Untangling threat perception in international relations: an empirical analysis of threats posed by China and their implications for security discourse

Abstract

In international relations, the concept of ‘threat perception’ is the key focus in research on war, deterrence, coercion, alliances and conflict. Although detecting and measuring the concept of threat is challenging, this field of study has become increasingly critical. To represent the intangible ‘threat notion’ more tangibly, this article selects ‘China Threat’ as the security reference object, merging constructivist and Copenhagen School principles of ‘Identity,’ ‘Power,’ and ‘Intention.’ This forms the basis for constructing questionnaires. An experimental observation and questionnaire survey was conducted on master’s and doctoral students from the Republic of China (Taiwan) who had previously taken part in a large-scale decision-making simulation. The research applies a ‘quasi-experimental approach’ and ‘non-participant observation.’ The regression analysis reveals that the perception of the difference in ‘power’ and the perception of ‘harmful intent’ are relatively correlated and predictive of threat perception. In other words, when China’s military power and harmful intention become more assertive, the group’s threat perception increases. These correlation findings serve as an effective indicator for future research and measurement of ‘perceived strength of enemy threat,’ strengthening awareness of enemy threats within the group by addressing textual issues and the establishing the logic of securitization discourse.

Keywords:

• Securitization
• threat perception
• identity
• power
• intention
• experimental method

REVIEWING EDITOR:

• Samuel Adu-Gyamfi, History and Political Studies, Kwame Nkrumah University of Science and Technology Collegeof Arts and Social Science, Kumasi, Ghana

SUBJECTS:

• Military & Strategic Studies
• Asian Politics
• Security Studies - Pol & Intl Relns

Questions about security cannot be separated from the most fundamental questions of political theory.

- Walker (1997, p. 3)

Introduction

The two World Wars of the twentieth century inflicted permanent global scars on many countries and people. To avoid such conflicts and losses, academics in the discipline of International Relations began to focus on how nations responded to external and internal threats, developing a study of global security after the war. During the Cold War, Security Studies were primarily concerned with national security and threats from outside the country at a military level. Conflicts and wars are the topics at hand. What are these threats? What is security exactly? Who determines or releases them, and where do they originate?

Realism concerns itself with power, and especially this striving for power-political advantage, put states at odds with each other, and the power struggle is a significant source of conflict in and of itself (Trachtenberg, 2003, p. 156); classical realists, such as Carr, and Morgenthau and offensive realists share the assumption that states seek to maximize power – that states are relentless seekers of power and influence (Zakaria, 1998, p. 19). For offensive realists, states maximize military power, influence, and wealth to become more secure in a world of anarchy, and the best way for a state to increase its odds of survival is to become the most potent state; their primary focus is on the causes of threats and how to respond to them (Lobell, 2017, p. 2). Neo-realism and neo-liberalism have traditionally focused on the security of state Centrism (Chatterjee, 2003, pp. 125–126). Regarding International Relations, they highlighted that countries would protect themselves at any cost if national security or interests were challenged. Therefore, it was difficult for them to move away from their own power and military methods. Most studies have concentrated on effective deterrence, weapons control and strategic relations between nations within the International System (Booth, 2007, p. 96).

Following the Cold War, the international community transitioned to a multipolar structure. The Copenhagen School broadened the origins and targets of threats, while also including regional security analyses. The Speech Act Theory in linguistic theory was used in the Epistemology and Methodology. The core of security analysis uses discourse to develop and place threat problems on the political agenda for acceptance by key political players. Consequently, rather than being an objective phrase, security is a subjective, self-referential behavior of the elite (Booth, 2007, p. 108). Furthermore, the Securitization method of viewing situations as threats from a specific group differs from the threat analysis of the Copenhagen School. It provides a distinct analytical framework for securitization issues (Buzan & Hansen, 2009, p. 36). This approach is commonly used to study health, cross-national crimes and international institutions. Considering the Copenhagen School’s Securitization Theory is founded on Austin’s speech-act theory, the primary emphasis is on the function of language.

Speech-act theory is not limited to the creation of linguistic structures. It considers non-verbal communicative situations, but Austin focuses on the relationship between language and actions (concerning words as ‘performatives’), where certain statements can describe a given fact and some practical actions. According to Ole Waver, actors can grasp the social situation in which they find themselves through speech and, more crucially, demonstrate an intentional attitude toward existing systems. Hence, language plays an integral role in the securitization process, and the effectiveness of securitization depends on the persuasiveness of the discourse (Collins, 2016a, 2016b, p. 172). Floyd defined securitization theory as follows: Securitization equals Security Measures (Discourse and Action) and Security Practices (Emergency Measures; Lo, 2015, p. 63).

Moreover, survival is a concern of security studies in the conventional subject of International Relations because of the ‘existential danger’ posed to the addressee – a threat whose uniqueness allows nations to use security as a pretext to justify using force. Consequently, security is a tool for transcending political rules and systems. Securitization is an extreme kind of ‘politicization.’ Special abilities and opportunities are accessible when securitization utilizes the rhetoric of existential danger. Consequently, the presence of existential danger justifies securitization to obtain control and possibilities by violating traditional standards. This implies that when an event occurs, actors’ views shift from ‘issue-based to politicized to securitized,’ depending on the many underlying components.

In recent years, the People’s Liberation Army (PLA) has dispatched military aircraft and warships to circumnavigate the Taiwan Strait, conducting military exercises against Taiwan on many occasions. However, members within Taiwan hold different views on this ‘existential threat.’ According to the survey on ‘Taiwan Public Opinion in the Cross-Strait Military Crisis’ conducted by the Taiwan Public Opinion Foundation in November 2021, only 28.1% of the respondents agreed that ‘China will attack Taiwan with force sooner or later.’ This indicates that the majority of the people do not believe that China will attack Taiwan with force. Despite China frequently taking alarming military actions against Taiwan, many people still believe that there will be no conflict or war between the two sides of the Taiwan Strait (Taiwanese Public Opinion Foundation, 2021). These results illustrate that different groups harbor varying perceptions of external threats, explaining the ambiguity and unpredictability of security content.

As security status is directly tied to threat perception, Rousseau discovered that elements such as Shared Identity and Power have varying effects on Threat-perception (Rousseau & Garcia Retamero, 2007, p. 744). Threat perception is also affected by an opponent’s intention. Singer presented a quasi-numerical threat assessment model in which threat perception = capacity × intention (Salmon & Imber, 2000, p. 55). This model stresses the relationship between a possible adversary’s weapon system and military-level capabilities, and their proclivity to utilize these capabilities to affect us. Some researchers claim that ‘Culture’ is essential in forming ‘Identity’ and has a solid instrumental tendency – various cultures project ideals to the outside world (Ding-Guo, 2004, pp. 140–141). For example, David L. Rousseau deconstructed the link between institutional structures and political norms in international debates using statistical analyses, historical case studies and laboratory research (Rousseau & Garcia Retamero, 2007, p. 766).

The Copenhagen School focuses on a regional level of analysis and explores the security generated by state interactions. Simultaneously, Securitization Theory emphasizes that the formation of security issues depends on how they are presented and the degree to which they are accepted. Therefore, securitization focuses on the discourse of ‘existential threats’ and the formation of substantive policies through political discourse, also known as the ‘speech-act’ securitization move.

In recent years, owing to the rapid growth of China’s national power, the ‘China threat theory’ has become an issue of global concern and has gradually influenced the foreign policy direction of countries around the world toward China. In Taiwan, since China has never announced that it would give up military aggression against Taiwan, one can often read newspaper headlines, such as ‘China military drills pose a serious threat to Taiwan’s national security,’ ‘China hegemony threatens Taiwan’s security,’ and ‘China’s disinformation attempts to undermine the core values of democracy,’ etc. However, whether or not these ‘existential threats’ constructed by discourses can effectively persuade the public or the political elites of different political parties or factions is related to the issue of ‘Threat perception,’ which is an unmentioned part of the ‘Securitization Theory.’ This part is not mentioned in the ‘Securitization Theory’ and has not been analyzed in domestic studies; therefore, it needs to be further discussed.

Hence, this study captures the perspectives of constructivism and realism on the source of threats through theoretical exploration and conducts empirical analyses using questionnaire surveys and experimental methods to understand the level of threat perception of China among audiences, with the hope of grasping the key factors affecting threat perception. This will not only fill in the limits and deficiencies of individual theories but also provide a cross-theoretical dialogue in the field of security research. In addition to compensating for the limitations and deficiencies of individual theories and providing a cross-theoretical dialogue in the field of security studies, the findings of this study will serve as a reference for security actors to apply to security referent objects in terms of the topic level, use of language, strength of wording and choice of policies to enhance the persuasive power of the securitization discourse.

The connotation and definition of the concept of threat

In the social sciences, security refers to the national capacity to protect a country’s internal values and keep it safe from threats. Humans require security in order to survive. It is vital to be free of harm, including unseen threats, intimidation and violence (Heywood, 2018, p. 298). However, it is evident that the abstract concept of security needs to be revised and contentious, and that it is impossible to define it objectively in political science (Smith, 2005, pp. 27–28). It is also true for the relative nature of ‘threat.’ To that end, we investigate the four dimensions of ‘security research and threat definition,’ ‘threat construct analysis,’ ‘threat construction perspective in security theory,’ and ‘the influences of Identity, Power, and Intention on Threat perception.’

Security research and threat definition

Security and threats are two distinct, opposing ideas. Individuals’ emotions and deeply ingrained ideals are reflected in their sense of security (Kolodziej, 2018, p. 7). Any survival issue was considered a security issue. Collins presented classic scholarly definitions of ‘security’ in Contemporary Security Studies (Collins, 2016a, 2016b, p. 3).

For national security, a threat is an act or series of events that first threatens lives drastically and degrade people’s quality of life in a brief period. Second, it is a substantial threat forcing a national government or a domestic private, nongovernmental entity (individual, group, or corporation) to narrow their choices of policies. (Ullman, 1983, p. 133)

Dr. Mo argued that security was initially undefined. Scholars define security as a goal, objective or protection against external threats and objective values (Da-Hua, 1998, pp. 20–22). Dr. Chen also thinks most security experts viewed it as ‘being free from threats (Chen, 2009, p.103).’ Thus, it can be found that security research focuses on ‘threats’ and the ability to respond to and recognize threats.

