more than one issue, they were asked to indicate which issue was most important. The subsequent questions were related to the issue that was indicated as most important. Respondents who indicated that they had experienced an issue were then asked if they had voiced this issue to their supervisor. We used this information to measure worker voice. For this question, a score of 1 indicated that a respondent had expressed worker voice, and a score of 0 indicated that a respondent had experienced an issue but had not expressed worker voice.

Voice efficacy

Voice efficacy was measured in the first wave. We adapted items from the political efficacy scale developed by Craig and Maggiotto (1982) and Craig et al. (1990) to measure voice efficacy in the workplace environment. These political efficacy scales have been validated in political science research, and capture multiple aspects of the process of speaking up that can be applied to worker voice. We did not adapt the entire scale due to limited space within the Work and Politics Survey. Voice efficacy was therefore measured using the following four items:

EFF1: I immediately recognize a violation of my labor contract

EFF2: Employees like me are very capable of pursuing their interests as workers

EFF3: Employees like me are very capable of contributing to important organizational decisions

EFF4: It is worth listening to the opinions of employees like me about labor conditions

Respondents were asked to indicate the extent to which they agreed with these items on a five-point Likert scale that ranged from 1 (strongly disagree) to 5 (strongly agree). We used Cronbach’s alpha (α) to measure the internal reliability of our constructs. Voice efficacy showed good internal reliability (α = .81).

Perceived supervisor responsiveness

We measured workers’ expectations about the responsiveness of the supervisor to voice in the first wave. We adapted two items on the responsiveness of political leaders (Craig and Maggiotto, 1982; Niemi et al., 1991) to the workplace environment, resulting in the following items:

RES1: Generally, my supervisor does not really care about my interests

RES2: Generally, my supervisor listens to the problems of employees like me

Similar to voice efficacy, we used a five-point Likert scale for these items. Furthermore, we reversed the scores of RES1 so that a higher score corresponds to higher levels of perceived supervisor responsiveness. The Cronbach’s alpha coefficient of this construct was relatively low (α = .56). We considered this score to be just acceptable, and decided to proceed with using the scale for several reasons. First, the scale only consists of only two items, which can lead to a lower Cronbach’s alpha coefficient (Eisinga et al., 2013). In addition, Cronbach’s alpha has rather restrictive assumptions and therefore generally underestimates true reliability (Eisinga et al., 2013; Graham, 2006; Trizano-Hermosilla and Alvarado, 2016). Second, despite the low alpha, the content of the items accurately reflects what we want to measure and this scale has been successfully used in previous research (Geurkink et al., 2022).

Vicarious experience

We measured vicarious experiences of coworker voice in the first wave via the following item: ‘In the past three years, have you witnessed or heard about one or more of your coworkers voicing an issue to their supervisor/employer?’ Those who had a vicarious experiences of coworker voice in the past three years scored 1 on this item, while those who had not scored 0.

Supervisor response to coworker voice

To measure the response of supervisors to coworker voice, the respondents were asked the following question in the first wave: ‘After your coworker(s) voiced the issue, did that action result in the following response by your supervisor/employer?’ Respondents could choose from 12 different supervisor responses and could select multiple responses. Two items were used to measure supportive supervisor responses to coworker voice (e.g., ‘The supervisor helped the coworker(s) to fix the issue’). Supervisor support was ranked as 1 when a respondent indicated that their supervisor had responded in one or in both ways, and was ranked as 0 when a respondent indicated that the supervisor had not responded in a supportive way. Nine items were used to measure suppressive supervisor responses to coworker voice (e.g., ‘The supervisor bullied, threatened or intimidated the coworker(s)’). Supervisor suppression was ranked as 1 when a respondent indicated that the supervisor had responded in at least one of these ways, and 0 when a respondent indicated that the supervisor had not responded in a suppressive way. One item was used to measure passive supervisor responses (‘The supervisor responded that he/she could not change much about the issue’). This item was not used in this study because we wanted to focus on vicarious experiences of either supervisor support or suppression. Respondents could also describe their own specific observations. The supportive and suppressive supervisor responses from this item were labeled accordingly. The full list of possible supervisor responses from which respondents could choose can be found in Appendix C.

To test the effect that vicarious experiences of these different supervisor responses can have, we combined the scores of the item corresponding to vicarious experience and the scores of the items corresponding to supervisor response to coworker voice into one categorical variable, namely, Vicarious experience and supervisor response; this variable had the following categories: ‘No vicarious experiences of coworker voice’, ‘Vicarious experiences of supervisor support of coworker voice’, and ‘Vicarious experiences of supervisor suppression of coworker voice’.

