Trust and Credibility in Health Crises: Comparative Insights From Government and Non-Government Spokespersons in Indonesia and Australia

The global pandemic due to the novel coronavirus disease (COVID-19) caused the World Health Organization (WHO) to declare a public health emergency During the pandemic, the “infodemic” referred to public discussion concerning the disease’s origin, tests for virus detection, and preventive action (WHO, 2020b) Government requires effective public outreach to enforce healthcare policies and risk management. Governments must release clarification and guidance to the public to accelerate risk communication to avoid worse situations, especially when conspiracy theories are rapidly spreading and influencing society (Chan et al., 2021). Given the health risks associated with COVID-19, accurate information is important, as misinformation can be a matter of life and death (Krause et al., 2020). During a pandemic, social media presents new difficulties for authorities (Komuhendo & Otchere, 2025). Reliable information is crucial, especially considering the abundance of unclear information about the virus. Furthermore, governments, represented by their spokespersons, play a crucial role in mitigating communication risks caused by misinformation.

When the COVID-19 “misinfodemic” was evolving, risk communication studies were paramount as references for key risk communicators, highlighting the significant role of effective spokespersons (Chang & Jiao, 2020). Some research has highlighted why robust risk communication is crucial in pandemic management (Tahir et al., 2025), revealing how spokesperson-led communication positively influences protective and preventive behaviors of the society, mediated by risk perception (Heydari et al., 2021).

In the field of risk communication, studies examining spokesperson optimization and the outcomes are limited. We propose a comparative analysis of spokespersons in Indonesia and Australia to highlight variations in health and risk communication approaches. While both countries experienced significant challenges during the COVID-19 pandemic, their outcomes and communication environments diverged considerably. Australia reported cases amounting to approximately 41% of its population, whereas Indonesia recorded only 2.4% (Mathieu et al., 2020). In terms of cumulative excess deaths per 100,000 people, Indonesia’s rate reached 365.14 compared to just 90.85 in Australia. These disparities underscore not only differences in health outcomes but also the varying effectiveness of public communication and trust-building strategies.

Rather than adopting a strict most-similar or most-different systems design, this study employs an exploratory contrastive comparative case approach. Indonesia and Australia were selected because they represent substantially different communication environments, institutional capacities, and sociopolitical contexts during the COVID-19 pandemic. Comparing these contrasting settings allows the study to examine how spokesperson credibility, communication strategies, and public responses emerge under different public health governance structures and media ecosystems. The comparison is therefore intended to generate analytical insights into contextual variation in crisis communication rather than causal generalization across national systems.

This study examines how spokesperson background, communication style, and institutional affiliation were associated with different forms of public response and communication patterns during the COVID-19 pandemic in Indonesia and Australia.^[1]

Literature Review

Risk Communication Framework

Risk communication can be defined as the real-time exchange of information, advice, and opinions between experts or officials and individuals facing potential life, health, or socioeconomic risks due to a hazard (WHO, 2020a). Furthermore, the success of any crisis communication is intertwined with the audience’s perception of the communicator.

Communicators must be trusted sources of accurate information. Specifically, perceived trust involves the communicator’s competency, lack of bias, as well as empathy and commitment. Credibility is another characteristic cited in prior research (Özoran & Yıldız, 2025); communicators must possess credibility, including professional experience, background, competency, and responsibility. When communicators are credible, the messages they convey become more trustworthy. The World Health Organization (WHO) provides the framework of best practices for how information and content education should be presented and be available widely to the public when the health crisis occurs. It covers the elements of trust, transparency, and the role of empathy in communication (WHO, 2020a). Therefore, several researchers (Butler, 2021) developed health communication and measured the effectiveness of risk communication, which depends on the perception of the audience and their beliefs, which were influenced by a credible and authoritative spokesperson.

Although diverse opinions exist among various groups, governments must engage with the public through a clear and consistent public engagement strategy. In public health emergencies, conflicting views, misinformation, and uncertainty can create disagreement and distrust toward official health messages. Therefore, building trust and credibility between communicators and audiences dispute resolution is essential for strengthening public acceptance of health communication (Hong et al., 2025). Given that vaccination-related communication varies in framing, emphasis, and interpretation, Xu and Loffelholz (2024) suggest that public health communication should be adapted to the sociocultural and institutional contexts in which messages are delivered to offer sufficient incentives to the public. As community relationships play a pivotal role in risk and crisis management, social media can contribute to real-time discourse, inspire public action, and foster trust in risk and crisis communication related to health issues (Arief et al., 2018). Governments, through their spokespersons, utilized various channels to disseminate news about vaccination and COVID-19-related information.

