The nexus of scam and fake news: An exploratory study of cases of fake news-scams in Singapore

Abstract

The increasing prevalence of fake news as well as scams has been a cause for concern for authorities in Singapore in recent years. In particular, the concurrent rise in fake news and online scams has shown signs of convergence in the form of fake news-scams, where false information is manipulated and employed in scams. However, literature examining this emerging phenomenon is scarce. As such, this study aims to deepen our understanding of this novel issue by exploring how scammers employ fake news in terms of type, channels of transmission, and persuasion techniques. Drawing from local case studies (n = 90) collected from 2016 to 2021, it was found that authority appeals were the most common theme employed in fake news-scams, and persuasion cues of authority were the most prevalent. In addition, Facebook was identified as the most common channel of transmission in the spread of fake news-scams. These findings serve to inform anti-scam prevention measures by local authorities and suggestions are made as to how various stakeholders (e.g. authorities, companies, and individuals) can be involved in dealing with fake news-scams.

Keywords

Fake news fraud misinformation scam Singapore

Introduction

Scams or deceptive schemes aimed at obtaining money or personal information from individuals, have increasingly plagued our digitised world. Since 2020, online scams have been particularly rampant in Singapore, with statistics revealing that 6 in 10 people have encountered a scam (Ang, 2021). In 2022, scams constituted over half of crime reports in Singapore, marking a concerning trend (Choo, 2022). The situation has only intensified, as evidenced by the first half of 2023, where scam cases continued to abound. The top five types of scams during this period were identified as job scams, e-commerce scams, fake friend call scams, phishing scams, and investment scams (Tang, 2023). This has become an issue of concern as beyond financial losses, victims undergo psychological and emotional stress, including stress and relationship issues arising from these scams (Button et al., 2014).

The COVID-19 pandemic has further exacerbated the issue, with scams on the rise due to increased social media usage (Ma and McKinnon, 2021; Rao, 2020), leading to a noticeable shift from physical to online crimes locally as reported by the Singapore Police Force (2022). Concurrently, the digital landscape has seen a rise in the propagation of fake news – fabricated information created to deceive. This has brought to the forefront the intersection of two phenomena – fake news and scams. Although traditionally associated with political manipulation, fake news articles and deepfake videos were increasingly being used to promote various types of financial scams, costing victims significant losses. For instance, in Australia, scammers created fake news articles and deepfake videos featuring celebrities endorsing fraudulent online investment trading platforms (Lyons, 2024). In Singapore, fake news articles have also falsely claimed that politicians in Singapore endorsed specific cryptocurrency trading platforms (CNA, 2022), subsequently soliciting personal information from unsuspecting individuals. This intersection underscores the complex challenges posed by the evolving landscape of digital scams in society.

Scams

Within scam literature, there is extensive research on the various factors associated with scams. Past studies on scams have focused on the victims’ perspective (e.g. Kerley and Copes, 2002; Norris et al., 2019). Other studies have explored the financial motivations from the scammers’ perspective (Burrell, 2008), which is unsurprising given that scams are characterised by the intent to deceive victims into an exchange of money, personal information, or goods and services (Chen et al., 2020). Another factor is social media – the rise of fake news and the increase in scams can be attributed to the widespread use of social media, which has significantly expanded their reach and influence. This could be attributed to the ease with which scammers can create hoax accounts while remaining anonymous (US Securities and Exchange Commission, 2022). According to Lu et al. (2020), scams leverage trust and susceptibility to lower the guards of victims, which can be done either by (1) appearing authentic or (2) using emotional appeals. Authenticity can be feigned by including logos of credible, renowned organisations, while emotional appeals target an individual innate motivation to fulfil short-term emotional needs like acceptance (Lu et al., 2020). Thus, for this study, scam-related factors such as the combination of techniques employed by scammers and their channels of transmission are relevant variables of interest.

Fake news

Both fake news and scams share similar psychological tactics, intentions, technological pathways, and persuasion outcomes. For instance, there are several definitions for fake news in the current literature looking at the intentions behind fake news. Tandoc et al. (2018) posit that fake news can be distinguished based on two dimensions: levels of facticity and deception. Furthermore, monetary gain have been consistently identified in the literature (Chen et al., 2020; Kim et al., 2021; Ong and Cabañes, 2018).

Even distribution methods can be similar to scams, as Soon and Goh (2017) argued that social media not only facilitates the propagation of false information across a wider audience but also shifts news consumption to online and social media platforms, where information is usually more unmediated and unverified (Ceron, 2015).

The psychological mechanisms for being persuaded on the authenticity of the content is similar. On an individual level, the heuristic cues (e.g. visual appeal of websites, number of likes) were found to increase susceptibility to believing fake news, especially salient on social media where popularity ratings are made visible to users (see Nedelcu and Balaban, 2021; Soon and Goh, 2017; Tandoc et al., 2018). Thus the spillover of fake news from political advertising into direct scams is an unsurprising trajectory given the similar platforms, heuristics and news consumption both phenomena share.

Fake news-scams

Therefore, this problematic convergence of scams and fake news represents a complex challenge in the digital domain, giving rise to a dual category: ‘fake news-scams’. These are schemes that amalgamate the deceptive elements of fake news with the financial motives of traditional scams, exploiting the rapid dissemination capabilities of social media to achieve fraudulent ends. This is consistent with the literature on how fake news also imitates the features and formats of authentic news (Tandoc, 2019).

Theories on source credibility and persuasion explain why deceptive tactics often involve the imitation of credible sources to enhance the persuasiveness of the message (Fischer et al., 2013; Rubin, 2019). For instance, there are instances such as fake cash giveaways, where scammers circulate fabricated news of events promising cash rewards to individuals who comply with their requests. An illustration is the ‘Covid-19 Compensation Lottery Prize’ scam in London, wherein scammers disseminated fake news about a World Health Organization-hosted event during the pandemic, prompting victims to provide personal information in anticipation of receiving cash prizes (World Health Organization, 2021). Another prevalent type involves fake endorsements, wherein scammers disseminate false news of celebrities endorsing or promoting fraudulent investment opportunities. For instance, a case involved the fabrication of news suggesting Australian entrepreneur Dick Smith’s endorsement of a ‘get rich quick’ cryptocurrency trading service. Victims, enticed by the promise of monthly earnings, were duped into providing credit card information (Fraudwatch, 2021).

These examples underscore the breadth of deceptive practices encapsulated by the term ‘fake news-scams’. This exploration seeks to shed light on the patterns and mechanisms underlying typical cases of fake news-scams, particularly given their recent emergence and the notable challenge highlighted by a recent study indicating that around 67.5% of Singaporeans struggle to distinguish between fake and real news (Soon and Goh, 2021). One of the reasons, as Soon and Goh (2021) argued, was that participants tended to fall for the manipulated news article because they relied on source cues (e.g. the inclusion of a logo of a Singapore mass media company) to ascertain the veracity of the information. Consequently, this could be seen in local varieties of fake news-scams where fake online articles would claim that local politicians in Singapore, such as Senior Minister Tharman Shanmugaratnam, endorsed certain cryptocurrency auto-trading programmes. These articles would redirect users to websites offering investments on cryptocurrency trading that subsequently solicit their personal information (CNA, 2022).

