In December 2019, cases of an acute respiratory disease of unknown cause emerged in Wuhan, China. A novel coronavirus was identified as the cause of the disease on the 7th of January 2020 and is now referred to as “COVID‐19.” The virus spread rapidly across the globe, leading the World Health Organization (WHO) to pronounce the outbreak as a pandemic on the 11th of March 2020 (World Health Organization, 2020a). As of mid‐September 2020, there have been over 28.6 million reported cases of the virus and 917,000 deaths globally—a major public health crisis (World Health Organization, 2020b). In response to the pandemic, and in the absence of biomedical interventions, a global public health campaign has been launched to prevent the spread of the virus by encouraging physical distancing, frequent handwashing, following respiratory hygiene, seeking medical care early, avoiding face touching, and following advice given by health authorities (World Health Organization, 2020a). Until biomedical protection against the virus is available (e.g., a vaccine), several social and behavioural interventions have been implemented to curtail its spread (Anderson, Heesterbeek, Klinkenberg, & Hollingsworth, 2020).
While biomedical disease control and prevention is acutely needed, the WHO has also stressed the importance of behavioural science in informing pandemic response measures (Kluge, 2020). During the relatively recent severe acute respiratory syndrome (SARS) outbreak in 2003 which affected many countries in Asia—referred to as the first pandemic of the 21st century (LeDuc & Barry, 2004)—social psychologists contributed to theory and research concerning the pandemic. For example, in the Asian Journal of Social Psychology’s special issue “The Psychology of SARS,” published in March 2004, social psychologists focused on the role of psychological factors in people’s responses to the SARS outbreak (Cheng & Tang, 2004). The special issue highlighted how social psychological theory and research might inform public health policies and communication in future pandemics by providing an understanding of how cognitive and affective factors (e.g., beliefs and attitudes) implicate risk perceptions and adoption of preventative health behaviours (Cheung, 2004). For example, Tam, Lau, and Chiu (2004) showed how nonpractisers of preventative behaviours (e.g., wearing a face mask) tended to underestimate the prevalence and importance of prosocial concerns in the preventative behaviours commonly adopted among practisers. Furthermore, Lee‐Baggley, DeLongis, Voorhoeave, and Greenglass (2004) found that people who responded empathically to others as a coping strategy engaged in effective precautionary behaviours to a greater extent (e.g., handwashing).
Turning to the present, many journals have called for psychologists to conduct research concerning COVID‐19, promoting the contributions that psychology can make to informing social and behavioural responses (see Arden & Chilcot, 2020; Jonas & Cesario, 2020; Smith & Gibson, 2020; Sokolowska, Ayton, & Brandstätter, 2020). Extensive expositions of theory and research highlighting how the field can support responses to the pandemic have also been coordinated (Van Bavel et al., 2020; see also Rosenfeld et al., 2020). As of mid‐September 2020, over 580 manuscripts concerning COVID‐19 had been uploaded on PsyArXiv (a preprint service for the psychological sciences; www.psyarxiv.com), with over 210 indexed as “social and personality psychology” (Syed, 2020).
Although Van Bavel et al. (2020) offered valuable insights into how the social and behavioural sciences can inform responses to the COVID‐19 pandemic, it does not focus on establishing consensus about the contributions that the social and behavioural sciences can make to informing pandemic responses, and research priorities to this end. The current research was designed with this specific purpose in mind. The Delphi method was employed as it not only garners expert opinion about directions for future‐orientated research but also establishes consensus about priorities (Rowe & Wright, 1999). Consensus, or collective agreement, is particularly important when it comes to informing and managing novel and pressing issues because it may consolidate research efforts, even among researchers with limited expertise of a given issue, and thereby contribute towards the advancement of a cohesive evidence‐base when the impact on policy and practice is potentially high and immediate. While strategic insight and recommendations of this nature and scope arguably should be formulated by scientific and professional organisations, there were no such guidelines available at the time the current research was conceptualised. To our knowledge, there are no such guidelines available for social psychology to date.