Dimensional analysis of threats

In security studies, a threat is both a component and an indicator of security. What are its components? ‘Security dilemma’ studies have focused on military challenges faced by governments (Collins, 2016a, 2016b, p. 225). After the Cold War, the total country power, military development, equipment and GDP did not apply to all threats. Thus, the threat interpretation is biased. Buzan notes that no political event, material force or academic theory explains the evolution of security studies (Buzan & Hansen, 2011, p. 6). Hence, this study attempts to obtain different and feasible complementary perspectives from various theories as essential indicators for measuring threats.

Threat construction from a realist perspective

International Relations Theories often evaluate a country’s ‘power’ by population and geographical size, resources, economic capacity, military strength and political stability; however, different Schools of Realism view power differently. Classical Realists view a nation’s power struggles as a human nature. They argue that power is both a means and an end for the nation and that a balance of power is essential to the international order (Morgenthau & Thompson, 1985, p. 187). Unlike traditional realism, neo-realism (or Structural Realism) maintains that foreign and domestic systems fundamentally differ. The domestic political systems are hierarchical. Unit equivalence underpins international politics. Thus, international regulatory power is limited to national domains (Waltz, 1979, pp. 81–93). The threat is ‘the ability or intention of one actor or group to impose any negative situation on another actor or group.’ Both classical and structural realists argue that the perception of threat by a group depends on the asymmetric power function between actors (Garcia-Retamero et al., 2012, pp. 179–180).

Threats occur when an individual or group negatively impacts another person. Classical and structural realist theorists have stated that power asymmetry affects group danger. In an international system, the weaker party fears being exploited and the stronger party fears being intimidated by subsequent powers and in fringing on its interests. Such a theoretical proposition is the first possible explanation for the ‘threat perception’ in this study.

Constructivist perspectives on threat construction

The Soviet Union imploded unexpectedly during the early 1990s. People have questioned the inadequacy of standard theories in foreseeing the Cold War’s evolution after 40 years. Since then, it has steadily evolved into a prominent research avenue in International Relations theory and politics (Da-Hua, 2003, p. 72). Constructivism holds that our knowledge of the universe is not objective and is built only on our shared understanding from a specific perspective (Collins, 2016a, 2016b, p. 131). According to Alexander Wendt, the three points of ‘enter the constructive relationship between agents and structures,’ are essential positions of Constructivism, making it a realistic approach to studying International Relations and security. This applies the concept of constructive thinking to the supplied concept of rationalism (Wendt, 1987, pp. 355–361).

In defining ‘identity,’ the subject understands and affirms oneself, and the interaction between actors and systems creates collective self-interest. As a result, the ‘collective societal identity’ emerges. In summary, constructivism offers a nonmaterialistic explanation for establishing the identity and origins of threat perceptions.

Threat construction of instrumentalization orientation

War, deterrence, coercion, alliances and conflicts in international relations revolve around threat perception and military capabilities. When intention is proposed as a military threat, detecting signals and verifying intent are the biggest issues in threat assessment and analysis (Huddy et al., 2013, p. 2). States’ early threat assessments conceal their objectives from their opponents out of self-interest (Rosato, 2015, p. 48). Constructivism, like Realism, academically, recognizes that states are unable to grasp the possible intentions of the opponent (Schuurman & Eijkman, 2015, p. 215).

However, constructivism fosters shared perceptions and behavioral practices, arguing that knowing the intents of others in actor interactions can lead to a greater understanding of each other, minimizing ambiguity and altering threat perceptions. Intentions are a part of unfinished actions; therefore, defining aggressive or offensive intent is difficult. Researchers use ‘Indicators’ to clarify words, behaviors and signs to improve prediction. Markers of aggressive goals may explain competing threat perceptions.

Hybrid analysis framework

The new realism paradigm, famed for its nuclear deterrence and defense–offense balance, was steadily questioned after the Cold War (Dannreuther, 2013, p. 43). The Constructivist movement challenged concepts, such as anarchy, sovereignty and the inability of war. According to Wendt, anarchy is a subjective status established by countries via interaction, mutual perception and understanding, allowing the field of security studies in International Relations to shift gradually (Wendt, 1992, p. 391).

International relations have a branch called ‘security studies.’ The changes brought about by the Cold War caused security studies to transcend existing traditions, giving rise to new ideas and views. The Copenhagen School, which includes European scholars, such as Ole Waever and Jaap de Wilde, and the Copenhagen Peace Research Institute (COPRI) have proposed the concepts of ‘Securitization’ and ‘De-securitization,’ arguing that the existence of the security issue is the result of policymakers’ escalation and expansion of ‘threats’ (Chang-Ching, 2019, p. 32). Politicization that functions in a ‘Speech-Act’ paradigm is described as securitization. Declaring ‘existential threats,’ taking ‘emergency action,’ and ‘disrupting established threats’ constitute successful securitization. Three steps are required for successful securitization: ‘declaring existential threats,’ ‘performing emergency action,’ and ‘breaking the established norms to influence the relationship of internal units’ that allow ‘Threats’ to be accepted or understood through discourse (Buzan et al., 1998, pp. 23–31).

Furthermore, to understand how threat perceptions are formed and influence policy actions, this study draws on constructivism’s emphasis on identity and philosophy, which may generate varying levels of threat perceptions and hostility toward external groups when differentiating themselves from others. The emphasis on group threat perceptions in realism is based on ‘power asymmetry functions.’ International security and conflict studies identify adversaries’ or foes’ aggressive intentions using indicators. We investigated the relationship between security discourse and identity, power inequalities, and goals through experimental and observational studies of the content of each linguistic aspect of the text by merging the views above into a ‘complex analytical framework.’

Hypothesis: identity, power and intention influence threat perception

The concept of threat is often utilized in numerous disciplines or levels of description. Although it appears impartial and objective, it is not convincing when we dig further into the concept. Therefore, this study explores the process and context of its formation. Second, after the unexpected collapse of the Soviet Union, International Relations shifted from forecasting to describing the status quo. However, the rise of constructivism brought a new viewpoint to many issues in International Relations, especially security studies, after the Cold War (Da-Hua, 2002, pp. 303–323).

In particular, constructivism places a premium on interpretive relationships. It investigates the role of human consciousness in international relations, given that awareness shapes the words and acts of both human and government actors. Understanding and analyzing their relationships might reveal the motivations underlying their words and behaviors (Da-Hua, 2006, pp. 74). The primary focus of this research is the extent to which actors perceive threats as a result of ‘identity/similarity,’ ‘power difference degree,’ and ‘harmful Intention-Perception degree.’

In March 2019, Chinese warplanes crossed the centerline of the Taiwan Strait for the first time in 20 years. In May 2021, the British magazine The Economist described Taiwan as The Most Dangerous Place on Earth (The Economist, 2021), and since 2020 the PLA has increased the scale and frequency of military aircraft and warships harassing Taiwan. In addition, the Chinese Communist Party (CCP) launched military exercises around Taiwan in August 2022 in retaliation for the visit of the U.S. Speaker of the House of Representatives Nancy Pelosi to Taiwan, leading to the outbreak of the so-called ‘Fourth Taiwan Strait Crisis’ and in recent years, with the continuous increase of China’s military operations and gray-zone conflict attacks on Taiwan.

How do Taiwanese people perceive China as a looming threat? How do Taiwanese people perceive the looming threat from China, and how do they react to military threats from China? The design of this study centers on Taiwan’s current dilemma of being threatened by ‘China’ and uses four analytic categories: identity level, power perception, intention perception and threat perception. ‘Perception’ and ‘Threat- Perception’ assume that they are linear relationships for analysis; after the literature and theoretical discussions, the research hypotheses were further developed (Figure 1).

Figure 1. Conceptual diagram of the developing. Source: Author.

Many previous studies have confirmed that ‘identification’ is an essential component of group formation; the people concerned share a perception of the same external ‘others’ as threats (Matonytė & Morkevičius, 2013, p. 53), and that a positive intergroup encounter improves intergroup attitudes and behaviors (Bagci et al., 2018, p. 3); since Taiwan and China have similar historical backgrounds and cultural traditions, but now belong to different political systems (democracy and authoritarianism), do the related identity differences affect Taiwanese people’s perceived threat from China? This study hypothesizes that the difference between the two systems will affect the threat perception of Taiwanese people toward China. Therefore, Hypothesis 1 was that the higher the identification of Taiwanese people toward China, the lower the threat perception toward China and vice versa.

Second, a country’s Power Index is often measured by a variety of criteria, such as the number of military units, workforce, resources and financial situation, and is often seen as an essential tool for achieving national political and diplomatic objectives (Slantchev, 2011, p. 65), and of course, as a specific indicator of threat. After China rolled out its military reform plan in 2015, it began conducting frequent military drills around Taiwan, including live drills, such as joint beachhead exercises, joint air and sea drills and large-scale joint military exercises. The Presidential Office of the Republic of China (Taiwan) has stated that the China has never renounced the use of force against Taiwan and that it is the greatest security threat to democratic Taiwan (Lu, 2023), but whether the official threat definition of the China also affects the threat perceptions of the public in general remains to be seen, hence, this study aimed to investigate whether the ‘public’s perception of the threat from China’ is affected by other variables, with the second assumption being that the ‘power perceptions’ of the respondents are changed. The second hypothesis is that by changing the ‘power perception’ of the respondents, the ‘threat perception’ can be changed.

Finally, threat perception is central to the theories of war, deterrence, alliances and conflict resolution. The serious consideration of intention as a source of threat independent of military capabilities (Stein, 2013, p. 1) and the rapid growth of China’s national power in recent years, the modernization of its military and the rate of increase in its armaments have prompted the U.S. to file key official reports over the years on its military capabilities. It has also prompted the United States to define China in major official reports over the years as ‘the only competitor with both the intent to reshape the international order and, increasingly, the economic, diplomatic, military, and technological power of the United States.’ ‘Diplomatic, military, and technological power to advance that objective’ (Liptak, 2021) and the CCP’s aggressive intentions toward Taiwan are clearly revealed in many official documents, therefore this study assumes that the ‘enemy’s intention to attack,’ also affects the level of threat perception, while ‘identity level,’ ‘power perception,’ ‘intention perception,’ and ‘threat perception’ are the most important. Therefore, the final hypothesis of this study states that there is a correlation between ‘identity level,’ ‘power perception,’ ‘intention perception’ and ‘threat perception’ (Figure 1).

Figure 2. A research framework. Source: Author.

In summary, the hypotheses of the study are based on the effects of the developed experimental text on the subjects’ danger perceptions under the categories of ‘external structural environment’ and ‘internal factors.’ As a result, three major hypotheses are proposed: (as shown in Figure 2).