Role of the Spokesperson in Crisis Communication

The role of a spokesperson is pivotal in crisis communication, as spokespersons represent the voice and credibility of governments and health institutions, requiring them to present accurate and consistent messages to the intended audience (Coombs & Holladay, 2010). In the context of health emergencies, spokespersons aim to connect with audiences, reduce psychological barriers, and garner support for protective measures, ultimately helping to reduce harm and improve crisis response (CERC, 2014). Selecting the right spokesperson is a complex and critical process, demanding qualities outlined in communication guidelines (Demeshko et al., 2022). Spokespersons should be able to convey audience-centric key messages supported by evidence, understand audience perspectives, articulate core messages, communicate confidently in media and online settings, respond effectively to difficult questions, and follow relevant organizational guidelines they represent for social media communication.

Likeability and Perception

The concept of likeability exerts a profound influence across various domains, encompassing social interactions, consumer behavior, and communication. Likeability, defined as the degree to which an individual is favored or appreciated, plays a central role in shaping initial impressions and subsequent perceptions. In the realm of mediated and digital communication contexts, Capurro et al. (2022) highlight how people swiftly assess others’ warmth and competence and how warmth, which is linked to likeability, significantly impacts initial judgments. The World Health Organization (WHO) provides a risk communication framework as guidance for spokespersons, saying they should maintain their credibility to gain trust and influence public compliance during a health emergency (WHO, 2020a). In health crisis communication, the likeability of the spokesperson and the organization they represent create a positive association that resonates with the audience.

Accordingly, our study addresses the following research questions:

RQ1: How do government and non-government health spokespersons differ in their communication strategies during the COVID-19 pandemic in Indonesia and Australia?

RQ2: How does a spokesperson’s background relate to patterns of public sentiment across different sociopolitical contexts?

RQ3: What types of educational content and communication campaigns are most frequently used by spokespersons to address COVID-19 information and misinformation?

Method

Data Collection

This study employed a comparative mixed-method design integrating qualitative content analysis and quantitative sentiment analysis of Indonesian and Australian spokespersons. Data collection for sentiment analysis aimed to thoroughly explore prevalent community topics related to COVID-19 education. Collaboration with the Information Technology Team of the Big Data Laboratory, played a pivotal role in employing data mining techniques across social media platforms based on Twitter (X) and national news outlets to systematically curate relevant information.

Moreover, individuals who had repeatedly broadcast educational content concerning the COVID-19 pandemic during the study period were included in this study. The sentiment analysis was based on robust scientific standards (Arief & Pangestu, 2021). Keywords such as “juru bicara” (spokesperson), "Juru bicara – Covid Vaksin" (spokesperson – COVID vaccine), and "Edukasi" or “Education” in English were carefully selected to reflect key pandemic discourse. “Spokesperson” includes key opinion leader communication, “spokesperson COVID vaccine” targeted vaccine-related conversation, and “education” pinpointed information dissemination.

Using these keywords, the analysis refined the information and presented the outcomes in Excel; hyperlinks of the social media posts were added. The posts that contained related variables, clearly defined temporal parameters (mentioned below), and inclusion criteria based on engagement metrics to ensure methodological consistency and reproducibility (Rufai & Bunce, 2020). The content underwent thematic categorization to identify predominant communication patterns, while sentiment analysis involved structured coding of message tones into predefined categories, supported by transparent and standardized procedures to guarantee analytical robustness and validity.

We conducted data collection, content analysis, and sentiment analysis for this study using systematic procedures described in the following sections. The timeframe chosen for this study consisted of two distinct intervals: Phase 1 (April - July 2022) and Phase 2 (August - December 2022). These intervals refer to the publication dates of the content included in the dataset. The analysis also included public mentions of the selected spokespersons within the defined study period.

Selecting Spokesperson Sample

The selection of spokespersons was based on established criteria drawn from the literature on public health communication and source credibility (Buckley et al., 2024). These criteria encompassed factors such as:

(1) repeated public communication during the COVID-19 pandemic, (2) substantial presence in national media and social media platforms, and (3) representation of diverse professional backgrounds, including government officials, medical experts, and public figures (Durau et al., 2022).