Control variables

Data for all the control variables were collected in the first wave. We control for gender, age, level of education (low-, middle-, and high-level education), whether the respondents held a supervisor position, and whether the respondents had experienced an issue but had not voiced it. Research on the effects of gender on voice has shown inconclusive results in terms of whether male or female workers are more likely to express voice to their supervisor (LePine and Van Dyne, 1998; Morrison, 2011; Young, 1978). Therefore, although we had no expectations about the role of gender in our model a priori, we included gender as a control variable. Furthermore, it is plausible that older respondents, respondents with a higher level of education, and respondents who hold a supervisory position may have had more opportunities to learn how to voice, leading to more experience with voice; thus, these respondents may be more likely to have higher prior levels of voice efficacy, which could mitigate the effects of vicarious learning. Finally, respondents who experienced an issue but remained silent can have lower prior levels of voice efficacy or existing expectations that their supervisor is less responsive to voice compared to respondents who did not experience an issue. This difference could influence the effects of vicarious learning.

Analyses

Table 1 shows the descriptive statistics for all the variables used in this study, including the reference categories for the categorical variables Vicarious experience and supervisor response and Education. A correlation matrix with all the variables can be found in Appendix D.

Table 1.

Descriptive statistics.

	Min	Max	M/%	SD
Worker voice (T2)	0.00	1.00	64.41%
Voice efficacy (T1)
EFF1	1.00	5.00	3.77	0.94
EFF2	1.00	5.00	3.73	0.90
EFF3	1.00	5.00	3.70	0.90
EFF4	1.00	5.00	3.84	0.81
Perceived supervisor responsiveness (T1)
RES1	1.00	5.00	3.45	1.02
RES2	1.00	5.00	3.50	0.96
Vicarious experience and supervisor response (T1)
Reference: No vicarious experience	0.00	1.00	76.60%
Vicarious experience of support	0.00	1.00	14.96%
Vicarious experience of suppression	0.00	1.00	8.44%
Education (T1)
Reference: Low-level education	0.00	1.00	10.86%
Middle-level education	0.00	1.00	49.70%
High-level education	0.00	1.00	39.45%
Supervisor position (T1)	0.00	1.00	22.07%
Respondent had an unvoiced issue (T1)	0.00	1.00	27.86%
Gender (T1) (0=Female)	0.00	1.00	50.90%
Age (T1) (0=15 years)	0.00	50.00	23.64	13.29

Note: N = 829.

We used structural equation modeling (SEM) to test our model and hypotheses. SEM allows us to simultaneously test all of the hypothesized relationships in our model while determining whether the operationalized measures represent our theoretical constructs, thus eliminating the need to performing separate analyses (Williams et al., 2009). We performed confirmatory factor analysis (CFA) to identify our latent variables: voice efficacy and perceived supervisor responsiveness to voice. We then incorporated these latent variables into our structural regression model: the vicarious voice model. To perform CFA and SEM, we used the Lavaan package in R (Rosseel, 2012). The diagonally weighted least squares (DWLS) method was used to estimate the parameters in our model. DWLS is a robust estimator that does not assume multivariate normality in the data. The dependent variable in our model, voice, is a binary variable. Therefore, our model does not have multivariate normality, so DWLS is an appropriate method to use (Mîndrilă, 2010).

Results

Measurement model

We performed confirmatory factor analysis to test the measurement model of our study. The results of the standardized estimated factor loadings are displayed in Figure 2. The unstandardized estimates and standard errors can be found in Appendix E. All the factor loading scores are greater than 0.4; indicating that all the items are good indicators of our latent constructs (Saris et al., 2009). One modification was made to our model. We added a residual correlation between EFF3 and EFF4. The items EFF3 (‘Employees like me are very capable of contributing to important organizational decisions’) and EFF4 (‘It is worth listening to the opinions of employees like me about labor conditions’) are both related to the concept of workers’ participation in the policymaking of an organization; therefore, this positive correlation qualifies as a logical modification to the model.

Figure 2.

Path diagram of the measurement model.

The modification significantly improved the fit of the measurement model. All the fit indices are presented in Table 2. The chi-square test compares the covariance matrices of the theorized model and the data. The results shown in Table 2 demonstrate that the chi-square of our measurement model is significant, indicating a poor model fit. However, chi-square is dependent on sample size and should therefore always be interpreted with caution (Kline, 2011). This significant result of the chi-square test could be due to the relatively large sample size of our study (N=829). The RMSEA is independent of sample size, and this value indicates a good model fit (RMSEA <.06). The other commonly reported fit indices also demonstrate a good model fit to the data (CFI >.95, TLI >.95, SRMR <.05) (Schermelleh-Engel et al., 2003). Based on these results, we conclude that voice efficacy and perceived supervisor responsiveness are both coherent constructs and that the measurement model is valid.

