Solidifying Fear Appeal Effect on E-Commerce Customer Purchase Intention during Covid-19 Pandemic through Digital Public Relation Persuasion

The aim of this study is to measure the effect of fear appeal and digital public relation persuasion respectively on e-commerce customer purchase intention and to solidify the effect of fear appeal on e-commerce customer purchase intention through digital public relation persuasion, during the Covid-19 pandemic. For this survey study, the data processed with SPSS and Amos programs is collected from Indonesian e-commerce customers, and taken using a snowball and non-probable sampling method. All hypotheses are supported. The constructs of this study can elaborate the theory of reasoned action since fear appeal and digital public relation persuasion function to solidify e-commerce customer purchase intention. The implication is e-commerce must develop digital public relation persuasion to solidify the effect of fear appeal on e-commerce customer purchase intention


I. INTRODUCTION
Covid-19 outbreak affects global economy.The 2020 World Economic Forum data reveals that from January to March 2020, the world economy contracted by 12 percent (WEF, 2020).The epidemic causes most businesses to close, decreases stock markets and government revenues, and drastically changes many aspects of life (Miller, 2006).In America, the outbreak decreases economic activity to an all-time high (Barker, et.al, 2020).Even when the fear of being affected by the outbreak has increased online transactions for personal protective equipment in Singapore, Australia (Insead, 2020& BBC, 2020), and in Wuhan China (Addo et al., 2020), Malaysian ecommerce revenues is actually sharply decreased.The decline is not only for the personal protective equipment, but also for food and clothing transactions.The condition causes a decrease in the imports of goods and affects economic activity.According to Hasanat, et., al, (2020), the decline is suspected since the goods sold by the Malaysian e-commerce originate from China, the country that is the source of the Covid-19 virus.
The different phenomenon demonstrated by e-commerce customers in Singapore, Australia, Wuhan and Malaysia shows the inclusiveness of the effect of fear appeal on e-commerce customers purchase intention during the Covid-19 pandemic.However, previous research findings also find the inclusiveness of the effect of fear appeal on customers purchase intention.When Kohn et al., (1982) and O'Hegarty et al., (2006) find the effect, Ruiter et al., (2001) do not.An effective approach to behavior changes is needed to solidify the fear appeal effect on e-commerce customer purchase intention (Glock & Kneer, 2009).The behavior changes can be realized by controlling customer acceptance toward the products offered by e-commerce (Toerien et al., 2016).The control can be carried out through relational marketing activities (Weiss, 2008) with a collaborative relationship approach (Riofita & Dimasadra, 2022;and Toerien et al., 2016).E-commerce must do persuasion to take the activities and approach to decrease and even eliminate the fear (Keller & Block, 1996).Considering that e-commerce uses an online platform (Purba & Setiyaningrum, 2022;Hertaswari & Dewi, 2021), the persuasion must be carried out digitally (Riofita, 2022;and Riofita & Iqbal, 2022).E-commerce needs an effective digital media for the persuasion.For the reason, e-commerce can empower digital public relation persuasion.Through the digital public relation persuasion, e-commerce can persuade the customers in interactive communication, even trigger and organize the information that hinders the customer's desire to do shopping and then provide the solutions that can attract the customers to do the shopping (Riofita & Dimasadra, 2022).The digital public relation persuasion works to represent e-commerce convincingly to build customer confidence, trust and loyalty (Addo et al., 2020).Even for accurate results, the digital public relation persuasion can be carried out massively.To succeed the works, the digital public relation persuasion acts as: (1) communication technician, namely handling the relationship with e-commerce customers through attractive promotion media designed to form their unpleasant emotional in a persuasive message to arouse their fear and to trigger them then to have purchase intention; (2) communication facilitator, namely; representing customers to inform e-commerce about their needs to address their fears, and representing e-commerce to inform customers about the products they can consume to do so; (3) expert prescriber, namely assisting customers to find the solutions of their fears by offering them to consume products provided by e-commerce; and (4) problem solver, namely assisting e-commerce to make decisions in determining the right product to address customer fears (Heerden & Rensburg, 2005).The activities show that the digital public relation persuasion is proper to work in online communities for the interest of e-commerce during the Covid-19 pandemic restricting social interaction (Wright & Hinson, 2009;and Constantinides & Stagno, 2011).
The lesson from the Covid-19 pandemic and discussions about the inclusiveness of the findings on the effect of fear appeals on e-commerce customer purchase intentions should provide the breakthroughs that can address the inclusiveness.Therefore, several questions are proposed to be answered through research results.The questions are: (1) Does fear appeal have a significant effect on e-commerce customer purchase intention?(2) Does fear appeal have a significant effect on digital public relation persuasion?(3) Does digital public relation persuasion have a significant effect on e-commerce customer purchase intention?(4) Does digital public relation persuasion mediate the effect of fear appeal on e-commerce customer purchase intention significantly?Therefore the aim of this study is to measure the effect of fear appeal and digital public relation persuasion respectively on e-commerce customer purchase intention and to solidify the effect of fear appeal on e-commerce customer purchase intention through digital public relation persuasion, during the Covid-19 pandemic.

