Skip to main content
NCBI home page
Wiley Open Access Collection logoLink to Wiley Open Access Collection
. 2024 Jul 11;45(1):214–222. doi: 10.1111/risa.15073

On the use of the term “real risk”

Roger Flage 1,✉, Terje Aven 1, Ingrid Glette‐Iversen 1
PMCID: PMC11735339  PMID: 38991782

Abstract

The term “real risk” and variations of this term are commonly used in everyday speech and writing, and in the scientific literature. There are mainly two types of use: i) in statements about what the real risk related to an activity is and ii) in statements about the risk related to an activity being real. The former type of use has been extensively discussed in the literature, whereas the latter type has received less attention. In the present study, we review both types of use and analyze and discuss potential meanings of type ii) statements. We conclude that it is reasonable to interpret a statement about the risk being real as reflecting a judgement that there is some risk and that the knowledge supporting this statement is relatively strong. However, such a statement does not convey whether the risk is small or large and needs to be supplemented by a characterization of the risk.

Keywords: real risk, risk characterization, risk communication, risk science

1. INTRODUCTION

A wide range of terms, expressions, and concepts are used to describe and communicate risk, in both scientific and non‐scientific contexts. A significant part of risk‐related research has been devoted to clarifying the meaning and interpretation of such concepts, terms, and expressions, with the main purpose being to provide a scientifically rigorous foundation for their understanding and use (e.g., Aven, 2013; Flage & Aven, 2015; Boholm et al., 2016; Jansen et al., 2019).

An expression that has garnered attention is “real risk,” which is extensively used in both everyday discourse and scientific literature. However, despite its widespread usage, the term is rarely provided with a clear understanding or definition. In scientific discourse, scholars have raised concerns about the lack of a clear and unified interpretation of the term. For instance, Cox (2007) addresses the use of the term “real risk” as a criterion for invoking the precautionary principle, emphasizing the need for a consistent understanding. Furthermore, Cox discusses the term “real risk” in relation to so‐called “false positives,” contrasting the labeling of some risks as “real” with those judged as such erroneously. Thompson (1990) provides a discussion of what constitutes a “real risk,” suggesting two ways of interpreting the term; one so‐called ‘causal’ interpretation, stating that “risks are real when there is sufficient reason to suspect the presence of a causal sequence that would produce the unwanted event” (Thompson, 1990, p. 12). Following this line of interpretation, the term is linked to the degree to which the potential occurrence of an event can be supported by reasoning or evidence. According to the other interpretation, referred to as ‘probabilistic’, “risks are real when there is a measurable probability of harm” (Thompson, 1990, p. 13). However, as there are different ways of interpreting probability, Thompson highlights that according to this notion of what constitutes a “real risk,” the meaning of the term needs to be seen in light of the adopted perspective on probability.

In everyday discourse, the term is commonly employed to inform the general public about risks through news media. However, labeling a risk as “real” is a strong rhetorical move, and from a professional risk analysis perspective, it is necessary to rely on a well‐founded rather than intuitive understanding of the term in order to ensure a prudent communication of risk. Clarifying the meaning and interpretation of the term “real risk” is essential not only for risk analysts and researchers but also for policymakers, practitioners, and the general public who rely on a clear understanding of risk in order to support decision‐making.

Preliminary studies of the use of the term “real risk” (or similar phrases such as “the risk is real”) in scientific and everyday discourse show that there are essentially two ways to use the term:

  • i)

    In statements about what the real risk (level) related to an activity is, for example, “the real risk is so close to zero that…” (Wallace, 1994, p. 144) or “The real risk is probably underestimated” (Navarro et al., 1998, p. 251).

  • ii)

    In statements about the risk related to an activity being real, for example, “the climate crisis will only raise the risk of future events. ‘This shows us the risk is real’” (Readfearn et al., 2022), or “there is ‘a real risk for a new armed conflict in Europe’” (Rankin et al., 2022).

The first type of use presupposes that a real risk (level) exists, and the term is used when making a statement about the size, extent, or degree of the risk. The second type is used when classifying a risk as real, suggesting that other risks—or, perhaps, the current “real risk” under different conditions—may be classified as non‐real.

The idea of real, true, or objective risk (level) as implied by type i) use is extensively discussed in the scientific literature (e.g., Aven et al., 2011; Aven & Renn, 2009; Freudenburg, 1988; Hermansson, 2012; Holton, 2004; Howarth, 1988; Kozine, 2018). On the other hand, while statements of type ii) are common in a wide range of contexts, the idea that some risks are “real” appears to have received far less attention. We have not been able to find conceptual discussions of this idea, and there is thus a need to clarify its meaning and implications.

In the present study, we review both type i) and type ii) uses of the term “real risk” and analyze and discuss potential meanings of type ii) statements. The meaning of the term clearly depends on the risk perspective adopted, that is, on how risk is understood and conceptualized. The central part of our analysis is thus a conceptual analysis, framing the “real risk” term in light of different risk perspectives. The analysis also examines the relationship between type i) and type ii) statements. If a real risk (level) exists, in the sense of objective, true or underlying risk, it obviously makes sense to talk about a risk being real. On the other hand, if such a notion of “real risk” is rejected, then it is not immediately clear that we can talk about there being a real risk of something; and even if such labeling can be shown to make sense, the meaning needs to be clarified. Does it then simply mean that the assessed probability of an event, seen in relation to its (potential) consequences, has reached a critical or unacceptable level? Or does it instead have to do with the knowledge basis of the risk characterization, that any risk can be labeled as real so long as the justification for doing so is sufficiently strong? And how can we determine the required probability or “strength of knowledge” level for either of these interpretations? These are the types of questions that the present study aims to answer.