In Singapore, there exist several measures to counter either scams or fake news. For instance, legal measures to address fake news and scams encompass the Protection from Online Falsehoods and Manipulation Act (POFMA) and the recent Online Criminal Harms Act (OCHA) passed in 2023 (Devaraj, 2023). During the COVID-19 pandemic, Singapore’s authorities employed POFMA correction orders to address misinformation, with approximately 60% of documented POFMA cases as of March 2022 being related to COVID-19. These correction orders were targeted at individuals and organisations spreading false information, specifically concerning the effectiveness of vaccines against the virus.

Scam-fighting efforts involved the formation of the Anti-Scam Command in March 2022 to heighten and synergise efforts to investigate, intervene, enforce, and gather intelligence on scams alongside partner organisations like banks (Tan, 2023a). Likewise, there have been public education efforts to improve media and digital literacy such as those by Meta, the Media Literacy Council and the National Crime Prevention Council to help people avoid falling prey to scams (Seow et al., 2022).

In delving into the intricacies of fake news-scams within the context of Singapore, the findings of this study could be a pivotal resource to better inform public education initiatives and policy development.

Types of fake news-scams

Although past studies have attempted to classify scams and fake news into their respective categories, work in untangling the relationships between fake news-scams is currently an understudied area. Traditionally, the typology of scams (e.g. phishing, loan scams, romance scams) is done based on their various modus operandi (e.g. Singapore Police Force, 2022), while fake news typology (e.g. satire, clickbait, to fabricated content) is classified based on content of varying levels of deception (Wardle and Derakhshan, 2017). Others argued that the intersection between crime and fake news can be divided into either (1) fake news-centric crime, (2) fake news which harms public order, or (3) fake news-assisted crime (Chen et al., 2020). Relevant to fake news-scams is the fake news-assisted crime category, which can be further classified as either get-rich-quick scams, phishing scams, or impersonation scams (Chen et al., 2020).

Therefore, the creation of a typology for fake news-scams holds significant merit, contributing to a deeper comprehension of this relatively unexplored intersection of phenomena. Some suggest that the continuous prevalence of both fake news and scams heightens online users’ vulnerability to emerging cyber deceits (Sarno and Black, 2023). Moreover, establishing such a typology could potentially benefit public education efforts by enhancing awareness and preparedness against evolving forms of deception and possibly reducing victimisation.

Channels of transmission

Similarly, while the channels of transmission for both fake news and scams are well-studied in their various domains, the channels of transmission for fake news-scams are not as well understood. In the case of scam research, it is well understood that different types of scams (e.g. crypto-scams, mobile gaming scams, love scams) thrive on different platforms due to their affordances like coordinating on social media platforms for ‘pump and dump’ schemes (Mirtaheri et al., 2021) or even tricking users to access scam websites through social media links (Phillips and Wilder, 2020). For fake news, there has been considerable attention on the role of social media in facilitating the spread of false information (e.g. Allcott and Gentzkow, 2017; Barfar, 2019; Ermakova et al., 2020). To the authors’ knowledge, only Chen et al. (2020) have identified Facebook and local citizen journalism sites as the main channels for circulating fake news-scams in Singapore. Thus, this study builds on this by examining the typical channels of transmission for fake news-scams.

Persuasion techniques

Cialdini’s (2016) six persuasion techniques, which include reciprocity, social proof, liking, consistency, authority, and scarcity (see Table 1) have been used in both fake news and scam research. For instance, phishing scams were found to mainly use consistency and scarcity (Zielinska et al., 2016). Some have even operationalised Cialdini’s persuasion techniques in computation models for the purpose of detecting the presence of such techniques in phishing emails (Chatterjee and Basu, 2021). Within fake news research, there are some efforts to incorporate Cialdini’s persuasion techniques. Some argued that false content from politicians like Donald Trump have persuasive appeal due to social proof from a large number of his supporters (Balnaves, 2020). Others found that the incorporation of authority cues (e.g. impersonating a credible news outlet and the format of reporting) in fake news messages can increase the credibility and appeal of the message (e.g. Nedelcu and Balaban, 2021). In addition, it was also found that reciprocity or a ‘give-and-take relationship’ was not as effective as a persuasive appeal (Chen et al., 2021). Hence, our study thus utilises Cialdini’s (2016) persuasion framework to examine whether the six persuasion techniques differ across the types of fake news-scams and their channels of transmission. Although Cialdini (2016) has included a seventh technique – unity, the existing six have been more rigorously tested in the fake news and scams settings, which informed the study’s design.

Table 1.

Overview of Cialdini’s persuasion techniques.

Technique	Definition	Example
1. Reciprocity	Providing an initial benefit in the hopes that individuals will return the favour by complying with a request due to feeling indebted	Giving victims a discount first, before asking for their personal information
2. Social proof	Using proof of norms or peer support to influence behaviours	Claiming that other users have signed up for a fake cryptocurrency programme
3. Liking	Appealing to individuals by tapping on or increasing the likeability of the subject	Impersonating a famous celebrity
4. Consistency	Using individuals’ desire to be consistent with past behaviours to influence behaviours	Having victims sign up for a fake membership first, before asking for their personal information
5. Authority	Using details and cues to signal expertise, credibility and/or position of power to influence behaviours	Impersonating government authorities
6. Scarcity	Limiting the supply of desired benefits	Claiming that victims were the only ones who won a fake cash giveaway

Taken together, the following research questions are

RQ1. How do persuasion techniques differ with the types of fake news-scams in Singapore?

RQ2. How do persuasion techniques differ with the types of channels of transmission for fake news-scams in Singapore?

Methods

Data collection

Newspaper articles and posts in English involving fake news in scams in Singapore were identified from 1 January 2016 to 31 May 2021 through Google News Search and scam-related Facebook groups. English-language content was used as Singapore, despite being a multilingual society, uses it commonly in various domains, including online communication and media reporting. Specific keywords (‘circulate’ OR ‘circulating’ OR ‘fake’ OR ‘false’ OR ‘news’ OR ‘phish’) AND (‘Singapore’) AND (‘scam’) were used in the search process on Google News. The start of the data collection was fixed in 2016 as the history of Facebook groups discussing scams showed that netizens on these groups were more active around 2016. Regarding the data sources, the decision to use Google News and scam-related Facebook groups was based on accessibility in the public domain and relevance to scams in Singapore.

A total of 637 articles/posts were aggregated into cases for further analysis. In the event that there are multiple articles reporting about the same incident, these separate articles are analysed as a single Fake News-Scam case.

Next, the cases selected for analysis were selected if they fulfil all of the following criteria: (1) involves a scam, (2) involves the use of fake news, (3) Singapore-related (see Table 2). In total, 90 cases of Fake News-Scams were identified through this process.

Table 2.

Inclusion and exclusion criteria.

