Brief report - “Every little thing gonna be all right” (at least for me): Dispositional optimists display higher optimistic bias for infection during the Italian COVID-19 outbreak

Abstract

Dispositional optimism (DO) and optimistic bias (OB) in risk perception are two distinct phenomena and previous studies about their reciprocal relationship report contrasting results. In the present study, we focused on the relationship between DO and OB when reporting the personal and the other persons' risk about COVID-19. We hypothesized that, when facing a largely uncontrollable risky situation (like the recent pandemic), dispositional optimists would defensively increase their OB about the current risks. A convenience sample of 414 Italian participants aged 18 or older were recruited. They completed a questionnaire investigating past protective behaviors, DO, perceived personal and other persons' COVID-19-related risk. Results of the mixed regression model showed that more optimistic people were more likely to underestimate their COVID-19 personal risk over the other's person risk. These results shed light on the relationship between different forms of optimism and provide useful insight about the potential implications of risk communication approaches to face the current pandemic.

Highlights

• •Dispositional optimism and optimistic bias are two distinct phenomena.
• •They are both linked to risk perception.
• •More dispositionally optimistic people report higher optimistic bias for COVID-19.

1. Introduction

Some people are more optimistic than others. Dispositional optimism (DO) is a personality construct that refers to generalized outcome expectancies that good things will happen even in the face of obstacles (Carver et al., 2010). DO is linked to decision behavior, motivation and self-regulation (Armor & Taylor, 1998); dispositional optimists usually perceive lower risks than pessimists do (e.g., Scheier et al., 1994; Monzani et al., 2015).

When people evaluate the risk that negative events will happen, they might differ in estimating the risk for themselves vs the risk for other individuals. In this regard, optimistic bias (OB) can be defined as a mistaken belief that makes us think that our chances of experiencing a negative event are lower than those of our peers (Weinstein, 1980). OB is often associated with increase controllability (Ruthig et al., 2007) and is stronger for negative events (Shepperd et al., 2002) than for the positive ones, leading to more risky behaviors (Shepperd et al., 2002; Masiero et al., 2015; Masiero et al., 2018). In this sense, even if they are related, DO and OB are considered distinct phenomena, with different impact on cognition, well-being and behaviors (Radcliffe & Klein, 2002).

Regarding the relationship between DO and OB, previous studies report contrasting results. Overall, DO generally leads to lower risk perception but not always to more OB (Radcliffe & Klein, 2002). For example, DO showed a positive effect on OB in people facing hurricanes Katrina and Rita (Trumbo et al., 2011), while in other conditions like HIV/AIDS, their relationship appeared quite weak (e.g., Radcliffe & Klein, 2002).

In the context of the COVID-19 pandemic, OB has been investigated by two recent studies (Kuper-Smith et al., 2020; Raude et al., 2020) showing that people estimated the probability of getting infected lower for themselves than for someone similar to them. However, no studies have investigated the relationship between DO and OB in the context of the recent pandemic.

In the present work, we suggest that the two phenomena are not completely independent. Indeed, we argue that, under severe circumstances, DO may be a fertile ground for OB. Specifically, we hypothesize that, when facing largely ineliminable and uncontrollable risks (like the COVID-19 pandemic), dispositional optimists, rather than admitting the existence of a generalized risk of negative events, would judge their risk lower than the others' risk. In other words, when thinking about the risks related to COVID-19, optimists would show a stronger OB preserving at least their control over the situation and part of the original optimistic perspective.

2. Method

2.1. Participants and procedure

A convenience sample of 446 participants was recruited in Italy from 12th March to 3rd April 2020. Data collection started almost three weeks after Italian Patient One being tested positive for COVID-19 (i.e., 21st February) and two days after the national quarantine (i.e., 10th March). The study and the link to participate were advertised on social networks (e.g., Facebook); the survey was administered online through Qualtrics. Thirty-two participants were excluded from analysis due to incomplete data. The remaining 414 participants were mainly women (70.3% female) with a mean age of 39.03 years (SD = 16.00, range = 18–99).

The study was conducted in compliance with the Declaration of Helsinki ethical standards. Informed consent was obtained from all participants.

Self-report measures are reported in Supplemental material A. DO was computed as the mean of the responses to the six items of the Italian version of the Life Orientation Test Revised (Scheier et al., 1994; Italian version: Steca et al., 2015). Participants indicated the extent to which they agreed with each of the items on a 5-point Likert-type scale ranging from strongly disagree to strongly agree (α = 0.83).

Individual perceived risk (IPR) was assessed with four items asking participants to indicate their short- (i.e., in the following two weeks) and long-term risk (i.e., in the following six months) of getting infected by the new coronavirus, as well as their own likelihood of having severe health consequences, such as pneumonia and being hospitalized due to COVID-19. For each statement, participants were asked to indicate their own likelihood on a 0–100 slider scales ranging from “It is very unlikely” to “It is very likely”. IPR was computed as the mean of the four ratings (α = 0.72).

Other person's risk (OPR) was measured by asking participants to think about an average person of their age and sex and then estimate this person's risk by considering the same four statements used for measuring IPR. OPR was computed as the mean of the four ratings (α = 0.76). The order of the assessment of IPR and OPR was randomized across participants.

