• 1. How are employee voice and organizational silence conceptualized in empirical organizational studies?

• 2. What internal communication conditions enable or constrain employee voice and organizational silence?

• 3. What theoretical perspectives are used to explain the relationship between internal communication conditions, employee voice, and organizational silence?

This study used a systematic review design to synthesize empirical evidence on how internal communication conditions shape employee voice and organizational silence in organizational settings. The review was conducted and reported with reference to the PRISMA 2020 reporting framework (Page et al., 2021). A systematic review design was selected because the study aimed to identify, select, appraise, and synthesize empirical studies using explicit and reproducible procedures.

The review was guided by the following research questions:

For the purpose of this review, internal communication conditions were defined as the relational, structural, and contextual features of organizational communication that shape whether employees perceive speaking up as safe, valued, legitimate, and consequential. These conditions include leadership communication, supervisor communication, communication climate, psychological safety, trust, leader–member exchange, feedback systems, formal speak-up channels, hierarchy, guanxi, and power distance.

This review was not prospectively registered in PROSPERO, and no separate review protocol was published before the review was conducted. The review procedures, including the search strategy, eligibility criteria, screening decisions, full-text decision log, data extraction matrix, methodological appraisal, and thematic synthesis files, were documented during the review process and are available as underlying and extended data in Zenodo: https://doi.org/10.5281/zenodo.21385364.

Studies were eligible for inclusion if they met all of the following criteria: they were peer-reviewed journal articles, written in English, published between 2010 and 2026, indexed in Scopus and retrieved through the final Scopus search, presented empirical findings using qualitative, quantitative, or mixed-methods designs, examined an organizational or workplace context, focused on employee voice, employee silence, organizational silence, speaking up, safety voice, voice behavior/behaviour, or upward communication, and addressed at least one internal communication condition such as leadership communication, supervisor communication, communication climate, psychological safety, trust, leader–member exchange, feedback climate, feedback systems, formal speak-up channels, hierarchy, guanxi, or power distance.

Studies were excluded if they did not focus on employee voice, employee silence, organizational silence, speaking up, safety voice, voice behavior/behaviour, or upward communication; were not situated in an organizational or workplace context; did not examine any relevant internal communication condition; were not empirical primary studies; or were review articles, conceptual papers, editorials, commentaries, conference papers, dissertations, book chapters, or reports. Studies were also excluded if they focused primarily on citizen voice, consumer voice, political voice, electoral communication, public participation, social activism, external public relations, corporate branding, marketing communication, external stakeholder communication, or whistleblowing only as a legal or compliance issue without examining internal communication or speaking-up processes within organizations. Duplicate records were also excluded.

Review articles and conceptual papers were not included as empirical evidence in the synthesis. When closely relevant, they were used only as background sources and not as primary data for thematic synthesis. The full eligibility criteria and exclusion reason codes are available in the extended data.

The bibliographic search was conducted in Scopus. Scopus was selected because it provides broad interdisciplinary coverage across communication, management, organizational behavior, psychology, business, healthcare management, education, and social science journals. No subject-area restriction was applied because employee voice, speaking up, organizational silence, and internal communication conditions are studied across multiple research areas.

The final Scopus search was conducted on 6 July 2026. The search was applied to the article title, abstract, and keyword fields. The search strategy combined three concept blocks: employee voice/silence, internal communication conditions, and organizational or workplace context. The publication window was set from 2010 to 2026, covering records available in Scopus up to the search date. The final search string was:

The search was limited to English-language journal articles published between 2010 and 2026. Only journal articles were included at the database search stage. Conference papers, book chapters, editorials, notes, and other non-article document types were excluded. The final Scopus search identified 463 records. The complete search log and Scopus export metadata are available as underlying data in Zenodo.

Records retrieved from Scopus were exported in RIS format and imported into Rayyan for record management and screening (Ouzzani et al., 2016). Rayyan was used to organize records, identify possible duplicates, and support the screening process. One duplicate record was removed, leaving 462 records for title and abstract screening.

Title and abstract screening was conducted by the lead author using the predefined eligibility criteria. Records were excluded when they clearly did not examine employee voice, employee silence, organizational silence, speaking up, safety voice, voice behavior/behaviour, or upward communication in organizational or workplace settings, or when they did not address relevant internal communication conditions. Screening decisions were made conservatively, and records were retained for full-text assessment when the title or abstract suggested possible relevance but did not provide sufficient detail for a final exclusion decision.

After title and abstract screening, 28 full-text records were retrieved and assessed for eligibility. One record was excluded because it was a discursive paper rather than an empirical primary study. Twenty-six empirical studies met all inclusion criteria and were included in the main thematic synthesis. One additional empirical study was retained as secondary contextual evidence because it examined employee voice in relation to organizational trust and culture but primarily focused on public sector performance outcomes rather than on the internal communication conditions shaping voice or silence.

The final evidence base therefore consisted of 26 studies in the main thematic synthesis, one secondary contextual study, and one full-text exclusion. Screening decisions and full-text reasons for exclusion were documented in a decision log, which is available as underlying data. The decision log was reviewed with the author team during manuscript preparation to support consistency and transparency in eligibility decisions. The study selection process is reported in the PRISMA flow diagram.

A structured data extraction matrix was developed to ensure consistency across included studies. Extracted information included bibliographic details, country or geographical context, sector or organizational setting, study design, sample size and participant characteristics, data collection and analysis methods, core voice or silence concept, internal communication condition examined, theoretical framework, key findings, relevance to the review questions, preliminary descriptive codes, analytical theme, and methodological notes.

For quantitative studies, extraction focused on the core constructs, hypothesized relationships, study design, sample, analytical strategy, and main supported findings. For qualitative studies, extraction focused on study context, participants, data collection, analytical approach, author-reported themes, and interpretive findings. For mixed-methods studies, qualitative and quantitative components were extracted separately where relevant and then integrated at the synthesis stage.

The long-form mixed-methods study on Freedom to Speak Up Guardians was treated as an empirical mixed-methods study. Only empirical findings from the interview, case study, implementation, and cross-case analysis sections were extracted for synthesis. Embedded literature review material within that report was not treated as primary evidence.

