The outbreak of the new coronavirus has caused tremendous concerns to public health, which are impacting human lives both physically and psychologically. The rise in coronavirus cases has led to the propagation of control measures for its prevention. This study aims to investigate the factors enhancing the coronavirus preventive behavior among the respondents.
To understand the coronavirus preventive behavior, the study is based on the value–belief–norm (VBN) theory. Data for the study has been collected through a survey of 319 respondents in New Delhi, India. The study uses structural equation modeling (SEM) to understand the factors impacting preventive behavior. For analysis, the study uses SEM to examine direct and indirect relationships and Hayes’ PROCESS macro SPSS module for moderating effects.
The results show that egoistic values have a negative impact on belief while altruistic values have a positive impact on the belief about the coronavirus outbreak. Belief is recorded to have a positive and significant impact on preventive behavior. Also, personal norms positively mediate the relationship between belief and preventive behavior. Additionally, the impact of awareness of preventive behavior is positively moderated by the symptomatic profile. Furthermore, the interaction effect is found to be conditioned positively with age and level of education.
To the best of the authors’ knowledge, no other work in the existing literature was found to apply the VBN theory to determine coronavirus preventive behavior. Further, the extensive moderation analysis done in this study is expected to be a significant contribution to the literature.
Ansari, Z., Zaini, S.H.R., Parwez, M. and Akhtar, A. (2021), "Effect of socioeconomic differences on COVID-19 preventive behavior among working adults in India", International Journal of Ethics and Systems, Vol. 37 No. 2, pp. 263-280. https://doi.org/10.1108/IJOES-08-2020-0136
Emerald Publishing Limited
Copyright © 2020, Emerald Publishing Limited
In December 2019, the first case of the novel coronavirus disease, commonly known as COVID-19, was reported from Wuhan, China (Harapan et al., 2020). Among other symptoms, the disease is known to cause breathlessness among humans. The virus causing COVID-19 is known as the severe acute respiratory syndrome coronavirus 2. In India, the first case of COVID-19 was reported on January 30, 2020, after which the disease spread at an alarming rate, and while writing this paper, the total cases of COVID-19 had surged past 17.3 million worldwide and 1.6 million in India (as of July 30, 2020). The human–human transmission of the disease occurs through the transmission of infected particulate matter via the respiratory organs (Xie et al., 2020). In the absence of an approved vaccine for the disease, the common measures suggested by various agencies include social distancing, washing hands frequently, covering nose and mouth, sanitizing things before using and others. The coronavirus pandemic has caused a severe impact on the economy throughout the world. Schools and educational institutions have been closed, factories and offices are shut and travel restrictions have been imposed. The demand for commodities, manufactured products and costly items has decreased substantially. All this has led to a reduction in the workforce across different sectors and millions have already lost their jobs (Nicola et al., 2020). On the other hand, demand for medical supplies has increased. Because of concerns over the availability of goods, people started panic buying and storing items in bulk leading to a rise in food and essential commodities (Chronopoulos et al., 2020).
Just as the coronavirus pandemic has posed severe challenges and difficult times for businesses across the globe (Baker et al., 2020), businesses in India too have been affected adversely. During these critical times, the national and international borders were sealed; shops of non-essential items were closed, most of the e-commerce companies were also not operating during initial days of lockdown and citizens were advised staying home (Maliszewska et al., 2020). Many companies have had to close their business either temporarily or permanently because of high operating costs and almost no demand. Companies are faced with an unforeseen threat to their existence. On one side, people are worried as the disease does not seem to end in the near future, and on the other side, they are concerned about their financial security because of huge job losses and reduction in salaries as well as the erosion of savings (Ho et al., 2020). The grave economic situation also bears consequences for consumers as it affects their attitude, perceptions and behavior.
The threat of COVID-19 has compelled people to make changes to their routine activities to control it from spreading (Mann, 2020). Lockdowns and social distancing procedures are commonly prescribed measures. As a result, people are now working from home, students are attending classes online and overall social activity has been reduced to a minimum. The rationale of the present study is to explore COVID-19 preventive behavior among the study group of working adults in India. Literature review in the section that follows helps in designing the framework for the study and developing hypotheses. The methodology is explained in the subsequent section after which analysis is conducted. Finally, the results are depicted and conclusions are drawn.
Stern et al. (1999) proposed the theory of value–belief–norm (VBN), which is an extension of norm activation theory propounded by Schwartz (1977). In the VBN model, the whole idea revolves around mapping specific human values toward the formation of pro-environmental behavior (PEB), which Kollmuss and Agyeman (2002) define as any human behavior specifically concerned with minimizing one’s negative action toward the natural and built environment. Identification of the antecedents of PEB is complex and there exist some reported inconsistencies in measuring it (Coelho et al., 2017; Markle, 2013). Sawitri et al. (2015) have proposed the social and cognitive perspective to understand PEB. Another study by Ramus and Killmer (2007) finds that the PEB includes the pro-social behavior which is directed toward and performed to promote the welfare of individuals or groups. In pro-social agency theory, individuals engage in the behavior of emotional attachment by sharing, helping and looking after one another (Caprara and Steca, 2005).
Environmental problems are caused by human behavior and can be resolved by changing the behavior. The coronavirus pandemic being a major problem directly affecting the lives of the masses, PEB needs to be reinforced to control the spread of the virus. Several guidelines have been issued by the World Health Organization (WHO) advising preventive measures such as the use of sanitizers and face masks, and social distancing to safeguard from COVID-19. The present study thus considers COVID-19 preventive behaviors as environmentally proactive.
VBN model considers the relationship of values, beliefs, norms and behavior in the causal chain. Values are subjective judgment about stimuli such as environment, people, object or anything (Dalvi-Esfahani et al., 2017; Brunsø et al., 2004). De Groot and Steg (2008) have developed an instrument that measures biospheric, altruistic and egoistic value orientation. These value components are considered vital precursors of environmental belief (Schultz, 2000). Biospheric values include attributes related to the environment and ecosystem (Stern et al., 1999). Altruistic values are more socially motivated and emphasize the importance of other people and beliefs which instigate them to involve in pro-environmental concerns (Stern, 2000). On the other hand, egoistic values that relate more toward pro-self-concept and prioritizing oneself may also emerge as a PEB (Stern and Dietz, 1994). However, egoism is self-enhancement value orientation and mostly has a negative impact on environmental behavior (Cavagnaro and Staffieri, 2015). Sony and Ferguson (2017) argue that self-concerns derive greater motivation than social ones. Wang et al. (2020) have conducted a detailed inquiry into the relationship between values and behavioral intention by applying the VBN theory.