The aims of this study were thus to gauge opinion and explore consensus in a panel comprising experts of social psychology on (a) how social psychological theory and research can contribute to inform knowledge and management of pandemics, (b) what research topics social psychology and social psychologists should prioritise to inform knowledge and management of pandemics, and (c) what strategies social psychology and social psychologists should implement to conduct research that will inform knowledge and management of pandemics.
The Delphi Process
The Delphi method involves an iterative process to synthesise expert opinion and, if possible, reach consensus on a topic, in two or more rounds, while also highlighting differences in opinion. The method is often employed when examining a complex topic where there is little or no definitive evidence, and where opinion is important, to inform theory, research, policy, and/or practice (Hasson, Keeney, & McKenna, 2000; Hsu & Sandford, 2007; Thangaratinam & Redman, 2005; Wilkes, 2015). A two‐round e‐Delphi study was conducted between the 31st of March and 6th of May 2020 using an expert panel of social psychologists. The survey rounds were hosted on the online survey platform Qualtrics (www.qualtrics.com). The study was preregistered in the Open Science Framework (https://tinyurl.com/ycru8zje) and received ethical approval from the Psychology Faculty Research Ethics Committee at Keele University (Reference No. PS‐200126).
In Round 1, participants were asked to rate their agreement with two statements on 5‐point Likert scales (1 = Strongly disagree, 2 = Disagree, 3 = Undecided, 4 = Agree, 5 = Strongly agree): (a) Social psychological theory and/or research can make a significant contribution to knowledge about pandemics; and (b) Social psychological theory and/or research can have a significant impact on how pandemics are managed. Participants were also asked to respond to three open‐ended questions: (a) How can social psychological theory and/or research contribute to inform knowledge and management of pandemics? (b) What research topics should social psychology/social psychologists prioritise to (further) inform knowledge and management of pandemics? (c) What strategies should social psychology/social psychologists implement to conduct research that will (further) inform knowledge and management of pandemics? The wording of the open‐ended questions does not assume that social psychology is positioned to inform knowledge about virological and immunological dimensions of pandemics; the reference to knowledge pertains to the multidisciplinary body of knowledge concerning pandemics, including its social and behavioural dimensions.
Free‐text responses were made available to participants at the end of the survey round to encourage comments and suggestions concerning the content. Participant demographics were also collected, including age, gender, country of residence, year of PhD award, year of professorship, institutional affiliation, and research expertise.
In Round 2, participants rated their agreement with 100 statements derived from the responses to the three open‐ended questions in Round 1 regarding contributions, priorities, and strategies. Responses were recorded using 5‐point Likert scales (1 = Strongly disagree, 2 = Disagree, 3 = Undecided, 4 = Agree, 5 = Strongly agree). Participants were again given the opportunity to provide comments and suggestions concerning the survey content. Participants were given up to 15 days to respond to each round.
As the Delphi method is a group decision mechanism, statistically significant samples are not required (Okoli & Pawlowski, 2004). A purposive sample of 252 international experts in social psychology was identified through editorial boards of prominent journals and executive committees of prominent academic societies in the field; all were listed as professors on the websites of their affiliated universities.1 Particular consideration was given to ensure that the invited panel was balanced in terms of gender and geographical region (i.e., WEIRD vs. non‐WEIRD). The invitation e‐mail outlined the research aims and process with a link to the first survey round, which also contained the information sheet and consent form. To complete the Delphi process, participants were required to provide their informed consent and complete both rounds. Participants were informed that £10 would be donated to the WHO COVID‐19 Solidarity Response Fund (https://www.who.int/emergencies/diseases/novel‐coronavirus‐2019/donate) for each panel member who completed both rounds. The incentive was an attempt to increase the uptake of the study and reduce attrition rates between rounds—a common impediment to the Delphi method (Hsu & Sandford, 2007).
To transform responses to the open‐ended questions in Round 1 into statements to be used as survey items in Round 2, the responses were processed, organised, and analysed using content analysis in line with guidelines provided by Keeney, Hasson, and McKenna (2011) and Graneheim and Lundman (2004). This involved identifying and categorising similar responses together into themes, or topic area; the process involved condensing and collapsing similar responses. The focus of the analysis was therefore on verbatim (or manifest) rather than latent meanings, and responses were aggregated under topic areas based on their verbatim content—the labels of the topic areas reflect the key contents of the respective responses. IBM SPSS Statistics (Version 26) predictive analytics software was used to aggregate and summarise participants’ demographic data and responses in both rounds. Consensus in this study was defined a priori as ≥70% of participants agreeing/strongly agreeing or disagreeing/strongly disagreeing with a statement. This level of agreement has been deemed appropriate in previous Delphi studies and guidelines (Balikuddembe, Tumwesigye, Wakholi, & Tylleskär, 2019; Sumsion, 1998; Vogel et al., 2019).