Another term closely related to “real risk” is “actual risk.” While the scope of the present study will be limited to focus on “real risk,” it is important to note that the similarities in usage and interpretation between these two expressions suggest that the analysis and subsequent conclusions drawn in the study are relevant also to the understanding and use of the term “actual risk.”

The remainder of the study is organized as follows. In Section 2, we review the use of the term “real risk” in the news media and scientific literature. Then, in Section 3, we perform a conceptual analysis of the term before discussing the results in Section 4. Finally, Section 5 concludes.

2. USE OF THE TERM “REAL RISK”

In this section, we review the uses of the term “real risk” and phrases including this term in the news media and scientific literature. The aim is not to present a complete review but to give examples of use and show that these are of either of the two types described in Section 1.

2.1. In the news media

Table 1 shows the results of a search performed on September 18, 2023, using the Norwegian news media archive Atekst from January 1, 1960 (year of first appearance of the term in the database), to December 31, 2022, and Google filtered for news. Atekst is a complete document search. Note that the search result number for the phrase “risikoen er reell” (“the risk is real”) in Atekst is exact, whereas the remaining Atekst and Google (news) numbers are approximate. The first row shows the number of search results for the Norwegian translation “reell risiko” in Atekst and for the term “real risk” in Google (news). The second row gives the number of search results for the Norwegian translation “risikoen er reell” in Atekst and for the phrase “the risk is real” in Google (news). The numbers show that the term gets considerable use in Norwegian and English‐language news media. The phrase “the risk is real” is used, though considerably less so than the term “real risk,” which is not unexpected given that the former phrase is more specific.

TABLE 1.

Number of search results for the term “real risk” (Norwegian: “reell risiko”) and the phrase “the risk is real” (Norwegian: “risikoen er reell”).

Terms and variants Atekst Google search filtered for news
“real risk” (Norwegian: “reell risiko”) ∼4932 ∼18,500,000
“the risk is real” (Norwegian: “risikoen er reell”) 181 ∼ 5,870,000
Open in a new tab

Source: Atekst and Google filtered for news.

To illustrate the various usages of the term “real risk” in the news media, some specific examples are highlighted in Table 2.

TABLE 2.

Examples of usages of the term “real risk” in the news media.

Source Use Type
Boseley, 2014 “Ebola virus: what is the real risk to the UK?” i)
Cook, 2020 “The real risk of downloading Chinese apps on to your phone” i)
Parker‐Pope, 2020 “(…) what's the real risk of catching Covid‐19 from a germy surface or object?” i)
Hanson, 2023 “Prison escapes in America: How common are they and what's the real risk?” i)
Kumar, 2013 “Of course, by any rational measure at all, this belief does not reflect the real risk level of funds.” i)
Rankin et al., 2022 “Nato chief warns of ‘real risk of conflict’ as talks with Russia over Ukraine end” ii)
Lagarias, 2024 “UK recession is a real risk—but ‘soft landing’ is more likely” ii)
McKie, 2022 “There is now a real risk of people drowning, particularly in basement flats if a major flash flood occurred in the middle of the night” ii)
Iversen, 2023 “There is a real risk that the Chinese economy will end up in a recession” ii)
Readfearn et al., 2022 “the climate crisis will only raise the risk of future events. ‘This shows us the risk is real’” ii)
Sridhar, 2021 “If we loosen restrictions too early, there is a real risk of a third wave in the UK” ii)
Storrs, 2015 “A false positive mammogram is a real risk factor for breast cancer, study finds” ii)
Open in a new tab

As seen from the examples presented in Table 2, the term “real risk” is used to describe various types of risk, including health risks, technological risks, financial risks, and risks related to climate change and extreme weather events. While interpretations of both type i) and type ii) can be found in the news media, type ii) interpretations appear to be more common.

2.2. In the scientific literature

Table 3 shows the results of a literature search performed on September 18, 2023, using the Scopus database from 1921 up to the present day and Google Scholar. Note that the Scopus numbers are exact, whereas Google Scholar only gives approximate search result numbers. Moreover, the Google Scholar search covers the complete document, whereas the Scopus search only covers the article title, abstract, and keywords. The top row shows the number of results for the term “real risk.” The bottom row shows the number of results for the phrase “the risk is real.” The latter is a type ii) phrase, while phrases that include the term “real risk” can be of either type i) or type ii). To see this, consider as an example “the real risk” and recall from Section 1 the statement, “The real risk is probably underestimated,” which is a type i) use. Then, consider the statement, “The real risk is that we could therefore soon find ourselves dealing with a series of compromised neurological cases” (Tartara et al., 2020, p. 1501), which is a type ii) use.

TABLE 3.

Number of search results for the term “real risk” and the phrase “the risk is real”.

Terms/phrases Scopus a Google Scholar
“real risk” 2621 ∼135,000
“the risk is real” 17 ∼2330
Open in a new tab
a

Article title, abstract, and keywords.

Source: Scopus and Google Scholar.

A selection of examples of the use of the term “real risk” in the scientific literature are presented in Table 4.

TABLE 4.

Examples of usages of the term “real risk” in the scientific literature.