The proposed framework states that egoistic and altruistic values are self-enhancement and prosocial behavior, respectively (Sosik et al., 2009), which guide individuals to contain the spread of COVID-19 by not letting themselves out and by maintaining social distance with others. Welfare and health concerns of others are of prime importance and irresponsible behavior can have damaging consequences. Therefore, it is anticipated that higher altruistic and egoistic values result in higher behavioral action to restrain COVID-19.
Information is an important component of successful behavioral change (Lorenzoni et al., 2007). The process of receiving information stimulates the norms, intentions and behavior (Antimova et al., 2012). Kaiser and Shimoda (1999) argue that beliefs reflect the information held with individuals and knowledge about environmental issues act as a stimulator to behave pro-environmentally. Therefore, beliefs are the cognitive processes of realizing the positive/negative effects of the particular event and realizing oneself equally responsible for the happening of the same event (Schwartz, 1977). Awareness of consequences and ascription of responsibility are the components of belief construct and the former is the precursor of the latter (Schwartz, 1977). Awareness of consequences is the knowledge of susceptible threats about the outcome of an event (De Groot and Steg, 2008). Similarly, the ascription of responsibility is a belief that emerges as human action toward promoting or preventing an event (De Groot and Steg, 2009). It can be described as human efforts to minimize the negative outcome of an event. According to the VBN model, values influence beliefs in the behavioral process (Stern, 2000). Jaini (2019) finds the positive impact of altruistic value on awareness of consequences. Similarly, Chua et al. (2019) evidence the positive influence of egoistic values on environmental beliefs. Hence, the proposed hypotheses are that individuals with high altruistic and egoistic values are likely to exhibit a higher awareness of consequences.
Altruistic values are positively related to the awareness of the consequences of COVID-19.
Egoistic values are positively related to the awareness of the consequences of COVID-19.
Once the individuals possess the values and beliefs for environment preservation, the pro-environmental sentiments lead to the development of norms in individuals. Norms are described as the feeling of moral obligation to actively promote the preservation of the environment (Kaiser and Shimoda, 1999). Norm leads individuals to engage in PEB and refrain from any activity that harms the environment.
Awareness of consequences, the ascription of responsibility and personal norms are components of the norm activation theory (Schwartz, 1977). Several studies in diversified areas have applied norm activation theory and established the relationships among the variables. Similarly, for the present case, awareness of COVID-19 and its consequences are well communicated by local and international agencies. The awareness of consequences stimulates individuals to accept responsibility to fight the disease. However, because of the unabated spread and fatal consequences of COVID-19, it is assumed that humans themselves are responsible for causing the spread. The hypotheses thus formed are:
Awareness of consequences positively influences the ascription of responsibility.
Awareness of consequences positively influences personal norms.
The ascription of responsibility positively influences personal norms.
Several previous studies have documented the impact of beliefs on personal norms (Abutaleb et al., 2020; Shin et al., 2018; Liu et al., 2020). However, fewer studies have examined the mediation effect of norms between beliefs and behavioral intention (Dalvi-Esfahani et al., 2017; Vaske et al., 2015). Jaini et al. (2019) finds a positive mediating effect of personal norms between beliefs and green behavior. The conceptual framework of the study postulates that the development of moral obligation leads individuals to commit in behavior that prevents them from causing the spread of the COVID-19 disease. Moreover, the researchers are interested to examine the direct effect of awareness of consequences and ascription of responsibility on behavior and indirect relationship with personal norms as mediators.
Common human behavior reflects that personality factors shape perceptions regarding threats and the corresponding responses (Gaygısız et al., 2012). Thus, preventive behavior may be exhibited more by those who feel threatened. Threats of illness emerge when people begin to show symptoms. Roy et al. (2020) have reported perceptions of behavioral dimensions in the wake of COVID-19 among Indian respondents. An inquiry into the adoption of preventive behavior is suggested by studying the role of moderators and influencing factors (Asmundson and Taylor, 2020; Jungmann and Witthöft, 2020; Blakey and Abramowitz, 2017). Stoltenberg et al. (2008) have examined the role of gender as a moderator between impulsivity and health-risk behavior and have thus called for more directed prevention efforts. Various studies have made inquiries based on age, gender and level of education into the adoption of preventive behavior practices including hand washing and wearing masks in times of pandemics (Lau et al., 2004; Leung et al., 2003; Tang and Wong, 2004). For a detailed review of demographic and attitudinal determinants of preventive behaviors during a pandemic, refer to Bish and Michie (2010). Based on extensive literature available, we too propose to test the moderating effects of gender, age and educational qualification individually between our main study dimensions. Additionally, we also apply a moderated moderation model by interacting with the realization of symptoms (RoS) with age, and with levels of educational qualification. The proposed hypothesis of moderation and moderated moderation are as follows:
Ascription of responsibility positively influences COVID-19 preventive behavior.
Awareness of consequences positively influences the COVID-19 preventive behavior.
The RoS positively moderates the effect of awareness of consequences on preventive behavior.
Differences in age groups positively moderates the effect of awareness of consequences on preventive behavior.
Differences in age groups positively moderates the combined effect of awareness of consequences and RoS on preventive behavior.
Differences in educational qualification positively moderates the effect of awareness of consequences on preventive behavior.
Differences in age groups positively moderates the combined effect of awareness of consequences and RoS on preventive behavior.
Personal norms positively influence COVID-19 preventive behavior.
Personal norms mediated the relationship between awareness of consequences and COVID-19 preventive behavior.
Personal norms mediated the relationship between ascription of responsibility and COVID-19 preventive behavior.
The present study is exploratory research which aims at examining the driving factors of the COVID-19 preventive behavior. For this purpose, data have been collected from individuals who are exposed to risk in the process of earning their livelihood. The sampling element thus comprises employed or business-holding individuals, who have been classified as working adults for the purpose of the study. The rationale for selecting this element is to purposefully look for the independent behavior of individuals for their business activities. The groups which are identified as most vulnerable to COVID-19 virus infection are studied on the basis of gender, age and occupation (WHO, 2020). Hence it is ensured that the sample includes all the demographic characteristics suggested by the WHO. The data has been collected from India’s capital city, New Delhi, as it is assumed that each type of sampling element resides here. Because of the prevailing COVID-19 conditions, the study uses a non-probabilistic sampling technique and data has been collected using Google Forms (Chakraborty, 2019; Dhir and Dutta, 2020). A total of 319 proper responses have been received appropriate for conducting the analysis.