• Hypothesis 1 (H1): The numerous factors of ‘threat perception’ vary significantly across respondents from different backgrounds

• Hypothesis 1–1 (H1-1): Local respondents experience danger perception factors more strongly than international respondents.

• Hypothesis 1–2 (H1-2): Military personnel have a stronger perception of threat than other respondents.

• Hypothesis 2 (H2): Participants’ danger perceptions are influenced by their reading of the experimental text (Quantitative/Qualitative analysis)

• Hypothesis 3 (H3): There are correlations between identity level, power perception, intention perception and threat perception

• Hypothesis 3–1(H3-1): Positive correlations exist among identity level, power perception, intention perception and threat perception.

• Hypothesis 3–2(H3-2): Threat perception can be predicted by identity level, power and intention.

Research limitation and suggestion

This study focused on the sources of threats from different theories. It incorporated the analytical framework of securitization theory, integrating quantitative and implementation research methods better to understand the nature and components of ‘existential threats.’ However, owing to the researcher’s resources and time, the following limitations were encountered:

Limitation and suggestion #1: Problems of sampling method and inference

In this study, 80 master’s and doctoral students from the politics and international relations departments were invited to participate in the decision-making activities. However, this sampling method limits the scope of inference of the results. However, this sampling method may also limit the scope of the inferences. In the future, provided that there are sufficient funds for the study, it is suggested that future researchers commission a large-scale telephone or internet survey to improve the accuracy and reliability of the study and gradually improve the inferences.

Limitation and suggestion #2: the experimental environment is characterized by several interfering factors

Through a questionnaire survey and thesis analysis, this study explored the key factors that may affect threat perception and simulated how political elites may secure the threat and take corresponding foreign policy actions during the decision-making process. However, there are still many interfering factors in the actual experimental environment, which invalidates some of the research hypotheses. Based on this study, future researchers can further revise the research hypotheses, exclude other possible interfering factors, differentiate between political elites and the general public, measure sources of threat perceptions, and revise the model to make the study more convincing.

Research methods and implementation

A quasi-experimental method and non-participant observation were used to measure and explore the interrelationships between threat perception concept formation and identity, power differences and intentions to determine whether the securitization framework influenced actors to transform their perceptions of China into threat perceptions.

Experimental method

Experimentation measures the causal relationships between variables by manipulating one or more variables and controlling for the research environment. Therefore, the participants must be homogeneous to ensure accuracy and precision of the research. Therefore, this study was conducted in cooperation with the National Defense University (Taiwan). Therefore, in order to coordinate with the annual ‘Cross-National Decision-Making Simulation Training’ academic activity of the National Defense University, this study invited a total of 80 master’s degree and doctoral degree students from Taiwan and other countries who are studying in Taiwan in the fields of international relations and strategic studies to participate in the experiment using intention sampling. Eighty graduate students were randomly and proportionally divided into eight groups, using a matching method. This sampling strategy ensured that each group of participants had similar origins, backgrounds and ages, thereby eliminating the possibility of systematic bias. According to the research design, these eight groups were separated into two parts, the ‘control group’ and the ‘experimental group.’ The control group comprises Group-China, Group-International, Group 3-Republic of China (Taiwan) #3 and Group 5-Republic of China (Taiwan) #5. The experimental group consists of Group-USA, Group 1-Republic of China (Taiwan) #1, Group 2-Republic of China (Taiwan) #2 and Group 4-Republic of China (Taiwan) #4.

Furthermore, variables were changed during the experiment in a ‘Between-subject design,’ allowing different participants to receive alternative experimental arrangements to eliminate the maturation effect, instrumental effect and other external interfering variables. The questionnaire was designed using a model constructed by Rousseau, an international relations scholar, in a 2006 study and was further developed using four dimensions, including ‘identity, power perception, intention, and threat perception.’ Participants were asked to rate the content of the questionnaire on a scale of 1 (strongly disagree) to 5 (strongly agree).

To ensure the reliability and validity of the questionnaire, after the design was completed, ten people were selected by intentional sampling to implement the ‘pretest,’ and the ‘method of internal consistency reliability’ was used to implement the item analysis as the basis for the formal questionnaire development. In Table 1, only the t-values of ‘A4,’ ‘A11’ and ‘A18’ did not reach the significant level (p < 0.05) and were deleted, while all other questions reached the significant level. The ‘t-value’ of ‘A4,’ ‘A11’ and ‘A18’ did not satisfy the significance level (p < 0.05) after assessing the pretest results, so these three questions were deleted. All other questions were significant. For the design of the "post-test" questionnaire, the methodology employed mirrored that of the initial testing phase. Upon completion of a comprehensive analysis involving all participants, the t-values for each of the 20 items were found to be statistically significant (p < .05). Consequently, all items were retained in the "post-test" questionnaire, as detailed in Table 2.

Table 1. Pre-test summary of threat-perception.

Dimension	Pre-test question	t-Value	Dimension	Pre-test question	t-Value	Dimension	Pre-test question	t-Value
Identity-level	A 1 A 2 A 3 A 4 A 5 A 6 A 7 A 8 A 9 A 10 A 11 A 12	3.563** 3.413** 6.504** 1.377 2.526** 3.413** 5.003** 3.987** 4.055** 4.274** 1.012 5.606**	Power-perception	A 13 A 14 A 15 A 16 A 17	2.334** 3.633** 6.465** 5.215** 3.274**	Threat- perception	A 25 A 26 A 27 A 28 A 29 A 30	4.840** 4.352** 3.612** 2.483** 5.141** 4.083**
			Intention-perception	A 18 A 19 A 20 A 21 A 22 A 23 A 24	1.415 5.117** 4.112** 3.256** 2.445** 5.367** 2.372**

Source: Author.

This study’s questions were divided into two categories: ‘Categorization Questions’ and ‘Sequential Questions.’ The background variables of the participants are the Category Questions, and the Sequential Questions are the questions of ‘identity level, power perception, intention perception, and threat perception.’ The five-point Likert scoring method was used. Cronbach’s alpha was used to ensure the legitimacy and trustworthiness of the questions. According to Guilford, it is extremely trustworthy when > 0.7, medium credible when between 0.7 and 0.35, and acceptable when greater than 0.3 (Guilford, 1965, pp. 3–8).

As a result, a validity analysis was performed on the ten returned pretest questionnaires, and the overall value was 0.955. A closer look at the values of each dimension revealed that the identity level was 0.878, power perception was 0.838, intention perception was 0.897 and threat perception was 0.855. As shown in Table 3, the credibility of these queries is very high.

Table 2. Post-test summary of threat-perception.

Dimension	Pre-test question	t-Value	Dimension	Pre-test question	t-Value	Dimension	Pre-test question	t-Value
Identity-level	B 1 B 2 B 3 B 4 B 5 B 6 B 7 B 8	3.462** 3.232** 5.764** 2.483** 3.025** 3.336** 4.365** 5.116**	Power-perception	B 9 B 10 B 11 B 12	4.216** 4.334** 3.366** 6.015**	Threat-perception	B 15 B 16 B 17 B 18 B 19 B 20	5.652** 4.515** 2.317** 2.483** 5.261** 4.022**
			Intention-perception	B 13 B 14	3.521** 3.134**

Note: **p < 0.01; *p < 0.05.

Source: Author.

Table 3. Cronbach’s α coefficients of the pre-test for threat-perception.

Category	Dimension	Cronbach’s α
Pre-test questionnaire	Identity-level	0.878
	Power-perception	0.838
	Intention-perception	0.897
	Threat-perception	0.855
Post-test questionnaire	Identity-level	0.936
	Power-perception	0.964
	Intention-perception	0.945
	Threat-perception	0.922

Source: Author.

Second, the Kaiser–Meyer–Olkin (KMO) and Bartlett Test of Sphericity were conducted. KMO is an index for comparing the simple and partial correlation coefficients among the variables; the larger the KMO value, the more common the factors are among the variables, and the more suitable it is to carry out factor analysis. Bartlett Test of Sphericity was performed using the chi-squared test. If significant, there is a significant correlation between the variables, and factor analysis can be conducted. After analysis, the KMO of the pretest questionnaires was 0.846 which proved that the analysis was feasible, and the value of Bartlett’s Test of Sphericity was 2100.262, with significance = 0.000 = 0.01, which showed that the questionnaire data of the pretest questionnaires were suitable for factor analysis and passed validation.

Each dimension’s variance was 69.37% for ‘approval level,’ 73.25% for ‘power-perception,’ 74.69% for ‘intention- perception’ and 70.36% for ‘threat-perception’ (see Table 3). Following the validity test, the pretest and posttest questionnaires were finalized with 27 and 20 questions, respectively, to constitute the official questionnaire.

Each element was investigated using a quasi-experimental grouping method. Following the first round of questions, two major groups (four experimental and four control groups) began the simulation. The experimental groups were given one of three predesigned contextual texts. The test was performed prior to the conclusion of the simulation activity to examine the change in threat perceptions of the different groups. The experimental design is shown in Table 4. Following administration of the questionnaire, the data were coded and analyzed using SPSS statistical analysis software. The hypotheses were examined using descriptive statistics, Test of Independence, t-test, one-way ANOVA, Pearson’s correlation and ANCOVA.

Table 4. Quasi-experimental design table.

Groups	Controlled variables (Pre-test)	Dependent variables (experimental input)	Independent variables (post-test)
Four experimental groups	O1	X1	O3
Four controlled groups	O2		O4

Source: Author.

Observational method

Another important validation method used in this study was nonparticipant observation. We assigned scorers to function as ‘observers.’ Without participating in the training group, we collected data on subjects’ threat perception alterations and decision-making behaviors under various experimental scenarios. These data were examined to see if there were any changes in the dimensions of ‘identity level,’ ‘power perception,’ ‘intention perception’ and ‘threat perception’ following the experimental arrangement (reading the text), and if so, whether these changes resulted in variations in their simulated foreign policy decisions.

This scenario is predicated on the fact that countries in the Indo-Pacific region confront strategic competition from the United States and China. This competition impacted diplomacy, military affairs, the economy, and trade to varying degrees. As the United States and China entered a period of strategic competition, the United States’ stance toward China rapidly shifted, marking the most profound and far-reaching shift in the international environment since the Cold War. During the preparatory stage, experts and scholars in the domains of International Relations and military affairs were invited to examine the event’s processes, issues and design to develop the six scenarios for this study.