The complete data extraction matrix is available as underlying data in Zenodo.

The methodological quality of included studies was appraised using the Mixed Methods Appraisal Tool (MMAT) because the review included qualitative, quantitative, and mixed-methods studies (Hong et al., 2018, 2019). MMAT was selected because it allows different empirical designs to be appraised using design-appropriate criteria within a single review.

All included studies were first assessed using the two MMAT screening questions: whether there were clear research questions and whether the collected data allowed the research questions to be addressed. Studies were then classified according to their design and assessed using the relevant MMAT category. Quantitative non-randomized studies were assessed with attention to sampling strategy, representativeness, measurement appropriateness, risk of non-response bias, and appropriateness of statistical analysis. Qualitative studies were assessed with attention to coherence between research questions, data collection, analysis, interpretation, and substantiation of findings. Mixed-methods studies were assessed according to the appropriateness of the qualitative and quantitative components and the degree of integration between them.

The MMAT appraisal was not converted into a single numerical score. Instead, studies were appraised at the criterion level and summarized as strong methodological confidence, moderate methodological confidence, include with caution, secondary contextual evidence, or excluded/not appraised. No empirical study was excluded solely on the basis of methodological appraisal. The appraisal was used to guide interpretation and weighting of evidence in the synthesis. The complete MMAT appraisal table is available as underlying data in Zenodo.

The review used thematic synthesis to integrate findings across qualitative, quantitative, and mixed-methods studies, informed by established methods for thematic synthesis in systematic reviews (Thomas & Harden, 2008). A meta-analysis was not conducted because the included studies were heterogeneous in terms of design, sector, country, conceptualization of voice and silence, measurement instruments, theoretical frameworks, and statistical models. Accordingly, no pooled effect measures were estimated.

The synthesis followed three stages. First, relevant findings from each included study were coded line by line or concept by concept. In quantitative studies, reported associations, mediation pathways, moderation effects, and supported hypotheses were translated into textual findings for coding. In qualitative studies, author-reported themes, participant-based interpretations, and conceptual categories were coded directly. In mixed-methods studies, qualitative and quantitative findings were coded separately and then compared for convergence.

Second, similar codes were grouped into descriptive themes. These themes captured recurring patterns across the evidence base, including leadership support, psychological safety, trust in supervisors, leader–member exchange, power distance, formal speak-up channels, communication climate, safety climate, organizational response, and silence as withholding.

Third, descriptive themes were interpreted into higher-order analytical themes. The final thematic synthesis generated six analytical themes: leadership communication as a signal for voice; psychological safety and trust as relational mechanisms; hierarchy, power distance, and the production of silence; voice infrastructure and organizational response; silence as active withholding; and voice and silence in digital and high-risk organizational contexts.

These themes were then integrated into a conceptual synthesis model showing how internal communication conditions shape employees’ interpretation of whether speaking up is safe, valued, legitimate, and consequential, which in turn shapes employee voice, speaking up, silence, withholding, or knowledge hiding. The thematic synthesis workbook and theme evidence matrix are available as underlying and extended data in Zenodo.

A formal statistical assessment of reporting bias or publication bias was not conducted because the review did not include meta-analysis or pooled effect estimation. The included studies were methodologically and conceptually heterogeneous, and the synthesis was thematic rather than statistical. To reduce selection bias within the review process, the search strategy, eligibility criteria, screening decisions, full-text decisions, data extraction matrix, methodological appraisal, and thematic coding structure were documented and made available as underlying or extended data.

A formal certainty-of-evidence assessment such as GRADE was not conducted because the review synthesized heterogeneous qualitative, quantitative, and mixed-methods evidence and did not estimate pooled intervention effects. Instead, confidence in the synthesis was considered through convergence across studies, methodological appraisal, relevance to the review questions, and consistency of thematic patterns.

One study was retained as secondary contextual evidence rather than included in the main thematic synthesis. The study was empirical and Scopus-indexed, and it examined employee voice in relation to organizational trust and organizational culture. However, its primary focus was public sector performance rather than the internal communication conditions that enable or constrain voice or silence. It was therefore used only to contextualize the broader organizational relevance of employee voice and was not used as a primary basis for generating analytical themes.

The review process is documented through the PRISMA flow diagram, PRISMA 2020 checklist, Scopus search log, Scopus export metadata, full-text decision log, data extraction matrix, MMAT appraisal table, thematic synthesis workbook, and theme evidence matrix. These materials are available as underlying or extended data in Zenodo: https://doi.org/10.5281/zenodo.21385364. The completed PRISMA checklist and PRISMA flow diagram are available alongside the review as reporting guideline files.

The Scopus search identified 463 records. After duplicate removal, 462 records were screened at the title and abstract stage. Records were excluded when they did not examine employee voice, employee silence, organizational silence, speaking up, safety voice, voice behavior, or upward communication in an organizational or workplace context, or when they did not address relevant internal communication conditions such as leadership communication, psychological safety, trust, leader–member exchange, communication climate, hierarchy, power distance, guanxi, or formal speak-up systems.

After title and abstract screening, 28 full-text records were retrieved and assessed for eligibility. One record was excluded at the full-text stage because it was a discursive paper rather than an empirical primary study. Twenty-six empirical studies met all inclusion criteria and were included in the main thematic synthesis. One additional empirical study was retained as secondary contextual evidence because it examined employee voice in relation to organizational trust and culture but primarily focused on public sector performance outcomes rather than on the communication conditions shaping voice or silence.

The final evidence base therefore consisted of 26 studies in the main thematic synthesis, one secondary contextual study, and one full-text exclusion. The study selection process is presented in Figure 1.

The 26 studies included in the main synthesis were published between 2018 and 2026. Although the search strategy covered studies published from 2010 to 2026, all studies that met the final eligibility criteria and were included in the main synthesis were published from 2018 onward. This indicates that empirical research linking employee voice or silence with internal communication conditions has developed most visibly in recent years.