The present study is survey-based research. The survey instrument consists of a questionnaire that is designed in three sections. The first section inquires into demographic and socioeconomic characteristics of respondents on a nominal scale. It also includes the responses regarding whether respondents have realized any kind of symptoms such as fever, dry cough, body ache or breathlessness during the past months. The RoS is designed as a dichotomous variable and responses have been received as “yes” for symptoms realized during last month or “no” for no realization of any symptom at all. The third section (refer to Appendix) is inclusive of study variables on the COVID-19 preventive behavior and responses have been received for this section on a seven-point Likert scale (“1” for “strongly disagree” and “7” for “strongly agree”). The measures of study constructs with their respective items are presented in Table 2. The measures of altruistic values and egoistic values (Hwang et al., 2020; Stern et al., 1999; Schwartz, 1992), awareness of consequences, ascription of responsibility and personal norms (Onwezen et al., 2013; Ghazali et al., 2019) and preventive behavior (Abdelhafiz et al., 2020) have been adopted and designed based on COVID-19 pandemic.
To examine the relationships between the different study elements, multivariate analysis technique, structural equation modeling (SEM), is used. SEM is useful for testing theories that can be represented in multiple equations involving dependence relationships (Hair et al., 2019). SEM models error in measurement model for observed variables and thereby provides more robust results (Chin, 1998). Research studies with complex models find SEM as an appropriate tool. SEM is very popular and extensively used in behavioral and social sciences (Raykov and Widaman, 1995). Moreover, using the PROCESS Macro, the integrated model of moderation and mediation can be estimated using latent variable-based SEM framework (Hayes and Preacher, 2013; Sardeshmukh and Vandenberg, 2017). For the purpose, bootstrapping confidence interval is required to be constructed for inference. Such inferences from bootstrapping confidence intervals and conditional process analysis with models including interaction between latent variable can be difficult (Hayes, 2018). Hence, the current study uses the Hayes’ PROCESS to analyze the interaction effect. Moreover, the benefits of PROCESS module are based on both the methodological and mathematical reasons (Shkoler and Kimura, 2020). The current research reviews some previous research studies (Hwang et al., 2020; Saleem et al., 2018) for the purpose of methodological reference.
To assess the interaction effect of COVID-19 symptoms and personal norm on COVID-19 preventive behavior, Hayes’ PROCESS Macro module (Version 3.5) for SPSS has been adopted (Hayes, 2013). The moderating variable, w, is described as a third variable which causes variation in the effect of independent variable X on dependent variable Y (Hayes, 2009; Baron and Kenny, 1986). Generally, moderator is considered as a situational variable that strengthens/weakens relationship between predictor and the outcome variable (Edwards and Lambert, 2007). Therefore, two-way interaction effect of X and w on Y is examined and if the unstandardized effect of X on Y changes with the function of w, moderating effect is said to be present (Aiken, 1991). Similarly, according to Lam et al. (2019) and Aiken et al. (1991), a three-way interaction occurs when the two-way interaction effect of X and M on y is conditional on third variable (z). Current study conducts single and multiple moderation analysis to examine the variation in the effect of awareness of consequences (X) on preventive behavior (Y). For two-way interaction, RoS, age and education are considered as moderating variables (w) which cause the variation in the effect size of awareness of consequences on preventive behavior. Similarly, for moderated moderation analysis, RoS has been considered as a primary moderator (w) and demographic variables (age and education) are included as secondary moderators (z), which are expected to have conditional effect on preventive behavior.
Profile of the respondents
Demographic and socioeconomic classification of the respondents is made on the basis of gender (male = 203; female = 116), age (18–33 years = 111; 34–49 years = 117; 50 and above = 91), monthly income (INR0–25,000 = 139; INR25,001–50,000 = 128; above INR50,000 = 52), educational qualification (pre-graduates = 80; graduates = 158; post-graduates = 81) and their realization of COVID-19 symptoms (yes = 133; no = 186). Male respondents, middle-aged, lower-income graduates have largely participated in the study. Respondents with no RoS at all share higher proportion of the total sample than those who have realized at least one of the symptoms.
Table 1 presents the descriptive statistics, Pearson correlation coefficients and the square root of the average variance extracted (AVE). The predicted power (depicted in Figure 1) of the impact of awareness of consequences and ascription of responsibility on personal norms (R2 = 0.38) and preventive behaviour (R2 = 0.54) explains a higher amount of variance. Descriptive statistics exhibits dispersion in responses to the variables of the study. Normality analysis has also been performed with the measures of skewness and kurtosis. Values for skewness and kurtosis below 2 (Curran et al., 1996) and kurtosis below 3 (Westfall and Henning, 2013) are indicative of the normality of the data. The results of the Pearson correlation show that variables are significantly correlated with one another at 99% confidence interval except egoistic values with altruistic values (r = −0.120; p < 0.05), personal norm (r = −0.144; p < 0.05) and preventive behavior (r = −0.130; p < 0.05).
Measurement model: validity, reliability and fitness
Validity is concerned with how well a concept is defined by the measures. Confirmatory factor analysis does not validate the summate scale rather it computes the latent construct scores for each response and establishes the relationships between constructs, hence, to be corrected for the amount of error variance that exists in construct measures (Hair et al., 2019). Construct validity has been ascertained based on suggested measures (Hair et al., 2019). For convergent validity, factor loadings, AVE and composite reliability (CR) have been estimated. High factor loadings greater than 0.7 CR, Cronbach’s alpha (CA) above 0.7 and AVE higher than 0.5 reflect no validity concerns for the measurement model. Likewise, discriminant validity has been assessed using the square root of AVE greater than the correlation between the constructs. The results for factor loadings are depicted in Table 2 and for other validity measures are provided in Table 1. Each construct confirms convergent and discriminant validity as suggested by Fornell and Larcker (1981).