The reading materials were obtained from trustworthy Chinese and international media and periodicals. They were correctly revised based on news facts without changing the location of the reports’ content, implying explicit signals, such as ‘different political systems (autocracy/democracy),’ ‘power differences (military development)’ and ‘intentions (military exercises).’ The first piece was an excerpt from BBC NEWS titled ‘Different Political Systems.’ This shows how China is increasing its grip on speech and is eventually controlling people’s beliefs and behaviors through various auditing procedures. The second piece, ‘Differences Between Powers,’ was published in November 2020 and based on the United States’ ‘China’s Military Power Report.’ This indicates that China’s PLA currently boasts of the world’s largest naval fleet and has surpassed the United States in terms of land-based missiles and integrated air defense systems. The final piece, ‘Differences Between Powers,’ is taken from BBC NEWS and focuses on the ‘most dangerous period’ for the PLA to attack Taiwan from 2020 to 2030 (see Annexes 1–3).

Participants in the simulated decision-making activity were considered as the observation group. Participants in this exercise replicated the decision-making and discourse-releasing processes by role-playing as top national security decision-makers. This simulation represents the interaction and reaction between countries. The World Event Interaction Survey (WEIS) produced by McClelland and the WEIS-Goldstein Scale developed by Goldstein (1992, pp. 369–385) were used to study the semantic conflict and cooperation of decision-making groups.

Each group observed the decision-making strategy, discussion substance, decision-making process and practicality of the plans. Following the observation, they assigned the ‘Conflict/Cooperation Index’ based on the WEIS-Goldstein Scale in order to undertake a further qualitative analysis of ‘threat-perception’ versus ‘decision-making.’ The observers received basic training one week before the simulation activity to consistently analyze the observed indicators collected by the observers and collect qualitative data objectively from the simulation activity. The training emphasizes stimulating activity management, procedures, reading materials, surveys and observational records.

Statistical validation of the China threat-perception questionnaire

After reviewing the obtained questions, invalid questionnaires, such as those with missing items, were filtered out, leaving 66 valid questionnaires. The vast majority of responses (63.6%) were from men. Most respondents (92.4%) held master’s degrees. Most respondents (69.7%) had yet to gain managerial experience. None (including students) accounted for 60.6% of all occupations. Table 5 contains the pertinent information.

Table 5. Basic data of research subjects (n = 66).

Basic information	Valid sample	Percentage (%)	Basic information	Valid sample	Percentage (%)
Nationality			Management experience
Taiwan	54	81.8	None	46	69.7
Others	12	18.2	5 years+	8	12.1
Gender			6–10 years	4	6.1
Male	42	63.6	11 years+	8	12.1
Female	24	36.4	Occupation
Education			None (include student)	40	60.6
Master’s student	61	92.4	Military	13	19.7
PhD student	5	7.6	Public servant/Teacher	4	6.1
			Service	3	4.5
			Finance/insurance	1	1.5
			Others	5	7.6

Source: Author.

Analysis of the current situation of the threat-perception questionnaire

Significant differences were observed in the various components of threat perception among subjects with different backgrounds.

t-Test for each element in threat-perception: locals versus foreigners

In this study, the construct of ‘threat perception’ can be divided into three dimensions: level of identity, power perception and intention perception. A five-point scale was used in the survey, with one indicating ‘strongly disagree’ and five indicating ‘‘strongly agree.’ The mean and standard deviation are used to illustrate the general situation of the three dimensions of perceived threat: level of identification, perception of power and perception of intention. t-Tests were conducted on the three threat perception facets separately, and the results showed that the ‘power perception’ of native speakers was much higher than that of foreigners. In contrast, there is no significant difference in ‘identity level’ and ‘intention perception.’ The results are summarized in Table 6.

Table 6. Independent variable t-tests for each dimension of developing threats: Taiwanese versus foreigners (n = 66).

Dimension	Nationality	Sample	Mean	Standard deviation	t-Test
Identity-level	Taiwanese	54	1.82	0.42	−1.82
	Foreigner	12	2.06	0.39
Power-perception	Taiwanese	54	4.12	0.70	0.32*
	Foreigner	12	4.00	1.31
Harmful-intention	Taiwanese	54	3.88	0.62	1.19
	Foreigner	12	3.65	0.66

Note: *p < 0.05.

Source: Author.

Independent t-test for each component in threat-perception: Military personnel versus others

The t-test was used to analyze the differences in the three dimensions of threat perception, that is ‘identity level,’ ‘power perception’ and ‘intention perception,’ between military personnel and other respondents from different occupational groups. The study found that there is a difference in the construct of ‘power perception’ between the local and the foreign respondents who reached a significant level, the scales of domestic respondents were significantly higher than that of the foreigners. Table 7 presents the results of the study.

Table 7. Independent variables t-tests for each dimension of developing threat – Military Personnel versus others (n = 66).

Dimension	Occupation	Sample	Mean	Standard deviation	t-Test
Identity-level	Military personnel	13	1.76	0.55	−1.01
	Others	53	1.89	0.38
Power-perception	Military personnel	13	3.51	1.31	−1.97
	Others	53	4.25	0.61
Harmful-intention	Military personnel	13	4.10	0.70	1.65
	Others	53	3.78	0.60

Note: *p < 0.05; **p < 0.01.

Source: Author.

Threat-perception of subject is influenced after reading securitization reading text

The first phase of this hypothesis was to assess the effects of reading the text on threat perception using a dependent t-test. Identity, power perception and intention perception pre- and post-test scores were independent variables. The pre-test and post-test danger perception ratings were the dependent variables. Second, to rule out variations in the pre- test between the two groups, as well as influences on the post-test, a one-way ANOVA was used to compare the two groups in terms of altering dimensions. The independent variable was the group, the covariate was the pre-test score, and the dependent variable was the post-test score.

t-Test: pre-test and post-test of the experimental and controlled groups

The experimental group’s pre-test and post-test scores differed significantly (t = −3.18, df = 35, p = 0.003.01), and the pre-test scores (3.21) differed from the post-test scores (3.43). Significant variations were also identified between the pre- and post-test scores of the control group (t = −4.34, df = 29, p = 0.000.001) and the pre-test (3.34) differed from the post-test (3.59). The preceding data revealed that after the simulation, the difference between the pre- and post-test scores was significant. Table 8 presents the results of the study.

Table 8. Paired samples t-test after accepting experimental inputs (n = 66).

Groups	Tests	Mean	Standard deviation	t-Test	p
Experimental groups (n = 36)	Pre-test	3.21	0.38	−3.18	0.003**
	Post-test	3.43	0.465
Controlled groups (n = 30)	Pre-test	3.34	0.33	−4.34	0.000***
	Post-test	3.59	0.39

Note: *p < 0.05; **p < 0.01; ***p < 0.001.

Source: Author.

The identity level and intention perception pre-test and post-test scores showed significant variations. Identity: t = −2.66, df = 35, p = 0.012.05, t = −2.07, df = 35, p = 0.046.05, intention perception t = −2.07, df = 35, p = 0.046.05. There were no significant differences between the experimental group’s pre-test (identity level: 1.84 and intention perception: 3.85) and post-test (identity level: 2.08 and intention perception: 4.22). The difference in power perception ratings between the pre-test and post-test was not significant (t = −0.48, df = 35, p = 0.637 > 0.05), and the pre-test score (3.93) differed from the post-test score (3.97). The results demonstrate that variations in identity level and intention perception were substantial but that for power perception was not.

The differences in the identity level and intention perception scores between the pre-test and post-test for the control group were also significant (Identity Level: t = −3.82, df = 29, p =.001.001; intention perception: t = −2.35, df = 29, p = 0.026.05). The pre-test (identity level: 1.90 and intention perception: 3.83) and post-test (identity level: 2.19, intention perception: 4.20) scores differed. Power perception pre-test and post-test scores did not differ significantly (t = −0.81, df = 29, p = 0.423 > 0.05), and the difference between the pre-test and post-test scores (4.31) was modest. The aforementioned findings reveal that variations in identity level and intention perception were substantial but that for power perception was not. Table 9 presents the results of the study.

Table 9. Paired samples t-test after accepting experimental inputs – three dimension (n = 66).

Groups	Dimension	Mean	Standard deviation	t-Test	p
Experimental groups (n = 36)	Identity-pevel pre-test	1.84	0.43	−2.66	0.012*
	Identity-pevel post-test	2.08	0.64
	Power-perception pre-test	3.93	1.02	−0.48	0.637
	Power-perception post-test	3.97	0.88
	Intention-perception pre-test	3.85	0.69	−2.07	0.046*
	Intention-perception post-test	4.22	1.07
Controlled groups (n = 30)	Identity-level pre-test	1.90	0.42	−3.82	0.001***
	Identity-level post-test	2.19	0.54
	Power-perception pre-test	4.31	0.47	−0.81	0.423
	Power-perception post-test	4.37	0.47
	Intention-perception pre-test	3.83	0.57	−2.35	0.026*
	Intention-perception post-test	4.20	0.81

Note: *p < 0.05; **p < 0.01; ***p < 0.001.

Source: Author.

One-way ANCOVA

Descriptive statistics – experimental and control groups

The questionnaire contained 27 questions, including 11 on identity level, four on power perception and 12 on intention perception. Threat perception was graded on a 5-point scale: Serious Threat, Threat, No Opinion, No Threat and Completely No Threat, with scores ranging from 5 to 1. Table 10 shows that the experimental group’s average pre-test score was lower than that of the control group (1.84 1.90). The experimental group’s average pre-test score on power perception was lower than that of the control group (3.93 4.31). The experimental group’s average pre-test score in intention perception was higher than that of the control group’s (3.85 > 3.83). The experimental group’s total average pre-test score was lower than that of the control group (3.21 3.34). The average post-test scores of both the experimental and control groups were higher than the average pre-test scores, indicating an improvement in both groups’ perceptions across all dimensions. The questionnaire was subjected to a covariate analysis to determine whether the differences were due to the securitization text.

Table 10. Descriptive statistics analysis–pre-test and post-test of the questionnaires for each dimension versus developing threats (n = 66).