Criteria	Description
Scam	A scam is coded when the case involves the intent to deceive an individual into an exchange of at least one of the categories below: • Money (e.g. cash, e-payment) • Personal information (e.g. emails, home addresses, preferences) • Goods and services (e.g. promised delivery of goods which did not materialise)
Fake News	The scam must involve the use of fake news, which is defined as • Inaccurate, misleading or fabricated content about current events • Content that is distributed through communication channels such as print, text messages, social media or websites (X. Chen et al., 2020; X. Chen and Goh, 2019) Fake news must also involve the mass circulation of inaccurate information, which typically employs media-related methods (e.g. advertisements and social media) and/or word-of-mouth (through messaging applications and social media) to reach a wider audience. Mass circulation of inaccurate information can occur through: • Receiving and/or sharing inaccurate information from one’s social circle through messaging applications and/or social media platforms • Receiving inaccurate information from an advertisement or post on social media platforms
Singapore-related	For a case to be classified as Singapore-related • It involves current events in Singapore, people in Singapore and/or external/foreign incidents where Singapore is implicated • It is disseminated among members of Singapore’s community • Members of Singapore’s community are targeted in these cases
Exclusion criteria	An incident will be excluded from review if the following criteria apply to it: • Supporting evidence about the fake news employed in the incident is found to be removed from the source post or article • Email phishing scams: ▪ For example, emails containing links to websites that request for personal information under the pretext of registering for the COVID-19 vaccination have been sent to Singaporeans. Although this is a phishing scam, it lacks the element of fake news, as it does not fulfil the criteria of attaining mass circulation. • Bank SMS phishing cases: ▪ For example, SMS that appears to be from a bank is sent to individuals, and could contain a suspicious link that requests for personal information or a one-time password (OTP). This scam case lacks the element of fake news

Data coding and analysis

A content analysis approach was employed to analyse the cases according to three types: type of Fake News-Scam, persuasion techniques used, and channels of transmission used.

Four independent coders coded the 90 fake news-scam cases using a codebook that was developed to facilitate the coding of the data along the domains of the type of Fake News-Scam and persuasion techniques used in that scam. The persuasion techniques used for analysis are as follows: Authority, Liking, Social proof, Consistency, Scarcity, and Reciprocity (Cialdini, 2016). Krippendorff’s alpha (α) was used to assess the reliability of the analysis and the values were acceptable (i.e. all values of α were >.80, with the exception of 1 variable where α = 0.758).

The 90 case studies were grouped together based on similarities in the type of fake news using open, axial, and selective coding (Strauss, 1987), resulting in a total of eight types of fake news-scams as seen in Table 3 below.

Table 3.

Overview of types of fake news-scams.

Types of fake news-scams	Description
Fake promotions	Fake promotional events from established brands
Fake endorsement on news article	Fake news article with a politician/celebrity being impersonated and quoted extolling fake investment platforms
Official impersonation	Impersonating a government official
Fake cash giveaway	Fake events where cash is given out to individuals who comply with scammers’ requests
Fake job opportunity	Fake job opportunities being advertised
Fake investment opportunity	Fake investment opportunities being advertised
Fake tax refund	Fake news on tax refund by the government
Donation scam	Appeal for donation with a fabricated backstory

Association rule mining

For this study, association rule mining (ARM) was used to discover relationships between characteristics of these scams (i.e. type, channel of transmission used, and persuasion tactics used) so as to understand the relationships within such fake news-scams. ARM is used in a variety of contexts, such as mining the rules between medical symptoms against age, sex, chronic condition, and mortality status (Tandan et al., 2021) and identification of phishing sites (Tripathi et al., 2017), and even for crime pattern analysis (Roy et al., 2021).

ARM was originally developed for the business setting where it is used to identify all the rules that would predict an item’s occurrence based on the occurrence of other items (Agrawal et al., 1993; Hipp et al., 2000). For example, in a supermarket context, an association rule relating flour to eggs with a confidence value of 0.7 suggests that a consumer is 70% likely to buy eggs if they buy flour.

The measures of effectiveness of the given rule are as follows:

Support indicates how frequently the if/then relationship appears in the database. For example, given two items X and Y in a database with n records, where X is the precedent and Y is the antecedent:

Support = P(X ∩Y) = number of times X and Y occur together ÷ number of records in a database

Confidence indicates the number of times these relationships have been found to be true, where the percentage of all records satisfying X that also satisfy Y:

Confidence = Conf(X ∙Y) = P(Y|X) =number of times X and Y occur together ÷ number of times X occurs

Lift is the confidence of the rule divided by the expected confidence, assuming that X and Y are independent of each other. Value of lift determines the relationship between X and Y; independent (=1), positive related (>1), negative related (<1). Greater lift values indicate stronger association and as such, it functions as a measure of the importance of a given rule:

Lift =Lift(X ⇒Y) = Support (X ∩Y) ÷( Support (X) × Support (Y) )

This study aimed to explore associations among types of fake news-scams, their transmission channels, and the persuasion techniques used. The data were analysed using the Apriori algorithm from the MLxtend library in Python. Adapting the targeted association rules approach from Rong et al. (2012), the study focused on associations where transmission channels and persuasion techniques precede the scams. Only rules with high support and confidence were considered, and each target rule was manually reviewed before the final analysis. This method effectively groups and identifies patterns among fake news-scam types, persuasion techniques, and transmission channels.

Results

Persuasion techniques

Table 4 shows the frequency of fake news-scam cases by persuasion techniques used. Overall, the most common persuasion techniques used in fake news-scams were authority (92.2%), liking (73.3%), and social proof (31.1%). Examples of fake news-scams that use authority include impersonating a local hospital to offer fake employment opportunities (Qi, 2021). Fake news-scams using liking include fake cash giveaways hosted by renowned local celebrities, which claim to offer cash in return for guessing a number from a video (Kua, 2021). Contrarily, fake news-scams involving social proof include visible comments of users who claimed to receive free fast food meal boxes under a fake promotion post (Jamal, 2020), which creates an illusion of others successfully receiving the ‘benefits’ of these scams.

Table 4.

Overview of persuasion techniques in fake news-scams.

Persuasion technique	Frequency
Authority	83 (92.2%)
Liking	66 (73.3%)
Social proof	28 (31.1%)
Scarcity	12 (13.3%)
Consistency	4 (4.44%)
Reciprocity	1 (1.11%)

The least common techniques were scarcity (13.3%), consistency (4.44%), and reciprocity (1.11%). Fake news-scams containing scarcity cues include faked sales of dried meat (bak kwa) from a Malaysian brand, intended to dupe victims into paying for non-existent bak kwa, as the brand had already sold out its stock of bak kwa in Singapore at that time (Ismail, 2021). Cases with consistency included a fake promotion case where the victim bought a few crabs that were sold at an extremely low price, since he had previously purchased from the brand successfully (Cheung, 2021). However, the case involving reciprocity involved an e-commerce scam where scammers provided refunds to victims only for the first few transactions, before claiming to face issues with refunds (Cheng, 2021).