The included studies covered diverse geographical and organizational contexts. A substantial proportion of the evidence came from Asian organizational settings, including China, Indonesia, Vietnam, Taiwan, Malaysia, Pakistan, the Republic of Korea, and Iran. Other studies were conducted in South Africa, England, Portugal, Greece, Nigeria, Albania, and cross-cultural virtual teams involving participants from 23 countries. This geographical distribution is important because hierarchy, power distance, guanxi, leadership authority, and cultural expectations appeared repeatedly as conditions shaping voice and silence.

The studies represented several organizational sectors, including healthcare and nursing, hospitality, telecommunications, manufacturing, finance, education, aviation, ICT, SMEs, public sector organizations, and virtual teams. Healthcare and nursing studies formed a particularly visible cluster, reflecting the importance of speaking up, safety voice, psychological safety, and error reporting in safety-critical environments. Studies in hospitality, finance, education, technology, and aviation extended the evidence base beyond healthcare and showed that voice and silence are also relevant to service quality, innovation, burnout, knowledge sharing, and organizational learning.

Methodologically, the evidence base was dominated by quantitative studies. Twenty-three studies used quantitative designs, including cross-sectional survey designs, time-lagged surveys, multi-source designs, matched leader–follower dyads, structural equation modelling, mediation and moderated mediation models, and network analysis. Two studies used qualitative designs, and one study used a mixed-methods design. The qualitative studies examined organizational silence and knowledge-related withholding, while the mixed-methods study examined the implementation of Freedom to Speak Up Guardians in NHS acute and mental health trusts.

The main concepts examined across the studies were employee voice, employee silence, organizational silence, speaking up, safety voice, promotive voice, prohibitive voice, voice–silence conversion, and knowledge-related withholding. The most common internal communication conditions were psychological safety, trust in supervisors or leaders, organizational trust, leader–member exchange, leadership style, safety climate, power distance, guanxi, communication skills, voice systems, and organizational response mechanisms.

A summary of included study characteristics is provided in Table 1.

The methodological quality of included studies was appraised using the Mixed Methods Appraisal Tool. The appraisal was used to guide the interpretation of the strength of evidence rather than to exclude studies mechanically. No empirical study was excluded solely on the basis of MMAT appraisal.

Overall, most included studies demonstrated moderate to strong methodological confidence. Stronger studies tended to use larger samples, multi-source or dyadic data, time-lagged designs, longitudinal elements, mixed-methods implementation analysis, or network analytic approaches. For example, the Freedom to Speak Up Guardians study used a mixed-methods design involving interviews with 87 Guardians and 116 interviews with stakeholders, workers, and national stakeholders, making it especially useful for understanding voice infrastructure and organizational response (Jones et al., 2022). Lu et al. (2025) used data from 812 nurses and network analysis to examine the relationships between voice behavior and team psychological safety climate.

The most common methodological limitations were reliance on cross-sectional survey designs, non-probability or convenience sampling, limited evidence of representativeness, self-report measurement, and restricted capacity to infer causality. These limitations were especially relevant when interpreting studies that tested mediation or moderated mediation models using cross-sectional data. Accordingly, the synthesis emphasizes convergence across studies rather than treating individual statistical associations as causal proof.

The MMAT appraisal summary is presented in Table 2.

The thematic synthesis generated six analytical themes: leadership communication as a signal for voice; psychological safety and trust as relational mechanisms; hierarchy, power distance, and the production of silence; voice infrastructure and organizational response; silence as active withholding; and voice and silence in digital and high-risk organizational contexts. These themes are summarized in Table 3.

Together, the themes show that employee voice and organizational silence are not merely individual communication choices. They are communicative responses shaped by leadership behavior, psychological safety, trust, leader–member relationships, power asymmetries, formal voice systems, and the perceived consequences of speaking up.

Leadership communication emerged as a central condition shaping whether employees perceived voice as safe, legitimate, and useful. Across the included studies, leadership appeared not only as a formal managerial role but also as a communication signal. Employees interpreted leader behavior as evidence of whether suggestions, criticism, concerns, or warnings would be welcomed, ignored, or punished.

Several studies showed that supportive leadership styles were associated with employee voice through relational and psychological mechanisms. Inclusive leadership was linked to employee voice through psychological safety and affective commitment in small hospitality firms (Fatoki, 2024). Servant leadership was associated with employee voice through trust in leaders among SME employees in Pakistan (Hassan et al., 2025). Ethical leadership shaped employee willingness to speak out through leader–member exchange, psychological safety, and psychological empowerment in a Chinese telecommunications context (Hu et al., 2018). Ambidextrous leadership also promoted employee voice in the Vietnamese medical device industry, with intrinsic motivation and power distance shaping this relationship (Tran et al., 2025).

Other studies extended this leadership theme by showing that voice is shaped by how leaders manage interpersonal distance, authority, and exchange quality. Paternalistic leadership styles influenced employee voice through trust in supervisors and self-efficacy (Dedahanov et al., 2022). Leader self-deprecating humour reduced employee silence through leader–member exchange, suggesting that informal and relational forms of leadership communication can reduce employees’ tendency to withhold ideas or concerns (An et al., 2023). In a more direct modelling pathway, leader prohibitive voice shaped follower prohibitive voice through psychological safety, indicating that leaders can normalize risk-oriented voice by demonstrating such behavior themselves (Tian et al., 2025).

Taken together, the evidence suggests that leadership communication affects employee voice not only by encouraging employees verbally, but by signalling the interpersonal and organizational consequences of speaking up. Employees are more likely to use voice when leaders communicate openness, humility, fairness, support, and responsiveness; they are more likely to remain silent when leadership signals distance, risk, or low receptiveness.

Psychological safety and trust formed the most recurrent relational mechanisms in the review. Across sectors, employees were more likely to speak up when they perceived the communication environment as safe for interpersonal risk-taking and when they trusted leaders, supervisors, teams, or organizations to respond constructively.

Psychological safety was linked to multiple forms of voice, including promotive voice, prohibitive voice, speaking up, safety voice, and general employee voice. In Chinese technology innovation companies, psychological safety climate was positively associated with both promotive and prohibitive voice, while guanxi with team leaders moderated this relationship (Ok et al., 2025). Among Chinese nurses, team psychological safety climate was positively connected with voice behavior, and network analysis identified respect for suggestions, receptiveness to new ideas, and collaborative skill development as important bridge nodes linking psychological safety and voice (Lu et al., 2025). In safety improvement contexts, psychological safety and joint problem-solving orientation were associated with speaking up and taking action (Bahadurzada et al., 2024).