Of the 252 experts invited to participate in this Delphi study, responses to Round 1 were received from 52 experts (response rate = 20.63%) from 25 countries; 35 of these experts participated in Round 2 (response rate = 67.31%). The size of the panel was within the range of what is typical of Delphi studies (i.e., 15–60; Hasson et al., 2000). Table 1 summarises demographic characteristics of the Delphi panel in both rounds.
|Round 1||Round 2|
|Male||25 (48%)||15 (43%)|
|Female||27 (52%)||20 (57%)|
|M (SD)||54.12 (8.66)||53.51 (7.93)|
|Years since obtaining PhD|
|M (SD)||24.15 (8.23)||22.80 (6.97)|
|Years since becoming Professor|
|M (SD)||11.08 (7.75)||10.28 (7.20)|
The two closed‐ended statements in Round 1 demonstrated high levels of consensus. “Social psychological theory and/or research can make a significant contribution to knowledge about pandemics” reached consensus at 92.30% (M = 4.40, SD = 0.77), and “Social psychological theory and/or research can have a significant impact on how pandemics are managed” reached consensus at 90.30% (M = 4.38, SD = 0.84).
One‐hundred statements were extrapolated and condensed based on the content analysis of the open‐ended responses of the expert panel in Round 1 and were organised into 11 broad topic areas. Eight of the 11 areas captured topics to which social psychology can contribute and that should be prioritised to inform knowledge and management of pandemics. These eight areas were labelled as follows, and are presented in random order: leadership, information and misinformation, health communication, social influence and persuasion, stress, coping, health,and social relations, prejudice and intergroup relations, compliance and policing, and societal and political processes. The remaining three areas concern strategies that social psychologists should implement to research to inform knowledge and management of pandemics, and were labelled as follows: theory and collaboration, methods and sampling, and external engagement. The statements categorised under the 11th topic area pertain primarily to societal engagement and knowledge transfer strategies rather than research strategies.
The statements were categorised into the topic areas based on coherence (i.e., internal homogeneity and external homogeneity). That is, a specific statement was categorised under a particular area when it corresponded to a greater extent with one as opposed to any other cluster of statements. However, many of the topics between areas are arguably highly interrelated. The topics and strategies are outlined in full in Tables S1 and S2 in the Supporting Information.
The results from Round 2 indicated high levels of consensus regarding the contributions that social psychology can make to inform knowledge and management of pandemics, with the expert panel agreeing with 55 of the 74 statements (74%). The consensus levels for the statements are also available in Tables S1 and S2 in the Supporting Information. However, while there were high levels of agreement about the contributions that social psychology can make to understanding and managing pandemics, there was less consensus about the topics that should be prioritised. For research priorities, consensus was reached for only 26 of the 74 statements (35%) across the 10 topic areas (consensus was not reached for any of the statements concerning compliance and policing). Consensus levels for the 26 statements that were agreed upon as research priorities are presented in Table 2.