A. E-Commerce Customer Purchase Intention
E-commerce customer purchase intention is customer willingness to make an online purchase through electronic shops during the Covid-19 pandemic (García-Salirrosas et al., 2022;and Chen, 2011).The purchase intention is determined by some factors (Purba & Setiyaningrum, 2022), such as customer's experience, preferences, external information, and alternative evaluation (Parasuraman et al., 1988;and Su & Li, 2019).During the Covid-19, the factors encourage customer to develop fear appeal since the factors can capture customer's mind (Ling et al., 2010;and Day., 1969).Through the factors, the fear appeal will function to predict the e-commerce customer purchase intention (Johnston & Warkentin 2010;and Xu et al., 2017 ).Since no conclusive results for the effect of the fear appeal on customer purchase intention (Kohn et al., 1982;O'Hegarty et al., 2006;and Ruiter et al., 2001), this study solidify the effect of the fear appeal on e-commerce customer purchase intention through digital public relation persuasion.The relationships between the constructs refer to the theory of reasoned action.Based on the theory, fear appeal and digital public relation persuasion are the triggers to predict customer behaviour in the form of e-commerce customer purchase intention (Fishbein & Ajzen, 1975).

B. Fear Appeal
Fear appeal is e-commerce customer unpleasant emotional designed by the e-commerce in a persuasive message to arouse their fear and to trigger them then to have purchase intention (Ruiter et al., 2014;and Morales et al., 2012).Through the fear appeal, e-commerce disseminates threat and the dangerous of the threat that will come to the customers, as well as the ways to save from the threat by buying the products sold by the e-commerce (Witte, 1992).The fear appeal is a motivator to assist the customers (Williams, 2012).Even customer needs are developed from the fear appeal.Through the fear appeal, e-commerce triggers the customer to have purchase intention then.It is in line with the previous studies conducted by Johnston & Warkentin (2010) and Xu et al., (2017 ).The proposed hypothesis is: H1: Fear appeal has significant effect on e-commerce customer purchase intention.
Fear appeal developed by e-commerce marketing strategy however can be responded by e-commerce customers in many ways.Some may avoid the messages disseminated through the fear appeal, some may minimize the threat revealed by the messages, some may pay selective attention to the messages, and some may deny the messages for personal relevance (Eagly & Chaiken, 1993).Such responses must be treated by ecommerce effectively.In social marketing, the treatment can be manifested in the form of persuasion.The persuasion should be built in interactive communication, trigger and organize the information preventing ecommerce customer purchase intention, and provide the solution attracting the customers to do shopping (Riofita & Dimasadra, 2022).Since e-commerce is in digital platform, e-commerce should apply digital public relation for the persuasion.The fear appeal will determine how the digital public relation persuasion works.The high-fear appeal will encourage the digital public relation persuasion to work more (Keller & Block, 1996).The proposed hypothesis is: H2: Fear appeal has significant effect on digital public relation persuasion.