Types of fake news-scams

Fake news-scams can be classified into the following three overarching types based on their appeals: (1) authority appeals, (2) financial appeals, and (3) empathy appeals. Table 5 contains the definition and frequency of fake news-scam cases according to each category. Authority appeals pertain more to being influenced by certain renowned figures, financial appeals tend to draw victims in with money-making opportunities, while empathy appeals target victims by appealing more to their emotions.

Table 5.

Overview of types of fake news-scams.

Categories	Types	Frequency
Authority appeal	1. Fake promotions	50 (55.6%)
	2. Fake endorsement on news article	23 (25.6%)
	3. Official impersonation	2 (2.22%)
Financial appeal	4. Fake cash giveaway	6 (6.67%)
	5. Fake job opportunity	5 (5.56%)
	6. Fake investment opportunity	2 (2.22%)
	7. Fake tax refund	1 (1.11%)
Empathy appeal	8. Donation scam	1 (1.11%)

To address RQ1, ARM was used to examine how characteristics of different types of fake news-scam cases frequently occur together. As the threshold for confidence was fixed at .70 (i.e. consequent follows antecedent 70% of the time), only rules for fake endorsement on news articles and fake promotions were obtained. Threshold for support was fixed at .05 (i.e. antecedents and consequent happen at least 5% of the time in the dataset), to ensure that rules obtained were prominent enough in the dataset. Obtaining rules only for these two types of fake news-scams may be attributed to both types comprising the most cases within our dataset (see Table 5). As a result, rules for the other types were likely to have a lower confidence score due to the consequent, or type of fake news-scam, being less frequent or common.

Fake endorsement on news article

A fake news-scam case was likely to be a fake endorsement on a news article, if it (1) used persuasion techniques of liking, social proof and authority, (2) occurred on a website and (3) did not use persuasion techniques of reciprocity, consistency or scarcity. A confidence score of 1.0 suggested that cases classified as fake endorsements on a news article always followed after the aforementioned characteristics in the dataset, suggesting a relatively strong association. Support was found to be 0.167, suggesting that the antecedents and consequent (i.e. fake endorsement on news article) made up a total of 16.7% items in the dataset.

Examples of fake endorsements on a news article which involve liking include fake investment opportunities purportedly endorsed by local celebrities (Kwok, 2019), while those with social proof typically include comments of bogus users claiming to have reaped the benefits of fake investments in cryptocurrency (Yang, 2019). Examples of fake endorsements involving authority include a fake investment platform using Singapore politicians like Prime Minister Lee Hsien Loong’s name and quotes suggesting the profitability of cryptocurrency (Lee, 2018).

Fake promotions

Two rules were obtained for fake promotions. First, we found that a fake news-scam case was likely to be a fake promotion if it (1) used liking and authority; (2) occurred on Facebook and (3) did not use reciprocity, consistency, scarcity and social proof. We also found that a case was likely a fake promotion if it (1) used liking and authority; (2) occurred on WhatsApp and (3) did not use reciprocity, consistency and scarcity. Both rules generated a high-confidence score of 1.0 while support was lower for both, where the first rule had a support value of .244 while the second rule had support of .089.

Fake promotion cases with liking cues include a fake promotion by an airline claiming to giveaway free plane tickets (Tee, 2019), thus leveraging on the brand’s reputability as a national carrier. Examples of fake promotions with authority cues include fake advertisements on mobile phone promotions by local banks, which solicit victims’ bank details (Ng, 2020).

Channels of transmission

Table 6 shows the breakdown of the channels of transmission used in the fake news-scam cases in the sample. Notably, a considerable proportion of fake news-scams occurred on Facebook (53.3%), WhatsApp (22.2%) and on websites (17.7%).

Table 6.

Overview of channels of transmissions for fake news-scams.

Channel of transmission	Frequency
Facebook	48 (53.3%)
WhatsApp	20 (22.2%)
Website	16 (17.7%)
Instagram	7 (7.78%)
Social media (not specified)	5 (5.56%)
Carousell	4 (4.44%)
Messaging apps (not specified)	3 (3.33%)
Unknown	1 (1.11%)

To address RQ2, the characteristics of fake news-scam cases across different channels of transmission were explored. Once again, a .70 threshold was selected for confidence, and a .05 threshold was selected for support. This generated rules for Facebook only, as other channels (e.g. website, Instagram) consisted of rules with confidence scores lower than .70. Once again, this could be due to the discrepancy in sample size across the different channels of transmission (see Table 6), with around half (53.3%) of fake news-scams occurring on Facebook.

Facebook

A fake news-scam was likely to have occurred on Facebook, if it (1) contained authority cues, (2) was a fake promotion, (3) also occurred on Instagram and (4) did not use techniques of consistency. Confidence was relatively high at .83, suggesting that the rule held true 83.3% of the time, out of all rules with the same antecedents. Support was relatively low at .056, indicating that there may be insufficient information on the relationship between the antecedents and consequent (i.e. Facebook scam). Examples of Facebook fake news-scams using authority cues include fake investment opportunities from major banks operating in Singapore (CNA, 2018).

Our findings suggest that most fake news-scams occur on Facebook, aligning with local reports (Singapore Police Force, 2022). While specific victim demographics for Facebook-related scams are unavailable, general data show that the majority of scam victims in Singapore are aged 20–40, closely matching the primary age group of Facebook users locally (Affableai, 2023). A survey suggested that younger people are less vigilant about scams, possibly due to over-confidence (Tan, 2023b). Although details like victims’ occupation or gender were not detailed, research indicates that factors such as gender and education do not predict scam susceptibility (Low, 2019), highlighting the importance of broad-reaching anti-scam measures.

Discussion

Persuasion techniques

Authority, liking, social proof

Authority (92.2%), liking (73.3%) and social proof (31.1%) were identified as the most common techniques used in fake news-scams. This could possibly be explained by the bandwagon effect, or the tendency to conform to beliefs based on others’ behaviour (Schmitt-Beck, 2015). This may account for why these techniques may be effective in Singapore which has a more collectivistic culture with higher power distance (Hofstede Insights, n.d.), suggesting a stronger emphasis on adhering to hierarchical structures and maintaining harmonious interpersonal relationships. Collectivism has been associated with a greater tendency to follow others’ behaviour (van Baaren et al., 2003) and conform to social norms (Iyengar and Hahn, 2009; H. Kim and Markus, 1999), accounting for why social proof may be effective. In higher power distance cultures, individuals or institutions of lower power expect to be told what to do by those of higher power (Hofstede, 2011). Consequently, this could explain why techniques like authority and liking may be effective in Singapore.

Scarcity, consistency, reciprocity

Scarcity (13.3%), consistency (4.44%) and reciprocity (1.11%) were found to be the least common. For scarcity, this may be due to its pressure-inducing nature, which may promote heightened emotional state and subsequently, negative emotions such as distress (Biraglia et al., 2021). For instance, limiting the number of people who can claim the benefits of a scam may also give rise to uncertainty given that the benefits are not guaranteed. This may in turn, counter-effectively disincentivise them from joining the fake event.