Trust appeared in several forms: trust in supervisors, trust in leaders, organizational trust, team trust, and trust embedded in leader–member exchange. For example, trust in supervisors mediated the link between high-performance work systems, stewardship climate, and employee voice among Nigerian telecommunications employees (Badru et al., 2024). Organizational trust was associated with voice behavior among NICU nurses through career resilience (Zhang et al., 2024). In another Chinese context, organizational trust and authenticity enhanced employee voice behavior, showing that voice is shaped by both relational confidence and the employee’s sense of being able to act authentically at work (Liu et al., 2022).

The synthesis indicates that psychological safety and trust are related but analytically distinct. Trust concerns the perceived reliability, benevolence, fairness, or integrity of leaders, supervisors, teams, or organizations. Psychological safety concerns whether the immediate communication environment is safe for interpersonal risk-taking. In practice, however, they frequently reinforce one another: trustworthy leaders and organizations help create psychologically safe climates, and psychologically safe climates make employees more willing to use voice.

Hierarchy and power distance shaped whether voice was perceived as legitimate, risky, or inappropriate. This theme was particularly visible in Asian organizational contexts, where speaking up may be interpreted as challenging authority, disrupting harmony, threatening face, or violating implicit expectations of deference.

In the Indonesian hospitality sector, workplace incivility was positively related to employee silence, and power distance strengthened the relationship between workplace incivility and silence. This suggests that hierarchical distance can intensify silence when employees experience disrespectful treatment from seniors or supervisors (Gustiawan et al., 2023). In the Vietnamese medical device industry, power distance moderated the indirect relationship between ambidextrous leadership and employee voice through intrinsic motivation, with the positive leadership–voice pathway becoming stronger in lower power-distance contexts (Tran et al., 2025).

Power distance also appeared as an individual or cultural boundary condition in studies of voice behavior. Zhou et al. examined workplace currencies, leader–member exchange, psychological safety, and power distance orientation in relation to employee voice, showing that power distance shapes how employees interpret relational exchange and voice opportunities (Zhou et al., 2020). Chang et al. examined power distance orientation, team-based self-esteem, team trust, and voice behavior among Taiwanese nurses, further indicating that hierarchy-related beliefs influence whether employees feel confident enough to speak up (Chang et al., 2019).

This theme shows that employee voice systems cannot be treated as culturally neutral. In hierarchical contexts, encouraging employees to speak up may be insufficient if organizations do not also address fear of authority, status asymmetry, face concerns, and the perceived cost of disagreeing with superiors. Voice is not only a matter of having something to say; it is also a matter of whether the organizational hierarchy permits employees to say it.

The review also found that employee voice requires organizational infrastructure. Individual willingness to speak up is unlikely to produce organizational learning when formal channels, response mechanisms, resources, or follow-up systems are absent or weak. This theme shifts attention from voice as an individual act to voice as part of an organizational communication system.

The mixed-methods study of Freedom to Speak Up Guardians provided the strongest evidence for this theme. Jones et al. found that speak-up systems depend not only on the presence of formal roles or policies but also on coherent implementation, time, resources, organizational support, monitoring, and a climate congruent with speaking-up values (Jones et al., 2022). This suggests that voice infrastructure must include both channels for expression and organizational capacity to respond.

Other studies showed that broader organizational systems also shape voice. High-performance work systems and stewardship climate were linked to employee voice through trust in supervisors, suggesting that HR systems and relational climates jointly shape employees’ willingness to communicate upward (Badru et al., 2024). High-performance work systems were also associated with promotive and prohibitive voice, which then related to innovative behavior, with psychological safety functioning as an important boundary condition (Miao et al., 2020). Safety climate was linked to safety voice through psychological safety, showing that employees’ willingness to raise safety concerns depends on both organizational climate and perceived interpersonal safety (Sun et al., 2022).

Together, these studies indicate that voice becomes consequential only when organizations provide credible routes for speaking up and demonstrate that employee input will be heard, processed, and acted upon. Without response capacity, voice risks becoming symbolic, and employees may learn that speaking up is futile.

The included studies challenge the assumption that silence is simply the absence of voice. Silence appeared as active and motivated withholding of ideas, concerns, knowledge, warnings, or evaluations. This withholding was shaped by fear, futility, burnout, incivility, weak psychological safety, low trust, communication barriers, organizational culture, and knowledge hiding.

Jafary et al.’s qualitative study identified several factors affecting organizational silence, including organizational communication, organizational culture, structure and process, management style, individual characteristics, and communication skills (Jafary et al., 2019). This supports the interpretation of silence as an organizational communication phenomenon rather than a purely individual disposition. Costa Pacheco et al. examined employee silence and psychological safety in an aviation company and showed that different forms of silence are related to psychological safety, upward communication, and broader organizational conditions (Costa Pacheco et al., 2025).

Silence also appeared as a response to resource depletion and stress. Among Greek teachers, psychological safety mediated relationships between burnout, engagement, and different silence motives, particularly acquiescent and quiescent silence (Kassandrinou et al., 2023). Similarly, the Indonesian hospitality study showed that workplace incivility can increase employee silence, especially under high power-distance conditions (Gustiawan et al., 2023). Leader self-deprecating humour, by contrast, reduced employee silence through higher-quality leader–member exchange in Chinese financial institutions (An et al., 2023).

The evidence therefore suggests that silence is multidimensional. Fear-based silence requires protection from retaliation and stronger psychological safety. Futility-based silence requires visible organizational responsiveness. Relational silence requires attention to interpersonal consequences and team norms. Knowledge-related silence requires trust, transparency, and better information-sharing systems. Treating silence as one undifferentiated outcome risks overlooking the different mechanisms that produce withholding.

The final theme concerns contexts in which the consequences of voice and silence are intensified: digital work, virtual teams, healthcare, aviation, and safety-critical organizations. In these settings, silence can affect not only employee participation but also patient safety, risk detection, service quality, innovation, and organizational learning.