The fitness of measurement model has been assessed with the goodness-of-fit (GOF) statistics which indicates how well theoretically proposed model fits to the reality as represented by the data. The fitness of measurement model is found to be statistically significant (χ2 = 308.885; df = 194; χ2/df = 1.592; p < 0.01). The statistics of alternate GOF indices such as absolute and comparative fit statistics [standardized root mean residual (SRMR) = 0.037; goodness-of-fit index (GFI) = 0.922; adjusted goodness-of-fit index (AGFI) = 0.899; root mean square error of approximation (RMSEA) = 0.043; PCLOSE = 0.896; normed fit index (NFI) = 0.936; relative fit index (RFI) = 0.924; Tucker-Lewis index (TLI) = 0.970; comparative fit index (CFI) = 0.975] are also found within the recommended values (Hu and Bentler, 1999; Browne and Cudeck, 1993).
Hypotheses testing with the structural model
Analysis of direct effects.
The hypotheses testing has been carried out using SEM with the maximum likelihood estimation method. The results exhibited in Table 3 show that awareness of consequences is positively and significantly influenced by altruistic values (β = 0.681; p < 0.01) while negatively influenced by egoistic values (β = −0.176; p < 0.01). Hence, H1 and H2 are accepted. The explanatory power of altruistic and egoistic values is recoded as (R2 = 0.30). Likewise, awareness of consequences is found to be positively influencing ascription of responsibility (β = 0.365; p < 0.05) personal norms (β = 0.359; p < 0.01) and preventive behavior (β = 0.296; p < 0.01). Therefore, H3, H5 and H7 are accepted. Moreover, the impact of ascription of responsibility on personal norms (β = 0.294; p < 0.05) and preventive behavior (β = 0.217; p < 0.01) is recoded as positive and significant. Hence, the H4 and H6 are also accepted. The personal norms positively and significantly influence preventive behavior (β = 0.333; p < 0.01). Hence, H8 is accepted.
Analysis of indirect effects.
The indirect effect of awareness of consequences and ascription of responsibility on personal norms and preventive behavior has been conducted using a bias-corrected bootstrapping method with 500 subsamples. The results of the indirect effect presented in Table 3 show that the indirect effect of awareness of consequences on preventive behavior with personal norms as mediating variable is positive and statistically significant (β = 0.119; p < 0.01). Similarly, the indirect effect of the ascription of responsibility on preventive behavior with personal norms as a mediating variable is found to be positive and statistically significant (β = 0.088; p < 0.01). The predictive power of the impact awareness of consequences and ascription of responsibility on personal norms (R2 = 0.38) and preventive behavior (R2 = 0.54) explains a higher amount of variance.
Assessing model fit.
The results for two-way interaction show that the interaction effect of awareness of consequences and symptoms (F [1, 315] = 14.475, p < 0.001), age (F [2, 313] = 4.839, p < 0.01) and educational qualification (F [2, 313] = 5.23, p < 0.01) are found to be significantly fitting the model, which accounts for 3, 2 and 12% of the explained variance. Hence, it can be opined that the interaction of awareness of consequences with symptoms, age and education is determined as a better fit model. Likewise, three-way interaction of awareness of consequences, symptoms and age (F [2, 307] = 16.293, p < 0.001) has accounted for 3% of the variance and three-way interaction of awareness of consequences, symptoms and educational qualification (F [2, 307] = 3.912, p < 0.001) recorded 0.8% of the explained variance.
Hypotheses testing of moderation effects using ordinary least squares regression.
The two-way interaction results of ordinary least squares (OLS) regression presented in Table 4 indicate the interaction effect of AC and RoS on PB is found to be positive and statistically significant (β = 0.36; SE = 0.07; p < 0.001) explaining 38% of the variance in PB. Hence, the H7a is accepted. The interaction effect of AC with the differences in middle-aged adults and young adults is found to be negatively but statistically significant (β = −0.259; SE = 0.08; p < 0.01). Similarly, differences in older adults and young adults negatively and significantly moderate the effect of AC on PB (β = −0.210; SE = 0.10; p < 0.05). Finding based on differences in age groups supports H7b, which hence is accepted. Moreover, the effect of AC on PB when moderated by young adults is greater than the moderation effect of middle-aged and older adults. Overall direct and interaction effect of AC on PB accounts for 39% of the variance. For educational groups, the impact of interaction of PB and differences in graduates and pre-graduates (β = −0.209; SE = 0.06; p < 0.01) and differences in post-graduates and pre-graduates (β = −0.216; SE = 0.10; p < 0.05) is estimated as negative and statistically significant and predicting preventive behavior with 64% of variance. Therefore results favor H7c, which is hence accepted. The interaction effect is larger with the pre-graduates and lower for the graduates and post-graduates.
The OLS regression results for three-way interaction show that the interaction effect of AC and realization of symptom on PB is positively and significantly moderated by the differences between middle-aged adults and young adults (β = 0.393; SE = 0.16; p < 0.01) and the differences between older adults and young adults (β = 0.814; SE = 0.20; p < 0.001). Findings support H7b1 and it is accepted. The three-way interaction effect of RoS with age accounts for 51% of the variance. Moreover, the three-way interaction effect is larger with the middle-aged and old adults than the young ones. Based on results, it can be summarized that when age groups have interacted with the realization of the symptoms, the PB enhances for middle-aged and older adults. Similarly, the differences between graduates and pre-graduates (β = 0.376; SE = 0.16; p < 0.05) and differences between post-graduates and pre-graduates (β = 0.579; SE = 0.22; p < 0.05) moderate the interaction effect of AC and RoS on PB. Hence, H7c1 is accepted. Resultantly, the RoS has a greater effect as the level of education increases. The three-way interaction model with AC, RoS and levels of education explains 66% of the variance.
Interaction effect of X and w on y conditioned at the values of z.