Groups	Dimension	Tests	Minimum	Maximum	Sum	Mean	Standard deviation
Experimental groups (n = 36)	Identity-level	Pre-test	1.00	2.86	66.14	1.84	0.43
		Post-test	1.00	4.00	74.45	2.08	0.64
	Power-perception	Pre-test	1.00	5.00	141.33	3.93	1.02
		Post-test	1.00	5.00	143.50	3.97	0.88
	Intention-perception	Pre-test	2.25	5.00	138.75	3.85	0.69
		Post-test	1.00	5.00	152.00	4.22	1.07
	Overall scores	Pre-test	2.45	3.81	115.41	3.21	0.38
		Post-test	2.33	4.50	123.42	3.43	0.46
Controlled groups (n = 30)	Identity-level	Pre-test	1.00	2.57	56.86	1.90	0.42
		Post-test	1.25	3.25	65.75	2.19	0.54
	Power-perception	Pre-test	3.67	5.00	129.33	4.31	0.47
		Post-test	3.50	5.00	131.00	4.37	0.47
	Intention-perception	Pre-test	2.50	5.00	114.75	3.83	0.57
		Post-test	2.00	5.00	126.00	4.20	0.81
	Overall scores	Pre-test	2.64	4.10	100.31	3.34	0.33
		Post-test	2.42	4.08	107.58	3.59	0.39

Source: Author.

Homogeneity test – internal regression analysis

Table 11 demonstrates that the Levene test of variances yields p = 0.389 > 0.05 for identity, p = 0.114 > 0.05 for power perception and p = 0.656 > 0.05 for intention perception. These figures indicate no statistically significant differences. As the Levene’s test results demonstrate the homogeneity of these dimensions, a regression analysis within the group may be used.

Table 11. Levene test: the questionnaires for each dimension versus developing threat (n = 66).

Dimension	Degree of freedom: 1	Degree of freedom: 2	F-test	Significance
Identity-level	1	64	0.751	0.389
Power-perception	1	64	2.574	0.114
Intention-perception	1	64	0.200	0.656

Source: Author.

Table 12 shows the homogeneity test findings for the regression coefficients (group pre-test) F = 0.001, p = 0.978 > 0.05 for the experimental and control groups’ pre-test and post-test on identity level. For power perception, the homogeneity test for the regression coefficients (group pre-test) was F = 0.150, p = 0.700 > 0.05. The homogeneity test for the regression coefficients of intention perception (group pre-test) was F = 0.234, p = 0.630 > 0.05.

Table 12. Intra-group homogeneity test: the questionnaires for each dimension versus developing threat (n = 66).

Dimension	Source of variance	Type-III square Sum (SS)	Degree of freedom (df)	Mean of squared sum (MS)	F-test (F)	Significance (Sig)
Identity-level	Group*Pre-test	0.000	1	0.000	0.001	0.978
	Error	15.103	62	0.244
Power-perception	Group*Pre-test	0.043	1	0.043	0.150	0.700
	Error	17.657	62	0.285
Intention-perception	Group*Pre-test	0.203	1	0.203	0.234	0.630
	Error	53.751	62	0.867
Overall	Group*Pre-test	0.026	1	0.026	0.201	0.656
	Error	8.041	62	0.130

Source: Author.

The intragroup homogeneity test (group pre-test) yielded F = 0.234, p = 0.630 > 0.05. The three aforementioned dimensions showed significant differences, indicating that there was no significant interaction between the two groups’ pre-test and post-test data. The experimental and control groups satisfied the hypothesis of intragroup homogeneity of the regression coefficients and were able to continue with the covariate analysis.

One-way ANCOVA

A homogeneity test of group regression coefficients was conducted. Results were p > 0.05, indicating that there were no significant differences. These results were consistent with intragroup regression coefficient homogeneity, implying that meeting the basic requirements of covariate analysis and executing covariate analysis were achievable. As a result, the groups were employed as independent variables and pre-test outcomes as covariates. The results are summarized in Table 13.

Table 13. ANCOVA: the questionnaires of controlled groups for each dimension versus developing threat (n = 66).

Dimension	Source of variance	Type-III square sum (SS)	Degree of freedom (df)	Mean of square sum	F-test (F)	Significance (Sig)	Influence
Identity-level	Group	0.074	1	0.074	0.311	0.579	0.005
	Error	15.103	63	0.240
Power-perception	Group	0.327	1	0.327	1.163	0.285	0.018
	Error	17.699	63	0.281
Intention-perception	Group	0.001	1	0.001	0.002	0.968	0.000
	Error	53.954	63	0.856
Overall	Group	0.063	1	0.063	0.492	0.486	0.008
	Error	8.067	63	0.128

Note: *p < 0.05.

Source: Author.

The F-value between identity level groups was 0.311 (p = 0.579 > 0.05) indicating that the difference was inconsequential, as shown in Table 13. The F-value between the power perception groups was 1.163 (p = 0.285 > 0.05), indicating that the difference was insignificant. The F-value difference between the intention and perception groups was 0.002 (p = 0.968 > 0.05), indicating that the difference was negligible. Likewise, the whole scale F-value is 0.492 (p = 0.486 > 0.05) likewise demonstrated insignificant differences. After controlling for the effects of the pre-test, no significant difference in the subjects’ post-test results was detected, despite the fact that they had read the securitization material. Table 15 shows that the effect sizes for all dimensions and the total scale were less than 0.1, indicating that the experiment had a minor effect (a small effect).

Table 14. The original and adjusted mean: the questionnaires for each dimension versus developing threat (n = 66).

Group	Dimension	Mean (the overall original post-test scores)	Standard deviation	Adjusted mean
Experimental groups (n = 36)	Identity-level	2.08	0.64	2.10
	Power-perception	3.99	0.88	4.09
	Intention-perception	4.22	1.07	4.22
	Overall scores	3.43	0.46	3.47
Controlled groups (n = 30)	Identity-level	2.19	0.54	2.17
	Power-perception	4.37	0.47	4.24
	Intention-perception	4.20	0.81	4.21
	Overall scores	3.59	0.39	3.53

Source: Author.

Table 15. Product moment correlation between identity level, power perception, intention perception and threat perception (n = 66).

Dimension	Identity-level	Power-perception	Intention-perception	Threat-perception
Identity-level	1	0.123	0.012	−0.118
Power-perception	0.123	1	−0.128	0.051
Intention-perception	0.012	−0.128	1	0.547**
Threat-perception	−0.118	0.051	0.547**	1

Note: **p < 0.001.

Source: Authors.

To eliminate differences in the pre-tests, the pre-test scores were included as covariates alongside the original average post-test scores and the modified total average scores. The experimental and control groups’ scores changed from (3.43, 3.59) to (3.47, 3.53). This demonstrates that the experimental group’s average post-test score was comparable to that of the control group’s average post-test score. The results are presented in Table 14.

Identity level, power perception, intention perception and threat perception are related

Correlation analysis between threat perception and other dimensions

This section explains the connections between threat perception, identity level, power perception and intention perception. Pearson’s product-moment correlation coefficient was used to investigate the relationships between two variables. The correlation coefficient was used to calculate the degree of correlation between variables. The assessment criteria are as follows: r < 0.40 is deemed low correlated; 0.40 ≦ r ≦ 0.70 is considered moderately correlated; and r > 0.70 is considered highly correlated.

In this study, the correlation coefficients of the four variables – identity level, power perception, intention perception and threat perception were examined individually. Table 17 shows that there is a certain degree of correlation between threat perception and intention perception (r = 0.547); second, according to the high and low values, it can be seen that the r value of 0.547 represents a ‘moderate positive correlation’ between intention perception and threat perception. The correlation coefficient between threat perception and ‘identity level’ and ‘power perception’ does not reach a significant degree. The correlation coefficients between ‘threat perception’ and ‘identity perception’ were not significant.

Table 16. Product moment correlation-power-perception versus intention-perception (n = 66).

Changing order	Multiple correlation coefficient (R)	Coefficient of determination (R^2)	Increased explanatory power (△R^2)	F-value	Standardized regression coefficients (β)	Tolerance	VIF
Power-perception	0.304	0.092	0.092	6.515	0.313	1.000	1.000
Intention-perception	0.390	0.152	0.059	4.417	0.244	0.999	1.001

Source: Author.

Table 17. Actions taken by all groups in Scenario 1.

Scenario	Controlled group	Experimental groups
⋆ Scenario 1(1330) Peking University builds Taiwan Strait Strategic Awareness Program Website providing real-time and dynamics reports to the situation of the waters surrounding Taiwan. In addition, according to the National Security Law, Chinese Ministry of Foreign Affairs announced that importing all weapons to Taiwan is Illegal and prohibited	Group-China: Taiwan is a part of China. The Taiwan Strait Strategic Situational Awareness Program Website is built to efficiently monitor the ships in China’s territorial waters based on maintaining PRC’s sovereignty and territorial integrity. Other countries have no right to interfere! Group-Internationals: Japan: This Website threatens security; Japan will establish a dedicated responding network to share information about the Taiwan Strait with neighboring countries. Philippines: The establishment of the Website is welcome. It will provide the countries requiring numerous practical and accurate oceanographic data. ROC-3: A national security conference meeting is initiated, and the General Secretary of the National Security Council analyzes the issue without taking the information security into account. ROC-5: Domestic universities and research institutions established a research project on the dynamics of the waters surrounding the Taiwan Strait.	Group-USA: The USA urges China not to become a peace-breaker in the region and is unpleased to see any unilateral changes to the status quo. ROC-1: The Website shows that China attempts to invade ROC (Taiwan). Given ROC is a sovereign and independent country accepting no intimidation, China is urged not to undermine the peace and stability in the Taiwan Strait and the region. ROC-2: ROC is a sovereign and independent country possessing Effective Exclusive Jurisdiction, and the international community has been calling to condemn this misconduct. ROC-4: The National Security Council has stated that the ROC military forces are fully aware of and constantly monitoring the movements in Taiwan Strait. In addition, free sailing in the waters around Taiwan and on the international sea lanes is protected by international law. Unilateral violation of international rules is not allowed.

Source: Authors.

Analysis of the predictive power of ‘identity level, power perception, and intention perception’ on threat perception

In this study, the mean threat perception score was summed as the dependent variable, and stepwise regression analyses were conducted to investigate the predictive power of the three dimensions of identification, power perception and intention perception. The correlation coefficients of the three dimensions, as shown in Table 15, in descending order, were ‘perception of intention’ (0.547), ‘perception of power’ (0.051) and ‘degree of identification’ (−0.118).

After putting the items into a multiple regression analysis in the above order, the results showed two predictive variables for threat perception: power perception (F = 6.515**) and intention perception (F = 4.417**), whereas identity level was insignificant. Table 16 presents the results of the study.