In cross-cultural virtual teams, relationship-oriented leadership influenced voice–silence conversion through psychological safety, while digital tool usage shaped this relationship (Dong & Xing, 2026). This indicates that digital tools do not automatically weaken communication; under some conditions, they can support relational leadership and voice. However, cultural differences and virtual team dynamics may also complicate how employees interpret speaking-up opportunities.

Healthcare and nursing studies repeatedly showed that speaking up is tied to safety and error prevention. Freedom to Speak Up Guardians were implemented to support workers and organizations in raising, responding to, and learning from speaking-up concerns (Jones et al., 2022). Studies of nurses in China and Taiwan showed that psychological safety, team trust, power distance orientation, and self-efficacy are important conditions shaping voice behavior in clinical settings (Chang et al., 2019; Lu et al., 2025).

Aviation and ICT studies extended this theme beyond healthcare. In aviation, employee silence was linked to psychological safety and upward communication, making silence relevant to high-reliability and safety-sensitive work (Costa Pacheco et al., 2025). In ICT firms, transparency, knowledge hiding, and employee silence were connected, suggesting that silence can also appear through knowledge concealment and weak information-sharing norms in knowledge-intensive settings (Grabocka, 2026).

Overall, this theme shows that in digital and high-risk contexts, voice is not merely a desirable participation behavior. It is part of organizational risk management, learning, innovation, and safety infrastructure.

Across the six themes, the evidence supports a communication-centered framework of employee voice and organizational silence. Internal communication conditions shape voice and silence through three interconnected layers: relational, structural, and contextual. Relational conditions include leadership communication, trust, leader–member exchange, psychological safety, authenticity, commitment, resilience, and motivation. Structural conditions include HR systems, speak-up channels, safety climate, joint problem-solving routines, feedback systems, and organizational response mechanisms. Contextual conditions include hierarchy, power distance, guanxi, digital mediation, virtual work, and high-risk professional settings.

The conceptual model in Figure 2 summarizes this synthesis. Internal communication conditions shape employees’ interpretation of whether voice is safe, valued, legitimate, and consequential. These interpretations influence communicative responses, including employee voice, speaking up, safety voice, employee silence, withholding, and knowledge hiding. These responses may then affect organizational consequences such as learning, innovation, safety improvement, problem detection, risk concealment, weak learning, or repeated silence.

This systematic review synthesized empirical evidence on how internal communication conditions shape employee voice and organizational silence. Across 26 studies included in the main thematic synthesis and one secondary contextual study, the review found that employee voice and silence are not simply individual communication choices. Rather, they are communicative responses to organizational conditions. Employees speak up when they perceive that voice is safe, legitimate, valued, and likely to produce a meaningful response. They remain silent when the communication environment signals interpersonal risk, futility, hierarchy, low trust, weak responsiveness, or possible retaliation.

The thematic synthesis generated six analytical themes: leadership communication as a signal for voice; psychological safety and trust as relational mechanisms; hierarchy, power distance, and the production of silence; voice infrastructure and organizational response; silence as active withholding; and voice/silence in digital and high-risk organizational contexts. These themes converge on a central argument: internal communication conditions operate through relational, structural, and contextual pathways. Relationally, employees interpret leader behavior, supervisor relationships, psychological safety, trust, and leader–member exchange as indicators of whether speaking up is safe. Structurally, they evaluate whether formal voice channels, HR systems, safety climates, feedback systems, and speak-up mechanisms can transform voice into organizational action. Contextually, their willingness to speak is shaped by hierarchy, power distance, guanxi, sectoral risk, and digital or virtual work arrangements.

The strongest and most consistent evidence concerned psychological safety and trust. Across healthcare, hospitality, manufacturing, technology, finance, education, aviation, and telecommunications contexts, employees were more likely to speak up when they perceived lower interpersonal risk and stronger relational confidence. Psychological safety and trust therefore appear to function as central relational mechanisms through which internal communication environments enable or constrain voice.

Leadership was also central. The review shows that leadership communication shapes voice not only through formal encouragement, but through signals embedded in everyday leader behavior. Inclusive leadership, servant leadership, ethical leadership, ambidextrous leadership, paternalistic leadership, leader prohibitive voice, and leader self-deprecating humour were not merely managerial styles; they functioned as communicative cues through which employees assessed whether upward communication was welcome or risky. This finding supports the argument that voice is partly produced through employees’ interpretation of leader responsiveness, humility, openness, and authority.

Hierarchy and power distance also emerged as important conditions. In high-power-distance or strongly hierarchical environments, voice may be interpreted as challenging authority, threatening face, or violating expectations of deference. This means that organizations cannot assume that formal voice opportunities will be experienced equally by all employees. In some settings, employees may remain silent not because they lack ideas or concerns, but because the social cost of speaking appears too high.

This review contributes to the literature in three main ways. First, it integrates employee voice and organizational silence within a single communication-centered synthesis. Previous studies often examine voice and silence separately, with voice framed as constructive expression and silence framed as withholding or withdrawal. This review shows that the two concepts are related but not simply opposites. A workplace may formally invite employee voice while still producing silence if employees perceive that speaking up is unsafe, unwelcome, or ineffective. Conversely, reducing silence requires more than increasing opportunities for expression; it requires changes in relational safety, leader response, trust, and organizational follow-through.

Second, the review shifts attention from individual willingness to internal communication conditions. Existing empirical work has examined individual-level mechanisms such as intrinsic motivation, self-efficacy, affective commitment, authenticity, and career resilience. These mechanisms are important, but the synthesis shows that they are embedded in broader communication environments. Employees do not decide whether to speak up in isolation. They interpret leader behavior, supervisor trustworthiness, peer norms, communication climate, hierarchy, formal channels, and past organizational responses before deciding whether voice is worth the risk.

Third, the review foregrounds silence as an active communicative process. Silence is not merely non-communication. It may involve defensive withholding, acquiescence, fear, futility, relational calculation, prosocial concealment, opportunism, or knowledge hiding. This distinction matters because different forms of silence require different organizational responses. Fear-based silence requires protection and psychological safety. Futility-based silence requires visible response and feedback. Relational silence requires attention to interpersonal consequences and team norms. Knowledge-related silence requires transparency, trust, and better knowledge-sharing infrastructure.