Table 2.

Fit indices for the measurement model and the hypothesized model.

	Measurement model	Vicarious voice model
	Value	Value
χ²	24.373 (p <.01)	107.625 (p <.01)
df	7	57
CFI	.988	.970
TLI	.974	.953
RMSEA	.055	.033
RMSEA 90% confidence interval	.032 – .079	.023 – .042
SRMR	.019	.032

Note: N=829.

Hypothesized model

Next, we analyzed our hypothesized model.³ The fit indices for our model are presented in Table 2. The chi-square is significant, indicating a poor model fit. Similar to the measurement model, this significant effect could be due to the relatively large sample size of our study (N=829). All the other fit indices indicate a good fit of our model to the data (CFI >.95, TLI >.95, RMSEA <.06, SRMR <.05). Therefore, we accepted our model and could test our hypotheses.

Testing of the hypotheses

In this section, we discuss the results of testing our hypotheses. The estimates of the vicarious voice model are presented in Table 3 and displayed in Figure 3. Our first hypothesis predicted that voice efficacy would have a positive effect on worker voice. Table 3 shows that this effect is indeed positive and significant (β=0.164, p <.01), which provides support for the first hypothesis. This finding suggests that workers with higher levels of voice efficacy are more likely to express worker voice compared to workers with lower levels of voice efficacy. Hypothesis 2 predicted that perceived supervisor responsiveness would have a positive effect on worker voice. However, Table 3 shows that this effect is not significant; therefore, the second hypothesis was rejected.

Table 3.

Estimates for the vicarious voice model.

	Estimate	Standardized estimate
Voice efficacy → Worker voice	0.130 (0.045)**	0.164
Perceived supervisor responsiveness → Worker voice	0.041 (0.055)	0.047
Vicarious experience of support → Voice efficacy	0.248 (0.048)***	0.147
Vicarious experience of suppression → Voice efficacy	0.108 (0.067)	0.050
Vicarious experience of support → Perceived supervisor responsiveness	0.209 (0.062)**	0.136
Vicarious experience of suppression → Perceived supervisor responsiveness	–0.351 (0.104)**	–0.178
Age → Worker voice	–0.000 (0.001)	–0.010
Gender → Worker voice	0.017 (0.036)	0.017
Middle-level education → Worker voice	0.033 (0.125)	0.035
High-level education → Worker voice	–0.014 (0.131)	–0.014
Supervisor position → Worker voice	0.023 (0.044)	0.020
Respondent had an unvoiced issue → Worker voice	–0.053 (0.045)	–0.049
Age → Voice efficacy	0.010 (0.001)***	0.220
Gender Voice efficacy	0.051 (0.039)	0.043
Middle-level education → Voice efficacy	0.223 (0.124)	0.185
High-level education → Voice efficacy	0.425 (0.128)**	0.345
Supervisor position → Voice Efficacy	0.165 (0.050)**	0.114
Respondent had an unvoiced issue → Voice efficacy	–0.250 (0.043)***	–0.186
Age → Perceived supervisor responsiveness	–0.002 (0.002)	–0.044
Gender → Perceived supervisor responsiveness	0.005 (0.048)	0.005
Middle-level education → Perceived supervisor Responsiveness	0.046 (0.161)	0.042
High-level education → Perceived supervisor responsiveness	0.208 (0.165)	0.185
Supervisor position → Perceived supervisor responsiveness	0.098 (0.059)	0.074
Respondent had an unvoiced issue → Perceived supervisor responsiveness	–0.411 (0.058)***	–0.336
R ²
Worker voice (T2)	0.046
Voice efficacy (T1)	0.183
Perceived supervisor responsiveness (T1)	0.207

Note: N = 829. Standard errors in parentheses. **p <.01, ***p <.001.

Figure 3.

Path diagram of the vicarious voice model.

Hypothesis 3 predicted that vicarious experiences of supervisor support of coworker voice would have a positive effect on voice efficacy. The results in Table 3 show that vicarious experiences of supervisor support of coworker voice do indeed have a significant positive effect (β=0.147, p <.001) on voice efficacy in our model. This finding provides support for Hypothesis 3, namely, that workers who observe or hear about supervisor support of coworker voice are more likely to be confident in their ability to voice compared to workers without a vicarious voice experience. Conversely, Table 3 shows that the effect of observing or hearing about supervisor suppression of coworker voice on voice efficacy is not significant. Therefore, we rejected Hypothesis 4 based on the results of our model.