The content of the tested personnel’s decision-making observation analysis

In addition to the quantitative data gathered from the experimental questionnaire, this study incorporated the concept of the China threat into the designed reading text by using the observer’s notes to collect respondents’ attitudes and responses after reading the text to test the decision-making and threat perception hypothesis. The study’s hypothesis investigates the notion that the higher China’s threat perception, the more conflicted the decision making.

Except for the participants in the China team of the control group, the semantics of the relevant decision content of each group were categorized and weighted using McClelland’s ‘WEIS Classification Criteria’ and Goldstein’s ‘Goldstein Scale for WEIS.’ They were not included in the WEIS because their role-playing matched Scenarios 1–6. The following section describes the unique implementation process. Participants in each group were invited to function as decision makers in a simulated national security meeting during the actual decision presentation segment. Within each scenario, a simulated scenario was released, allowing members to discuss and propose actions for the group.

Scenario 1 analysis

This Scenario was ‘Peking University builds Taiwan Strait Strategic Awareness Program Website providing real-time and dynamic reports to the Scenario of the waters surrounding Taiwan; additionally, the Chinese Ministry of Foreign Affairs announced that importing all illegal weapons to Taiwan is prohibited under the National Security Law.’

This scenario also represents the initial phase of threats, assuming that the groups’ threat perceptions of China are equal; hence, the response is less likely to be significantly contentious. The groups replied as expected in Scenario 1, and the answer communications from the control and experimental groups tended to state the viewpoint and attitude of the group’s representative country without harsher diplomatic vocabulary Table 17).

Analysis of Scenario 2 and 3

Scenario 2: ‘An aerial image of a US ship loaded with weapons sold to Taiwan is displayed on the Taiwan Strait Strategic Awareness Program Website.’ Scenario 3: ‘The Indo-Pacific Command invites Taiwan to participate as an observer in military exercises.’ According to the Chinese media, the Military Affairs Commission holds closed-door sessions with theater commanders to improve battle preparation. Assuming that reading the text increases threat perception in the experimental group, the group will respond to the message more forcefully than the control group.

The replies of the control group were largely similar to those of the preceding condition, and the reactions involved conveying positions, explanations and appeals. In contrast, the experimental group’s reactions became stronger with wording, primarily denial and emphasis (see Table 18).

Table 18. Actions taken by all groups in Scenario 2.

Scenario	Controlled group	Experimental groups
⋆ Scenario 2 (1333) The Taiwan Strategic Situational Awareness Program Website shows an aerial image of a USA boat ‘LNGPOWERED’ loaded with weapons sold to Taiwan. CMA CGM in San Diego owns the boat.	Group-China: No foreign parties are allowed to undermine the unity of the Chinese. The USA has violated the 817 Communiqué committing to gradually decrease arms sales to Taiwan. The arms sale from the USA to Taiwan is strongly condemned, given the stability and security of the Taiwan Strait are negatively impacted. Group-Internationals: Japan is interpreting the images shared by the Taiwan Strait Strategic Awareness Program. ROC-3: The arms sales from the USA to Taiwan always follow the Taiwan Relations Act regulating that only necessary defensive weapons are provided to Taiwan. China does not need to overreact in regular arms sales. ROC-5: The information released on the Taiwan Strategic Situation Awareness Program Website is not helpful to the peaceful development between China and Taiwan. It is suggested that China does not need to be overly concerned and speculative.	Group-USA: The USA deeply regrets that China’s self-initiated messages are not to maintain the stability of the Taiwan Strait. The USA emphasizes that any attack on USA vessels will be considered an act of war. ROC-1: Because the reported claims are inconsistent with the facts, this is a general business transaction with the USA for transporting civilian goods; China does not need to be over speculating. ROC-2: Given the unidentified sources and inaccurate information on the Website, the international society is suggested not to be misled by the CCP, and excessive interpretations and speculations are unnecessary. ROC-4: The Ministry of Foreign Affairs holds a press conference to emphasize that Taiwan does not belong to PRC and China has no right to interfere in any matters related to the sovereignty of the ROC.

Source: Author.

In Scenario 3, the responses of the control group were more substantial than in the previous two circumstances, and the wording of the responses began to express military reflexes and appeals. In contrast, the experimental group’s replies maintained higher perceptions of the prior condition to make matching decisions against China’s military threats, whereas the reactions of the US group and teams 1 and 4 of the R.O.C. group displayed expressions of caution (as shown in Table 19).

Table 19. Actions taken by all groups in Scenario 3.

Scenario	Controlled group	Experimental groups
⋆ Scenario 3 (1335) The Indo-Pacific Command formally invites Taiwan to participate as an observer in the Rim of the Pacific Exercise. China’s Global Times reports: The Central Military Commission convenes closed-door meetings with theater commanders to escalate combat readiness.	Group-China: China strongly condemns Taiwan’s participation in this exercise; this action violates the One-China principle of the Shanghai Communiqué and Communiqué on the Establishment of Diplomatic Relations. The closed-door meeting to safeguard the integrity of China’s sovereignty is a necessary action against foreign threats. Group-Internationals: Japan, Philippines and Indonesia do not oppose Taiwan’s participation in the exercise as an observer and remain vigilant about China’s movements. They urge China to be calm and restrained and not to become a peace-breaker in the region. ROC-3: Appreciate USA in offering the opportunity of participating in military exchanges. Enhancing Taiwan’s defense capabilities is beneficial to the peace of the region. To cope with China’s closed-door meeting, Taiwan will monitor the status of the air and sea surrounding Taiwan to prepare for responding a crisis based on relevant information. ROC-5: We thank the USA for the invitation and believe that it will help our country learn the skills to work together with multinational forces in combat. We also urge China not to make military provocations with wrong judgments to undermine regional peace and stability. We hope to restore the mechanism of cross-strait dialogue.	Group-USA: Inviting Taiwan to participate in RIMPAC as an observer is to stabilize the peaceful development in the region and free the democratic countries from oppression. To cope with China’s closed-door meeting, we enhance military mobilization and combat readiness, increase patrols of our frigate fleets in the South China Sea, and pay attention to China’s dynamics in the region. ROC-1: We thank the USA’s invitation, Taiwan is willing to assume responsibility for maintaining peace in the Asia-Pacific region. We reaffirm the sovereignty of the ROC, urge China to remain calm and restrained to avoid escalating conflicts impacting regional security and condemn any disruption of peace in the Taiwan Strait. ROC-2: We are grateful for the support from friendly countries in the RIMPAC. We will learn from this experience and continue maintaining the region’s stability. In response to China’s closed-door meeting, maintaining regular trainings, enhancing the awareness of air fighters and combat vessels, and comprehensively collecting intelligence. ROC-4: The invitation to the RIMPAC is a manifestation of Taiwan-USA relations and a milestone in security cooperation that highlights the role of Taiwan in the USA’s Indo-Pacific strategy. Taiwan and China should continue maintaining communication channels. We hope that China will not overreact to this event and weaken the relationship between the people on both sides.

Source: Author.

Analysis of Scenario 4 and 5

Scenario 4: ‘The CCP announced armed exercises in Taiwan’s southeastern waters, and all vessels were barred from entering.’ An aerial image on the Website shows that ‘US ships arrived in Hawaii and Guam, and US and Japanese fleets appeared in Bashi Channel.’ Scenario 5: ‘By messaging Taiwan’s Ministry of National Defense, the CCP’s Eastern Theater Command invites the Chief of General Staff (CGS) to their ship next to the Pinnacle Islands’ (MND). They hope to establish communication channels to avert unintentional disputes that could lead to war. At this point, the experimental group was given the second reading text, ‘CCP’s Intention Maximum Possibility of Attacking Taiwan Before publishing Scenario 4, the experimental group read the text.’ We anticipated that reading the second text would increase the experimental group’s perception of China’s threats. The control group’s answers indicated a general warming trend and were stated sensibly. The experimental group’s replies in this scenario were more muscular, with greater caution and threats in phrasing and semantics. Team 1 of the ROC group used expanded military exercises to respond (see Table 20).

Table 20. Actions taken by all groups in Scenario 4.

Scenario	Controlled group	Experimental groups
⋆ Scenario 4 (1401) The Maritime Safety Administration of PRC announced that armed exercises are immediately launched within the triangle area (north of 21°N; east of 121°30’ E; west of 122°E) in the southeastern waters of Taiwan, and entering this area is prohibited. On the Taiwan Strait Strategic Awareness Program Website, aerial images showed that the LNGPOWERED was advancing to the middle of Hawaii and Guam. The Japanese frigate Izumo (DDH183) and the USS Ronald Reagan CVN-76 appeared together in the Bashi Channel.	Group-China: It is our natural right to exercise in our waters, and other countries are not allowed to interfere. The USA and Japan are trying to disrupt the regional order. China strongly condemns such a provocative act and immediately urges to correct the mistaken provocation. So as not to damage regional peace and stability, China will take necessary measures to defend its sovereignty and security. Group-Internationals: Japan, the Philippines, and Vietnam strongly condemned China for undermining regional security. The Philippines and Vietnam stressed and urged China to act consistently with its self-commitment to its responsibilities as a superpower. They called on the USA, China and Japan to assume their duties as superpowers and not escalate conflicts easily. ROC-3: Adopt protests through diplomatic approaches to address the peace across the Taiwan Strait is threatened. PRC should immediately stop the exercises in this region. The Executive Yuan convened an inter-ministerial meeting and requested the relevant ministries prohibit this area’s access. ROC-5: The MND uses surveillance to be aware of the current situation, commands air fighters and surface ships to monitor the area and maintains close communication with the USA and Japan.	Group-USA: We strongly condemn China’s act of sending a destroyer to the periphery of the potential conflict area for routine training. We clarify the exercise with Japan is a routine and ask China not to over-interpret. We reiterate that the Asia-Pacific region’s situation is stable and will not sit back and watch any attempts or acts that undermine regional stability! ROC-1: We condemn China’s live-fire exercise considered as military aggression. Taiwan will expand the scale of annual exercises, calling on China not to underestimate the determination of defending our territory to the end. Monitor activities in Bashi Channel and work closely with the USA to participate in the USA and Japan joint exercise. ROC-2: The military forces immediately upgrade their combat readiness. They closely monitor the situation and exchange information with the USA and Japan. They urge China to be self-restraint, maintain regional peace and stability, and avoid any measures escalating the situation. They also closely liaise with the USA and Japanese combat fleets around the Bashi Channel and deepen the military exchanges with the USA and Japan. ROC-4: Taiwan is comprehensively monitoring the dynamics of China’s maritime exercises, adopting preventive deployment, and appeal China not to take the opportunity to cause trouble. We also appeal to the region’s countries to share the responsibility of maintaining peace and stability in the Indo-Pacific region, keeping track of and monitoring the ship movements in the Bashi Channel. We dispatch naval vessels to patrol the area to deal with unexpected situations.