The main conceptual contribution of this review is the framing of internal communication conditions. This term captures the relational, structural, and contextual features of organizational communication that shape whether employees perceive voice as safe, valued, legitimate, and consequential. Rather than treating internal communication narrowly as information transmission from management to employees, this review conceptualizes it as the communicative environment within which employees evaluate risk, authority, trust, responsiveness, and consequence.

The synthesis suggests three layers of internal communication conditions. The first layer is relational. This includes leadership communication, supervisor–employee interaction, psychological safety, trust, leader–member exchange, team respect, and mutual support. These relational conditions shape whether employees feel interpersonally safe enough to express ideas, concerns, disagreement, warnings, or criticism.

The second layer is structural. This includes formal speak-up systems, reporting channels, HR systems, feedback systems, safety climates, joint problem-solving routines, and organizational response mechanisms. Structural conditions determine whether employee voice has a credible route and whether the organization can convert voice into learning, safety improvement, innovation, or change.

The third layer is contextual. This includes hierarchy, power distance, guanxi, cultural norms, virtual work, digital communication tools, high-risk professional settings, and sector-specific risks. Contextual conditions shape the meaning of voice itself. In low-hierarchy settings, voice may be interpreted as constructive participation. In high-power-distance or safety-critical contexts, the same act may be interpreted as challenging authority, threatening face, or exposing organizational failure.

This three-layer framework clarifies why interventions that simply ask employees to “speak up” are insufficient. Voice depends on whether employees perceive speaking up as safe, valued, heard, and consequential. Therefore, internal communication should not be understood only as message delivery, but as the design of relational and structural conditions that make employee voice possible and meaningful.

The findings have several implications for organizational communication and human resource management. First, organizations should not assume that employees will speak up simply because voice channels exist. Formal channels are necessary, but they are not sufficient. Employees also need to believe that the organization will listen, protect them from negative consequences, and act on the information they provide. This is especially important in healthcare, aviation, and other high-risk sectors where silence can prevent organizations from detecting problems early.

Second, leadership development should include communication behaviors that make voice safer and more credible. The included studies suggest that leaders promote voice when they listen, respond, model openness, communicate humility, provide support, and demonstrate that critical or risk-oriented communication is legitimate. Leader prohibitive voice is especially important because it shows that leaders can model the very behavior they expect from employees. When leaders themselves raise concerns about harmful practices, they reduce uncertainty about whether such communication is acceptable.

Third, organizations need to diagnose the type of silence they are dealing with. Silence caused by fear requires different interventions from silence caused by futility, burnout, relational concern, or knowledge hiding. If employees are silent because they fear retaliation, the priority is protection, psychological safety, and leader accountability. If they are silent because they believe nothing will change, the priority is response visibility and feedback loops. If silence is connected to knowledge hiding, the priority is transparency, trust, and knowledge-sharing norms.

Fourth, organizations operating in high-power-distance or hierarchical contexts need culturally sensitive voice systems. Encouraging voice without addressing status asymmetry may expose lower-level employees to risk. In such settings, voice mechanisms should include confidential channels, clear non-retaliation policies, leader training, collective problem-solving routines, and visible follow-up. The goal is not only to invite employee voice, but to reduce the hierarchical costs of speaking up.

Fifth, internal communication and HR practitioners should treat voice as part of organizational learning infrastructure. Voice should be connected to decision-making, safety improvement, innovation, and feedback systems. When voice is disconnected from action, employees may learn that speaking up is symbolic rather than meaningful. The synthesis suggests that organizations should design voice systems not only around expression, but also around response.

The review also identifies several methodological implications. The evidence base is dominated by quantitative survey studies, many of which are cross-sectional. Although these studies provide useful associations and theory-testing evidence, they are limited in their ability to establish causal relationships. Future research should use more longitudinal, time-lagged, experimental, diary, qualitative, ethnographic, and mixed-methods designs to capture how voice and silence develop over time.

There is also a need for more studies that examine actual communication processes rather than only self-reported perceptions. Many studies measure psychological safety, leadership style, trust, or voice behavior through survey scales. These measures are useful but may not fully capture how voice happens in meetings, digital platforms, reporting systems, or supervisor–employee interactions. Future research could examine actual communication episodes, message content, silence moments, response patterns, and feedback loops.

Another methodological issue concerns the tendency to test mediation models using cross-sectional data. Several included studies used mediation or moderated mediation designs, but not all had temporal separation or multi-source data. Future studies should align research design more closely with causal claims, especially when proposing that leadership produces psychological safety, which then produces voice. Time-lagged, longitudinal, or multi-source designs would strengthen such claims.

The review also suggests a need for comparative and contextualized research. Many included studies were conducted in Asian contexts, which is a strength for understanding hierarchy and power distance. However, future work should compare how internal communication conditions operate across cultures, sectors, and institutional settings. Studies that compare high- and low-power-distance organizations, remote and co-located teams, or healthcare and non-healthcare settings would add important explanatory depth.

This review has several limitations. First, it was limited to Scopus-indexed, English-language journal articles. Although Scopus provides broad interdisciplinary coverage across management, organizational behavior, psychology, communication, health services research, and social sciences, relevant studies indexed only in other databases or published in non-English outlets may have been missed. Therefore, the review should be interpreted as a systematic review of Scopus-indexed studies rather than a comprehensive review of all possible literature on employee voice and silence.

Second, the review did not conduct a meta-analysis. The included studies varied substantially in design, sector, country, theoretical framework, measurement approach, and outcome conceptualization. This heterogeneity made thematic synthesis more appropriate than statistical pooling. However, the absence of meta-analysis means that the review does not estimate pooled effect sizes.

Third, the evidence base was dominated by quantitative survey studies. This creates a risk that the synthesis reflects what is most easily measured rather than the full complexity of organizational communication. Qualitative and mixed-methods studies were fewer, even though they are particularly useful for understanding how voice and silence are experienced, negotiated, and interpreted in everyday organizational life.