|Topic Areas and Statements||Consensus Level||M (SD)|
|How leaders can promote compliance with public health regulations during pandemics||77%||4.03 (0.86)|
|How leaders can promote cooperation and solidarity (across segments of society) during pandemics||83%||4.09 (0.82)|
|How decision‐making processes about public health policy occur among leaders during pandemics||74%||3.97 (0.99)|
|Information and misinformation|
|How information about pandemics is disseminated and transformed||77%||4.11 (0.90)|
|How rumours about pandemics originate and spread||71%||4.17 (0.92)|
|How conspiracy theories about pandemics originate and spread||71%||4.06 (0.87)|
|How to counter misinformation and fake news about pandemics||83%||4.46 (0.78)|
|How social norms can be invoked in health communication to increase compliance with public health regulations during pandemics||86%||4.26 (0.70)|
|How cultural value dimensions can be invoked in health communication to increase compliance with public health regulations during pandemics||74%||3.94 (0.97)|
|Social influence and persuasion|
|How and what persuasion/social influence strategies promote attitude and behaviour change to increase compliance with public health regulations during pandemics||91%||4.37 (0.65)|
|How and what persuasion/social influence strategies promote altruism, prosocial behaviour, and cooperation during pandemics||91%||4.37 (0.65)|
|How group membership and social identity processes affect altruism, prosocial behaviour, and cooperation during pandemics||80%||4.26 (0.85)|
|Stress, coping, health, and social relations|
|How risk perceptions are appraised by individuals and social groups during pandemics||77%||4.03 (0.79)|
|How people cope with stress caused by physical (social) distancing during pandemics||89%||4.43 (0.78)|
|How physical (social) distancing affects health outcomes (psychological and/or physiological) during pandemics||74%||4.09 (0.98)|
|How uncertainty and precarity affect health outcomes (psychological and/or physiological) during pandemics||77%||4.06 (0.94)|
|How to promote positive health outcomes during physical (social) distancing during pandemics||83%||4.17 (0.71)|
|How social relations are affected by physical (social) distancing during pandemics||77%||3.97 (0.92)|
|How people interact and connect with one another during physical (social) distancing during pandemics||77%||3.97 (0.96)|
|How different modes of social contact (online vs. face‐to‐face) during physical (social) distancing affect health outcomes (psychological and/or physiological) during pandemics||74%||3.94 (0.91)|
|Intergroup relations and prejudice|
|How pandemics affect intergroup conflicts||80%||4.18 (0.92)|
|How the experience of threat and fear affect intergroup relations during (and after) pandemics||80%||4.12 (0.82)|
|How prejudice between groups is affected by pandemics||77%||4.18 (0.81)|
|How groups believed to be sources of the spread of viruses are stigmatised during (and after) pandemics||71%||4.12 (0.86)|
|How to reduce prejudice towards individual and/or groups who are stigmatised during (and after) pandemics||74%||4.12 (0.89)|
|Societal and political processes|
|How to understand and address social inequalities resulting from pandemics||86%||4.53 (0.72)|
The research priorities for which consensus was reached are as follows: (a) leadership during pandemics—how leaders can promote cooperation, solidarity, and compliance with public health regulations, and how decision making processes about public health policy during pandemics occur; (b) information and misinformation during pandemics—how information is disseminated and transformed, how rumours and conspiracy theories originate and spread, and how to counter misinformation and fake news; (c) health communication during pandemics—how social norms and cultural value dimensions can be invoked to increase compliance with public health regulations; (d) social influence and persuasion during pandemics—how to promote attitude and behaviour change to increase compliance with public health regulations, and also altruism, prosocial behaviour, and cooperation; (e) health and social relations during pandemics—how risks and stress are appraised, the transformation of social relations and social contact, and its effect on health, how uncertainty and precarity affect health, and how to promote positive health outcomes; (f) intergroup relations and prejudice during pandemics—how threat and fear shape perceptions and responses to pandemics, how intergroup relations, prejudice, and conflict are affected, how groups believed to be the sources of viruses are stigmatised, and how to reduce prejudice against those that are stigmatised; and (g) social inequalities during pandemics—how to understand and address social inequalities that emerge.
The statements regarding research strategies, and societal engagement and knowledge transfer strategies, also reached high levels of consensus, with the expert panel agreeing with 19 of the 26 statements (73%). The consensus levels for the strategies that were agreed upon are presented in Table 3.