Source: Authors.

The control groups continued the exercise, and all three groups agreed with China’s proposal for Cross-Strait Military Dialog. The experimental groups, in contrast, finally rejected CCP’s plan. The United States group issued warnings. Groups 2 and 4 of the ROC rejected uneven discussions, whereas Group 1 took the most aggressive countermeasures by withdrawing the diaspora from China (Table 21).

Table 21. Actions taken by all groups in Scenario 5.

Scenario	Controlled group	Experimental groups
⋆ Scenario 5 (1403) The CCP’s Eastern Theater Command invites the Chief of the General Staff (CGS) to their ship next to Pinnacle Islands by messaging Taiwan’s Ministry of National Defense (MND). They aim to build a communication channel to avoid accident conflicts that may escalate to war	Group-China: Taiwan authorities should carefully consider ‘the 1992 Consensus’ and the long-lasting ‘One country, two systems’ policy. Making wrong choices will lead to irreparable mistakes. Besides exchanging with the CGS, all local governments of counties and cities are also welcome to take the initiative to discuss with the mother country to avoid hurting innocent Taiwanese. Group-Internationals: In order to avoid the expansion and escalation of conflicts, Japan, the Philippines, and Vietnam are willing to provide venues for third-party discussions. They also appealed to the USA, China and Taiwan to manage the conflicts peacefully. ROC-3: Taiwan is optimistic about inviting the CGS to the Pinnacle Islands for communication onboard the landing ship. Our country advocates responding to sovereignty disputes peacefully and rationally and maintaining regional security in East Asia by conducting multilateral talks with all parties on sovereignty, sea lanes of communication, and natural resources to support our national interests. ROC-5: Based on the Cross-Strait interaction and reciprocity principle, we will not assign the CGS to the meeting but appoint the commander of the Third Combat Theater instead. We will also invite the Chief of Liaison Section in AIT, the Japanese Military Attaché in Taiwan, to form an international delegation and invite international and domestic media reporters to witness the meeting.	Group-USA: The USA does not allow unilateral changes to the status quo. If China acts irrationally, the USA will assist Taiwan. We negotiate and appeal to China to be calm and restrained. If China withdraws, the USA will also withdraw its actions. ROC-1: In response to the tense situation on both sides of the Taiwan Strait, the Mainland Affairs Council suggested activating the evacuation mechanism to withdraw Taiwanese from China using chartered flights or subsidies. ROC-2: Taiwan is open to dialogue conducive to cross-strait stability through reciprocal channels. We regret that the invitation was from the PRC Eastern Theater Commander, and exchanging with the GCS belittles the sovereignty of Taiwan. ROC-4: Taiwan has always agreed to maintain a smooth channel of peaceful communication. It has the sincerity to reduce possible conflicts in the region. Still, all dialogues should be comparable to the Koo-Wang talks in 1993 and 1998. Both sides met on an equal footing and invited the USA and Japan to negotiate and build a multi-party meeting mechanism to reduce the possibility of regional conflicts.

Source: Author.

Analysis of Scenario 6

Scenario 6: ‘In Bashi Channel, CCP’s KJ-500 Air Early Warning (AEW), a KJ-200 Anti-submarine Aircraft, and a 052 C Destroyer (Haikou-171) with a Z-9 Anti-submarine Helicopters conduct a Cruising Mission.’ Before publishing Scenario 6, the experimental group was given a third reading text: ‘Power Threat – China’s Military Development Report 2020 demonstrates that the capabilities of the CCP’s Navy and Nuclear Weapons surpass the United States.’

Nonetheless, the control group’s answer was sensible, with China promising to stop its military cruise. In this Scenario, the experimental group maintained a strong attitude and the words and semantics of the answers were primarily forceful denunciations and criticisms. The ROC Group-1 reaction was still strong, and the accompanying action was ‘conducting an inventory check of combat resources and funding and condemning the war’ (Table 22).

Table 22. Actions taken by all groups in Scenario 6.

Scenario	Controlled group	Experimental groups
⋆ Scenario 6 (1426) A Cruising Mission is conducted by CCP’s KJ-500 Air Early Warning, a KJ-200 Anti-submarine Aircraft, and a 052C Destroyer (Haikou-171) with a Z-9 Anti-submarine Helicopters onboard in Bashi Channel. The third reading text: ‘Power Threat’ – China’s Military Development Report 2020 illustrates that the capabilities of CCP’s Navy and Nuclear Weapons surpass the United States	Group-China: The PLA’s patrol mission is necessary to defend the national unity and territorial security. Group-Internationals: Japan: We reiterated our call for China to stop unnecessary military cruises from building confidence in China among regional countries. Philippines: China’s long-standing attitude has forced the countries in the region to worry about its intentions. We will strengthen cooperation with the USA and other countries to deal with China’s threat. ROC-3: In response to the PRC’s actions in the Bashi Channel, Taiwan Air Force fighters were dispatched to track enemy aircraft. The Taiwan Navy keeps tracking the maritime movements of the enemy to ensure Taiwan’s overall security. ROC-5: We appeal to China not to be a peace-breaker and immediately stop the military interference and threats in our surroundings because the Taiwanese will not accept such an act and will not succeed.	Group-USA: We condemn China’s repeated aircraft patrols in the vicinity of the Bashi Channel to intimidate the countries in the region. We reaffirm the USA’s commitment and responsibility to protect regional stability and security and relevant allies from Chinese military threats. ROC-1: Taiwan’s government has implemented an inventory check of war resources and funding. We have formulated policies to maintain people’s living stability and asked the Taiwanese to maintain regular life without extra worries. We denounce China’s ongoing cruise missions in the region as a way to disintegrate peace. ROC-2: The MND, and related diplomatic and intelligence units continue to monitor PRC’s developments and appeal to China not to harm Cross-Strait relations and undermine regional peace and stability. ROC-4: We keep monitoring and concerning the PRC’s patrol missions using air and sea surveillance. A press release from the Mainland Affairs Council appeals that China has the right and obligation to integrate and participate in regional peace-making policies, not to be the factor of discord and destabilization.

Source: Author.

Therefore, after assigning WEIS weights, the conflict weights for both groups were clearly visible in the succeeding table. An analysis of the data disclosed significant disparities in the fluctuations of values. The conflict weights for the experimental group were considerably higher than those seen in the control group, as indicated in Table 23.

Synthesis and summaries

Following the aforementioned study, the hypotheses were validated to determine the impact of threat perceptions. Table 24 summarizes the variance and association between the variables, validation of the hypotheses and findings.

Table 23. WEIS values in different scenarios.

Groups		WEIS values in different scenarios
		Scenario 1	Scenario 2	Scenario 3	Scenario 4	Scenario 5	Scenario 6
Controlled group	Group-China (not included)	Discussion/0.0	Condemn/−3.4	Condemn/−3.4	Exercise/−7.6	Deny/−1.1	Threat/−7.0
	Group-Internationals	Action/−0.1	Comment/0.0	Urge/−0.2	Condemn/−3.4	Facilitate/−0.1	Criticize/−2.2
	ROC-3	Discussion/−0.2	Urge/−0.2	Monitor/−1.1	Objection/−2.4	Agree/3.0	Monitor/−1.1
	ROC-5	Action/−0.2	Urge/−0.2	Urge/−0.2	Monitor/−1.1	Consult/2.8	Criticize/−2.2
	Mean-Controlled Groups	−0.16	−0.13	−0.5	−2.3	1.9	−1.83
Experimental group	Group-USA	Comment/−0.1	Warning/−3.0	Exercise/−7.6	Threat/−4.4	Warning/−3.0	Condemn/−3.4
	ROC-1	Comment/−0.2	Reject/−0.9	Condemn/−3.4	Exercise/−7.6	Evacuate /−4.1	Condemn/−3.4
	ROC-2	Comment/−0.2	Reject/−0.9	Monitor/−1.1	Warning/−3.0	Reject/−0.1	Criticize/−2.2
	ROC-4	Comment/−0.2	Sovereignty /−2.4	Criticize/−2.2	Warning/−3.0	Reject/−0.1	Criticize/−2.2
	Mean-experimental groups	−0.18	−1.8	−3.6	−4.5	−1.8	−2.8

Source: Author.

Table 24. Research hypotheses results and analysis.

Research hypotheses	Results
H1: The various dimensions of ‘threat perception’ differed significantly among respondents from different backgrounds
H1-1: The dimensions of threat perception were more strongly perceived by local respondents than by foreign respondents	Partially support
H1-2: Military personnel perceive the threat perception dimensions more strongly than other respondents	Reject
H2: Participants’ reading of the experimental text will affect their threat perceptions (quantitative/qualitative analysis)	Partially support
H3: Identity level, power perception, intention perception andthreat perception are correlated	(Quantitative statistics-reject/qualitative observations-support)
H3-1: Identity level, power perception, intention perception andthreat perception are positively correlated	Partially support
H3-2: Identity level, power and intention are capable of predicting threat perception	Partially support

Source: Author.

Hypothesis 1 is grounded in the constructivism theory’s concept of ‘identity.’ It assumes that the more similar the ‘identity’ is, the lower the perceived threat will be (H1-1). Moreover, perceived threat varies among different occupational groups (military and general) (H1-2). Upon examination, a significant difference (t-test 0.32*) was identified in the three dimensions of ‘power perception,’ ‘Identity’ and ‘Intention’ between domestic and foreign respondents. This indicates that domestic respondents possess a more robust perception of China’s development of comprehensive national power and military expansion compared to foreign respondents.

Second, military personnel and other occupations were analyzed by independent sample t-tests across three dimensions of ‘Identity,’ ‘power perception’ and ‘intention.’ No significant differences were found between military personnel and the other subjects. However, the overall mean score for the ‘perceived threat from China’ questionnaire was 3.94 (with a minimum of 1 and a maximum of 5). The mean score for this item was 4.10, which remained higher than the other subjects. In comparison, the mean score of this item for military personnel was 4.10, still surpassing that of other respondents.