Source Use Type
Cirillo & Taleb, 2020, p. 606 “We argue that a natural and empirically correct framework for assessing (and managing) the real risk of pandemics is provided by extreme value theory (…)” i)
Cadar, 2017, p. 3 “However, the real risk of dementia is probably underestimated (…)” i)
Côté et al., 2024, p. 3 “This framework focuses on evaluating a single risk to provide an estimate, although practitioners must keep in mind that the real risk is probably higher.” i)
Mitchell et al., 2010, p. 266 “The study, however, demonstrates an important point: the real risk of drowning is markedly higher than estimated using only population‐based rates.” i)
Galatolo & Lazzeri, 2016, p. 308 “Notwithstanding, high safety factors sometimes were not able to avoid failures because the real risk level is not known (…)” i)
Wallace, 1994, p. X “the real risk is so close to zero that…” i)
Navarro et al., 1998, p. 251 “The real risk is probably underestimated” i)
Nartea et al., 2019, p. 178 “Pregnancy and postpartum period represent a key moment in women's lives in which the risk of obesity is real.” ii)
Brito & Griffiths, 2016, p. 55 “The risk of loss is real; for example, Australian and British AUVs have been lost under ice sheets and one team maintained a lightweight tether to an AUV when operating under sea ice.” ii)
Gleick, 2006, p. 497 “The fact that there are numerous examples of actual and planned attacks on water systems in the past suggests that the risk is real.” ii)
Lu et al., 2022, p. 511 “It cannot be disputed that, if the current system remains as it is going forward, there is a real risk of an aluminium crisis” ii)
Omran & Almaliki, 2020, p. 1196 “Elderly COVID‐19 patients are at a real risk of complications due to impaired immune function (…)” ii)
Merkow & Breithaupt, 2014, p. 72 “a risk is real when there is a presence of a threat, a vulnerability that the attacker can exploit, and a high likelihood that the attacker will carry out the threat” ii)
Tartara et al., 2020, p. 1501 “The real risk is that we could therefore soon find ourselves dealing with a series of compromised neurological cases” ii)
Sağir et al., 2021, p. 174 “Our results suggest that although vitamin D deficiency is not a real risk factor, it is a helpful factor in explaining increased mortality rates.” ii)
Open in a new tab

Similar to the examples found in the news media, the use of the term “real risk” in scientific literature also indicates that both type i) and type ii) interpretations of the term are used, with the latter type being more common. Furthermore, we find that the term is used in various fields and disciplines, including health, finance, environmental sciences, and engineering.

2.3. Review of the use of the term “real risk”

As seen from the examples in the above sections, we find both type i) and type ii) uses of the term “real risk” in the news media and scientific literature, and the term is referred to in relation to a wide range of topics and fields. A closer look at the various applications of the term indicates that the distinction between the two interpretations arises from the emphasis placed on the term “real risk” (or “real risk factor/level”) as a distinct entity (type i) versus the characterization of the risk itself as “real,” highlighting the adjective “real” (type ii).

When the focus is on the noun part of “real risk,” reflecting a type i) interpretation, the use of the term presupposes the existence of an objective, true value of the risk related to a specific activity or event. In these cases, the term “real risk” refers to this underlying, true value. Following this line of interpretation, there is a gap between the objective “real risk” and judgments of this risk. In news media, this gap is often reflected as the contrast between the “real risk” and people's perceptions or beliefs about the risk. For example, when discussing the risk of prison escapes, the “real risk” is compared to public concerns by questioning “what level of concern should the public genuinely have?” (Hanson, 2023). Similarly, Kumar (2013) contrasts people's perception of risk related to investments with “the real risk level” of funds, where the former represents what people “think they know about funds,” whereas the latter is referred to as “the actual characteristics of different investments.”

In the scientific literature, type i) interpretations of the term “real risk” are often associated with attempts to assess, estimate, or evaluate the underlying, true, objective risk value, using various methods and approaches. For example, Cirillo and Taleb (2020) suggest that “the real risk” of pandemics can be found using extreme value theory. In other cases, the term is used to underscore that there exists a true, objective risk value for which we cannot provide an accurate estimate. Examples include referrals to “the real risk” as “not known” (Galatolo & Lazzeri, 2016, p. 308), “underestimated” (Cadar, 2017, p. 3), or “higher than estimated” (Mitchell et al., 2010, p. 266).

For the cases indicating a type ii) interpretation, the perspective shifts from focusing on “real risk” as a term reflecting an objective quantity or value to the description of a risk as “real.” Thus, the term is used to describe a property of the risk (or risk factor), not as a reference to the risk per se. Both in news media and in scientific literature, type ii) interpretations are often employed to emphasize the significance of the risk related to some activity or event. For example, Lagarias (2024) warns of the “real risk” of a UK recession, while Readfearn et al. (2022) assert that climate change will “raise the risk of future events” and emphasize that “the risk is real.” These statements underscore that the risk associated with these events warrants a certain level of concern. A similar use is found in the scientific literature, where the term “real risk” is used to highlight that there is a significant potential for negative consequences associated with a specific event or scenario. For example, Omran and Almaliki (2020) state that elderly COVID‐19 patients are at “a real risk” of complications due to impaired immune function, while Lu et al. (2022) warn of “a real risk” of an aluminum crisis if current practices persist.

While the characterization of risk as being “real” is extensively used in both news media and scientific literature, it is rarely supported by a clear understanding of what “real” actually means. An exception is Merkow and Breithaupt (2014), who provide an explicit interpretation of a risk as “real” when a threat (event/scenario) has been identified, and there is a high likelihood that the event will occur. For the cases where no explicit interpretation is provided, the use of the term “real risk” indicates the risk being of a certain extent or magnitude. How large the extent of the risk is, however, varies. For example, the term is linked to different descriptions and characteristics, such as the risk being “statistically significant” (Sağir et al., 2021, p. 173), a “slightly higher risk” (Storrs, 2015), or an event having a 25% probability of occurring (Lagarias, 2024). In some cases, the use of the term “real risk” is associated with current knowledge and historical data. For instance, Gleick (2006) and Merkow and Breithaupt (2014) refer to the risk as “real” by pointing to knowledge/evidence about the occurrence of similar events in the past.

3. ANALYSIS

In this section, we will examine possible interpretations of the “real risk” term, focusing on the type ii) line of interpretation in which the term is used in statements about the risk related to an activity being characterized as “real.” As mentioned in Section 1, the interpretations will depend on the risk perspective, the way we understand the risk concept. First, in Section 3.1, we introduce two main categories of such risk perspectives to provide a basis for the analysis coming in Section 3.2.

