Motives Affecting Fashion Consumers Behavior Toward Online Shopping During Covid-19 Pandemic

COVID-19 has begun to penetrate Indonesia and has many impacts so that people can begin to adapt to this virus. People have experienced changes in their shopping behavior during this pandemic, especially in the fashion industry, many of them prefer to shop through online shops to avoid crowds and apply government regulations, with this policy requires adaptation. To sustain competitiveness, the fashion industry must understand customer behavior during this global crisis. This study aims to measure the influences between factors towards fashion consumption among the Z generation during the pandemic. Subjective norms and hedonic motivation were the independent variables in this research, perceived usefulness as a mediating variable, while the intention to purchase fashion online shopping during the pandemic was the dependent variable. This research was quantitative and surveys with 369 valid respondents. This study investigates how technology acceptance model (TAM) factors and social factors determine customer purchase intention. The research was conducted using questionnaires and analyzed using SEM, with SmartPLS Software. The finding suggests that perceived usefulness and subjective norms positively and significantly influenced intention to purchase fashion products online.


INTRODUCTION
The Covid-19 pandemic has been running for a year and a half in Indonesia. Covid-19 has had a significant impact on the people of Indonesia, especially in terms of lifestyle changes (Azmi & Genoveva, 2020). Social restrictions cause customers to switch to online purchases (Lestari & Genoveva, 2021). Reducing shopping activities during the pandemic becomes minimize the risk of transmission (Genoveva & Syahrivar, 2020). This change, makes every individual must be ready to face a cultural shift where every fulfillment of needs must go hand in hand with technological developments. In regions that apply PSPB (social restrictions by local government), the online purchases is increase, in the period January -February 2021, there is an increase in the frequency of orders by 4.4%, then it increases dramatically in the March-April period to 53.7%. In the next period, namely May-June, it increased again to 79.3% and increased significantly in the July-August period to 109.3% (Ekarina, 2021).
The best-selling online product according to Aprilia, an SEO (Search Engine Optimization) content in 2021, is fashion, where 55% of online shop buyers buy clothes (Aprilia, 2021). Similar to a number of retailers, especially fashion, have closed a number of their outlets during the Covid-19 pandemic, however, they sales have shifted to online. Senior Vice President of MAP Fashion said that before the pandemic, office clothes sold the most. During the pandemic the trend changes, the types of clothes that are sold are casual and sports clothes (Kompas.com, 2020). On the other hand, fashion companies use social media networks to remain innovative and productive, resulting in a variety of new and exciting campaigns. The world's fashion brands recognize the opportunities presented by this epidemic and continue to hold online fashion shows, stay-at-home campaigns, and fundraise (Kulsum, 2020). Therefore, to survive, clothing companies that have been relying on retail stores have turned to online sales.
The argument from (Vahdat, 2020) state that online technology has a major impact on customer shopping processes and behavior in this digital era. As a result, both large and small businesses want to enter the world of the internet. Technology Acceptance Model (TAM), on the other hand, is the most commonly recognized paradigm for understanding customer attitudes about mobile technology adoption. The important difference here is that the determinants of attitude, which are determined individually, are adapted from the theory of reasoned action. Subjective norms are one of the factors that influence consumer purchasing decisions, because they are based on a person's perception of what he should or should not do. It is assumed that customers will be more likely to make online purchases if they see their peers supporting them (Garcia et. al., 2020). Meanwhile, customers are attracted to fulfill their demands because of their shopping motivation. Consumer purchase motivation is divided into two categories: utilitarian and hedonic. Hedonic purchase motivation refers to consumer behavior that is motivated by pleasure, entertainment, fantasy, and sensory input (Tyrväinen, 2020). Consumers tend to buy utilitarian goods from offline stores and hedonic products from online stores. Both of the above will be effective if supported by perceived usefulness. (Syakir & Setiyanto, 2019) define that perceived usefulness when consumers find it easy to interact with e-commerce pages, including searching for information about products and making payments online, they will consider shopping online. The indicators that can be used for these variables include easy to learn, can be controlled, clear and easy to understand, flexible, easy to become skilled and easy to use. A customer's buying decision is influenced by perceived usefulness, so if a customer perceives a high level of benefit from shopping on a social media site like Facebook, their chances of making a real transaction will increase.
Based on the problems above, the authors assume that the shift of fashion entrepreneurs from retail stores to online in order to maintain the sustainability of their business. By understanding the factors related to consumer perceptions, namely subjective norms and hedonic motivation, they can encourage consumers intention. In addition, aspects of digital technology related to ease of usefulness, this combination will encourage consumers to intention of purchase.
To our best knowledge, this is the first study of customer behavior in buying fashion through e-commerce during the Covid-19 pandemic. In addition, our respondents are also exclusive, namely the Z generation. We choose the Z generation because they are the most technology literate (Szymkowiak, et.al., 2021).