The results of the conditional effect (visualized in Figure 2) of AC and RoS interaction at the values of the different age groups and levels of education are also estimated. The results for age groups show that for young adults, the effect of those with the RoS (β = 0.777; SE = 0.07; p < 0.001) is larger than for those without RoS (β = 0.433; SE = 0.07; p < 0.001). The effect of AC on PB is significantly moderated by the differences in RoS [θxw→Y | (z = young) = 0.344, F (1, 307) = 10.205, p < 0.01]. Similarly for middle-aged adults [θxw→Y | (z = middle) = 0.74, F (1, 307) = 35.749, p < 0.001] and older adults [θxw→Y | (z = old) = 0.47, F (1, 307) = 7.475, p < 0.01], the effect of AC on PB is moderated by the differences in RoS. The effect of middle-aged adults with the RoS (β = 0.742; SE = 0.09; p < 0.001) is larger than for those without RoS (β = 0.005; SE = 0.08; p > 0.05), and for older-adults, the effect of those with the RoS (β = 0.204; SE = 0.11; p > 0.05) is smaller than for those without RoS (β = 0.673; SE = 0.13; p < 0.001). Likewise, in case of education qualification, the effect of graduates with RoS (β = 0.361; SE = 0.07; p < 0.001) is greater than for those without RoS (β = 0.071; SE = 0.06; p > 0.05). The RoS significantly moderates the effect of AC on PB among graduates [θxw→Y | (z = graduates) = 0.29, F (1, 307) = 8.870, p < 0.01]. Those who have completed post-graduation [θxw→Y | (z = post-graduates) = 0.49, F (1, 307) = 7.828, p < 0.01], greater effect has been recorded for those with RoS (β = 0.433; SE = 0.12; p < 0.001) than those without RoS (β = −0.06; SE = 0.12; p > 0.05). The negative and insignificant effect is recorded for pre-graduates [θxw→Y | (z = pre-graduates) = −0.09, F (1, 307) = 0.387, p > 0.05].
The current study identifies the factors influencing the coronavirus preventive behavior among working adults in India. For this purpose, the study adopts the VBN theory to develop a structural framework for the study. Hence, the impact of VBN has been examined using the SEM approach. The results signify that the value belief and norms have a positive and significant impact on preventive behavior except egoistic values which have a negative but significant impact. Moreover, the study has examined the relationship between awareness of consequences because of coronavirus and preventive behavior is moderated by the RoS, age and education. Moreover, it can be argued that the individuals facing symptoms, preventive behavior increases as their age and levels of education increases. Mixed findings are reported for the individuals who have not faced any symptoms during the sample period.
As new social contracts evolve, individuals are faced with ethical issues. Outbreak of COVID-19 has instigated the emergence of a new normal which has forced individuals to reframe the norms and ethics which used to be the commonplace of any society. The present study identifies the development of belief and norms as antecedents of COVID-19 preventive behavior. Results of the study record the positive influence of altruistic values, beliefs and norms on preventive behavior. In the case of socioeconomic characteristics of working adults, respondents who have realized at least one symptom during past few days are found to exhibit greater preventive behavior than those who have not realized any of the symptoms at all. However, the effect is found to be statistically significant for both sets of respondents exhibiting the realization as well as the non-realization of symptoms.
The policy implication of the study pertains to the enhancement of the value system, especially altruistic values so that the idea of togetherness can be brought into action. More education and awareness of COVID-19 is required to ensure preventive behavior. Respondents are found following the emerging norms for ensuring safety and security. Moreover, as the people grow older, they become more health conscious. The relationship can be moderated by the level of education as it can be argued that even at younger age, individuals with higher level of education are more concerned about health and, hence, adopt preventive behavior. Previous studies also advocate the impact of education on level of awareness (Komolafe et al., 2020).
As the current study has been conducted amid COVID-19 outbreak and safety concerns are of primary importance for everyone, the process of data collection using Google Forms is a potential limitation of the study. Because of a lack of prior knowledge about the characteristics of the respondents, data collected through online mode raises concerns. However, the researchers have tried to minimize the effect by collecting more data than required. In accordance with suggestions provided by Hair et al. (2019) and Cohen (1988) regarding determination of sample size, it is assumed that the data is appropriate and results can be generalized to the population. The direction of future research depends on several aspects of the research. From a theoretical perspective, other sociological, psychological and behavior theories can be used to analyze health behavior. Other elements can also be included in the study using psychographic segmentation. Cross-cultural and cross-country analysis can also be a potential future direction.
Descriptive statistics correlation estimates
Diagonal values are the square root of AVE; below diagonal values are the correlations between the constructs; above diagonal values are the Pearson correlation coefficients. *Correlation is significant at the 0.05 level (two-tailed). **Correlation is significant at the 0.01 level (two-tailed)
Constructs and validity measures
|Awareness of consequences|
|Ascription of responsivity|
Results of hypothesis testing using SEM
|H1||AV → AC||0.681||0.084||0.557||0.813**|
|H2||EV → AC||−0.176||0.061||−0.279||−0.066**|
|H3||AC → AR||0.365||0.054||0.259||0.455*|
|H4||AR → PN||0.294||0.053||0.11||0.376*|
|H5||AC → PN||0.359||0.05||0.255||0.486**|
|H6||AR → BI||0.217||0.051||0.103||0.328**|
|H7||AC → BI||0.296||0.053||0.212||0.397**|
|H8||PN → BI||0.333||0.065||0.202||0.474**|
|H9||AC → PN → CPB||0.119||0.036||0.070||0.198**|
|H10||AR → PN → CPB||0.088||0.32||0.037||0.152**|
*p-value < 0.05; **p-value < 0.01; LLCI = lower level confidence interval; ULCI = upper level confidence interval
Results of OLS regression for two-way and three-way interaction effects
|AC × Sym||0.36||0.07||4.551***|