For consistency and reciprocity, both techniques rely on executing scams with multiple stages. The former leverages on victims’ past or previous behaviour, while the latter involves giving initial benefits to elicit a sense of indebtedness among victims. Having multiple stages may not only involve greater effort for victims to follow through, but for scammers, it would require deliberate and extensive planning and execution, which could account for their low frequency in the dataset. As a result, such techniques may not be suitable or even scalable on social media platforms. In addition, Chen et al. (2021) also suggest that reciprocity may give rise to more scepticism than appreciation towards the initial benefits provided by scammers. Therefore, scarcity, consistency and reciprocity may not be cost-effective strategies for scammers.

Types of fake news-scams

Fake endorsements on a news article

Fake endorsements on news articles tend to contain liking, social proof and authority cues. As fake endorsements involve the credibility or likeability of renowned and authoritative figures, the associations with the use of techniques of liking and authority are grounded in existing theory. According to the heuristic-systematic model, having a more credible source (i.e. politician or celebrity) reduces users’ cognitive load in determining the credibility of online information, thus making said information more convincing (S. Chen et al., 1999; Kang et al., 2011). The use of social proof may once again be due to the bandwagon effect, where including comments of others signing up for hoax investment programmes further boosts the persuasiveness of the credible source.

Fake endorsements on news articles also usually occur on websites. This may be due to the nature of news articles being uploaded on online websites, thus explaining why such endorsements would be found on such channels rather than social media. Alternatively, this may also be due to the general distrust towards the credibility of social media, with close to 90% of Singaporeans believing that social media contained widespread misinformation (Yong, 2022). Taken together, fake endorsements may have been uploaded on websites to further boost its persuasiveness.

Finally, fake endorsements on news articles usually do not use reciprocity, consistency or scarcity. As aforementioned, this may be attributed to the amount of effort required to execute and follow through scams requiring multiple steps, as well as the negative affect and uncertainty associated with having restricted benefits. Fake news-scams use various cues to signal authenticity, which aim to mislead and deceive victims. For fake celebrity endorsements, an example would be Singaporean billionaire Peter Lim’s fake Bitcoin endorsement (Choo, 2019), which used cues of authenticity involving the source of information and having a credible expert. First, the article included logos of credible news channels. Another cue would be the inclusion of an image of well-known local news anchor, adding credibility to the fake article’s claim that Peter Lim was interviewed by the anchor, thus enhancing credibility among Chinese viewers. Finally, the scammers chose Peter Lim, a famous local billionaire, as the celebrity to endorse the fake investment. This would likely increase the persuasiveness of this scam.

Fake promotions

Fake promotions tend to contain liking and authority cues, similar to fake endorsements on news articles. Since fake promotions leverage on the reputation of well-known brands, this is expected as victims are generally drawn to the credibility and likeability of involved brands to be incentivised enough to join these ‘promotional events’.

Fake promotions were also more likely to occur on social media platforms, specifically Facebook and WhatsApp. A likely explanation could be due to the nature of these fake promotions being more ‘social’ in nature (e.g. brand giveaways), as compared to fake endorsements where investment programmes tend to be more confidential to each individual. For instance, the fast food meal box scam required victims to share the link with 10 of their friends to claim a meal voucher (Jamal, 2020). Hence, these promotions may be held on social media as it may facilitate the ‘sharing’ of such news across a wider audience more quickly.

Finally, fake promotions tend to not use techniques of reciprocity, consistency and scarcity and social proof. Once again, this could be due to the amount of effort required on both the scammer’s and victim’s end, as well as the heightened emotional arousal associated with scarcity, both of which may prove counter-effective. Interestingly, fake promotions do not use social proof, despite the ‘social’ nature of such scams. One possible explanation is that the nature of these scams being forwarded by friends and family in itself is sufficient to prove that these fake promotions are endorsed by one’s peers, thereby eliminating the need for the post or message itself to state this explicitly, which may have counter-effectively raised scepticism.

Consider also the case of a deceptive ‘$200 coupon giveaway’ phishing post that posed as a local supermarket (Sheng Siong, 2020). The cues of authenticity in this instance predominantly revolved around the source of information. The perpetrators included the supermarket’s logo within the fraudulent survey, creating a semblance of legitimacy. Moreover, they incorporated an image featuring regular customers at the supermarket intending to enhance the post’s credibility by showcasing it as popular and well-regarded. This strategic use of authenticity cues aimed to make the deceptive promotion appear genuine, thus heightening its overall persuasiveness.

Channels of transmission

Facebook

Overall, fake news-scams on Facebook tend to use the technique of authority. This may be due to the heuristic-systematic model, where authority cues from a politician or celebrity boost the credibility of such scams, making them more convincing.

Fake news-scams on Facebook were also likely to be fake promotions. This is aligned with the findings for fake promotions, where Facebook was likewise identified as a more frequent channel of transmission. This could be due to Facebook’s ability to proliferate news quickly, which may also explain why Facebook scams tend to occur concurrently on Instagram, due to the ease of being able to reach different audience groups quickly on both platforms.

Finally, fake news-scams on Facebook tend to not use consistency. Again, this could signal that victims of such scams generally do not follow through with multiple steps involved and that scammers may not be willing to devote large amounts of effort to execute such scams.

Implications and recommendations

This study builds on existing literature by examining the underlying patterns of fake news-scams, an understudied area. Moreover, findings may guide the development of more specific anti-scam measures such as prebunking and debunking. Furthermore, this study is also uniquely positioned to help infer how new technologies like generative AI can create fake news-scams that are appearing in our current landscape.

Prebunking measures

Drawing from the inoculation theory of social psychology, prebunking involves building cognitive and psychological resilience against malicious persuasion attempts first through repeated exposure to examples of fake news-scam cases (Compton, 2013; Roozenbeek et al., 2020). The findings suggest practical steps for organisations and governmental bodies to better allocate resources against fake news-scams. Brands, especially those offering giveaways, should educate consumers to recognise their official accounts, perhaps by enhancing their online presence and securing verification badges on social media. Similarly, government entities should instruct the public on identifying authentic sources to distinguish them from scams. Moreover, collaborating with social media platforms could enhance the dissemination of anti-scam messages or enable proactive monitoring and public alerts about potential scams. Given the prevalent use of authority cues in scams, individuals should be cautious and verify the credibility of websites claiming authority, such as by checking their URLs.

Debunking measures

Debunking involves retroactively correcting online misinformation after it has been encoded (Tay et al., 2022). As fake news is primarily circulated through social network technologies (e.g. Facebook, WhatsApp), individuals within one’s social circle play an imperative role in propagating fake news. Hence, individuals should ensure that information is verified before further warning their social network so as to increase awareness and to reduce the likelihood of more victims falling for these scams.

Generative AI and the future of fake news-scams

The democratisation of Generative AI has the potential to enable unprecedented labour efficiencies in production of text, image, audio and video. In turn, it has also enabled the proliferation of fake news and scam tactics. For example, in December 2023, several deepfakes that involved public figures like Singapore’s Prime Minister Lee Hsien Loong and Deputy Prime Minister Lawrence Wong were used to promote investment scams (Chia, 2023), which relied on using political figures as authority cues to persuade the target audience to take part in the investment scam. These deepfake videos will become even harder to spot with the advancement of technology and will prove to be a worrying issue in the near future.