Purchase Intention
A customer's desire to make a purchase is referred to as purchase intention. To put it another way, purchase intention refers to a buyer's probability of purchasing a thing after considering it (Rehman, 2019). While, (Kian et al., 2017) arque that purchase intention is also regarded as an element of a consumer's cognitive behavior that reveals how an individual intends to buy a particular brand. Purchase intention can be used to predict what items or brands customers will purchase next time they go shopping. A positive purchase intention reflects a consumer's loyalty to a specific product or brand and indicates the probability of purchasing. Customer knowledge, consumer perceptions, product packaging or design, and celebrity endorsement, to name a few, all impact buy intent (Younus, Rasheed, & Zia, 2015). Consumers go through a process to satisfy their desires, which includes making a purchase decision (Leiwin & Genoveva, 2020).

Subjective Norms
Subjective norms are based on one's perception of what one should or should not do depending on the potential reward or punishment for doing it (Garcia et al., 2020). While (Nikita, 2015) and (Hasbullah et al., 2015) share a similar view, they define subjective norms as perceived pressure, as exerted by neighbors, co-workers, classmates, and others who participate and have an impact on the customer's behavior, either directly or indirectly. Then, (Noor et al., 2020) highlighted from the aspect of online shopping, subjective norms influenced by public opinions and expectations from specifics groups. Subjective norms refer to ''perceived pressures on a person to perform a given behavior and the person's motivation to comply with those pressures.'' Thus, subjective norms reflect how the customer is affected by the perception of some significant referents (e.g., family, friends, and colleagues, among others) of his/her behavior. Subjective norms have been considered to be a key factor of perceived usefulness in several psychological research (Kim, Kim, & Shin, 2009). In the same result come from (Andoh, 2018) he found that, subjective norms influence on perceived usefulness. Besides that, Noor et al (2020) also concluded that, subjective norms have an impact on purchase. This result in line with (Koch, Frommeyer, & Schewe, 2020), they state that subjective norms are found to have a positive effect on e-commerce purchase intentions. Therefore, our first and second hypothesis as follows: H1: During the pandemic, consumers' subjective norms are positively influence to perceived usefulness.
H2: During the pandemic, consumers' subjective norms are positively influence to purchase intention

Hedonic Motivation
Customers are attracted to markets to meet their demands because of their shopping motivations. Consumer purchase motivations are divided into two categories: utilitarian and hedonic. Hedonic purchasing motivation refers to consumer behavior that is motivated by pleasure, amusement, fantasy, and sensory inputs. In the context of technology, Hedonic Motivation is defined as the enjoyment or pleasure derived from the use of technology (Tyrväinen, 2020). Consumers tend to buy utilitarian goods from offline stores and hedonic products from online stores. The competitive advantage of online stores allows it to better meet consumer needs (Shen, Cai, & Guo, 2016). Hedonic motivation influences the interests of the purchase process, such as sights, pleasure, enjoyment, and distraction, and is utilized as a self-centered requirement such as freedom of shopping atmosphere, fanciful needs, escape from life, and distraction, rather than pragmatism (Rajan, 2020).
In consumer behavior research, hedonic motivation is defined as a significant predictor of purchase intention, and it has recently been linked to customer happiness and loyalty. In the context of technology, Hedonic Motivation is defined as the enjoyment or pleasure derived from the use of technology (Tyrväinen, 2020). The most often used model for the acceptance of information technology and information systems is the technology acceptance model (TAM). TAM, on the other hand, is limited in its ability to handle intrinsic motivation, particularly cognitive absorption -a person's sensory/cognitive curiosity, control, pleasure and enjoyment, engagement, and temporal dissociation. HMSAM is an HMS (Hedonic -Motivation System) acceptance model that serves as a complement to other theoretical perspectives such as the TAM (Technology Acceptance Model) and the unified theory of technology acceptance and use (Oluwajana et.al, 2019) Based on the discussion above, our second and fourth hypothesis as follow: H3: During the pandemic, consumers' hedonic motivation is positively influence to perceived usefulness H4: During the pandemic, consumers' hedonic motivation is positively influence to purchase intention