|AC × w1||−0.259||0.08||−2.955**||−0.209||0.06||−3.029**|
|AC × w2||−0.210||0.10||−1.975*||−0.216||0.10||−2.101*|
|F||[3, 315] 63.684***||[5, 313] 39.789***||[5, 313] 111.811***|
|Variables||Sym × Age||Sym × Edu|
|AC × Sym||0.345||0.10||3.195**||−0.086||0.13||−0.622|
|AC × z1||−0.428||0.11||−3.778**||−0.422||0.09||−4.566***|
|AC × z2||0.24||0.15||1.566||−0.553||0.13||−3.984**|
|Sym × z1||−2.664||0.80||−3.299**||−2.086||0.68||−3.061**|
|Sym × z2||4.139||1.07||3.866***||−3.4925||1.10||−3.151**|
|AC × Sym × z1||0.393||0.16||2.399*||0.376||0.16||2.224*|
|AC × Sym × z2||0.814||0.20||−4.014***||0.579||0.22||2.586*|
|F||[11, 307] 39.179***||[11, 307] 55.868***|
Sym = symptoms; Edu = educational qualification; w1 and w2 = dummy variables in two-way interaction; z1 and z2 = dummy variables in three-way interaction for age and education. Both w1 and z1 represents middle-aged adults and graduates in their respective category. Both w2 and z2 represents old-aged and post-graduates in their respective category. For age and education, young adults and pre-graduates are considered as reference category, respectively. *p < 0.05; **p < 0.01; ***p < 0.001
Measures of constructs and their factor loadings
|Awareness of consequences|
|AC1||I am sure that the novel coronavirus is a real threat||0.866|
|AC2||There is a constant rise in the number of deaths because of COVID-19||0.859|
|AC3||Society has suffered because of novel coronavirus||0.851|
|AC4||COVID-19 has caused huge losses to the economy||0.875|
|Ascription of responsibility|
|AR1||I feel jointly responsible for the spread of coronavirus||0.812|
|AR2||I feel jointly responsible for violating social distancing norms||0.778|
|AR3||I feel jointly responsible for taking the coronavirus pandemic lightly||0.87|
|AR4||I feel jointly responsible for neglecting hygiene||0.877|
|PN1||I feel a moral obligation to act as per guidelines issued to fight COVID-19||0.824|
|PN2||I feel a moral obligation to restrict travelling during the pandemic COVID-19||0.808|
|PN3||I feel a moral obligation to inform others about safety measures against COVID-19||0.737|
|PN4||I would feel morally obliged to ensure wearing a face mask if I have to go out||0.845|
|PB1||I frequently use hand sanitizer/wash my hands to protect against COVID-19||0.794|
|PB2||I wear a face mask when I go outside||0.869|
|PB3||I have contributed money in the fight against COVID-19||0.849|
|PB4||I have limited my social interactions because of the prevailing COVID-19 conditions||0.868|
Abdelhafiz, A.S., Mohammed, Z., Ibrahim, M.E., Ziady, H.H., Alorabi, M., Ayyad, M. and Sultan, E.A. (2020), “Knowledge, perceptions, and attitude of Egyptians towards the novel coronavirus disease (COVID-19)”, Journal of Community Health, pp. 1-10.
Abutaleb, S., El-Bassiouny, N.M. and Hamed, S. (2020), “A conceptualization of the role of religiosity in online collaborative consumption behavior”, Journal of Islamic Marketing.
Aiken, L.S., West, S.G. and Reno, R.R. (1991), Multiple Regression: Testing and Interpreting Interactions, Sage, Newbury Park.
Antimova, R., Nawijn, J. and Peeters, P. (2012), “The awareness/attitude‐gap in sustainable tourism: a theoretical perspective”, Tourism Review, Vol. 67 No. 3, pp. 7-16.
Asmundson, G.J. and Taylor, S. (2020), “How health anxiety influences responses to viral outbreaks like COVID-19: what all decision-makers, health authorities, and health care professionals need to know”, Journal of Anxiety Disorders, Vol. 71, p. 102211.
Baker, S.R., Farrokhnia, R.A., Meyer, S., Pagel, M. and Yannelis, C. (2020), How Does Household Spending Respond to an Epidemic? Consumption during the 2020 Covid-19 Pandemic (No. w26949), National Bureau of Economic Research.
Baron, R.M. and Kenny, D.A. (1986), “The moderator–mediator variable distinction in social psychological research: conceptual, strategic, and statistical considerations”, Journal of Personality and Social Psychology, Vol. 51 No. 6, pp. 1173-1182.
Bish, A. and Michie, S. (2010), “Demographic and attitudinal determinants of protective behaviours during a pandemic: a review”, British Journal of Health Psychology, Vol. 15 No. 4, pp. 797-824.
Blakey, S.M. and Abramowitz, J.S. (2017), “Psychological predictors of health anxiety in response to the Zika virus”, Journal of Clinical Psychology in Medical Settings, Vol. 24 Nos 3/4, pp. 270-278.
Browne, M.W. and Cudeck, R. (1993), “Alternative ways of assessing model fit”, in Bollen, K.A. and Long, J.S. (Eds), Testing Structural Equation Models, Sage, Newbury Park, CA, pp. 136-162.
Brunsø, K., Scholderer, J. and Grunert, K.G. (2004), “Closing the gap between values and behavior – a means–end theory of lifestyle”, Journal of Business Research, Vol. 57 No. 6, pp. 665-670.
Caprara, G.V. and Steca, P. (2005), “Self–efficacy beliefs as determinants of prosocial behavior conducive to life satisfaction across ages”, Journal of Social and Clinical Psychology, Vol. 24 No. 2, pp. 191-217.
Cavagnaro, E. and Staffieri, S. (2015), “A study of students’ travellers values and needs in order to establish futures patterns and insights”, Journal of Tourism Futures, Vol. 1 No. 2.
Chakraborty, U. (2019), “The impact of source credible online reviews on purchase intention”, Journal of Research in Interactive Marketing, Vol. 13 No. 2.
Chin, W. (1998), “Commentary: Issues and opinion on structural equation modeling”, MIS Quarterly, Vol. 22 No. 1, pp. 7-16.
Chronopoulos, D.K. Lukas, M. and Wilson, J.O. (2020), “Consumer spending responses to the COVID-19 pandemic: an assessment of Great Britain”, Available at SSRN 3586723.
Chua, K.B., Quoquab, F. and Mohammad, J. (2019), “Factors affecting environmental citizenship behaviour”, Asia Pacific Journal of Marketing and Logistics, Vol. 32 No. 1, pp. 86-104.
Coelho, F., Pereira, M.C., Cruz, L., Simões, P. and Barata, E. (2017), “Affect and the adoption of pro-environmental behaviour: a structural model”, Journal of Environmental Psychology, Vol. 54, pp. 127-138.
Cohen, J. (1988), Statistical Power Analysis for the Behavioral Sciences, Lawrence Erlbaum, Mahwah, NJ.
Curran, P.J., West, S.G. and Finch, J.F. (1996), “The robustness of test statistics to normality and specification error in confirmatory factor analysis”, Psychological Methods, Vol. 1 No. 1, p. 16.
Dalvi-Esfahani, M., Ramayah, T. and Rahman, A.A. (2017), “Moderating role of personal values on managers’ intention to adopt green IS”, Industrial Management and Data Systems, Vol. 117 No. 3.