Our fifth hypothesis predicted that vicarious experiences of supportive supervisor responses to coworker voice would have a positive effect on perceived supervisor responsiveness. Table 3 shows a positive and significant relationship (β=0.136, p <.01) between vicarious experiences of supervisor support of coworker voice and perceived supervisor responsiveness. This result is in line with our fifth hypothesis and suggests that workers who observe or hear about supervisor support of coworker voice are more likely to perceive their supervisor as open to voice compared to workers with no vicarious voice experience. Hypothesis 6 predicted that vicarious experiences of suppressive supervisor responses to coworker voice would have a negative effect on perceived supervisor responsiveness. Table 3 shows that this effect is negative and significant (β=−0.178, p <.01), which supports Hypothesis 6. This finding indicates that workers who have vicarious experiences of suppressive supervisor responses to coworker voice are less likely to perceive their supervisor as open to voice than workers who have not had a vicarious experience. An overview of all the hypotheses is shown in Table 4.

Table 4.

Overview of the results of all the hypotheses.

Hypothesis	Vicarious voice model
H1: Voice efficacy → + Voice	Supported
H2: Perceived supervisor responsiveness → + Voice	Rejected
H3: Vicarious experience of support → + Voice efficacy	Supported
H4: Vicarious experience of suppression → - Voice efficacy	Rejected
H5: Vicarious experience of support → + Perceived supervisor responsiveness	Supported
H6: Vicarious experience of suppression → - Perceived supervisor responsiveness	Supported

In terms of the overall effect on the dependent variable, Table 3 shows that the vicarious voice model explained 4.6% of the variance in worker voice. Furthermore, the explained variance of voice efficacy and perceived supervisor responsiveness is 18.3% and 20.7%, respectively.

Control variables

The estimates for the control variables in our model are presented in Table 3. Age is positively related to voice efficacy. With regard to level of education, respondents with a middle- and high-level education are more likely to have higher levels of voice efficacy than respondents with a low-level education. Furthermore, respondents who hold a supervisory position are more likely to have higher levels of voice efficacy. We also find that the presence of unvoiced issues is negatively related to voice efficacy and perceived supervisor responsiveness.

Conclusion and discussion

We aimed to investigate the extent to which vicarious voice and supervisor responses influence worker voice. We hypothesized that workers learn from their coworkers’ voice experiences through vicarious experiences. By observing the voice of others, workers gain confidence in their own capabilities to speak up and gain information about social norms regarding voice within the organization. Moreover, they are able to observe the consequences of voice through these experiences, i.e., whether their supervisor supports or punishes their coworker’s voice. When distinguishing between positive and negative vicarious experiences, we find that vicarious experiences involving supervisor support of coworker voice are positively related to voice efficacy. Supervisor suppression of coworker voice, however, is unrelated to voice efficacy. This suggests that a worker’s confidence in their own capabilities to voice issues is higher when coworkers are rewarded for voice; however, this confidence is unaffected by supervisors’ suppressive responses. These findings are consistent with the existing research on the effects of direct experiences of supervisor responses to worker voice on voice efficacy (Geurkink et al., 2022). Two common psychological mechanisms are consistent with this finding, namely, self-serving biases and defensive attribution. Self-serving biases mean that people attribute success to themselves and failure to external factors, such as, in our case, a supervisor (Campbell and Sedikides, 1999). The effect found in this study may also be indicative of defensive attribution, in which people blame the victim of a situation and distance themselves from that victim so that they will feel less vulnerable to becoming a victim themselves (Shaver, 1970). Our study suggests that workers’ confidence in their capabilities to voice is not affected by suppressive vicarious experiences, indicating that such mechanisms might also apply to vicarious voice experiences.

With regard to workers’ expectations regarding the responsiveness of their supervisors, we find that vicarious experiences of coworker voice that are met with supervisor support are associated with higher expectations regarding the responsiveness of workers’ supervisors, while suppressive responses are associated with lower expectations. In terms of worker voice, we find that voice efficacy has a positive effect on worker voice. However, we did not find the expected positive effect of perceived supervisor responsiveness on worker voice.

Our results show the ways in which the voice of workers sets an example for their peers, and they also suggest that observations of voice are associated with a higher likelihood that workers will express voice. This finding supports the idea of a spiral of voice, in which the voice of one worker stimulates the voice of others, when voice is supported by supervisors. Interestingly, vicarious experiences of voice suppression do not affect voice efficacy. Therefore, we found no support for the idea that suppression has a silencing effect on those observing these responses, at least not in regard to these workers voicing their own issues.