As for the questionnaire on ‘identification with China,’ the overall mean score of identification with China is 1.825 (with a minimum of 1 mark and a maximum of 5 marks). The mean score of this item for military personnel is 1.76, indicating a lower level than that of other occupations. This result shows that, despite military personnel being responsible for national security duties for a long time and frequently dealing with conflicts in the gray areas against China, they still exhibit a higher level of ‘threat perception’ and a lower level of ‘identity’ with China.

Summarizing the above analysis, in the test section of H1-1: The dimensions of threat perception were more strongly perceived by local respondents than by foreign respondents; among the three dimensions, ‘perception of power’ reaches a significant level, while the rest of ‘perception of power’ and ‘level of identification’ are not significant, so the hypothesis is partially supported. H1-2: Military personnel perceive the threat perception dimensions more strongly than other respondents and is classified as ‘Rejected.’

Hypothesis 2 (H2): ‘Subjects receiving securitized reading text will experience threat perception.’ This part is based on the Copenhagen School of Securitization Theory, which emphasizes the use of ‘words’ and the design of the threat text to enhance the measurement of China’s threat perception, and to observe whether the experimental group will describe the threat through words, and respond to the threat with a corresponding intensity of foreign policy actions.

In the design, news events were rewritten to include topics, such as ‘strict censorship of information in communist countries,’ ‘China’s significant military growth,’ and ‘the PLA’s continuous training in Taiwan,’ and the texts were further distributed to the experimental group to test whether subjects would affect threat perception and take higher levels of conflictive actions in the decision-making process after reading differently oriented ‘securitized’ messages.

In addition to t-test, variance analysis, and covariance analysis, a qualitative observation method was adopted. The semantics of the corresponding decision contents of the experimental/control groups were classified and weighted through observers, following McClelland’s ‘Classification Criteria for WEIS’ and Goldstein’s ‘Goldstein’s Scale for WEIS.’ The semantics were then probed, and numerical values were assigned to compare the differences between the two groups.

Regarding quantitative analysis, it was found that although the covariate analyses of ‘degree of identification’ and ‘perceived intention’ of the two groups were not significant after the release of the experimental text, a comparison of the pre-test scores with the post-test scores revealed that the post-test means of the experimental group increased in all dimensions. This indicated that the threat perception of the subjects increased in all dimensions.

Second, qualitative observation of the analysis revealed that the experimental group began to show significant differences in the values of diplomatic decision-making actions after reading the text inputs. The ‘conflicting values’ of the experimental group toward China were significantly higher than those of the control group, and the test group generally took more assertive action in the situation. The wording and semantics of the responses had a stronger warning and threatening connotation, suggesting that the higher the perceived threat level of China, the more conflicted the decision making. This finding is consistent with the results of the literature review on the realism, indicating that ‘power differentials affect threat perceptions’ (see Figure 3 above).

Figure 3. Average WEIS values in different Scenarios. Source: Author.

Therefore, in Hypothesis H2, participants’ reading of the experimental text was expected to affect their threat perceptions (quantitative/qualitative analysis). The quantitative statistics did not hold, but the qualitative analysis did; thus, partially supporting the hypotheses of this study.

Hypothesis 3 uses the ‘threat perception’ paradigm to examine the effects of identity, power and intention on threat perception. Intent and fear of risk were positively linked, and threats increased with intent. Power- and motivation-predicted risks, while group danger increased with power differences and malice. This test may help to identify correlations, and this is partially supported by the evidence. With the growth of China’s comprehensive national strength, China has changed its foreign policy from ‘keep a low profile and concentrate on self-improvement (韜光養晦)’ to ‘Wolf warrior diplomacy.’ In recent years, China’s military expenditure has continued to rise, with a total amount reaching 17 times that of Taiwan’s defense budget. Internationally, in addition to intimidating Taiwan by force, the CCP has continued to expand the ‘grey zone conflict’ toward Taiwan, such as information attacks, ‘paramilitary’ actions, and military aircraft and warships harassing Taiwan. Regarding infiltration into Taiwan, China uses funds, political orders, red media, unification gangsters and political agents to oppress Taiwan’s free and democratic politics in an all-around way. Some groups in Taiwan have been wooed. These ‘expansion and intimidation of military power’ and ‘intended infringement of the territory of other countries’ have significantly increased the Taiwanese’s perception of the China threat.

Hypothesis 3: Identity Level, Power Perception, Intention Perception and Threat Perception are correlated. The correlation between the variables is examined based on the threat perception model, and the most significant impacts are identified. The predictive effects of these three variables on threat perception were also investigated.

After using the cumulative correlation and multiple stepwise regression analysis, it is found that the correlation between ‘intention perception’ and threat perception is the most significant part, which is moderately positive, that is, the greater the intention perception of others, the more pronounced the threat perception is;

Second, in terms of predictive power, it is revealed that the two components of ‘power perception’ and ‘intention perception’ have predictive power on threat. In other words, when the power difference is more significant, and the intention to harm is more robust, the threat perception of the group will increase accordingly.

Since the study found no association between ‘identity’ and threat, this part differs from Rousseau’s (2007) finding that ‘shared identity reduces threat perception.’ The concept of ‘identity’ varies with different life experiences and political-social environments. In Taiwan, due to its pluralistic democratic values, generational replacement and the ‘threat from China,’ Taiwanese people’s identity with China has declined dramatically (Wang et al., 2023, p. 8). However, ‘identity’ is not significantly related to threat in this study because of factors, such as the methodology and the small sample size. The respondents’ mean score on this dimension was 1.94 (with one being the lowest and five the highest), indicating that most respondents still perceive similarity with China as very low.

Therefore, the hypothesis test results of H3: Identity level, power perception, intention perception and threat perception are correlated; H3-1 is ‘partially supported’; and H3-2 is also ‘partially supported.’

Conclusion

In security studies, threat perception is a fundamental concept, because it pertains to people’s cognition and evaluation of the security environment, influencing their security behaviors and decisions. Research on measuring threat perception holds both essential theoretical and practical value for improving people’s security awareness and capabilities, thereby improving the overall security level of society. Constructivism, which is the primary approach to security research in international relations, focuses on the interactive construction of concepts, identities, agents and structures. Identity as an overarching concept, integrates these three elements and stands as the core of constructivism. One of the concepts it provides is a non-material explanation of the source of the sense of threat − that is, where the threat originates – and it targets various types of threats, such as ‘cognitive warfare’ and ‘grey zone conflicts,’ which are challenging to perceive and evaluate accurately. This approach provides a method for measuring threat perception.

Therefore, it is necessary to explore deeply the impact of relevant variables on threat perception, and identify which variables are key to effective intervention in securitization during the process of threat perception. This study uses a cross-domain integrated research framework that incorporates the Copenhagen School Securitization Theory. Additionally, a composite analysis framework was adopted to capture perspectives from both realism and constructivism on different potential sources of threats. The concept of securitization was then employed to establish a more comprehensive analysis framework.

In terms of research design, we utilized designed materials to conduct group experimental observations, and the ‘WEIS-Conflict/Cooperation Index’ was used to measure the discourse usage, wording intensity and policy choices of security actors (different decision-makers in each group). Differences enable the analysis of the source of threats to move away from a purely materialistic orientation (realism believes that the source of group threat is based on ‘power asymmetry’). Hypothesis 1 is mainly aimed at verifying the primary variables of the population, while Hypothesis 2 and Hypothesis 3 conduct experimental methods and statistical regression analysis one by one to understand the impact of variables, such as ‘identity level,’ ‘power perception’ and ‘intention perception’ on ‘threat’ ‘Perception,’ the study found that ‘power perception and intention perception’ can indeed predict ‘threat perception.’

The main contribution of this study is to connect the realist concept of threat with the concept of ‘speech-action’ in the Copenhagen School’s securitization theory, emphasizing how to strengthen the connection between ‘language and threat perception.’ It uses the experimental method, designing the experimental materials to manipulate verbal expressions to observe whether there is a change in the audiences’ threat perception of China. Simultaneously, it combines linguistic and psychological methods to analyze the possible sources of threat perception and the role of speech in threat perception: specific impact mechanisms and the relationship between externally expressed responses and policy actions.

With the growth of China’s comprehensive national strength, it has changed its foreign policy from ‘keep a low profile and concentrate on self-improvement (韜光養晦)’ to ‘Wolf warrior diplomacy.’ In recent years, China’s military expenditure has continued to rise, reaching 17 times that of Taiwan’s defense budget. Internationally, in addition to intimidating Taiwan by force, the CCP has continued to expand the ‘grey zone conflict’ toward Taiwan through information attacks, ‘paramilitary’ actions and military aircraft and warships harassing Taiwan.

In the future, relevant government departments can refer to the results of this study and further strengthen awareness of enemy threats within the group by setting text topics and establishing security argument logic. The relevant suggestions are as follows.

1. Establishment of security argument logic: Faced with threats from China’s frequent military exercises, gray zone conflicts, and united front infiltration against Taiwan, the government should formulate a clear security argument logic to strengthen the ‘power difference’ and China’s intention to invade. It poses a substantial threat to Taiwan’s security.

2. Strengthen threat awareness and national identity within the group: Taiwan should actively promote its national identity in education, culture, communication and public psychology and increase media literacy education so that the public can empower the capacity to interpret false information – the ability to enhance psychological defense and security resilience in all dimensions.

‘Questions about security cannot be separated from the most fundamental questions of political theory.’ Understanding the source of threats can not only reduce the fear of the unknown but also allow us to examine our own weaknesses and shortcomings, strengthen the national psychological defense, and ensure national security.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Additional information

Notes on contributors

Chang-Ching Tu

Chang-Ching Tu is an associate professor at the Graduate Institute of Strategic Studies, College of International Security and National Defense Affairs, ROC National Defense University (Taiwan). He received his Ph.D. from the National Defense University (Taiwan) and has been teaching the International Conflict and Crisis Management course since 2017.

Han-Ping Tien

Han-Ping Tien was graduated from ROC National Defense University(Taiwan), where he worked as a research assistant at the Graduate Institute of Strategic Studies. He currently serves in the Army as a full Colonel.

Ji-Jen Hwang

Ji-Jen Hwang is an Associate Professor at Rabdan Academy in U.A.E. A native of Taiwan, he holds a Ph.D. in politics from the University of Newcastle upon Tyne in the U.K., as well as a Masters in Library Science & Information Studies from the University of North Carolina in U.S.A.

Annexes 1

Annexes 2

Annexes 3