Perceived Usefulness
Consumer perceptions of perceived usefulness can exchange the positive purchase intention in the context of e-commerce (Wang, Yeh, & Liao, 2013). In ecommerce, usefulness will be described as consumers' perceptions that purchasing through online stores will increase the outcome of their shopping experience (Garcia et al., 2020). According to (Keni, 2020) perceived usefulness as customers judgement in which they believe that the technologies that the customers to adopt will improve they activities. In this research purchase intention is directly impacted perceived usefulness. The degree to which a person believes that employing a given system would improve his or her job performance is referred to as perceived usefulness. Attitude, Subjective Norms, and Perceived Behavioral Control all have a direct impact on intention (Ha & Nguyen, 2019). From the several definitions above, the authors can conclude that perceived usefulness is how an online shopping platform provides benefits to consumers, both the products offered and information related to the product.
As compared to manual shopping systems, the perceived usefulness of an online shopping system can be felt as it can be used quickly and easily, as well as improve the job performance, increases productivity, enhance effectiveness and the system is useful (Venkatesh & Davis, 2000). The data is also supported by (Rahmiati & Yuannita, 2019) discovered that perceived usefulness had an impact on consumer purchase intentions. It can be concluded from these statements that perceived usefulness plays an important role in a person's intention to act towards a system. Several studies have shown that perceived usefulness has a significant effect on online shopping intention and is positively linked to online users' habits and attitudes. On the other hand, Yulianita (2020) found that perceived usefulness as the most influence on millennials generation in online shopping context. Therefore, our fifth hypothesis as the follow: H 5 : During the pandemic, consumers' perceived usefulness is positively influence to purchase intention.

METHODOLOGY
This research is based on a phenomenon that occurs among Z Generation who are already familiar with online shopping, which is the population in this study, they can become respondents if they have made a purchase of fashion products, at least once during the pandemic. Because it involves quite a lot of respondents, this research is a quantitative research (Sugiyono, 2018). Data was collected by using an online questionnaire distributed by non-probability sampling using the snowball sampling method. The total respondents who filled out the questionnaire validly were 369 people. The data is processed using SEM, with SmartPLS software.
This study uses four variables adapted from the theory of Technology Acceptance Model (TAM) proposed by Davis (1989). Those variables are Subjective Norms and Hedonic Motivation as the exogenous variables (Independent), Perceived Usefulness as the moderating variable and Purchase Intention as the endogenous (dependent) variable (Br. Ginting, 2009). The research framework as shown in figure  1 below.

Figure 1. Theoretical Framework
The total questions for the preparation of the questionnaire were 27 questions. The questions for the Subjectives Norms variables consist of 7 questions by adopting (Garcia et.al., 2020) theory. Meanwhile, Hedonic Motivation from (Gupta & Gentry, 2016) consisted of 7 questions. Then Perceived Usefulness consists of 7 questions with the adoption of theory from (Cho & Sagynov, 2015). The last is the Purchase Intention variable consisting of 6 questions using the theory of (Dachyar & Banjarnahor, 2017) and (Rodriguez & Fernandez, 2017).

Respondent Profile
The gender of respondents only has a few gaps between female and male. Female with 185 respondents (50.1%), while male with 184 respondents (49.9%), the respondent in this research was born in range of 1997-2012 with 100% (Z generation). Means that, the respondents are suitable with purchase intention, it can summarize that majority respondent in this research is university students with 284 respondents (77.0%), the worker 52 respondents (14%) and the rest are entrepreneurs and job seekers as much as 32 respondents (9%). It might conclude that these respondents are suitable enough with purchase behavior. The majority of the respondents are chosen Shopee, therefore, Shopee become the most used online shopping platform with 220 respondents (59.6%), the second is Tokopedia with 80 respondents (21.7%) and rest come from the mixture of Instagram, Facebook, Zalora, Lazada, official website of 69 respondents (18.7%).

Loading Factor (Outer Model)
Our research consists of 4 variables with 27 indicators. The evaluation of the measurement model is processed using the PLS Algorithm. Figure 2 is the corrected loading factor, which shows that from 27 indicators, there are 6 indicators that were removed because they did not meet the criteria > 0.7 so that they did not affect the path coefficient (Hair et. al.,2014).

Cross Loadings
Based on the results obtained in table 1 below, the cross loading of each latent variable indicator meets the requirements, where the cross-loading value is greater than the cross loading of other latent variables (shaded areas in gray). Based on these results, it can be concluded that the indicators used in this study have appropriate discriminant validity in constructing each variable.

Cronbach's Alpha
In table 2 below, it can be seen that the Cronbanch's Alpha value of each research variable is > 0.7. This conclusion shows that all research variables have met the requirements of Cronbanch's Alpha value, therefore, it can be said that all variables have a high level of reliability and the composite reliability value of all variables also shows > 0.7.  Table 3 shows that the R-square value of Perceived usefulness is 0.321, meaning that it has a moderate to weak correlation (Hair et.al., 2014), because the R-square between 0.25 -0.50 indicates that 32.1% Subjectives Norm and Hedonic Motivation contribute to Perceived Usefulness, while the rest, namely 67.9% was explained by other factors that did not take part in this study. The results also show that the R-square of Purchase Intention is 0.543, has a Moderate to Strong correlation (Hair et al., 2014), because the R-square is in the area of 0.50 -0.75 which means that the Perceived Usefulness variable contributes to Purchase Intention by 54.3%, the remaining 45.7% is influenced by other factors. which were not investigated in this study.  (Sarwono & Narimawati, 2015) suggest that relevant predictive criteria can be categorized into three groups, namely the value of 0.02 has a weak influence, while 0.15 has a sufficient influence and 0.35 has a strong influence. In table 4 it can be concluded that Subjective Norms and Hedonic Motivation on Perceived Usefulness have a sufficient influence with a value of 0.173, while Perceived Usefulness on Purchase Intention has a strong influence because it has a value of 0.307.