Sentiment analysis was conducted using systematic social listening and media monitoring to examine statements made by selected spokespersons. The analysis included spokespersons from different backgrounds, such as communities, opinion leaders, academician, public figure, religious figure, and government officials. The final sample consisted of 12 spokespersons, drawn from both countries: 8 Indonesian and 4 Australian spokespersons.

In the Indonesian context, spokespersons were identified through a two-stage process. First, government spokespersons were identified by mapping individuals who were officially appointed by the Ministry of Communication and Information and the National COVID-19 Task Force in COVID-19 communication. Second, non-government spokespersons were identified through media monitoring and social listening by selecting individuals who repeatedly participated in public discussions on COVID-19 during the study period. This process resulted in identifying eight Indonesian spokespersons who were prominent.

This approach ensures that the appointed Indonesian spokesperson is represented as the official spokesperson and the empirically dominant voice in the public discourse. For Australia, the selection follows a functional-institutional approach, which reflects a more centralized state communication structure.

The disparity in sample sizes between Indonesia and Australia reflects variations in national communication frameworks, wherein the quantity and diversity of spokespersons are influenced by institutional configurations and crisis communication protocols (Lovari & Bowen, 2020).

These actors can be systematically identified across national media outputs and official communication channels, ensuring consistency and comparability. Importantly, the difference in the number of spokespersons between Indonesia and Australia does not indicate an imbalance in sampling but rather reflects structural differences in the communication ecosystem. Indonesia’s multi-actor communication environment, involving both government and non-government actors, resulted in a wider distribution of influential voices, whereas Australia’s centralized system concentrates communicative authority within a limited set of institutional actors.

The Spokespersons Categorization

In Australia, the spokesperson categories were NGS (Non-Government Spokesperson), GS1 (Government Spokesperson 1), GS2 (Government Spokesperson 2), and GS3 (Government Spokesperson 3). Indonesia had more categories: NGS1 (Non-Government Spokesperson 1), NGS2 (Non-Government Spokesperson 2), GS1 (Government Spokesperson 1), GS2 (Government Spokesperson 2), and GS3 (Government Spokesperson 3). These classified the spokespersons based on their associations and backgrounds (1 = medical professionals, 2 = economic experts, and 3 = public figures from entertainment industry).

Understanding how these spokespersons created and shared content could offer insights into risk management. For each of the selected spokespersons, we sought one of their public campaign pieces shared on social media and assessed the media exposure, including comments, mentions, or engagement where available, of this educational content in light of the sentiment analysis. Finally, this research elucidates risk communication theory within the spokesperson’s content.

Data Analysis and Processing

The chosen social media platform for data collection is Twitter (now X), which involves accessing Twitter’s data through its public API URLs using the Tweepy library in Python. The initial implementation includes importing necessary libraries like Tweepy and TextBlob, with TextBlob being employed for sentiment analysis. The data obtained from Twitter comprises tweets containing alphabetic characters, as well as emoticons, laughter, and other emotional signs. The data collection spans a week, with daily data stored in separate CSV files, focusing on content and timestamps.

Sentiment Analysis

Sentiment analysis is common in natural language processing (NLP). Some models are available to examine the sentiment within text, paragraphs, or documents. Sentiment analysis was used to classify text as positive, negative, or neutral, depending on the language and context. Positive sentiments included expressions indicating satisfaction, enthusiasm, or approval. For example, “The new vaccine rollout is a big success.” Negative sentiments included expressions indicating disappointment, anger, frustration, or criticism. For example, “The government’s effort on dealing with the pandemic has been disastrous.” It indicated a negative opinion. Neutral sentiments include factual statements or descriptions, showing no strong emotions or opinions. For example, “The health ministry issued multiple new guidelines today.” It had no emotional tone. The analysis also focused on public reactions to specific spokesperson statements.

Results

The results are organized according to the three research questions to clarify how the findings address specific dimensions of spokesperson communication, public sentiment, and educational campaign patterns across Indonesia and Australia.