Our study contributes to the literature on workplace voice in several ways. Theoretically, we add vicarious learning to the repertoire of the antecedents of voice. By distinguishing between voice efficacy and perceived supervisor responsiveness, we were able to uncover whether and how negative and positive vicarious experiences influence workers’ confidence in their own capabilities as well as the confidence in their supervisor’s responsiveness. Furthermore, we tested whether workers’ beliefs regarding the responsiveness of their supervisor have an effect on worker voice. Our results indicate that this effect is absent, suggesting that workers do not consider the responsiveness of their supervisors when they express worker voice. This finding contradicts research on promotive voice, i.e., individual workers speaking up to stimulate organizational interests, which has established that perceived supervisor responsiveness is an important antecedent of individual voice (Detert and Burris, 2007; Saunders et al., 1992; Tangirala and Ramanujam, 2012; Van Dyne et al., 2008). Indeed, a recent study finds a positive effect of workers’ expectations regarding the responsiveness of their supervisor on promotive voice (Snoep-Delleman et al., 2024). One potential explanation for this difference in the effect of perceived supervisor responsiveness on worker voice and promotive voice may be that the antecedents of individual voice change depending on the content or motive of voice (cf. Liang et al., 2012). Certain attitudes may be more relevant when workers speak up about organizational interests compared to when they speak up for their own interests. These differences should be further explored in future research.

Empirically, our original two-wave panel data set enabled us to distinguish important parts of the causal chain underlying the expression of worker voice. Although two-wave panel data do not allow for establishing causality, the time ordering of our questions does allow for establishing that the hypothesized cause (vicarious experience with voice) precedes the hypothesized effect (worker voice). Nevertheless, it should be noted that the time frame of the data collection can have an impact on the results. Namely, due to the relatively long period of time (up to four years) between the vicarious voice experience and the expression of worker voice, the effect of the vicarious voice experience might have decreased somewhat. However, despite this potential concern, we still find significant effects.

Additionally, it is important to consider the specificities of our sample to get a better understanding of the generalizability of our results. For reasons of data availability – the key questions for our study were only asked to a subset of the original sample – our sample consists of workers who did not have an issue (roughly 80% of our sample) or who did not voice their issue (roughly 20% of our sample) in the first wave of the study, but who did have an issue in the second wave of the study. The fact that the vast majority of our sample did not have an issue prior to the first wave of the study may raise the suspicion that most of our respondents work in organizations with fewer problems than the average organization. At the same time, the fact that all of our respondents did have an issue in the second wave of the survey counters this concern. Given that we only have data on the individual level and not on the level of the organization, we are unable to empirically establish the work context of the respondents in our sample. An intriguing question for future research is whether vicarious voice experiences have more impact in work environments with many problems and/or more examples of voice suppression than in work environments with fewer problems and/or fewer examples of voice suppression.

Another important innovation of our study is the theoretical and empirical inclusion of supervisor responses to coworker voice. By studying such responses we are able to further investigate the role of the supervisor in stimulating or hampering voice at work. Furthermore, our study inspires interesting routes for future research. First, workers’ relationship and closeness with their coworker, as well as their identification with the organization, are likely to moderate the learning effect of vicarious voice (Mitchell et al., 2015). Second, coworker responses to voice shape working conditions and could help to explain how the social context at work affects workers’ ability to speak up. Another route for future research would be to examine the link between worker voice and employee voice, e.g., whether and when responses to (vicarious) worker voice stimulate or hamper workers’ expressions of ideas and suggestions in terms of improving the functioning of the organization (cf. Kaufman, 2015).

Finally, the findings of our study seem to convey an encouraging message to those who fear that increased worker vulnerability to retaliation may have a contagious effect that leads to a culture of silence; while workers who speak up and are supported by their supervisors encourage their peers to speak up as well, we find no evidence that suppressive supervisor responses discourage voice. However, suppressive responses to workers’ voice do not encourage their coworkers to speak up either. Thus, our study provides managers with valuable insights into the social mechanisms associated with the supervisor’s response to voice.

Footnotes

Appendices

Data availability statement

The data used in this study and all files, codes, and scripts used to obtain the results can be made available upon request.

Declaration of conflicting interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

This article was supported by the Netherlands Organization for Scientific Research (NWO) under project number 453-15-001.

Ethical considerations

This article was part of the VICI research project ‘Linking the Discontented Employee and the Discontented Citizen’ approved by the Netherlands Organization for Scientific Research (NWO). Therefore, there was no ethical approval necessary from the university.

Consent to participate

Respondents provided written consent to participate in the survey. The data used in this study have been anonymized.

Consent for publication

Not applicable.

ORCID iD

Arjuna Snoep-Delleman

Notes

Author biographies

Arjuna Snoep-Delleman is a PhD candidate at the Department of Economics, Institute for Management Research, Radboud University. His research interests include workplace socialization, leadership behavior, workplace inequality, and worker voice. He has recently published in International Labour Review.