Despite the rise of high-quality fake news-scams created by generative AI, it is crucial to focus on their primary goal: persuading individuals. This study emphasises the importance of grounding research on fake news-scams in Cialdini’s persuasive tactics – reciprocity, social proof, liking, consistency, authority, and scarcity. As technologies evolve, producing more sophisticated scams, understanding these persuasion methods provides essential insights for navigating and countering deception. Future research should concentrate on these strategies to develop effective responses to new forms of deceptive content in the digital realm.

Limitations and future directions

This study is not without its limitations, primarily due to the conservative scope of data collection which only included fake news-scam cases reported on Google News Search and Facebook, excluding other channels and unreported cases. In addition, only English-language cases were analysed, omitting incidents in other languages such as Malay, Tamil or Mandarin. Future research could broaden the scope by incorporating cases from varied channels and languages to enhance the dataset of fake news-scam cases in Singapore. Furthermore, adopting qualitative methods like focus group interviews with scam victims could provide deeper insights into unreported cases and more comprehensive patterns of fake news-scams in Singapore.

Conclusion

This study investigated the use of persuasion techniques in 90 fake news-scam cases across various types and transmission channels, employing association rule mining to identify common patterns. The most prevalent techniques – authority, liking and social proof – remained consistent across different scam types. Fake endorsements were more common on websites, while fake promotions predominated on social media platforms. Specifically, fake news-scams on Facebook often involved promotions that also appeared on Instagram and included cues of authority. This research addresses the underexplored area of fake news-scams, guiding the development of preventive and corrective measures against scams. Future research should explore unreported cases through qualitative methods or examine non-English fake news-scam cases to enhance the representativeness of the findings.

Footnotes

Declaration of Conflicting Interests

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

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

Ethics Approval

This is an observational study on open-source data. No ethics approval is required. This article does not contain any studies with human participants or animals performed by any of the authors.

Informed Consent

ORCID iD

Xingyu Ken Chen

Author biographies

Xingyu Ken Chen is a Senior Research Analyst in the Home Team Psychology Division, holding a Masters in Information Studies from Nanyang Technological University. His research interests include societal polarisations, online misinformation, information operations, strategic communications and natural language processing. Currently, he is pursuing his PhD at the Nanyang Technological University, researching the use of AI in misinformation detection.

Siew Maan Diong is the Director of the Research Directorate, Home Team Psychology Division. She is currently researching behavioural influences on individual attitudes and behaviours. Her other key research interests include resilience, crisis and trauma, and organisational assessment.

Bethany Leong holds a Bachelor of Social Sciences (Honours) in Psychology from the National University of Singapore. She is interested in criminal psychology and industrial/organisational psychology.

Rong Hui Tan is a Psychology and Media Analytics graduate from Nanyang Technological University. Her research interest lies in protective factors against misinformation and disinformation.

Shannon Su Yan Ng holds a Bachelor of Arts (Honours) in English with a minor in Psychology from Nanyang Technological University. She was a researcher turned practitioner in the field of crime and investigations.

Stephanie Chan graduated from the University of Leicester with a Masters in Forensic Psychology (First Class Honours). She is a lead psychologist in the Home Team Psychology Division, Ministry of Home Affairs. Her current research interest is in crime and forensic psychology in the law enforcement context.

References

Affableai (2023) Demographics of Facebook Users in Singapore as of June, by Age Group. Statista, June. Available at: https://www.statista.com/statistics/1398477/singapore-demographics-of-facebook-users-by-age-group/

Agrawal

Imieliński

Swami

(1993) Mining association rules between sets of items in large databases. In: Proceedings of the 1993 ACM SIGMOD international conference on management of data, Washington, DC, 26–28 May, pp. 207–216. New York: ACM.

Allcott

Gentzkow

(2017) Social media and fake news in the 2016 election. Journal of Economic Perspectives 31(2): 211–236.

Ang

(2021) 6 in 10 People in Singapore Have Encountered a Scam – Don’t Fall for the Scammers’ Tricks! Ministry of Home Affairs, 16 February. Available at: http://www.mha.gov.sg/home-team-news/story/detail/6-in-10-people-in-singapore-have-encountered-a-scam-don-t-fall-for-the-scammers-tricks!/

Balnaves

(2020) A return to the good old days: Populism, fake news, yellow journalism, and the unparalleled virtue of business people. In: Bowden

Muldoon

Gould

, et al (eds) The Palgrave Handbook of Management History. Cham: Palgrave Macmillan, pp. 935–949. Available at: https://link.springer.com.remotexs.ntu.edu.sg/referenceworkentry/10.1007/978-3-319-62114-2_45

Barfar

(2019) Cognitive and affective responses to political disinformation in Facebook. Computers in Human Behavior 101: 173–179.

Biraglia

Usrey

Ulqinaku

(2021) The downside of scarcity: Scarcity appeals can trigger consumer anger and brand switching intentions. Psychology & Marketing 38(8): 1314–1322.

Burrell

(2008) Problematic empowerment: West African internet scams as strategic misrepresentation. Information Technologies & International Development 4(4): Article4.

Button

Lewis

Tapley

(2014) Not a victimless crime: The impact of fraud on individual victims and their families. Security Journal 27(1): 36–54.

10.

Ceron

(2015) Internet, news, and political trust: The difference between social media and online media outlets. Journal of Computer-Mediated Communication 20(5): 487–503.

11.

Chatterjee

Basu

(2021) How vulnerable are you? A novel computational psycholinguistic analysis for phishing influence detection. In: Proceedings of the 18th international conference on natural language processing (ICON), pp. 499–507. Available at: https://aclanthology.org/2021.icon-main.61

12.

Chen

Duckworth

Chaiken

(1999) Motivated heuristic and systematic processing. Psychological Inquiry 10(1): 44–49.

13.

Chen

Xiao

Mao

(2021) Persuasion strategies of misinformation-containing posts in the social media. Information Processing & Management 58(5): 102665.

14.

Chen

Goh

(2019) Fake news: Five key things home team officers should know. Home Team Journal 8: 121–129.

15.

Chen

Goh

, et al. (2020) Crime-fake news nexus. In: Khosrow-Pour

(ed.) Encyclopedia of Criminal Activities and the Deep Web. Hershey, PA: Information Science Reference, pp. 52–65.

16.

Cheng

(2021) Police warn of new scam advertising fake jobs for E-commerce platforms. CAN, 12 May. Available at: https://www.channelnewsasia.com/singapore/new-scam-trend-advertising-fake-jobs-e-commerce-platforms-police-1358701

17.

Cheung

(2021) 2 Crabs for $9.90? Man Loses Over $18,000 after Falling for Scam. AsiaOne, 16 April. Available at: https://www.asiaone.com/singapore/2-crabs-990-man-loses-over-18000-after-falling-scam

18.

Chia

(2023) Public urged to be on guard as deepfake content will grow more sophisticated: Experts. The Straits Times, 29 December. Available at: https://www.straitstimes.com/tech/public-urged-to-be-on-guard-as-deepfake-content-will-grow-more-sophisticated-experts

19.