De Groot, J.I. and Steg, L. (2008), “Value orientations to explain beliefs related to environmental significant behavior: how to measure egoistic, altruistic, and biospheric value orientations”, Environment and Behavior, Vol. 40 No. 3, pp. 330-354.
De Groot, J.I. and Steg, L. (2009), “Morality and prosocial behavior: the role of awareness, responsibility, and norms in the norm activation model”, The Journal of Social Psychology, Vol. 149 No. 4, pp. 425-449.
Dhir, S. and Dutta, T. (2020), “Linking supervisor-support, person-job fit and person-organization fit to company value”, Journal of Indian Business Research, Vol. 12 No. 4.
Edwards, J.R. and Lambert, L.S. (2007), “Methods for integrating moderation and mediation: a general analytical framework using moderated path analysis”, Psychological Methods, Vol. 12 No. 1, pp. 1-22.
Fornell, C. and Larcker, D.F. (1981), “Structural equation models with unobservable variables and measurement error: algebra and statistics”.
Gaygısız, Ü., Gaygısız, E., Özkan, T. and Lajunen, T. (2012), “Individual differences in behavioral reactions to H1N1 during a later stage of the epidemic”, Journal of Infection and Public Health, Vol. 5 No. 1, pp. 9-21.
Ghazali, E.M., Nguyen, B., Mutum, D.S. and Yap, S.F. (2019), “Pro-environmental behaviours and value-belief-norm theory: assessing unobserved heterogeneity of two ethnic groups”, Sustainability, Vol. 11 No. 12, p. 3237.
Hair, J.F., Black, W.C., Babin, B.Y.J. and Anderson, R.E. (2019), Multivariate Data Analysis Techniques, Cengage Learning EMEA.
Harapan, H., Itoh, N., Yufika, A., Winardi, W., Keam, S., Te, H. and Mudatsir, M. (2020), “Coronavirus disease 2019 (COVID-19): a literature review”, Journal of Infection and Public Health, Vol. 13 No. 5.
Hayes, A.F. (2009), “Beyond Baron and Kenny: statistical mediation analysis in the new millennium”, Communication Monographs, Vol. 76 No. 4, pp. 408-420.
Hayes, A.F. (2013), “Introduction to mediation”, Moderation, and Conditional Process Analysis: A Regression-Based Approach, The Guilford Press, New York, NY.
Hayes, A.F. (2018), “Partial, conditional, and moderated moderated mediation: quantification, inference, and interpretation”, Communication Monographs, Vol. 85 No. 1, pp. 4-40.
Hayes, A.F. and Preacher, K.J. (2013), “Conditional process modeling: using structural equation modeling to examine contingent causal processes”, in Hancock, G.R. and Mueller, R.O. (Eds), Quantitative Methods in Education and the Behavioral Sciences: Issues, Research, and Teaching. Structural Equation Modeling: A Second Course, IAP Information Age Publishing, pp. 219-266.
Ho, J. Kim, A. and Yamakawa, N. (2020), “Survey: Asian consumer sentiment during the COVID-19 crisis”, Mckinsey and Company, available at: www.mckinsey.com/featured-insights/asiapacific/survey-asian-consumer-sentiment-during-the-Covid-19-crisis (accessed 26 July 2020).
Hu, L.T. and Bentler, P.M. (1999), “Cutoff criteria for fit indices in covariance structure analysis: conventional criteria versus new alternatives”, Structural Equation Modeling: A Multidisciplinary Journal, Vol. 6 No. 1, pp. 1-55.
Hwang, J., Kim, W. and Kim, J.J. (2020), “Application of the value-belief-norm model to environmentally friendly drone food delivery services: the moderating role of product involvement”, International Journal of Contemporary Hospitality Management, Vol. 32 No. 5, pp. 1775-1794.
Jaini, A., Quoquab, F., Mohammad, J. and Hussin, N. (2019), “Antecedents of green purchase behavior of cosmetics products”, International Journal of Ethics and Systems, Vol. 36 No. 2.
Jungmann, S.M. and Witthöft, M. (2020), “Health anxiety, cyberchondria, and coping in the current COVID-19 pandemic: which factors are related to coronavirus anxiety?”, Journal of Anxiety Disorders, Vol. 73, p. 102239.
Kaiser, F.G. and Shimoda, T.A. (1999), “Responsibility as a predictor of ecological behaviour”, Journal of Environmental Psychology, Vol. 19 No. 3, pp. 243-253.
Kollmuss, A. and Agyeman, J. (2002), “Mind the gap: why do people act environmentally and what are the barriers to pro-environmental behavior? ”, Environmental Education Research, Vol. 8 No. 3, pp. 239-260.
Komolafe, M.A., Olorunmoteni, O.E. and Fehintola, F.O. (2020), “Effect of health education on level of awareness and knowledge of Nigerian in-school adolescents on stroke and its risk factors”, Journal of Stroke and Cerebrovascular Diseases, Vol. 29 No. 5, p. 104757.
Lam, L.W., Chuang, A., Wong, C.S. and Zhu, J.N. (2019), “A typology of three-way interaction models: applications and suggestions for Asian management research”, Asia Pacific Journal of Management, Vol. 36 No. 1, pp. 1-16.
Lau, J.T., Yang, X., Tsui, H.Y. and Pang, E. (2004), “SARS related preventive and risk behaviours practised by Hong Kong-mainland China cross border travellers during the outbreak of the SARS epidemic in Hong Kong”, Journal of Epidemiology and Community Health, Vol. 58 No. 12, pp. 988-996.
Leung, G.M., Lam, T.H., Ho, L.M., Ho, S.Y., Chan, B.H.Y., Wong, I.O.L. and Hedley, A.J. (2003), “The impact of community psychological responses on outbreak control for severe acute respiratory syndrome in Hong Kong”, Journal of Epidemiology and Community Health, Vol. 57 No. 11, pp. 857-863.
Liu, Y., Lee, J.M. and Lee, C. (2020), “The challenges and opportunities of a global health crisis: the management and business implications of COVID-19 from an Asian perspective”, Asian Business and Management, Vol. 1.
Lorenzoni, I., Nicholson-Cole, S. and Whitmarsh, L. (2007), “Barriers perceived to engaging with climate change among the UK public and their policy implications”, Global Environmental Change, Vol. 17 Nos 3/4, pp. 445-459.