Agnes Akkerman is Professor of Regulation of Labour at the Amsterdam Institute for Advanced Labour Studies/Hugo Sinzheimer Institute (AIAS/HSI) at the University of Amsterdam and Professor of Labour Market Institutions and Labour Relations at the Radboud University. Her research interests include industrial conflict, co-determination and worker participation and voice. She has recently published her interdisciplinary research in the American Journal of Sociology, Work, Employment and Society, Political Psychology, Socio-Economic Review, and Comparative Political Studies.

Katerina Manevska is Assistant Professor in Empirical Political Science at Radboud University, Nijmegen. She is a cultural and political sociologist with a special interest in political and cultural change, employment relations, and interethnic relations (see: ). She has recently published her interdisciplinary research in Economic and Industrial Democracy, Socio-Economic Review, and Political Psychology.

Roderick Sluiter is a Postdoctoral Researcher at the Department of Economics and Lecturer at the Political Science Department, both at the Institute for Management Research, Radboud University. His research interests include workplace voice, precarious work, and political behavior. He has recently published in the American Journal of Sociology, Political Studies, and Socio-Economic Review.

References

Akkerman

Geurkink

Manevska

, et al (2018) Work and Politics Panel Survey 2018.

Akkerman

Manevska

Sluiter

, et al (2017) Work and Politics Panel Survey 2017.

Akkerman

Sluiter

Jansen

(2020) Temporary work and deviant behavior the role of workplace cohesion. The Sociological Quarterly 61(4): 678–702.

Aryee

Walumbwa

Mondejar

, et al (2017) Core self-evaluations and employee voice behavior. Journal of Management 43(3): 946–966.

Balch

(1974) Multiple indicators in survey research: The concept ‘sense of political efficacy’. Political Methodology 1(2): 1–43.

Bandura

(1986) Social Foundations of Thought and Action: A Social Cognitive Theory. Upper Saddle River, NJ: Prentice Hall.

Bandura

(1997) Self-Efficacy: The Exercise of Control. New York: Freeman.

Bandura

(2008) Observational learning. In: Donsbach

(ed.) The International Encyclopedia of Communication. Oxford: Wiley-Blackwell, pp. 3359–3361.

Barry

Wilkinson

(2016) Pro-social or pro-management? A critique of the conception of employee voice as a pro-social behaviour within organizational behaviour. British Journal of Industrial Relations 54(2): 261–284.

10.

Bashshur

(2015) When voice matters: A multilevel review of the impact of voice in organizations. Journal of Management 41(5): 1530–1554.

11.

Botero

Van Dyne

(2009) Employee voice behavior: Interactive effects of LMX and power distance in the United States and Colombia. Management Communication Quarterly 23(1): 84–104.

12.

Bowen

Blackmon

(2003) Spirals of silence: The dynamic effects of diversity on organizational voice. Journal of Management Studies 40(6): 1393–1417.

13.

Burris

(2012) The risks and rewards of speaking up: Managerial responses to employee voice. Academy of Management Journal 55(4): 851–875.

14.

Burris

Detert

Chiaburu

(2008) Quitting before leaving: The mediating effects of psychological attachment and detachment on voice. Journal of Applied Psychology 93(4): 912–922.

15.

Campbell

Grunn

Miller

(1954) The Voter Decides. Evanston, IL: Row, Peterson and Company.

16.

Campbell

Sedikides

(1999) Self-threat magnifies the self-serving bias: A meta-analytic integration. Review of General Psychology 3(1): 23–43.

17.

Cialdini

Trost

(1998) Social influence: Social norms, conformity and compliance. In: Gilbert

Fiske

Lindzey

(eds) The Handbook of Social Psychology. Boston, MA: McGraw-Hill, pp. 151–192.

18.

Cialdini

Kallgren

Reno

(1991) A focus theory of normative conduct: A theoretical refinement and reevaluation of the role of norms in human behavior. In: Zanna

(ed.) Advances in Experimental Social Psychology. San Diego, CA: Academic Press, Inc, pp. 201–234. Available at: https://linkinghub.elsevier.com/retrieve/pii/S0065260108603305.

19.

Craig

Maggiotto

(1982) Measuring political efficacy. Political Methodology 8(3): 85–109.

20.

Craig

Niemi

Silver

(1990) Political efficacy and trust: A report on the NES pilot study items. Political Behavior 12(3): 289–314.

21.

De Lange

(2015) Vrouwen en Flexwerkers Frauderen Steeds Meer. Het Financieele Dagblad. Available at: https://fd.nl/economie-politiek/1093128/vrouwen-en-flexwerkers-frauderen-steeds-meer.

22.

Dekker

Den Ridder

Van Houwelingen

(2013) Continu Onderzoek Burgerperspectieven: Kwartaalbericht 2013-4. Den Haag: Sociaal en Cultureel Planbureau.