|Areas and Statements||Consensus Level||M (SD)|
|Theory and Collaboration|
|Evaluate the relevance and applicability of social psychological theory and research for understanding and managing pandemics||86%||4.42 (0.66)|
|Evaluate multiple as opposed individual social psychological theories to identify theories most relevant for understanding and managing pandemics||77%||4.36 (0.96)|
|Focus on complex societal as opposed to simple individualistic processes to understand and manage pandemics||80%||4.30 (0.92)|
|Coordinate research efforts and collaborations within social psychology through academic societies and associations||77%||4.18 (0.88)|
|Engage in interdisciplinary collaborations to understand and address topics and issues relevant to pandemics from different and new perspectives||86%||4.61 (0.66)|
|Methods and sampling|
|Internet‐based research to accommodate for physical (social) distancing measures during pandemics||91%||4.53 (0.51)|
|Mixed‐methods and multimethod research||87%||4.55 (0.62)|
|Qualitative research||71%||4.09 (0.73)|
|Cross‐sectional research||74%||4.09 (72)|
|Longitudinal research||91%||4.61 (0.56)|
|Intervention studies||80%||4.24 (0.79)|
|Secondary data analysis||71%||4.09 (0.84)|
|Collect data from diverse samples, such as non‐WEIRD, community, and hard‐to‐reach populations||91%||4.73 (0.52)|
|Increase data‐sharing and transparency through open science during pandemics||83%||4.48 (0.71)|
|Engage with the media to help the public understand social behaviour during pandemics||86%||4.39 (0.75)|
|Engage with policymakers and practitioners to highlight the contributions that social psychology can make to understanding and managing pandemics||94%||4.64 (0.49)|
|Engage with policymakers and practitioners to ensure policies are evidence‐based during pandemics||94%||4.73 (0.45)|
|Engage with policymakers and practitioners to agree on research priorities during pandemics, and thereby increase the reach and uptake of research during pandemics||83%||4.39 (0.70)|
Strategies for which consensus was reached are as follows: (a) theory and collaboration—evaluating the relevance and applicability of social psychological theory for understanding and managing pandemics, with a focus on evaluating multiple theories and complex society‐level as opposed to individual‐level processes, coordination of research efforts and collaborations through academic societies, and interdisciplinary collaboration; (b) methods and sampling—mixed‐ and multimethod research, including qualitative, cross‐sectional, longitudinal, and intervention research, and secondary data analysis, Internet‐based data collection, data collection from diverse samples, and increased data sharing and transparency through open science practices; (c) external engagement—engagement with the media to help the public understand social behaviour, and with policymakers and practitioners to highlight the contributions that can be made by social psychology, ensure policies are evidence‐based, and agree on research priorities to increase its reach and uptake.
The motivation behind this Delphi study was to advance a consolidated, and consensual, account of the contributions that social psychology can make to informing knowledge and management of pandemics, and the research topics and strategies that should be prioritised to this end. The findings indicate that the expert panel identified and agreed upon 55 topics to which social psychology can contribute, 26 topics that should be prioritised, and 19 research strategies that should be implemented.
The contributions and research priorities identified and agreed upon by the expert panel can be distilled down to three broad domains. The first domain concerns improving, and thereby increasing compliance with, public health regulations during pandemics, but also promoting cooperation, solidarity, altruism, and prosocial behaviour—through leadership and the design of public health policy and communication. The second domain concerns the nature and dissemination of information and misinformation about pandemics, and how misinformation, through the modes of conspiracy theories and fake news, can be counteracted. The third domain concerns the consequences of pandemics—with regard to social relations and health, prejudice and intergroup relations, and social inequality. The research strategies identified and agreed upon, in turn, highlight theoretical and methodological diversity and adaptability, collaboration and coordination around research undertakings, both intra‐ and interdisciplinary, and engagement with external stakeholders.