Finally, control variables were assessed as well. First, official data (i.e., number of confirmed COVID-19 cases updated to the day of survey completion) from the Italian Prime Minister1 were retrieved. Second, past preventive behaviors (PPB) towards COVID-19 infection were assessed by asking people to rate their adherence to COVID-19 guidelines (e.g., the Italian Ministry of Health, 2020). Participants were asked to rate the frequency of 11 precautionary measures in the previous 2 weeks on a 5-point Likert-type scale ranging from never to always. PPB was computed as the mean of the responses to the 11 statements (α = 0.73). Participants' age and gender were also measured.

3. Results

Preliminary correlational analyses highlighted a strong and positive association between IPR and OPR (r = 0.74, p < .001). The remaining correlations were all weak or not significant (correlational analyses are presented in Supplemental material B).

We performed a multilevel regression model with Jamovi 1.1.9 (Gallucci, 2019; The Jamovi Project, 2019). In this model, risk estimation (IPR and OPR) was considered as the dependent variable, the target of the risk estimation (e.g., the participant or the average person) as a within-subject factor, and dispositional optimism as a between-subject predictor. We also included a random intercept and a second-order interaction term between DO and the target of the risk estimation allowing to test whether any differences between IPR and OPR (i.e., OB) would be conditional to the level of DO. Participant's age, gender and PPB, and the number of confirmed COVID-19 cases were entered in the model to control for any differences in risk estimation. All continuous independent variables (age, number of COVID-19 cases, and DO) were grand-mean centered.

The interaction between DO and the target of the risk estimation was related to levels of estimated COVID-19 risk (b = 1.83, SE = 0.87, t = 2.11, p = .036) (the overall multilevel regression model is reported in Supplemental material C). A simple slope analysis was performed to test the effect of the target of the risk estimation at different levels of DO. As reported in Fig. 1, people with low (b = −0.24, SE = 0.86, p = .778) or average DO (b = 1.04, SE = 0.61, p = .088) did not make different estimation of COVID-19 risk for themselves or the average person. On the contrary, more optimistic people (b = 2.32, SE = 0.86, p = .007) underestimate their COVID-19 IPR (M = 29.7, SD = 1.22) over the OPR (M = 32.0, SD = 1.22).

Fig. 1.

Illustration of the simple-slope analyses for COVID-19 risk for the target of the risk estimation-by-DO interaction.

4. Discussion

In this brief report, we investigated the relationship between DO and OB for the risks connected with the COVID-19 outbreak. Results demonstrated that people with more DO report OB about COVID-19, meaning that they would consider the risk lower for themselves than for other persons. On the contrary people with low or average DO do not show OB by displaying similar IPR and OPR. Our interpretation is that dispositional optimists, rather than admitting the pervasive risk of negative events, tend to defensively keep low their personal risk (if compared to the others). In other words, when thinking about the risks of COVID-19, dispositional optimists would use OB to preserve at least part of the control over the events. This phenomenon would be consistent with the adaptive controllability which accompanies OB (Ruthig et al., 2007).

Some limits should be recognized. First, as the study design was correlational, it is not possible to assess causal relationships between variables. However, our interpretation of the relationship between DO and OB is consistent with the general relationship between personality and situational constructs. Second, further studies are needed because the existing literature on OB demonstrated that findings may vary as a function of the specific health risks being examined (Nezlek & Zebrowski, 2001). Our study regards a specific context of low perceived controllability. We suggest that the interaction that we found between DO and OB could be present also in situations characterized by low controllability, but it is not generalizable to every context.

Despite such limitations, our contribution has some interesting implications particularly relevant during this pandemic: it shed light on some discussed aspects of the relationship between these two constructs, suggesting that under certain conditions DO may reinforce OB. From a practical point of view, it is important to implement effective communication and debiasing strategies to modify people's misconceptions about the risk of COVID-19. This should be especially relevant for more optimistic people.

Specifically, such results may have a practical implication on the development of future messages provided to the population. For example, in Italy, in March 2020 a bottom-up campaign adopted the optimistic hashtag “everything will be alright” (#andràttutobene), to give hope to people. However, it is necessary to remember that an excess of optimism, especially when it suggests that someone is less at risk than others, could prevent individuals from taking precautions against the infection spread, with severe consequences especially on vulnerable people around (Van Bavel et al., 2020).

Similarly, also the message that “COVID-19 may concern specific at-risk group” (like the elderly; see: Utych & Fowler, 2020), even if communicated with the goal of limiting panic, may strengthen the younger people's OB, with negative consequences on preventive behaviors and, consequently, on the pandemic spread. Three decades of HIV prevention teach us that promoting safe behaviors in the entire population should be preferred to the creation of “at-risk groups” (Mazzoni et al., 2017; Schiller et al., 1994) and our hope is that we have learned something from this lesson.

The following are the supplementary data related to this article.

Supplemental material A

Self-report measures assessing dispositional optimism, individual perceived risk, other person's risk, and past preventive behaviors towards COVID-19 infection.

Supplemental material B

Correlational analysis among DO, IPR, OPR, and PPB.

Supplemental material C

Results of the multilevel regression model.

CRediT authorship contribution statement

Dario Monzani: Conceptualization, Methodology, Investigation, Formal analysis, Data curation, Writing - original draft, Writing - review & editing. Alessandra Gorini: Conceptualization, Methodology, Investigation, Writing - original draft, Writing - review & editing. Davide Mazzoni: Conceptualization, Methodology, Investigation, Writing - original draft, Writing - review & editing. Gabriella Pravettoni: Conceptualization, Methodology, Supervision, Writing - review & editing.