Fourth, some concepts were unevenly represented. Psychological safety and leadership received substantial attention, whereas internal communication as an explicit construct was less commonly named. Many studies examined internal communication conditions indirectly through trust, leader–member exchange, leadership, safety climate, or power distance. This required conceptual interpretation in the synthesis.

Fifth, one secondary contextual study was retained but not used as a primary basis for theme generation because its focus was more on employee voice as a predictor of public sector performance than on conditions shaping voice or silence. This decision helped maintain conceptual focus, but it also means that the review gives more attention to antecedents and mechanisms than to downstream organizational outcomes.

Future research should develop a more explicit internal communication theory of employee voice and silence. Rather than treating voice only as an employee behavior or silence only as an outcome of fear, future studies should examine how communication systems, leader responses, informal norms, digital channels, and feedback loops produce or reduce voice and silence.

More qualitative and ethnographic work is needed to examine how employees decide whether to speak up in real organizational situations. Interviews, observations, meeting analysis, and digital communication analysis could reveal how silence is produced in practice. Such work would be especially valuable in hierarchical, high-risk, or culturally complex organizations.

Future research should also examine the role of digital communication technologies. Virtual teams, platform-based communication, anonymous reporting tools, internal social media, and AI-mediated feedback systems may change how employees interpret risk, visibility, and psychological safety. The current evidence base contains only limited work on digital and virtual contexts, even though these settings are increasingly central to organizational communication.

Another important direction is to examine organizational response. Many studies focus on whether employees speak up, but fewer examine what organizations do after employees speak. Future research should study how voice is received, categorized, escalated, ignored, acted upon, or fed back to employees. Without response, voice may not produce learning; without learning, employees may become silent again.

Finally, more research is needed on silence as a multidimensional phenomenon. Different forms of silence may have different causes and consequences. Future studies should distinguish fear-based silence, futility-based silence, prosocial silence, relational silence, deviant silence, and knowledge hiding. This would allow organizations to design more precise interventions.

The authors gratefully acknowledge the Indonesia Endowment Fund for Education (Lembaga Pengelola Dana Pendidikan/LPDP), Ministry of Finance of the Republic of Indonesia, for supporting the their postgraduate studies.