C. Digital Public Relation Persuasion
Digital public relation persuasion is digital interactive communication provided by e-commerce to do persuasion to decrease and even eliminate customer fear to do online shopping during the Covid-19 pandemic.The digital public relation persuasion triggers and organizes the information preventing the e-commerce customer purchase intention and provides an attractive solution by doing shopping on the e-commerce (Riofita & Dimasadra, 2022).The digital public relation persuasion works to represent e-commerce convincingly to build customer confidence, trust and even loyalty (Addo et al., 2020).Therefore, the digital public relation persuasion functions to trigger e-commerce customer purchase intention (Constantinides & Fountain, 2008).This is in line with the findings of Riofita & Dimasadra (2022), Bagwell (2016), and Hassani (2017).The proposed hypothesis is: H3: Digital public relation persuasion has significant effect on e-commerce customer purchase intention.

D. The Mediation of Digital Public Relation Persuasion
Many researchers have measured the effect of fear appeal on purchase intention, Some find the effect (Johnston & Warkentin 2010;and Xu et al., 2017 ) but some do not (Toerien et al., 2016;Hasanat, et., al, 2020).To provide more conclusive result, Glock et al., (2012), Cheah, et al., (2015) and Müller et al., (2016) suggest e-commerce to carry out an effective approach toward the customer behaviour.The customer behaviour should be controlled through the customer acceptance toward the products offered (Toerien et al., 2016).The control is carried out to minimize and even to eliminate customer fear through relational marketing activities (Weiss, 2008) with a collaborative relationship approach (Riofita & Dimasadra, 2022;and Toerien et al., 2016) in the form of persuasion (Keller & Block, 1996).For effective result, e-commerce should employ digital public relation to do the persuasion, since the e-commerce is digital platform (Riofita & Dimasadra, 2022).The digital public relation persuasion works to disseminate fear appeal and convince the e-commerce customers to have purchase intention (Huang & Hagan, 2011).Through the digital public relation persuasion, e-commerce bridges the information customers need with the information e-commerce provides to enhance purchase intention of the customers' (Constantinides & Stagno, 2011).By doing so, the digital public relation persuasion has the potential to solidify the effect of fear appeal on e-commerce customers purchase intention (Constantinides & Fountain, 2008).Therefore, the proposed hypothesis is:

A. Research Design and Respondents
This study is an online survey.The questionnaire is sent via a google form link to all of researcher social media contacts, such as Facebook, Whats Up and Instagram having experience to do shopping at Indonesian e-commerce during the Covid-19 pandemic.After the respondents filled in, researcher asks them to send the survey link to their friends who also have the experience.Therefore, the respondents are taken using a snowball and non-probable sampling method for convenience, judgment, quotas (Zikmund & Babin, 2010) and to minimize budget and time constraints (Toerien et al., 2016).

B. Research Instrument
Research instrument is a questionnaire containing closed questions.The answers are in a Likert scale from 1 -5, where 1 represents strongly disagree and 5 represents strongly agree.The questionnaire questions are adapted from the research variable indicators.For fear appeal variable, the indicators are adapted from Witte (1992), namely: fear, threat and efficacy.The indicators for digital public relation persuasion are adapted from Heerden & Rensburg (2005), namely: communication technician, communication facilitator; expert prescriber; and problem solving.Meanwhile, the indicators of e-commerce customer purchase intention are adapted from Salirrosas et al., (2022), namely: the quality of website, the design of web pages, and the display of product details.

C. Data Analysis
Data is processed using SPSS and Amos programs.Data normality is analyzed through the values of multivariate, CR Skewness and CR Kurtosis.The data will be declared normal if the values are in the range of ±2.548 < normality value < ± 2.548 (Hair et al., 2017).Model reliability is analyzed through α Cronbach value.The α Cronbach value of around 0.9 is excellent, around 0.8 is very good, around 0.6 to 0.7 is adequate, and below 0.5 is unreliable (Kline, 1998).Model validity is analyzed through Confirmatory Factor Analysis, namely factor loading, the values of Average Variance Extracted (AVE) and Composite Reliability (CR).The minimum value of factor loading and AVE is 0.5 and CR is 0.7 (Hair et al., 2017).Furthermore, the fit model is analyzed through the critical value of Chi-Square (CMIN/df): 1 to 3; Probability (P): ≥ 0.05; Root Mean Square Error Approximation (RMSEA): ≤ 0.08; Incremental Fix Index (IFI): ≥ 0.90; Goodness of Fit Index (GFI): ≥ 0.90; Comparative Fix Index (CFI) ): ≥ 0.95 and Tucker-Lewis Index (TLI): 0.95 The model will be declared fit if 1 criterion of the goodness of fits meets the critical value (Hair et al., 2017).
The hypotheses are analyzed by comparing the value of t statistics with t table.If the value of t statistics is greater than the value of t table, the hypothesis is declared partially supported, and vice versa.The simultaneous effect of fear appeal and digital public relation persuasion on e-commerce customer purchase intention is analyzed by comparing the F statistic value and F table.If the F statistic value is greater than F table value, simultaneously fear appeal and public relation persuasion are declared to have an effect on e-commerce customer purchase intention, and vice versa.The magnitude of the simultaneous effect is analyzed through the value of R square.Furthermore, for digital public relation persuasion mediation on the effect of fear appeal on e-commerce customer purchase intention is analyzed through the Sobel test value at the significance level of 0.05 or 5%.If Z value or mediation value is greater than the critical value of 1.96, the digital public relation persuasion is declared mediating the effect of fear appeal on e-commerce customer purchase intention, and vice versa.