Communication Strategies Across Countries (RQ1)

The data consisted of approximately 12,751 social media posts from Indonesian spokespersons and 2,954 social media posts/contents/news from Australian spokespersons. Table 1 shows the distribution of content and campaign categories across both Indonesian and Australian spokespersons. Among Australian spokespersons, policy and government was the most common content category, followed by public image-related content. Content related to COVID-19 cases appeared in fewer posts, while swab testing, countering hoaxes or fake news, and religiosity were not represented in the Australian sample. Among Indonesian spokespersons, posts were more likely to be COVID-19 case updates and vaccine education and healthy living content. Categories such as hoaxes/fake news and vaccine religiosity appeared only in the Indonesian sample and represented smaller proportions of the overall content distribution.

Selected statements were used to illustrate the types of issues addressed by spokespersons in each country. Australian spokespersons often addressed public health restrictions, hospital preparedness, mask use, and border requirements. Indonesian spokespersons frequently addressed vaccine-related misinformation and policy updates. For example, one Indonesian spokesperson clarified that there was no evidence linking Favipiravir use to claims that COVID-19 survivors’ nails glowed under ultraviolet light. Another explained that the booster vaccination program was accelerated to January 2022 following the emergence of the Omicron variant in Indonesia. Indonesian statements quoted in this section were translated into English by the authors.

Spokesperson Background and Public Sentiment (RQ2)

Table 2 represents the sentiment distribution of posts associated with Australian and Indonesian spokesperson categories. Among Australian spokespersons, spokespersons linked to the art and law category produced the largest share of positive sentiment, while spokespersons associated with the economic spokesperson category received the highest proportion of negative sentiment, and spokespersons associated with health institutions received neutral sentiments.

A similar pattern appeared in Indonesia, although the communication environment was more fragmented and involved a wider range of actors. Among Indonesian spokespersons, spokespersons related to food and drug control agency and pharma–corporate spokespersons had the largest shares of positive sentiment. In contrast, the spokesperson related to medical spokesperson/expert had the largest share of negative sentiment. Neutral sentiment was most prominent among the COVID-19 Response Acceleration Task Force celebrity spokespersons and the Authorities Body-Minister group.

These findings indicate that public reactions toward spokespersons were closely connected to institutional affiliation, public visibility, and the broader political climate surrounding COVID-19 communication.

Educational Campaign Patterns and Misinformation Response (RQ3)

As presented in Table 1, the analysis identified clear differences in the educational themes emphasized by spokespersons in Indonesia and Australia. Indonesian spokespersons devoted substantial attention to vaccine-related misinformation, religious concerns, and clarification of circulating hoaxes. Content categories such as hoax or fake news and vaccine religiosity appeared only in the Indonesian dataset, reflecting the local communication environment during the pandemic. Several spokespersons actively responded to rumors regarding vaccine safety, halal status, conspiracy narratives, and misleading claims circulating on social media platforms. The findings suggest pandemic communication in Indonesia was shaped by competing information flows and the rapid spread of unverified claims.

In contrast, Australian spokesperson communication focused more consistently on public policy measures, health regulations, hospital preparedness, and risk management updates. Government policy and institutional communication occupied a larger proportion of the Australian dataset, while misinformation-oriented messaging appeared less prominent. Australian spokespersons more frequently emphasized procedural guidance, including border control, mask regulations, testing arrangements, and vaccination schedules.

These findings of how the two countries differs suggest that spokesperson campaigns were shaped by distinct communication pressures, institutional structures, and public information needs during the pandemic period.

Discussion

Results from the study showed differences in how spokespersons communicated COVID-19-related information in two countries. In Indonesia, health authority spokespersons employed a multifaceted strategy, addressing hoaxes and religious aspects. On the other hand, Australian health authority spokespersons primarily emphasize education and government policy. The different approaches used by the spokespersons between Australia and Indonesia indicated that crisis communication strategies depend on context. Thus, designing communication strategies tailored to the particular culture, society, and informational needs is crucial (Bradley et al., 2019).

This study also found that the effectiveness of spokesperson strategies in educating the public regarding pandemic issues is different between the two countries. The variations were due to the focus on different communication approaches. Indonesian spokespersons addressed a broader range of locally relevant concerns, such as hoaxes and religious concerns, while Australian spokespersons emphasized education and government policy. This finding aligns with previous research that highlights that various communication approaches are needed and should be tailored to specific cultural contexts in risk communication (Butler, 2021; Komuhendo & Otchere, 2025).