3.1. Risk conceptualized and characterized

Risk can be conceptualized in different ways; see for example the overviews in Aven et al. (2011) and Aven (2012). For the purpose of the present study, we distinguish between two main categories of risk perspectives. The first we refer to as the “probability‐based perspective,” I, and the second, the “uncertainty‐based perspective,” II. The setup is the same. We consider an activity, for example an investment, a plane trip or life in a country or continent. This activity can lead to some consequences C, which include hazardous events and their effects with respect to some values (e.g., health and the environment). For the probability perspective I, there are presumed “objective” probabilities p associated with C, describing properties of the activity considered. The probabilities p referred to are frequentist probabilities or propensities. Their existence and justification are based on a construction (real or thought‐constructed) of a large population of similar situations to the one considered. The p is nearly always unknown and is estimated in risk assessments, leading to, for example, an estimate of the probability of a defined event or an estimate of the expected consequences. Uncertainties of these estimates need to be addressed in order to conduct a full risk assessment. According to this perspective, risk r is conceptualized as r = (C,p).

For the uncertainty‐based perspective, risk R is conceptualized as R = (C,U), where U refers to uncertainty associated with C. Looking into the future, the consequences C are unknown and hence uncertain. The magnitude of the risk is characterized in the risk assessment by specifying potential consequences and using a method for measuring or describing the uncertainties. This leads to a general characterization of the form (C’,Q,K), where C’ is the specified consequences, Q is a measure or description of the uncertainties, and K is the knowledge that C’ and Q are based on. The description Q is commonly represented by subjective (knowledge‐based) probabilities P and associated judgments of the strength of the knowledge supporting P, SoK. By distinguishing between the actual consequences C and those specified in the risk assessment, C’, risk related to potential surprises can be addressed. A surprise here refers to the occurrence of an event that was not identified in the risk assessment.

Note the difference between p and P. The former probability is a measure of variation, whereas the latter expresses the assessor's uncertainty or degree of belief (Aven & Reniers, 2013; Lindley, 2006).

3.2. Analysis

Starting from the basis established in Section 3.1, we now look into possible interpretations of the “real risk” term. The interpretations are derived from the review in Section 2 in which the term is associated with risks that have been identified, risks that are of a certain size/extent, or risks for which there exists some evidence/knowledge. Based on this, we consider three possible interpretations of the term: a) the risk exists, b) the risk is of a certain size/extent, and c) there is evidence/knowledge supporting the judgment of the risk being of a certain size/extent. To reflect the various understandings of what constitutes the risk being of a certain extent, three risk/probability levels are specified: i), ii), and iii). A risk/probability level of i) refers to the risk being not negligible, ii) refers to the risk being moderately large, while level iii) points to the risk being large. Table 5 summarizes the main findings of the analysis. We refer to Section 4 for a discussion of other potential interpretations.

TABLE 5.

Interpretations of the term “real risk” for the two risk perspectives, I and II.

Risk perspective Interpretations of real risk (and the risk is real) Risk perspective I probability‐based Risk perspective II uncertainty‐based Comments
a) The risk exists (or risk exists) The risk can be meaningfully interpreted, and p is not negligible for one or more serious events (where p refers to a frequentist probability) P is not negligible for one or more serious events, and there is relatively strong knowledge supporting P (where P refers to a subjective/knowledge‐based probability) An adjustment of a) is to express that there is evidence/knowledge that the risk exists; refer to the difference between b) and c)
There is a question regarding what “relatively strong” means
b) The risk is i) not negligible, ii) moderately large, or iii) large p is i) not negligible, ii) moderately large, and iii) large for one or more serious events P is i) not negligible, ii) moderately large, and iii) large for one or more serious events, and the knowledge supporting P is relatively strong There is a question then regarding what “relatively strong” means
c) There is evidence/knowledge supporting the risk being i) not negligible, ii) moderately large, or iii) large There is evidence/knowledge supporting p being i) not negligible, ii) moderately large, and iii) large for one or more serious events P is i) not negligible, ii) moderately large, and iii) large for one or more serious events, and there is evidence/knowledge supporting the probability judgment There is a question regarding how strong the evidence/knowledge has to be
Relatively strong evidence/knowledge could be used instead of just evidence/knowledge; then there is a question regarding what “relatively strong” means
Open in a new tab

Consider first the interpretation that real risk refers to “the risk exists” or “risk exists,” referred to as type a). In the case of the probability‐based perspective I, we suggest explaining this by expressing that p is not negligible for one or more serious events. It is tacitly assumed that p can be meaningfully defined. Think about the coronavirus risk in early 2020. If someone then stated that a pandemic risk exists, it seems meaningful to interpret this as saying that there exists a non‐negligible frequentist probability p of severe consequences of the virus. What non‐negligible means would then be an issue. The term indicates that the probability is not so small that it can be ignored, but since we use the term “exists,” it is difficult to argue that the probability needs to be large. To apply perspective I, it must be possible to justify the existence of a frequentist probability (propensity). As mentioned in Section 3.1 and thoroughly discussed in the literature (e.g., Aven & Reniers, 2013; Singpurwalla, 2006), such probabilities cannot be introduced for unique events. It must be possible to make repeated constructions of the situation considered. For an event like “a new armed conflict in Europe,” we would conclude that such constructions cannot be meaningfully defined. Similarly, it is a challenge to define frequentist probabilities of pandemic occurrence events. Still, looking at populations of people, it may be possible to give an interpretation of such probabilities if talking about the frequentist probability that a person becomes seriously ill or dies due to a pandemic virus within a specific period of time, as this probability refers to the fraction of people in this population that become seriously ill or die. It is common just to presume the existence of such underlying frequentist probabilities. However, to properly understand the concept of risk, including real risk, it is necessary to reflect on their meaning and justification.