•  An J, Di H, Yang Z, et al.: Does My Humor Touch You? Effect of Leader Self-Deprecating Humour on Employee Silence: The Mediating Role of Leader-Member Exchange. Psychol. Res. Behav. Manag. 2023; 16: 1677–1689. PubMed Abstract | Publisher Full Text | Free Full Text
•  Badru AF, Karadas G, Olugbade OA, et al.: Can employees’ trust their supervisor? The role of high-performance work systems and stewardship climate on employee voice. Heliyon. 2024; 10(19): e37795. PubMed Abstract | Publisher Full Text | Free Full Text
•  Bahadurzada H, Kerrissey M, Edmondson AC: Speaking Up and Taking Action: Psychological Safety and Joint Problem-Solving Orientation in Safety Improvement. Healthcare (Switzerland). 2024; 12(8). PubMed Abstract | Publisher Full Text | Free Full Text
•  Chang WY, Hsu CT, Yu PY: Confidence is the plant of slow growth: A moderated mediation model for predicting voice behavior among power distance orientation and team-based self-esteem in Taiwanese nurses. Psychol. Res. Behav. Manag. 2019; 12: 609–617. PubMed Abstract | Publisher Full Text | Free Full Text
•  Chao KI, Cheng JW, Yen HC, et al.: Should I speak up? How trust in leaders and leader–leader exchanges influence nurses’ voice behaviour. Nurs. Open. 2024; 11(2): e2101. PubMed Abstract | Publisher Full Text | Free Full Text
•  Costa Pacheco D, de Serpa Arruda Moniz AID , Nunes Caldeira S, et al.: Employee Silence and Psychological Safety: Insights from an Aviation Company with Broader Organizational Applications. Tourism and Management Studies. 2025; 21(2): 97–111. Publisher Full Text
•  Dedahanov AT, Fayzullaev AK u, Abdurazzakov OS, et al.: Paternalistic Leadership Styles and Employee Voice: The Roles of Trust in Supervisors and Self-Efficacy. Sustainability (Switzerland). 2022; 14(19). Publisher Full Text
•  Dong X, Xing L: How relationship-oriented behavior influences employee voice-silence conversion in cross-cultural virtual teams: the mediating role of psychological safety. Front. Psychol. 2026; 16PubMed Abstract | Publisher Full Text | Free Full Text
•  Edmondson A: Psychological safety and learning behavior in work teams. Adm. Sci. Q. 1999; 44(2): 350–383. Publisher Full Text
•  Edon JI, Koroh VFH, Tuhuteru A, et al.: Underlying and extended data for “Internal Communication Conditions Shaping Employee Voice and Organizational Silence: A Systematic Review of Scopus-Indexed Empirical Studies”. [Dataset]. Zenodo. 2026. Publisher Full Text
•  Fatoki O: Inclusive Leadership and Employee Voice Behaviour: Serial Mediating Effects of Psychological Safety and Affective Commitment. Administrative Sciences. 2024; 14(9). Publisher Full Text
•  Grabocka E: Enabling Innovation through Transparency: A Grounded Theory of Knowledge Hiding and Employee Silence in Albanian Information and Communication Technology Firms. Ianna Journal of Interdisciplinary Studies. 2026; 8(1): 981–996. Publisher Full Text
•  Gustiawan D, Noermijati AS, Indrawati NK: Workplace incivility to predict employee silence: Mediating and moderating roles of job embeddedness and power distance. Cogent Bus. Manag. 2023; 10(1). Publisher Full Text
•  Hassan N, Yoon J, Dedahanov AT: Servant Leadership Style and Employee Voice: Mediation via Trust in Leaders. Administrative Sciences. 2025; 15(3). Publisher Full Text
•  Hong QN, Pluye P, Fàbregues S, et al.: Mixed Methods Appraisal Tool (MMAT), version 2018: User guide. McGill University; 2018.
•  Hong QN, Pluye P, Fàbregues S, et al.: Improving the content validity of the Mixed Methods Appraisal Tool: A modified e-Delphi study. J. Clin. Epidemiol. 2019; 111: 49–59.e1. PubMed Abstract | Publisher Full Text
•  Hu Y, Zhu L, Zhou M, et al.: Exploring the influence of ethical leadership on voice behavior: How leader-member exchange, psychological safety and psychological empowerment influence employees’ Willingness to Speak Out. Front. Psychol. 2018; 9(SEP). PubMed Abstract | Publisher Full Text | Free Full Text
•  Iddrisu I, Mohammed B: Investigating the influence of employee voice on public sector performance: The mediating dynamics of organizational trust and culture. Social Sciences and Humanities Open. 2024; 10: 101096. Publisher Full Text
•  Jaaffar T, Samy NK: Investigating the complex relationships between leadership, psychological safety, intrinsic motivation, and nurses’ voice behavior in public hospitals using PLS-SEM. Belitung Nurs. J. 2023; 9(2): 165–175. PubMed Abstract | Publisher Full Text | Free Full Text
•  Jafary H, Yazdanpanah A, Masoomi R: Identification of factors affecting organizational silence from the viewpoint of middle managers of Shiraz University of Medical Sciences: A qualitative study. Shiraz E Medical Journal. 2019; 20(4). Publisher Full Text
•  Jones A, Maben J, Adams M, et al.: Implementation of ‘Freedom to Speak Up Guardians’ in NHS acute and mental health trusts in England: The FTSUG mixed-methods study. Health and Social Care Delivery Research. 2022; 10(23): 1–124. PubMed Abstract | Publisher Full Text
•  Kassandrinou M, Lainidi O, Mouratidis C, et al.: Employee silence, job burnout and job engagement among teachers: the mediational role of psychological safety. Health Psychol. Behav. Med. 2023; 11(1). PubMed Abstract | Publisher Full Text | Free Full Text
•  Liu X, Song L, Zheng J, et al.: When Chinese Employees Speak Up: The Experience of Organizational Trust and Authenticity Enhances Employees’ Voice Behavior. Int. J. Environ. Res. Public Health. 2022; 19(23). PubMed Abstract | Publisher Full Text | Free Full Text
•  Lu M, Xia R, Wang R, et al.: Network analysis of the relationships between voice behavior and team psychological safety climate among Chinese nurses. BMC Nurs. 2025; 24(1): 794. PubMed Abstract | Publisher Full Text | Free Full Text
•  Miao R, Lu L, Cao Y, et al.: The high-performance work system, employee voice, and innovative behavior: The moderating role of psychological safety. Int. J. Environ. Res. Public Health. 2020; 17(4). PubMed Abstract | Publisher Full Text | Free Full Text
•  Morrison EW: Employee voice behavior: Integration and directions for future research. Acad. Manag. Ann. 2011; 5(1): 373–412. Publisher Full Text
•  Morrison EW, Milliken FJ: Organizational silence: A barrier to change and development in a pluralistic world. Acad. Manag. Rev. 2000; 25(4): 706–725. Publisher Full Text
•  Ok C, Choi M, Kim HE: The Effect of Psychological Safety Climate on Voice Behavior in Chinese Technology Innovation Companies: Focusing on the Moderating Effect of Guanxi with Team Leaders. Sustainability (Switzerland). 2025; 17(15). Publisher Full Text
•  Ouzzani M, Hammady H, Fedorowicz Z, et al.: Rayyan—a web and mobile app for systematic reviews. Syst. Rev. 2016; 5: 210. PubMed Abstract | Publisher Full Text | Free Full Text
•  Page MJ, McKenzie JE, Bossuyt PM, et al.: The PRISMA 2020 statement: An updated guideline for reporting systematic reviews. BMJ. 2021; 372: n71. PubMed Abstract | Publisher Full Text | Free Full Text
•  Pinder CC, Harlos KP: Employee silence: Quiescence and acquiescence as responses to perceived injustice. Research in Personnel and Human Resources Management. 2001; Vol. 20. : pp. 331–369. Publisher Full Text
•  Song Y, Peng P, Yu G: I Would Speak Up to Live Up to Your Trust: The Role of Psychological Safety and Regulatory Focus. Front. Psychol. 2020; 10. PubMed Abstract | Publisher Full Text | Free Full Text
•  Sun Y, Yang H, Wu X, et al.: How Safety Climate Impacts Safety Voice—Investigating the Mediating Role of Psychological Safety from a Social Cognitive Perspective. Int. J. Environ. Res. Public Health. 2022; 19(19). PubMed Abstract | Publisher Full Text | Free Full Text
•  Thomas J, Harden A: Methods for the thematic synthesis of qualitative research in systematic reviews. BMC Med. Res. Methodol. 2008; 8: 45. PubMed Abstract | Publisher Full Text | Free Full Text
•  Tian X, Chae H, Song D, et al.: Leader prohibitive voice shapes employee voice through psychological safety moderated by self-efficacy and generational differences. Sci. Rep. 2025; 15(1). Publisher Full Text
•  Tran TVT, Nguyen TTQ, Alang T, et al.: Ambidextrous Leadership and Employee Voice in the Vietnamese Medical Device Industry: The Role of Intrinsic Motivation and Power Distance. SAGE Open. 2025; 15(4). Publisher Full Text
•  Van Dyne L, Ang S, Botero IC: Conceptualizing employee silence and employee voice as multidimensional constructs. J. Manag. Stud. 2003; 40(6): 1359–1392. Publisher Full Text
•  Van Dyne L, LePine JA: Helping and voice extra-role behaviors: Evidence of construct and predictive validity. Acad. Manag. J. 1998; 41(1): 108–119. Publisher Full Text
•  Zhang X, Huang X, Hu Y, et al.: The relationship between organizational trust and voice behavior among neonatal intensive care unit nurses in tertiary A hospitals in Sichuan Province: the mediating role of career resilience. Front. Public Health. 2024; 12. PubMed Abstract | Publisher Full Text | Free Full Text
•  Zhou Q, Huo D, Wu F: Different Workplace Currencies and Employee Voice: From the Multidimensional Approach of Leader–Member Exchange. Front. Psychol. 2020; 11. PubMed Abstract | Publisher Full Text | Free Full Text