Moreover, we compared how spokesperson strategies shaped public perceptions concerning health communication. The spokesperson’s role influences public perceptions of government and healthcare policies inferred from public replies, comments, and mentions related to spokesperson posts, with Indonesian spokespersons potentially experiencing higher positive sentiment due to their multifaceted approach. In contrast, Australian spokespersons may experience some difficulty in dealing with a broader spectrum of concerns. Spokesperson strategy included the type of messaging used and the sentiment reflected in the analyzed content.

Finally, our research also revealed the significant differences in the risk communication approaches between the two countries. Indonesian spokespersons employed a complex strategy while Australian spokespersons used an educational approach. The public’s compliance and participation in preventive measures are influenced by government policies and trust in the information delivered by spokespersons.

In Australia, where trust in government has varied according to political leanings, the pandemic response and associated communications have sometimes been interpreted through a partisan lens. Similarly, in Indonesia, varying political ideologies have influenced public trust in health communications, with differing degrees of skepticism and acceptance observed across the political spectrum. The results from the study showed that the spokesperson for Food & Drug Control Agency that has no political ideology has the highest positive sentiment (97%), followed by the spokesperson from professional non-government sector (83%), while the COVID-19 Response Acceleration Task Forces that relates with political ideology had the lowest (8%).

Importantly, the study also identified feedback dynamics between communication and public sentiment. Public responses are not passive reactions but actively shape the communication environment, influencing how messages are interpreted, shared, and contested. This dynamic interaction suggests that trust is continuously negotiated rather than statically held, reinforcing the need for adaptive communication strategies.

At the same time, the findings should be interpreted with caution. While sentiment analysis provides valuable insights into patterns of public response, it does not capture the full complexity of trust, which is influenced by deeper social, political, and experiential factors. The cross-linguistic nature of the dataset also presents challenges, although efforts were made to ensure consistency before sentiment classification.

Overall, this study contributes to the literature by advancing a relational understanding of credibility in health crises. Rather than viewing trust as a fixed attribute of spokespersons, the findings highlight the importance of context, interaction, and communication practice in shaping public perceptions. This has important implications for risk communication strategies, suggesting that effective communication requires not only technical accuracy but also contextual sensitivity, cultural awareness, and sustained engagement with public concerns.

Conclusion

This study addressed three research questions: (1) how government and non-government health spokespersons differed in their communication strategies during the COVID-19 pandemic in Indonesia and Australia; (2) how spokesperson’s background relate to patterns of public sentiment across different sociopolitical contexts; and (3) what types of educational content and communication campaigns were most frequently used by spokespersons to address COVID-19 information and misinformation.

The findings show that spokesperson strategies differed across the two countries. Australian communication was more institutionally oriented, focusing on government policy, public health restrictions, and official health guidance. Indonesian communication involved both government and non-government actors and addressed a broader range of issues, including case updates, vaccine education, misinformation, and religious concerns.

Spokesperson background was associated with different sentiment patterns across the two countries. In Australia, sentiment varied across arts and law, economic, and health authority categories, while in Indonesia it varied across a broader range of government, health, corporate, and public figure-related spokespersons. Overall, these patterns suggest that sentiment differed according to spokesperson background and institutional affiliation, although they should be interpreted as sentiment associated with spokesperson-related content rather than direct evidence of trust, credibility, or communication effectiveness.

This study also found that educational content and campaign patterns differed between Indonesia and Australia. Indonesian spokespersons addressed a wider range of locally specific issues, including vaccine education, misinformation, hoaxes, and religious concerns, while Australian spokespersons focused more on institutional and policy-oriented communication. These differences suggest that spokesperson campaigns reflected the distinct communication pressures, institutional structures, and public information needs of each country.

Limitation and Future Research Considerations

This study is not without its limitations. First, the research setting was limited to Twitter/X and selected online news or social media content related to selected spokespersons, which, when coupled with the exploratory nature of the data collection process, imposes certain restrictions on the generalizability of the findings across the entirety of the social media landscape. Second, this study did not directly measure spokesperson likeability. Positive sentiment should not be interpreted as equivalent to likeability, because likeability would require additional measures, such as audience surveys, interviews, or experimental assessment of how audiences perceive each spokesperson. Third, the study did not systematically detect bot accounts or apply strict geographic filtering, meaning that some sentiment patterns may include automated accounts or users outside Indonesia and Australia. While the study provides valuable insights into spokespersons and communication within the context explored, it may not encompass the full spectrum of nuances that different social media platforms present. Future research should combine sentiment analysis with audience-based data collection, bot detection, and geographic filtering where available.