In most cases, p is unknown, and then there is an issue with p being non‐negligible or with claims about p being non‐negligible based on some evidence/knowledge. In the latter case, the question is how strong this evidence/knowledge needs to be. The statement that the risk exists (is real) expresses a judgment based on some evidence/knowledge. The judgment is subjective and needs to be evaluated on the basis of the evidence/knowledge provided.

For risk perspective II, the statement that risk exists can be given different interpretations, depending on whether “risk exists” is related to the risk concept (C,U) or its characterization (C’,Q,K). In the former case, there is basically no epistemic component beyond the fact that there are uncertainties about what the consequences will be (Ylönen & Aven, 2023), making an interpretation based on (C,U) alone not very informative. In Table 5, we have instead adopted an interpretation based on the risk characterization (C’,Q,K), expressing the notion that P is not negligible for one or more serious events and that there is relatively strong knowledge supporting P. Using this interpretation, the statement that risk exists has an epistemic component, reflected by P and the supporting evidence/knowledge. Again, the issue of the meaning of “relatively strong” in relation to the strength of the evidence/knowledge can be discussed.

Consider now the interpretation that real risk refers to b): The risk is i) not negligible, ii) moderately large, or iii) large. For risk perspective I, this is understood as p being i) not negligible, ii) moderately large, or iii) large for one or more serious events. Again, as p is commonly unknown, statements about p being non‐negligible or similar would be claims, not facts, about p. Is real risk then about p being non‐negligible, moderately large or large? Nothing in the wording of the term “real risk” points to the risk being large. Real risk is more about existence; hence, non‐negligibility is easier to justify than the two other categories. However, “non‐negligible” is a rather imprecise term and can be given different interpretations: “some risk” may seem to be a more plausible understanding than a “tiny risk.” For the term “real risk” to be of interest, the stakes involved would need to be of a certain extent, hence the restriction of one or more serious events, though this does not imply that the risk is large. Taking into account the uncertainties about what the true p is means that we also need to reflect on the evidence and knowledge supporting the claims about p. We are led to interpretation c), which states that there is evidence/knowledge supporting the judgment that the risk (p in case I) is not negligible (moderately large, large). For c), the key is the evidence/knowledge justifying that p is non‐negligible (moderately large, large).

For the risk perspective II, similar arguments can be provided for why i) non‐negligible is a more plausible interpretation than ii) and iii) regarding P. Interpretation b) requires that the knowledge supporting P is relatively strong, whereas interpretation c) just states that there is evidence/knowledge supporting the probability judgment. However, these formulations could be changed to make them similar. Hence, for this risk perspective, there is basically no difference between b) and c).

4. DISCUSSION

The analysis in Section 3 revealed differences in the interpretation between the two risk perspectives, but these are not as large as may be initially expected. The reason is that for both risk perspectives, “real risk” statements in a practical context need to be seen as judgments about the risk, not the risk itself. Evidence/knowledge is a key component for these judgments, not only the probabilities and their assessments. A main challenge is determining what level of evidence/knowledge is required. Certainly, it should not be weak. There must be some data, information, model results, test results, argumentation, and so forth, providing confidence about the statements. As such, the evidence/knowledge should be relatively strong. What defines “relatively strong” evidence/knowledge is a judgment, and will depend on the assessor, as well as on the relevant context or situation at hand. Different types of schemes have been established for evaluating the strength of the evidence/knowledge; see, for example, Askeland et al. (2017), Flage and Aven (2009), and Berner and Flage (2016), which could be useful for this purpose.

In Section 3, we have considered some potential interpretations of the term “real risk” (and “the risk is real”). These interpretations have been generated by the authors of this study, using brainstorming techniques as well as systematic reviews of how people have actually applied the terms. We looked for ways to make the actual wording plausible, adopting a risk science framework with precision on concepts and terminology.

Section 1 referred to two ways of using the term “real risk.” The first type is used when making a statement about a presumed underlying real risk, whereas the second is used when classifying a risk as real, indicating that other risks could be non‐real. The analysis in Section 3 is restricted to the second type of use. When it comes to the former type, the framework is the risk perspective I, which is based on the idea that an underlying true, objective risk (C,p) exists. A statement like “the real risk is so close to zero that” (Wallace, 1994, p. 144) is to be understood as p being “so close to zero that …,” and a statement like “the real risk is probably underestimated” (Navarro et al., 1998, p. 251) as the estimate of p (in relation to an undesirable event) probably being too low.

The “real risk” term has particular relevance when seen in the context of the social amplification of risk framework (SARF). This framework, introduced by Kasperson et al. (1988), provides an understanding of how public perceptions of risk or risk events are subject to a complex interplay of social, psychological, and cultural factors that can serve to either amplify or attenuate the perceived risks associated with a particular event or hazard. According to the SARF, the amplification or attenuation of risk perceptions is significantly influenced by the media and public discourse. SARF explains how risks or risk events assessed by experts as low risks are amplified and produce public concern. The reference for the experts’ risk communication has typically been an objective, real risk (defined by “objective” probabilities p), in contrast to people's subjective risk perception. Using such a reference, amplification is expected, as risk cannot be properly conveyed without also reporting uncertainties and evidence/knowledge, as discussed above. People tend to dismiss social amplification using the argument that “it's about people's perceptions of risks, not real risks.”  However, as clarified in this study, real risk is also about judgments, and these judgments do not represent an objective truth. Using the insights of the present study, the risk communication can be improved by clarifying that to understand what real risk means and what the real risk is, one needs to also reflect on and address uncertainties, evidence and knowledge, and not only probabilities and probability estimates.