IV. RESULT/FINDINGS
The number of respondents who filled in the questionnaire is 254 respondents.However, only 247 respondents whose data can be processed as research data.Multivariate normality value is 0.222 and the skewness and kurtosis values for each indicator are in the range of ±2.548 < normality value < ± 2.548, therefore the research data is considered normal (Hair et al., 2017).α Cronbach values are in the range of 0.770 and 0.836, therefore model reliability is good (Kline, 1998).The minimum factor loading, AVE, CR values respectively are 0.555, 0.598 and 0.783.Therefore, based on the Confirmatory Factor Analysis, research model is valid.Furthermore, the research model is also fit since the values of Chi Square (CMIN/DF): 41.425/30 = 1.381;Probability: 0.080; RMSEA: 0.038; GFI: 0.970; AGFI: 0.945; CFI: 0.980; and TLI: 0.984 (Hair et al., 2017).Based on the hypothesis analyses, the effect of fear appeal on e-commerce customer purchase intention has a t-statistical value of 6.272 > 1.65 (t-table), with a p value of 0.000 or below 0.05.Therefore H1 is supported.The effect of fear appeal on digital public relation persuasion has a t-statistical value of 9.239 > 1.65 (t-table), with a p value of 0.000 or below 0.05.Therefore H2 is also supported.Meanwhile, the effect of digital public relation persuasion on e-commerce customer purchase intention has a t-statistical value of 7.062 > 1.65 (t-table), with a p value of 0.000 or below 0.05.Therefore H3 is supported.
The effect of digital public relation persuasion on online purchase intention has β value of 0.389, larger than the β value of the effect of fear appeal on e-commerce customer purchase intention (0.346).Even, the effect of the digital public relation persuasion together with fear appeal simultaneously has F-statistical value of 88.367 > 2.03 (F-table) with a p value of 0.000.Meanwhile, based on the value of R square, simultaneously the effect of fear appeal and digital public relation persuasion on e-commerce customer purchase intention is 0.405 or 40.5 %.The results reveal that the digital public relation persuasion is an important construct to develop e-commerce customer purchase intention during the Covid-19.Regression coefficient value of the effect of fear appeal on e-commerce customer purchase intention is 0.424, at a standard error value of 0.068 and a significance value of 0.000.Meanwhile, when mediating the effect of fear appeal on e-commerce customer purchase intention, the regression coefficient value of digital public relation persuasion on e-commerce customer purchase intention is 0.443, at a standard error value of 0.063 and a significance value of 0,000.Using Sobel Test, at a significance level of 0,05 or 5%, Z value or the mediation value of digital public relation persuasion on the effect of fear appeal on e-commerce customer purchase intention is 4.665, larger than the critical value of 1,96.Therefore H4 is supported.Note: significance level of 0.05 or 5%.