The topics, priorities, and strategies identified and agreed upon in this study arguably mirror and complement the content presented in the expositions offered by Van Bavel et al. (2020) and Rosenfeld et al. (2020). The strength of the Delphi method is that it utilises a consensus‐making process among experts to identify areas of particular importance and priority; the main premise of the method is that group opinion, as opposed to individual opinion, is more “valid” and “reliable” (Keeney et al., 2011). Rosenfeld et al. (2020) indeed noted that given the number of involved authors, “each of whom offers a unique set of experiences and viewpoints . . . not every author endorses every idea” (p. 3). To this end, the Delphi method enables experts to offer their viewpoints while also voicing their opinions about the viewpoints offered by other experts. On the other hand, the method does not generate findings that explicate how the topics, priorities, and strategies that are agreed upon can be operationalised into research undertakings to inform knowledge and management of pandemics. The expositions coordinated by Van Bavel et al. (2020) and Rosenfeld et al. (2020) provide richer accounts to this end, and it is recommended that they are considered in conjunction with the findings from this study. This is because contributors to these expositions were selected based on their respective areas of expertise that were a priori considered relevant for understanding and managing the behavioural dimensions of pandemics—contributors were therefore able to elaborate in greater detail about relevant theory and research. The fact that the findings from this Delphi study mirror many of the areas considered in the expositions (e.g., leadership, science communication, stress and coping, and social inequality) should be taken as a strength as it corroborates the discipline’s understanding of the contributions that it can make towards informing knowledge and management of pandemics. However, this Delphi study is still uniquely placed in that it unearthed expert agreement about the contributions that the discipline can make and priority areas for research; research priorities and strategies were, in fact, not explicated by Van Bavel et al. (2020).
The range of topics and strategies could, of course, have been more exhaustive and possibly even narrower if the expert panel had a different composition. Notably, the majority of expert panellists are based in the United Kingdom, with only two based in the United States. This is in stark contrast to the proportion of social psychological research that is produced and published by researchers based in the United States (Arnett, 2008; Henrich, Heine, & Norenzayan, 2010); although the initial expert invitation list was arguably more representative of the global composition of the discipline. However, with this in mind, it is important that research designed along the lines of the priority areas identified in this Delphi study takes into account context‐specific factors that may impact social and behavioural dimensions of pandemics. For example, nation‐level variation in responses to the COVID‐19 pandemic is likely to shape which and how social and behavioural dimensions are prioritised by social psychologists within a given context. Shifting and/or intensifying intergroup conflicts may be more salient in some contexts, and their nature is likely to be shaped by context‐specific historical and political processes. Likewise, misinformation about the origins of COVID‐19 will differ between contexts, making it important to account for specificities in the content, spread, and implications of misinformation across contexts (e.g., for prejudice and compliance with public health regulations).
Furthermore, marginal consensus (i.e., 60–69%) was reached for many topics and strategies. Often Delphi studies go on to a third round asking the expert panel to re‐assess statements that do not reach consensus to ensure that they are given due consideration before they are “discarded.” However, a decision was made not to extend the study to a third round given the time commitment it would involve for the expert panel under the current COVID‐19 circumstances. Moreover, and relatedly, although a 70% threshold was set to ascertain consensus, it does not necessarily mean that the topics and strategies that did not reach this level of consensus in this study are not irrelevant for social psychology and psychologists. Some social psychologists may have unique expertise that enables them to engage with external stakeholders to inform knowledge and management of particular dimensions of pandemics. For example, although the topics about policing did not reach consensus in this study, social psychologists have already been formulating and influencing policing strategy in the context of the COVID‐19 pandemic (Reicher & Stott, 2020).
The Delphi method is not commonly utilised in psychology (let alone social psychology) but has the potential to serve as a critical tool for identifying research priorities and consolidating research efforts, particularly in times of crisis. This can potentially ensure that research priorities are streamlined and that the evidence base for priority areas is systematically advanced (IJzerman et al., 2020). While the research reported herein does not represent an authoritative or definitive overview of research priorities for social psychology in relation to pandemics, it does represent a systematic attempt to reach consensus about the part that the field and its experts can and should play in addressing pandemics, both in the present and in the future. It is a methodological approach that should be adopted and implemented more widely, and arguably authoritatively by scientific and professional organisations, to cultivate consensus and calibrate focus in times of crisis; the Delphi method was, after all, developed for technological forecasting during the Cold War (Keeney et al., 2011). The method can also be adopted to pave directions for theorisation and research into societally pertinent issues that have not been attended to by social psychologists, particularly within the mainstream. The method may thus be particularly productive for attending to context‐specific issues facing Asian societies that have not yet been examined from a social psychological perspective. However, for now our hope is that the findings generated by this Delphi study provide a roadmap for how social psychologists and social psychology can contribute optimally to addressing one of the greatest challenges of the 21st century. We hope its findings will stimulate focus, discussion, ideas, and collaboration to the benefit of both science and societies.
Conflict of Interest
The authors have no conflict of interest to declare.