We have restricted attention to two main categories of risk perspectives, I and II. These two categories represent two fundamentally different views on how to conceptualize and understand risk. The probability‐based perspective, I, presumes the existence of a true, objective risk, which is to be estimated in the risk assessment. The alternative, the uncertainty‐based perspective, II, rejects the idea of such underlying true probabilities: uncertainty is a main component of risk, not probability. Probability is just a tool we use to express uncertainties, but it is not a perfect tool. The knowledge—and its strengths and weaknesses—supporting the probabilities is as important as the probabilities. Nonetheless, as discussed above, the analysis in Section 3 showed that the term “real risk” can be used with relatively similar interpretations for the two risk perspectives.

Other risk perspectives also exist, including expected value‐based perspectives and risk definitions based on events/consequences (refer to Aven, 2012 and Aven et al., 2011 and references therein). However, as discussed in Aven et al. (2011), these two perspectives cover a broad set of commonly used definitions. A category not included is perspectives, where risk is defined by its measurement, for example, through subjective probabilities P. We could refer to such a risk perspective as (C,P), using the terminology from Section 3. Interpretations of a), b), and c) can also be provided for such risk perspectives, but problems would occur if risk were limited to the probabilities and the evidence/knowledge were not given weight. To argue that a risk is real because a subjective probability P is specified as non‐negligible is not very convincing, unless the evidence/knowledge supporting P is not included in the argumentation.

5. CONCLUSION

The risk perspective II can always be used, and from the above analysis, we are led to the conclusion and recommendation that real risk should be interpreted as expressing that there is some risk, and the knowledge supporting this statement is relatively strong. Alternatively, using probability, the recommended interpretation is as follows: The probability P for one or more serious events is not negligible—it is noteworthy/significant—and there is relatively strong evidence/knowledge supporting this probability judgment. The first part of this interpretation is general and can also be used for risk perspective I. For this perspective, real risk can also be understood as: There is relatively strong evidence/knowledge supporting the judgment that p is not negligible (noteworthy/significant) for one or more serious events.

What constitutes a probability being noteworthy/significant is essentially a judgment, and closely related to discussions concerning risk acceptance and risk tolerance. For example, different stakeholders and groups may have varying thresholds for what constitutes an acceptable probability level based on their values, beliefs, and preferences. A probability that is deemed significant by one group may be considered tolerable or even negligible by another. Furthermore, as emphasized by Renn (1998), these judgements are also influenced by perceptions of risk, as “effects and likelihood are enriched by the perceived presence of situational and risk specific characteristics that depend on properties such as the degree of perceived personal control, the perception of a social rather than an individual risk, or the familiarity of the risk situation” (1998, p. 60).

Through the conceptual analysis of the term “real risk” in the present study, we have provided interpretations of the term that are grounded in risk science knowledge. However, while these interpretations can be justified and used to explain the term “real risk,” we are skeptical regarding the use of this term in both scientific contexts and elsewhere. The main issue is that the practical applications of the term often leave its underlying interpretation unspecified. Consequently, as the term is rarely supplemented with a clear interpretation, its use can easily lead to misunderstandings and poor risk communication. Thus, while our analysis offers a clear framework for interpreting “real risk,” the most prudent approach to describing and communicating risk is to use risk science knowledge, which provides precise and clear terminology for conceptualizing and characterizing risk. This also applies to statements about risk in news media and everyday language. In these contexts, the use of the term “real risk” could, in some cases, be employed for rhetorical effect rather than to denote a specific level of risk. Nevertheless, any statement about risk could have implications for the understanding, communication, and handling of the risk, and rather than dismissing the term as idiomatic, we argue that there is a need to establish a clear understanding of its meaning and use within the framework of risk science. If one would like to say that the risk is high, then this needs to be expressed using proper risk science theory. Similarly, if the goal is to express that the likelihood is small, large, and so forth and that the knowledge supporting these statements is based on strong or weak knowledge, then risk should be characterized using the theories and guidance for how to do this, using probabilities, knowledge strength judgments, so forth according to the standards provided by contemporary risk science knowledge.

ACKNOWLEDGMENTS

The authors are grateful to four anonymous reviewers for their useful comments and suggestions on earlier versions of this study.

Flage, R. , Aven, T. , & Glette‐Iversen, I. (2025). On the use of the term “real risk”. Risk Analysis, 45, 214–222. 10.1111/risa.15073