V. DISCUSSION
Partially fear appeal has significant effect on e-commerce customer purchase intention.The finding indicates that the fear appeal is effective to be e-commerce marketing material during the Covid-19 pandemic.The fear appeal provides social attendance in the forms of affection, acceptance and social information.The social attendance is a requirement to develop e-commerce customer purchase intention (Addo et al., 2020).Therefore the fear appeal is proper for e-commerce to attract customer purchase intention.The finding confirms the finding of Johnston & Warkentin (2010) and Xu et al., (2017 ).
Partially fear appeal has significant effect on digital public relation persuasion.The finding indicates that since fear appeal can be responded in many ways, such as avoiding the messages disseminated through the fear appeal, minimizing the threat revealed by the messages, paying selective attention to the messages, and denying the messages for personal relevance (Eagly & Chaiken, 1993), e-commerce must treat the responses effectively through the digital public relation persuasion.The high-fear appeal will encourage the digital public relation persuasion to work more (Keller & Block, 1996).
Partially, digital public relation persuasion has significant effect on e-commerce customer purchase intention.The finding indicates that the digital public relation persuasion is e-commerce media to carry out interactive communication with customers during the Covid-19 (Parise & Guinan, 2008).Through the media, e-commerce develop relational marketing activities with a collaboration relationship approach (Weiss, 2008;Klassen, 2002;and Gibbs, 2002).Through the activities and approach, the media represents e-commerce to convince e-commerce customers to have purchase intention by building confident, trust and even loyalty (Constantinides & Fountain, 2008;and Addo et al., 2020).The finding supports the findings of Riofita & Dimasadra (2022), Bagwell (2016) and Hassani (2017).
Digital public relation persuasion mediates the effect of fear appeal on e-commerce customer purchase intention significantly.As expected, as mediator, the digital public relation persuasion provides more conclusive results for the effect.The finding shows that the digital public relation persuasion can bridge the information the customers need with the information the e-commerce provides to solidify customer purchase intention during the Covid-19 pandemic (Constantinides & Stagno, 2011).Furthermore, an effect of digital public relation persuasion on e-commerce customer purchase intention is larger than the effect of fear appeal on e-commerce customer purchase intention.Even, simultaneously the effect of digital public relation persuasion and fear appeal on ecommerce customer purchase intention is good enough.The results reveal that the digital public relation persuasion can control customer behavior to minimize and even to eliminate customer fear.The digital public relation persuasion runs relational marketing activities (Weiss, 2008) with a collaborative relationship approach effectively (Riofita & Dimasadra, 2022;and Toerien et al., 2016).

VI. CONCLUTION AND RECOMMENDATION
The aim of this study is to measure the effect of fear appeal and digital public relation persuasion respectively on e-commerce customer purchase intention and to solidify the effect of fear appeal on e-commerce customer purchase intention through digital public relation persuasion, during the Covid-19 pandemic.The findings demonstrate that fear appeal and public relation persuasion respectively has significant effect on e-commerce customer purchase intention.Even digital public relation persuasion mediates the effect of fear appeal on ecommerce customer purchase intention significantly.Therefore, as expected, the digital public relation persuasion has important role as both the antecedent of e-commerce customer purchase intention and the mediator to solidify the effect of fear appeal on e-commerce customer purchase intention.The findings indicate that the constructs employed in this study can elaborate the theory of reasoned action in e-commerce context.Since the theory reveals that behaviour is predicted by trigger, fear appeal and digital public relation persuasion are the triggers for customer behaviour in the form of e-commerce customer purchase intention (Fishbein, & Ajzen, 1975).As the recommendation then, e-commerce must employ the digital public relation persuasion to solidify the effect of fear appeal on e-commerce customer purchase intention during the Covid-19.For the purpose, the digital public relation persuasion must function as: (1) communication technician, namely handling the relationship with ecommerce customers through attractive promotion media designed to form their unpleasant emotional in a persuasive message to arouse their fear and to trigger them then to have purchase intention; (2) communication facilitator, namely; representing customers to inform e-commerce about their needs to address their fears, and representing e-commerce to inform customers about the products they can consume to do so; (3) expert prescriber, namely assisting customers to find the solutions of their fears by offering them to consume products provided by e-commerce; and (4) problem solver, namely assisting e-commerce to make decisions in determining the right product to address customer fears (Heerden & Rensburg, 2005).

280H4:
Digital public relation persuasion mediates the effect of fear appeal on e-commerce customer purchase intention significantly.Based on the literature review and the development of hypotheses, this study develop conceptual framework as shown at figure 1.

Table 1 .
Instrument Measurement Validity

Table 2 .
Significance Test Results Table is 1.65; F-Table is 3.03; and Level of Significance is 0.05.