Choo

(2022) Scams account for over half of crime reports in first half of 2022, with victims losing S$228m. TODAY, 19 October. Available at: https://www.todayonline.com/singapore/scams-job-scams-crime-higher-1979131

20.

Choo

(2019) Advertising scam promoting investment endorsed by billionaire Peter Lim resurfaces. The Straits Times, May 25. Available at: https://www.straitstimes.com/singapore/advertising-scam-promoting-investment-endorsed-by-billionaire-peter-lim-resurfaces

21.

Cialdini

(2016) Pre-suasion: A Revolutionary Way to Influence and Persuade. New York: Simon & Schuster.

22.

CNA (2018) Beware of DBS, OCBC phishing scams: Police. CAN, 30 August. Available at: https://www.channelnewsasia.com/singapore/phishing-scams-dbs-ocbc-investment-sms-police-807146

23.

CNA (2022) Articles claiming Tharman Shanmugaratnam endorses crypto auto-trading programmes are fake: Police. CAN, 13 July. Available at: https://www.channelnewsasia.com/singapore/fake-online-articles-tharman-shanmugaratnam-endorse-crypto-bitcoin-2807421

24.

Compton

(2013) Inoculation theory. In: Dillard

Shen

(eds) The SAGE Handbook of Persuasion: Developments in Theory and Practice. New York: Sage, pp. 220–236.

25.

Devaraj

(2023) New law passed to remove online content that is criminal or harmful. The Straits Times, 5 July. Available at: https://www.straitstimes.com/singapore/parliament-passes-bill-that-allows-government-to-remove-online-criminal-content

26.

Ermakova

Nurbakova

Ovchinnikova

(2020) Covid or not Covid? Topic shift in information cascades on Twitter. In: Proceedings of the 3rd International Workshop on Rumours and Deception in Social Media (RDSM), Barcelona, Online, pp. 32–37. Available at: https://www.aclweb.org/anthology/2020.rdsm-1.3

27.

Fischer

Lea

SEG

Evans

(2013) Why do individuals respond to fraudulent scam communications and lose money? The psychological determinants of scam compliance. Journal of Applied Social Psychology 43(10): 2060–2072.

28.

Fraudwatch (2021) Expert explanation: Fake celebrity endorsements. They aren’t who they seem. Digital Brand Protection – Fraudwatch, 9 March. Available at: https://fraudwatch.com/blog/expert-explanation-fake-celebrity-endorsements-they-arent-who-they-seem/

29.

Hipp

Güntzer

Nakhaeizadeh

(2000) Algorithms for association rule mining – A general survey and comparison. ACM Sigkdd Explorations Newsletter 2(1): 58–64.

30.

Hofstede

(2011) Dimensionalizing cultures: The Hofstede model in context. Online Readings in Psychology and Culture 2(1): 2307–0919.

31.

Hofstede Insights (n.d.) Singapore Hofstede Insights 2023. 10 January. Available at: https://www.hofstede-insights.com/country/singapore/

32.

Ismail

(2021) Police warn of online bak kwa scams ahead of Chinese New Year. The Straits Times, 4 February. Available at: https://www.straitstimes.com/singapore/police-warns-of-bak-kwa-scams-ahead-of-chinese-new-year

33.

Iyengar

Hahn

(2009) Red media, blue media: Evidence of ideological selectivity in media use. Journal of Communication 59(1): 19–39.

34.

Jamal

(2020) McDonald’s Says Free Family Meal Box Giveaway Is a Hoax. The New Paper, 29 May. Available at: https://www.tnp.sg/news/singapore/mcdonalds-says-free-family-meal-box-giveaway-hoax

35.

Kang

Bae

Zhang

, et al. (2011) Source cues in online news: Is the proximate source more powerful than distal sources? Journalism & Mass Communication Quarterly 88(4): 719–736.

36.

Kerley

Copes

(2002) Personal fraud victims and their official responses to victimization. Journal of Police and Criminal Psychology 17(1): 19–35.

37.

Kim

Xiong

Lee

, et al. (2021) A systematic review on fake news research through the lens of news creation and consumption: Research efforts, challenges, and future directions. PLoS ONE 16(12): e0260080.

38.

Kim

Markus

(1999) Deviance or uniqueness, harmony or conformity? A cultural analysis. Journal of Personality and Social Psychology 77: 785–800.

39.

Kua

(2021) Fake celebrity Facebook giveaways ask users to ‘Guess the Number’. Facebook, March 3. Available at: https://www.facebook.com/groups/538229993008898/posts/1870113303153887/

40.

Kwok

(2019) Gossip Mill: Local celeb Bryan Wong falls victim to fake news – and other entertainment news this week. Asiaone, 22 December. Available at: https://www.asiaone.com/entertainment/gossip-mill-local-celeb-bryan-wong-falls-victim-fake-news-and-other-entertainment

41.

Lee

(2018) PM Lee Hsien Loong warns against online scam using his name to solicit bitcoin investments. The Straits Times, 22 September. Available at: https://www.straitstimes.com/singapore/pm-lee-hsien-loong-warns-against-online-scam-using-his-name-to-solicit-bitcoin-investments

42.

Low

(2019) Think you’ll never be a scam victim? Chances are you’re more likely to fall for one. TODAY, 10 July. Available at: https://www.todayonline.com/singapore/think-youll-never-be-scam-victim-chances-are-youre-more-likely-fall-one

43.

Chan

Chai

, et al. (2020) Examining the influence of emotional arousal and scam preventive messaging on susceptibility to scams. Crime Prevention and Community Safety 22(4): 313–330.

44.

Lyons

(2024) Scams promoted in fake news articles and deepfake videos cost Australians more than $8m last year. The Guardian, 1 March. Available at: https://www.theguardian.com/australia-news/2024/mar/01/scams-promoted-in-fake-news-articles-and-deepfake-videos-cost-australians-more-than-8m-last-year

45.

KWF

McKinnon

(2021) COVID-19 and cyber fraud: Emerging threats during the pandemic. Journal of Financial Crime 29(2): 433–446.

46.

Mirtaheri

Abu-El-Haija

Morstatter

, et al. (2021) Identifying and analyzing cryptocurrency manipulations in Social Media. IEEE Transactions on Computational Social Systems 8(3): 607–617.

47.

Nedelcu

Balaban

(2021) The role of source credibility and message credibility in fake news engagement. Perspectives from an experimental study. Journal of Media Research 14(3): 42–62.

48.

(2020) Police warn public of fake bank ads on mobile phone promotions. The Straits Times, 4 December. Available at: https://www.straitstimes.com/singapore/police-warn-public-of-fake-bank-advertisements-on-mobile-phone-promotions

49.

Norris

Brookes

Dowell

(2019) The psychology of internet fraud victimisation: A systematic review. Journal of Police and Criminal Psychology 34(3): 231–245.

50.