REFERENCES

  1. Askeland, T. , Flage, R. , & Aven, T. (2017). Moving beyond probabilities—Strength of knowledge characterisations applied to security. Reliability Engineering & System Safety, 159, 196–205. 10.1016/j.ress.2016.10.035 [DOI] [Google Scholar]
  2. Aven, T. (2012). The risk concept—Historical and recent development trends. Reliability Engineering & System Safety, 99, 33–44. 10.1016/j.ress.2011.11.006 [DOI] [Google Scholar]
  3. Aven, T. (2013). On the meaning of a black swan in a risk context. Safety Science, 57, 44–51. 10.1016/j.ssci.2013.01.016 [DOI] [Google Scholar]
  4. Aven, T. , & Reniers, G. (2013). How to define and interpret a probability in a risk and safety setting. Safety Science, 51(1), 223–231. 10.1016/j.ssci.2012.06.005 [DOI] [Google Scholar]
  5. Aven, T. , & Renn, O. (2009). On risk defined as an event where the outcome is uncertain. Journal of Risk Research, 12(1), 1–11. 10.1080/13669870802488883 [DOI] [Google Scholar]
  6. Aven, T. , Renn, O. , & Rosa, E. A. (2011). On the ontological status of the concept of risk. Safety Science, 49(8), 1074–1079. 10.1016/j.ssci.2011.04.015 [DOI] [Google Scholar]
  7. Berner, C. L. , & Flage, R. (2016). Comparing and integrating the NUSAP notational scheme with an uncertainty based risk perspective. Reliability Engineering & System Safety, 156, 185–194. 10.1016/j.ress.2016.08.001 [DOI] [Google Scholar]
  8. Boholm, M. , Möller, N. , & Hansson, S. O. (2016). The concepts of risk, safety, and security: Applications in everyday language. Risk Analysis, 36(2), 320–338. 10.1111/risa.12464 [DOI] [PubMed] [Google Scholar]
  9. Boseley, S. (2014, July 30). Ebola virus: What is the real risk to the UK? The Guardian . https://www.theguardian.com/society/2014/jul/30/ebola‐what‐is‐the‐real‐risk
  10. Brito, M. , & Griffiths, G. (2016). A Bayesian approach for predicting risk of autonomous underwater vehicle loss during their missions. Reliability Engineering & System Safety, 146, 55–67. [Google Scholar]
  11. Cadar, D. (2017). A life course approach to dementia prevention. Journal of Aging and Geriatric Medicine, 1(2), 1–6. [Google Scholar]
  12. Cirillo, P. , & Taleb, N. N. (2020). Tail risk of contagious diseases. Nature Physics, 16(6), 606–613. 10.1038/s41567-020-0921-x [DOI] [Google Scholar]
  13. Cook, J. (2020, August 3). The real risk of downloading Chinese apps on to your phone. The Telegraph. https://www.telegraph.co.uk/technology/2020/08/03/real‐risk‐downloading‐chinese‐apps‐phone/
  14. Côté, J. N. , Germain, M. , Levac, E. , & Lavigne, E. (2024). Vulnerability assessment of heat waves within a risk framework using artificial intelligence. Science of the Total Environment, 912, 169355. 10.1016/j.scitotenv.2023.169355 [DOI] [PubMed] [Google Scholar]
  15. Cox, L. A. (2007). Regulatory false positives: True, false, or uncertain? Risk Analysis, 27(5), 1083–1086. 10.1111/j.1539-6924.2007.00975.x [DOI] [PubMed] [Google Scholar]
  16. Flage, R. , & Aven, T. (2009). Expressing and communicating uncertainty in relation to quantitative risk analysis. Reliability & Risk Analysis: Theory & Application, 2(13), 9–18. [Google Scholar]
  17. Flage, R. , & Aven, T. (2015). Emerging risk—Conceptual definition and a relation to black swan type of events. Reliability Engineering & System Safety, 144, 61–67. 10.1016/j.ress.2015.07.008 [DOI] [Google Scholar]
  18. Freudenburg, W. R. (1988). Perceived risk, real risk: Social science and the art of probabilistic risk assessment. Science, 242(4875), 44–49. 10.1126/science.3175635 [DOI] [PubMed] [Google Scholar]
  19. Galatolo, R. , & Lazzeri, R. (2016). Experiments and model predictions for fatigue crack propagation in riveted lap‐joints with multiple site damage. Fatigue & Fracture of Engineering Materials & Structures, 39(3), 307–319. 10.1111/ffe.12354 [DOI] [Google Scholar]
  20. Gleick, P. H. (2006). Water and terrorism. Water Policy, 8(6), 481–503. 10.2166/wp.2006.035 [DOI] [Google Scholar]
  21. Hanson, T. (2023, September 18). Prison escapes in America: How common are they and what's the real risk? CBS News. https://www.cbsnews.com/news/prison‐escapes‐in‐america‐how‐common‐are‐they‐and‐whats‐the‐real‐risk/ [Google Scholar]
  22. Hermansson, H. (2012). Defending the conception of “objective risk”. Risk Analysis, 32(1), 16–24. 10.1111/j.1539-6924.2011.01682.x [DOI] [PubMed] [Google Scholar]
  23. Holton, G. A. (2004). Defining risk. Financial Analysts Journal, 60(6), 19–25. 10.2469/faj.v60.n6.2669 [DOI] [Google Scholar]
  24. Howarth, C. I (1988). The relationship between objective risk, subjective risk and behaviour. Ergonomics, 31(4), 527–535. 10.1080/00140138808966697 [DOI] [Google Scholar]
  25. Iversen, M. (2023). Det en reell risiko for at den kinesiske økonomien havner i en resesjon. DN.no. https://www.dn.no/utenriks/kina/‐det‐en‐reell‐risiko‐for‐at‐den‐kinesiske‐okonomien‐havner‐i‐en‐resesjon/2‐1‐1500867
  26. Jansen, T. , Claassen, L. , Van Kamp, I. , & Timmermans, D. R. M. (2019). Understanding of the concept of ‘uncertain risk’. A qualitative study among different societal groups. Journal of Risk Research, 22(5), 658–672. 10.1080/13669877.2018.1503614 [DOI] [Google Scholar]
  27. Kasperson, R. E. , Renn, O. , Slovic, P. , Brown, H. S. , Emel, J. , Goble, R. , Kasperson, J. X. , & Ratick, S. (1988). The social amplification of risk: A conceptual framework. Risk Analysis, 8(2), 177–187. [Google Scholar]