23.

Detert

Burris

(2007) Leadership behavior and employee voice: Is the door really open? Academy of Management Journal 50(4): 869–884.

24.

Detert

Treviño

(2010) Speaking up to higher-ups: How supervisors and skip-level leaders influence employee voice. Organization Science 21(1): 249–270.

25.

Dix

Forth

Sisson

(2008) Conflict at work: The pattern of disputes in Britain since 1980. NIESR Discussion Paper 316. London: National Institute of Economic and Social Research.

26.

Dorenbosch

Sanders

Beudeker

(2015) Multi-Jobbing: Wenselijke of Onwenselijke Arbeidsmarktdynamiek? In: Dynamiek Op de Arbeidsmarkt. De Focus Op Flexibilisering. Den Haag: Centraal Bureau voor de Statistiek.

27.

Duan

Kwan

Ling

(2014) The role of voice efficacy in the formation of voice behaviour: A cross-level examination. Journal of Management & Organization 20(4): 526–543.

28.

Eisinga

Te Grotenhuis

Pelzer

(2013) The reliability of a two-item scale: Pearson, Cronbach, or Spearman-Brown? International Journal of Public Health 58(4): 637–642.

29.

Festinger

(1954) A theory of social comparison processes. Human Relations 7(2): 117–140.

30.

Finkel

(1985) Reciprocal effects of participation and political efficacy: A panel analysis. American Journal of Political Science 29(4): 891–913.

31.

Gastil

Xenos

(2010) Of attitudes and engagement: Clarifying the reciprocal relationship between civic attitudes and political participation. Journal of Communication 60(2): 318–343.

32.

Geurkink

Akkerman

Sluiter

(2022) Political participation and workplace voice: The spillover of suppression by supervisors. Political Studies 70(2): 327–347.

33.

Gist

Mitchell

(1992) Self-efficacy: A theoretical analysis of its determinants and malleability. Academy of Management Review 17(2): 183–211.

34.

Graham

(2006) Congeneric and (essentially) tau-equivalent estimates of score reliability. Educational and Psychological Measurement 66(6): 930–944.

35.

Hebdon

(2005) Toward a theory of workplace conflict: The case of U.S. municipal collective bargaining. Advances in Industrial and Labor Relations 14: 33–65.

36.

Hebdon

Stern

(1998) Tradeoffs among expressions of industrial conflict: Public sector strike bans and grievance arbitrations. ILR Review 51(2): 204–221.

37.

Holland

Cooper

Sheehan

(2017) Employee voice, supervisor support, and engagement: The mediating role of trust. Human Resource Management 56(6): 915–929.

38.

Judge

Bono

(2001) Relationship of core self-evaluations traits – self-esteem, generalized self-efficacy, locus of control, and emotional stability – with job satisfaction and job performance: A meta-analysis. Journal of Applied Psychology 86(1): 80–92.

39.

Kaufman

(2015) Theorising determinants of employee voice: An integrative model across disciplines and levels of analysis. Human Resource Management Journal 25(1): 19–40.

40.

Kish-Gephart

Detert

Treviño

, et al (2009) Silenced by fear: The nature, sources, and consequences of fear at work. Research in Organizational Behavior 29: 163–193.

41.

Kline

(2011) Principles and Practice of Structural Equation Modeling. New York: Guilford Publications.

42.

LePine

Van Dyne

(1998) Predicting voice behavior in work groups. Journal of Applied Psychology 83(6): 853–868.

43.

Liang

Farh

CIC

Farh

J-L

(2012) Psychological antecedents of promotive and prohibitive voice: A two-wave examination. Academy of Management Journal 55(1): 71–92.

44.

Liu

Zhu

Yang

(2010) I warn you because I like you: Voice behavior, employee identifications, and transformational leadership. Leadership Quarterly 21(1): 189–202.

45.

Marsden

(2013) Individual voice in employment relationships: A comparison under different forms of workplace representation. Industrial Relations 52(s1): 221–258.

46.

Milliken

Morrison

Hewlin

(2003) An exploratory study of employee silence: Issues that employees don’t communicate upward and why. Journal of Management Studies 40(6): 1453–1476.

47.

Mîndrilă

(2010) Maximum likelihood (ML) and diagonally weighted least squares (DWLS) estimation procedures: A comparison of estimation bias with ordinal and multivariate non-normal data. International Journal for Digital Society 1(1): 60–66.

48.

Mitchell

Vogel

Folger

(2015) Third parties’ reactions to the abusive supervision of coworkers. Journal of Applied Psychology 100(4): 1040–1055.

49.

Morrell

(2003) Survey and experimental evidence for a reliable and valid measure of internal political efficacy. Public Opinion Quarterly 67(4): 589–602.