Ong

Cabañes

(2018) Architects of Networked Disinformation: Behind the Scenes of Troll Accounts and Fake News Production in the Philippines. Newton Tech4Dev Network. Available at: https://medium.com/@MediaManipulation/architects-of-networked-disinformation-283b02b6dfb8

51.

Phillips

Wilder

(2020) Tracing cryptocurrency scams: Clustering replicated advance-fee and phishing websites. In: 2020 IEEE international conference on blockchain and cryptocurrency (ICBC), Toronto, ON, Canada, 2–6 May. DOI: 10.1109/ICBC48266.2020.9169433.

52.

(2021) Fake SGH employment letter allegedly circulates in India, Hospital warns of scam, 21 April. Available at: https://mustsharenews.com/fake-sgh-employment-letter/ (accessed 10 January 2023).

53.

Rao

(2020) Commentary: It’s not just work and the economy. COVID-19 is also changing how we use Social Media. CNA, 14 April. Available at: https://web.archive.org/web/20221125173149/https://www.channelnewsasia.com/commentary/covid-19-coronavirus-has-changed-social-media-use-763841

54.

Rong

Law

, et al. (2012) A behavioral analysis of web sharers and browsers in Hong Kong using targeted association rule mining. Tourism Management 33(4): 731–740.

55.

Roozenbeek

Linden

van der Nygren

(2020) Prebunking interventions based on ‘inoculation’ theory can reduce susceptibility to misinformation across cultures. Harvard Kennedy School Misinformation Review 1(2): 008.

56.

Roy

Bordoloi

Das

, et al. (2021) Association rule mining on crime pattern mining. In: 2021 international conference on computational performance evaluation (ComPE), Shillong, India, 1–3 December, pp. 269–272.

57.

Rubin

(2019) Disinformation and misinformation triangle: A conceptual model for ‘fake news’ epidemic, causal factors and interventions. Journal of Documentation 75(5): 1013–1034.

58.

Sarno

Black

(2023) Who gets caught in the web of lies? Understanding susceptibility to phishing emails, fake news headlines, and scam text messages. Human Factors 66: 1742–1753.

59.

Schmitt-Beck

(2015) Bandwagon effect. In: Wessler

Barnhurst

Ikeda

, et al (eds) The International Encyclopedia of Political Communication. Hoboken, NJ: John Wiley & Sons, pp.1–5.

60.

Seow

Chia

Lim

, et al. (2022) Singapore businesses adapting their operations to fight fraud and scams. The Straits Times, 1 May. Available at: https://www.straitstimes.com/singapore/courts-crime/businesses-adapting-their-operations-to-fight-fraud-and-scams

61.

Sheng Siong (2020) Phishing Post on ‘$200 Coupon’ Giveaway. Sheng Siong Supermarket. Available at: https://corporate.shengsiong.com.sg/phishing-post-on-200-coupon-giveaway/

62.

Singapore Police Force (2022) Mid-year Crime Statistics 2022. Singapore Police Force. Available at: https://www.police.gov.sg/-/media/90E8A6D32BBE4E838025817C160E40CA.ashx

63.

Soon

Goh

(2021) Singaporeans’ Susceptibility to False Information (19, IPS Exchange). Singapore: Institute of Policy Studies.

64.

Soon

CWT

Goh

ZSS

(2017) What Lies beneath the Truth: A Literature Review on Fake News, False Information and More. Singapore: Institute of Policy Studies.

65.

Strauss

(1987) Codes and coding. In: Strauss

(ed.) Qualitative Analysis for Social Scientists. Cambridge: Cambridge University Press, pp.55–81.

66.

Tan

(2023a) Commentary: Singapore banks’ latest anti-scam measures may be inconvenient, but would you rather lose your life savings? CNA, 12 October. Available at: https://www.channelnewsasia.com/commentary/singapore-banks-anti-scam-measures-apps-restriction-money-lock-3836826

67.

Tan

(2023b) Young people say in MCI survey they can identify scams, but police crime statistics show otherwise. The Straits Times, 22 September. Available at: https://www.straitstimes.com/singapore/courts-crime/young-people-say-in-mci-survey-they-can-identify-scams-but-police-crime-statistics-show-otherwise

68.

Tandan

Acharya

Pokharel

, et al. (2021) Discovering symptom patterns of COVID-19 patients using association rule mining. Computers in Biology and Medicine 131: 104249.

69.

Tandoc

(2019) The facts of fake news: A research review. Sociology Compass 13: e12724.

70.

Tandoc

Jr Lim

Ling

(2018) Defining ‘fake news’: A typology of scholarly definitions. Digital Journalism 6(2): 137–153.

71.

Tang

(2023) More than S$330 million lost to scammers in first half of 2023; cases continue to rise. CNA, 13 September. Available at: https://www.channelnewsasia.com/singapore/police-scam-cybercrime-phishing-fake-friend-malware-first-half-2023-3766796

72.

Tay

Hurlstone

Kurz

, et al. (2022) A comparison of prebunking and debunking interventions for implied versus explicit misinformation. British Journal of Psychology London 113: 591–607.

73.

Tee

(2019) Singapore Airlines warns of phishing scam promising free plane tickets. The Straits Times, 13 January. Available at: https://www.straitstimes.com/singapore/courts-crime/singapore-airlines-warns-of-phishing-scam-promising-free-plane-tickets

74.

Tripathi

Nigam

Edla

(2017) A novel web fraud detection technique using association rule mining. Procedia Computer Science 115: 274–281.

75.

US Securities and Exchange Commission (2022) Social media and investment fraud – Investor alert. US Securities and Exchange Commission, 29 August. Available at: https://www.sec.gov/oiea/investor-alerts-and-bulletins/social-media-and-investment-fraud-investor-alert

76.

van Baaren

Maddux

Chartrand

, et al. (2003) It takes two to mimic: Behavioral consequences of self-construals. Journal of Personality and Social Psychology 84(5): 1093–1102.

77.

Wardle

Derakhshan

(2017) Information disorder: Toward an interdisciplinary framework for research and policy making. Council of Europe Report 2017: 109.

78.

World Health Organization (2021) Fraudulent ‘COVID-19 Compensation Lottery Prize’ Scam, Falsely Alleges Association with WHO and Others. Geneva: World Health Organization. Available at: https://www.who.int/news/item/06-08-2021-fraudulent-covid-19-compensation-lottery-prize-scam-falsely-alleges-association-with-who-and-others

79.

Yang

(2019) Film-maker Jack Neo files police report over fake bitcoin site with bogus remarks attributed to him. The Straits Times, 6 March. Available at: https://www.straitstimes.com/singapore/courts-crime/film-maker-jack-neo-files-police-report-over-fake-bitcoin-site-with-bogus

80.

Yong

(2022) Singaporeans’ trust in science remains high, but misinformation could affect this: Study. The Straits Times, 13 June. Available at: https://www.straitstimes.com/singapore/singaporeans-trust-in-science-remains-high-but-misinformation-could-affect-this-study

81.

Zielinska

Welk

Mayhorn

, et al. (2016) A temporal analysis of persuasion principles in phishing emails. Proceedings of the Human Factors and Ergonomics Society Annual Meeting 60(1): 765–769.