Maliszewska, M. Mattoo, A. and Van Der Mensbrugghe, D. (2020), “The potential impact of COVID-19 on GDP and trade: a preliminary assessment”.
Mann, C.L. (2020), “8 Real and financial lenses to assess the economic consequences of COVID-19”, Economics in the Time of COVID-19, Vol. 81.
Markle, G.L. (2013), “Pro-environmental behavior: does it matter how it’s measured? Development and validation of the pro-environmental behavior scale (PEBS)”, Human Ecology, Vol. 41 No. 6, pp. 905-914.
Nicola, M., Alsafi, Z., Sohrabi, C., Kerwan, A., Al-Jabir, A., Iosifidis, C. and Agha, R. (2020), “The socio-economic implications of the coronavirus and COVID-19 pandemic: a review”, International Journal of Surgery, Vol. 78.
Onwezen, M.C., Antonides, G. and Bartels, J. (2013), “The norm activation model: an exploration of the functions of anticipated pride and guilt in pro-environmental behaviour”, Journal of Economic Psychology, Vol. 39, pp. 141-153.
Ramus, C.A. and Killmer, A.B. (2007), “Corporate greening through prosocial extrarole behaviours – a conceptual framework for employee motivation”, Business Strategy and the Environment, Vol. 16 No. 8, pp. 554-570.
Raykov, T. and Widaman, K.F. (1995), “Issues in applied structural equation modeling research”, Structural Equation Modeling: A Multidisciplinary Journal, Vol. 2 No. 4, pp. 289-318.
Roy, D., Tripathy, S., Kar, S.K., Sharma, N., Verma, S.K. and Kaushal, V. (2020), “Study of knowledge, attitude, anxiety and perceived mental healthcare need in Indian population during COVID-19 pandemic”, Asian Journal of Psychiatry, p. 102083.
Saleem, M.A., Eagle, L., Yaseen, A. and Low, D. (2018), “The power of spirituality: exploring the effects of environmental values on eco-socially conscious consumer behaviour”, Asia Pacific Journal of Marketing and Logistics, Vol. 30 No. 4, pp. 867-888.
Sardeshmukh, S.R. and Vandenberg, R.J. (2017), “Integrating moderation and mediation: a structural equation modeling approach”, Organizational Research Methods, Vol. 20 No. 4, pp. 721-745.
Sawitri, D.R., Hadiyanto, H. and Hadi, S.P. (2015), “Pro-environmental behavior from a social cognitive theory perspective”, Procedia Environmental Sciences, Vol. 23, pp. 27-33.
Schultz, P.W. (2000), “Emphasizing with nature: the effects of perspective taking on concern for environmental issues”, Journal of Social Issues, Vol. 56 No. 3, pp. 391-406.
Schwartz, S.H. (1977), “Normative influences on altruism”, Advances in Experimental Social Psychology, Vol. 10 No. 1, pp. 221-279.
Schwartz, S.H. (1992), “Universals in the content and structure of values: theoretical advances and empirical tests in 20 countries”, Advances in Experimental Social Psychology, Vol. 25 No. 1, pp. 1-65.
Shin, Y.H., Im, J., Jung, S.E. and Severt, K. (2018), “The theory of planned behavior and the norm activation model approach to consumer behavior regarding organic menus”, International Journal of Hospitality Management, Vol. 69, pp. 21-29.
Shkoler, O. and Kimura, T. (2020), “How does work motivation impact employees’ investment at work and their job engagement? A moderated-moderation perspective through an international lens”, Frontiers in Psychology, Vol. 11, p. 38.
Sony, A. and Ferguson, D. (2017), “Unlocking consumers’ environmental value orientations and green lifestyle behaviors”, Asia-Pacific Journal of Business Administration, Vol. 9 No. 1, pp. 37-53.
Sosik, J.J., Jung, D. and Dinger, S.L. (2009), “Values in authentic action: examining the roots and rewards of altruistic leadership”, Group and Organization Management, Vol. 34 No. 4, pp. 395-431.
Stern, P.C. (2000), “New environmental theories: toward a coherent theory of environmentally significant behavior”, Journal of Social Issues, Vol. 56 No. 3, pp. 407-424.
Stern, P.C. and Dietz, T. (1994), “The value basis of environmental concern”, Journal of Social Issues, Vol. 50 No. 3, pp. 65-84.
Stern, P.C., Dietz, T., Abel, T., Guagnano, G.A. and Kalof, L. (1999), “A value-belief-norm theory of support for social movements: the case of environmentalism”, Human Ecology Review, pp. 81-97.
Stoltenberg, S.F., Batien, B.D. and Birgenheir, D.G. (2008), “Does gender moderate associations among impulsivity and health-risk behaviors?”, Addictive Behaviors, Vol. 33 No. 2, pp. 252-265.
Tang, C.S.K. and Wong, C.Y. (2004), “Factors influencing the wearing of facemasks to prevent the severe acute respiratory syndrome among adult Chinese in Hong Kong”, Preventive Medicine, Vol. 39 No. 6, pp. 1187-1193.
Vaske, J.J., Jacobs, M.H. and Espinosa, T.K. (2015), “Carbon footprint mitigation on vacation: a norm activation model”, Journal of Outdoor Recreation and Tourism, Vol. 11, pp. 80-86.
Wang, L., Wong, P.P.W. and Alagas, E.N. (2020), “Antecedents of green purchase behaviour: an examination of altruism and environmental knowledge”, International Journal of Culture, Tourism and Hospitality Research, Vol. 14 No. 1.
Westfall, P. and Henning, K.S. (2013), Understanding Advanced Statistical Methods, CRC Press.
WHO (2020), Population-Based Age-Stratified Seroepidemiological Investigation Protocol for Coronavirus 2019 (COVID-19) Infection, 26 May 2020 (No. WHO/2019-nCoV/Seroepidemiology/2020.2), World Health Organization.
Xie, X., Huang, L., Li, J.J. and Zhu, H. (2020), “Generational differences in perceptions of food health/risk and attitudes toward organic food and game meat: the case of the COVID-19 crisis in China”, International Journal of Environmental Research and Public Health, Vol. 17 No. 9, p. 3148.
Naja, F. and Hamadeh, R. (2020), “Nutrition amid the COVID-19 pandemic: a multi-level framework for action”, European Journal of Clinical Nutrition, pp. 1-5.