  28. Kozine, I. (2018). On a true value of risk. International Journal of General Systems, 47(3), 263–277. 10.1080/03081079.2017.1415896 [DOI] [Google Scholar]
  29. Kumar, D. (2013, October 28). Is investing in mutual funds a risky affair? The Economic Times . https://economictimes.indiatimes.com/opinion/et‐commentary/is‐investing‐in‐mutual‐funds‐a‐risky‐affair/articleshow/24804638.cms?from=mdr
  30. Lagarias, G. (2024, February 9). UK recession is a real risk—But ‘soft landing’ is more likely. Evening Standard. https://www.standard.co.uk/business/uk‐recession‐gdp‐inflation‐soft‐landing‐prices‐jobs‐economy‐2024‐stagflation‐b1138146.html
  31. Lindley, D. V. (2006). Understanding uncertainty. Wiley‐Interscience. [Google Scholar]
  32. Lu, X. , Zhang, Z. , Hiraki, T. , Takeda, O. , Zhu, H. , Matsubae, K. , & Nagasaka, T. (2022). A solid‐state electrolysis process for upcycling aluminium scrap. Nature, 606(7914), 511–515. 10.1038/s41586-022-04748-4 [DOI] [PubMed] [Google Scholar]
  33. McKie, R. (2022, February 20). London flooding poses ‘significant risk’ unless immediate action taken. The Guardian. https://www.theguardian.com/uk‐news/2022/feb/20/london‐flooding‐poses‐significant‐risk‐unless‐immediate‐action‐taken
  34. Merkow, M. S. , & Breithaupt, J. (2014). Information security: Principles and practices. Pearson Education. [Google Scholar]
  35. Mitchell, R. J. , Williamson, A. M. , & Olivier, J. (2010). Estimates of drowning morbidity and mortality adjusted for exposure to risk. Injury Prevention, 16(4), 261–266. 10.1136/ip.2009.024307 [DOI] [PubMed] [Google Scholar]
  36. Nartea, R. , Mitoiu, B. I. , & Nica, A. S. (2019). Correlation between pregnancy related weight gain, postpartum weight loss and obesity: A prospective study. Journal of Medicine and Life, 12(2), 178–183. 10.25122/jml-2019-0015 [DOI] [PMC free article] [PubMed] [Google Scholar]
  37. Navarro, F. , Taourel, P. , Michel, J. , Perney, P. , Fabre, J. M. , Blanc, F. , & Domergue, J. (1998). Diaphragmatic and subcutaneous seeding of hepatocellular carcinoma following fine‐needle aspiration biopsy. Liver, 18(4), 251–254. 10.1111/j.1600-0676.1998.tb00161.x [DOI] [PubMed] [Google Scholar]
  38. Omran, H M. , & Almaliki, M S. (2020). Influence of NAD+ as an ageing‐related immunomodulator on COVID 19 infection: A hypothesis. Journal of Infection and Public Health, 13(9), 1196–1201. 10.1016/j.jiph.2020.06.004 [DOI] [PMC free article] [PubMed] [Google Scholar]
  39. Parker‐Pope, T. (2020, May 28). What's the risk of catching coronavirus from a surface? The New York Times . https://www.nytimes.com/2020/05/28/well/live/whats‐the‐risk‐of‐catching‐coronavirus‐from‐a‐surface.html
  40. Rankin, J. , Harding, L. , & Borger, J. (2022, January 12). Nato chief warns of ‘real risk of conflict’ as talks with Russia over Ukraine end. The Guardian. https://www.theguardian.com/world/2022/jan/12/nato‐chief‐warns‐of‐real‐risk‐of‐conflict‐as‐talks‐with‐russia‐over‐ukraine‐end
  41. Readfearn, G. , Evershed, N. , & Nicholas, J. (2022, March 1). What caused the ‘rain bomb’? How the unprecedented Queensland and NSW 2022 floods unfolded. The Guardian. https://www.theguardian.com/australia‐news/2022/mar/01/how‐the‐unprecedented‐queensland‐and‐nsw‐floods‐unfolded
  42. Renn, O. (1998). Three decades of risk research: Accomplishments and new challenges. Journal of Risk Research, 1(1), 49–71. 10.1080/136698798377321 [DOI] [Google Scholar]
  43. Sağir, M. , Kaplan, M. , Tanoğlu, A. , & Demirel, F. (2021). Relationship between vitamin D levels and mortality rates of critically ill patients in intensive care unit. Anatolian Current Medical Journal, 3(2), 171–175. 10.38053/acmj.901628 [DOI] [Google Scholar]
  44. Singpurwalla, N. D. (2006). Reliability and risk: A bayesian perspective. John Wiley & Sons. [Google Scholar]
  45. Sridhar, D. (2021, May 17). If we loosen restrictions too early, there is a real risk of a third wave in the UK. The Guardian. https://www.theguardian.com/commentisfree/2021/may/17/third‐wave‐risk‐uk‐vaccinated‐covid‐variant
  46. Storrs, C. (2015, December 02). A false positive mammogram is a real risk factor for breast cancer, study finds . CNN. https://www.cnn.com/2015/12/02/health/false‐positive‐mammogram‐cancer‐risk/index.html [Google Scholar]
  47. Tartara, F. , Cofano, F. , Zenga, F. , Boeris, D. , Garbossa, D. , & Cenzato, M. (2020). Are we forgetting non‐COVID‐19‐related diseases during lockdown? Acta Neurochirurgica, 162(7), 1501–1501. 10.1007/s00701-020-04385-8 [DOI] [PMC free article] [PubMed] [Google Scholar]
  48. Thompson, P. B. (1990). Risk objectivism and risk subjectivism: When are risks real symposium: Public participation in risk management. Risk: Issues in Health and Safety, 1(1), 3–22. [Google Scholar]
  49. Wallace, A. (1994). Soil science, pesticides, and risk analyses. Communications in Soil Science and Plant Analysis, 25(1–2), 143–148. 10.1080/00103629409369021 [DOI] [Google Scholar]
  50. Ylönen, M. , & Aven, T. (2023). A framework for understanding risk based on the concepts of ontology and epistemology. Journal of Risk Research, 26(6), 581–593. 10.1080/13669877.2023.2194892 [DOI] [Google Scholar]

Articles from Risk Analysis are provided here courtesy of Wiley

RESOURCES