50.

Morrison

(2011) Employee voice behavior: Integration and directions for future research. The Academy of Management Annals 5(1): 373–412.

51.

Morrison

Milliken

(2000) Organizational silence: A barrier to change and development in a pluralistic world. The Academy of Management Review 25(4): 706–725.

52.

Mowbray

Wilkinson

Tse

HHM

(2015) An integrative review of employee voice: Identifying a common conceptualization and research agenda. international journal of management reviews 17(3): 382–400.

53.

Myers

(2018) Coactive vicarious learning: Toward a relational theory of vicarious learning in organizations. Academy of Management Review 43(4): 610–634.

54.

Niemi

Craig

Mattei

(1991) Measuring internal political efficacy in the 1988 National Election Study. American Political Science Review 85(4): 1407–1413.

55.

Pohler

Luchak

Harmer

(2020) The missing employee in employee voice research. In: Wilkinson

Donaghey

Dundon

, et al. (eds) Handbook of Research on Employee Voice. Cheltenham: Edward Elgar Publishing, pp. 177–201.

56.

Roberts

(2010) Vicarious learning: A review of the literature. Nurse Education in Practice 10(1): 13–16.

57.

Rosseel

(2012) Lavaan: An R package for structural equation modeling. Journal of Statistical Software 48: 1–36.

58.

Salancik

Pfeffer

(1978) A social information processing approach to job attitudes and task design. Administrative Science Quarterly 23(2): 224–253.

59.

Saris

Satorra

Van der Veld

(2009) Testing structural equation models or detection of misspecifications? Structural Equation Modeling: A Multidisciplinary Journal 16(4): 561–582.

60.

Saunders

Sheppard

Knight

, et al (1992) Employee voice to supervisors. Employee Responsibilities and Rights Journal 5(3): 241–259.

61.

Schermelleh-Engel

Moosbrugger

Müller

(2003) Evaluating the fit of structural equation models: Tests of significance and descriptive goodness-of-fit measures. Methods of Psychological Research 8(2): 23–74.

62.

Schwarzer

Hallum

(2008) Perceived teacher self-efficacy as a predictor of job stress and burnout: Mediation analyses. Applied Psychology 57(1): 152–171.

63.

Shaver

(1970) Defensive attribution: Effects of severity and relevance on the responsibility assigned for an accident. Journal of Personality and Social Psychology 14(2): 101–113.

64.

Sluiter

Manevska

Akkerman

(2022) Atypical work, worker voice and supervisor responses. Socio-Economic Review 20(3): 1069–1089.

65.

Snoep-Delleman

Akkerman

Sluiter

, et al (2024) You better watch out: How the supervisor response to worker voice affects promotive voice. International Labour Review 163(2): 247–269.

66.

Spouse

(2002) Professional Learning in Nursing. Oxford: Blackwell Science.

67.

Stajkovic

Luthans

(1998) Self-efficacy and work-related performance: A meta-analysis. Psychological Bulletin 124(2): 240–261.

68.

Stanojevic

Akkerman

Manevska

(2020a) Good workers and crooked bosses: The effect of voice suppression by supervisors on employees’ populist attitudes and voting. Political Psychology 41(2): 363–381.

69.

Stanojevic

Akkerman

Manevska

(2020b) The oppressive boss and workers’ authoritarianism: Effect of voice suppression by supervisors on employees’ authoritarian political attitudes. Contemporary Politics 26(5): 573–595.

70.

Tangirala

Ramanujam

(2012) Ask and you shall hear (but not always): Examining the relationship between manager consultation and employee voice. Personnel Psychology 65(2): 251–282.

71.

Trizano-Hermosilla

Alvarado

(2016) Best alternatives to Cronbach’s alpha reliability in realistic conditions: Congeneric and asymmetrical measurements. Frontiers in Psychology 7: 1–8.

72.

Valentino

Gregorowicz

Groenendyk

(2009) Efficacy, emotions and the habit of participation. Political Behavior 31(3): 307–330.

73.

Van Dyne

Kamdar

Joireman

(2008) In-role perceptions buffer the negative impact of low LMX on helping and enhance the positive impact of high LMX on voice. Journal of Applied Psychology 93(6): 1195–1207.

74.

Williams

Vandenberg

Edwards

(2009) 12 structural equation modeling in management research: A guide for improved analysis. The Academy of Management Annals 3(1): 543–604.

75.

Withey

Cooper

(1989) Predicting exit, voice, loyalty, and neglect. Administrative Science Quarterly 34(4): 521–539.

76.

Young

(1978) The subordinate’s exposure of organizational vulnerability to the superior: Sex and organizational effects. Academy of Management Journal 21(1): 113–122.