Model Fit
Take a look to the NFI value in Table 5 below, it shows the result of model fit. The model used in this research is 84% fit. Model fit value of 0.84 means that the diversity in the study is 84%. The results of this study indicate that the observed values have been reconstructed well and have a predictive relevance model, because the Q-square value is > 0 (Sarwono & Narimawati, 2015).

Hypothesis Testing
Our first hipothesis (H1) is accepted within the t-statistic value 5.688 which is more than 1.96, and p-value 0.000, which is < 0.05 (table 6). Therefore, Subjective Norms has a positive significant effect towards customers Perceived Usefulness. Subjective norm refers to a person's perception of whether or not others think he or she should engage in a particular action. The subjective norm indicators, which will inquire from family, influence, and colleague impact (Nikita, 2015). It is mainly about a person's perception of other people's opinions about whether or not he or she should engage in the action, namely friends and relatives (Kim, Kim, & Shin, 2009;Teravest et al, 2020).  Our second hypothesis (H2) is accepted within the t-statistic value 2.190 which is more than 1.96, and p-value 0.029, which is < 0.05. Therefore, Subjective Norms has a positive significant effect towards customers Purchase Intention. This result in line with the previous research from (Noor N.M. et al., 2020) and (Koch, Frommeyer, & Schewe, 2020) in the context on online purchase intention.
Our third hypothesis (H3) is rejected within the t-statistic value 1.954 which is lower than 1.96, and p-value 0.051, which is >0.05. Therefore, Hedonic Motivation has not significant effect towards customers Perceived Usefulness. Consumers tend to purchase utilitarian goods from offline stores and hedonic products from online stores (Shen et al, 2016). Our assumption of this result is that Z Generation enjoys shopping with friends more offline than online. In addition to shopping, they can enjoy other activities after or during shopping such as eating together, watching movies and other togetherness events.
Our fourth hypothesis (H4) is accepted within the t-statistic value 7.852 which is more than 1.96, and p-value 0.000, which is <0.05. Therefore, Hedonic Motivation has significant effect towards customers intention. Hedonic motivation influences the interests of the purchase process, such as freedom of shopping atmosphere, fanciful needs, escape from life, and distraction, rather than pragmatism (Rajan, 2020). During the covid-19 pandemic, customers limit themselves to shopping outside, online shopping solutions are an option and can be assumed as part of entertainment for consumers.
Our fifth hypothesis (H5) is accepted within the t-statistic value 2.875 which is more than 1.96, and p-value 0.004, which is < 0.05. According (Garcia et al., 2020) perceived usefulness is influence to consumers' perceptions after they have used something; therefore, in e-commerce, usefulness will be described as consumers' perceptions that purchasing through online stores will increase the outcome of their shopping experience. The positive relationship between perceived usefulness and consumers' behavioral intentions to buy online has a lot of empirical support (Koch, Frommeyer & Schewe, 2020). The result of this research is in line with the previous research by (Dachyar & Banjarnahor, 2017) that stated perceived usefulness proved to be significant towards purchase intention in Company B.

CONCLUSION
Our contribution through this research are, first, subjective norms in Z Generation are very important, in this case the psychological influence of peers, family, and the people around them influence Z Generation on purchase intentions. Second, hedonic motivation during the pandemic, apparently has no effect on Z generation in perceived usefulness. Our assumption of this result is that Z Generation enjoys shopping with friends more offline than online. In addition to shopping, they can enjoy other activities after or during shopping such as eating together, watching movies and other togetherness events. Third, perceived usefulness affects purchase intentions, meaning that sellers can use this strategy as part of a strategy to increase sales. Enjoying and being entertained when making the purchase process requires online business managers to be more creative in order to engage with customers.
The limitation in this study is that the respondents are exclusively in Z generation, comparison with other generations can be an alternative for future research. In addition, combining quantitative and qualitative research will be able to reveal other aspects that cannot be known from quantitative data. Using other variables such as self-efficacy, personality factors, ease of use, and social media such as influencer, the content and delivery can add to the wealth of future research.