The Effect of Hedonic and Utilitarian Values on Online Impulse Buying: Mediating Role of Browsing

. The Effect of Hedonic and Utilitarian Values on Online Impulse Buying: Mediating Role of Browsing. Journal of Madani Society , 2 (1), 17-25


Introduction
Online transactions in Indonesia continue to increase annually by around 20 trillion (Barata, 2019;Rumata & Sastrosubroto, 2020). However, in 2018, the increase occurred massively, which in 2017 was around 100 trillion to 140 trillion. It indicates that every year the use of the Internet continues to grow, as well as the use of online shopping platforms. The ease of internet access makes people who initially only have the intention of seeing the product buy it, even those who have no intention of buying the product online because of the ease of access. The marketing strategy of providing various promos creates an atmosphere where consumers feel FOMO (Fear of Missing Out) (Çelik et al., 2019). FOMO is a condition where consumers buy products spontaneously and without a plan at a certain time just because they are afraid of missing an event such as a discount or flash sale or because a product is a limited edition (Reagle, 2015), even though initially they did not have the plan to purchase a certain product. This kind of thing is called impulse buying. Impulse buying is defined as an act that was not previously recognized consciously due to a consideration or purchase intention formed before entering the store (Firmansyah, 2018).
One of the factors that influence impulse buying is consumer perceived value. Chen & Dubinsky (2003) explain that consumer perceived value has become important in facilitating and predicting consumer behavior. In the context of impulsive shopping, consumer perceived value is categorized into two dimensions: hedonic and utilitarian. Consumers with hedonic values are those who shop on the basis of personal pleasure. Hedonic personality gives consumers little rationality because they will buy a product if they feel like the product without considering the price and other information. Consumers with utilitarian values tend to only shop for what they need. For example, they only look for certain products by considering the information that the product (Babin & Attaway, 2000).
Research on impulse buying has been done a lot at this time. One of them was researched by Zhang et al., (2018) regarding the effect of hedonic and utilitarian values on online impulse buying with a sample of students in China. Researchers conducted a similar study but with a different sample, namely self-employed people with control over their finances. This research will also be carried out in Bengkulu Province, considering that Bengkulu people are quite consumptive in online shopping. It was proved by an increase in online transactions which increased by 75% in the third quarter of November 2020 (Antaranews Bengkulu, published on April 10, 2021). In addition, according to CNBC Indonesia (2021), Bengkulu is included among the 10 poorest provinces in Indonesia, with a percentage of 15.30%. The phenomenon that occurs is interesting to study because it relates to the consumer behavior of the Bengkulu people. Therefore, researchers examine the effect of hedonic and utilitarian value on online impulse buying mediated by browsing.

Impulse Buying
Impulse buying is part of an unplanned purchase which is different from the planning of a consumer's spending. According to Mowen & Minor (2002), impulse buying is an act carried out without having previous problems or buying intentions or intentions that were formed before entering the store. According to Semuel (2006), some people think that shopping activities can be a tool to relieve stress, and spending money can change a person's mood significantly. In other words, money is a source of strength. The ability to spend money makes a person feel powerful. Unplanned buying means activities to spend money that is not controlled, mostly on items that are not needed. Items purchased unplanned (impulse products) are more likely to be desired to buy, and most of these items are not needed by customers. According to Kosyu et al. (2014), impulsive behavior is driven by a strong desire from consumers to fulfill their own needs at that time. When shopping, someone will have positive emotions to buy the product without prior planning in the form of a shopping list. According to Park et al. (2015), when a person's shopping experience becomes a goal to meet the satisfaction of hedonistic needs, the product chosen to be purchased is not based on the initial plan when going to the store but rather on an impulsive purchase caused by the fulfillment of hedonistic needs or because of positive emotions. Unplanned purchases (impulse buying) can be classified into four types: planned impulse buying, reminded impulse buying, suggestion impulse buying, and pure impulse buying (Beatty & Elizabeth Ferrell, 1998).
1. Pure Impulse buying 2. Pure Impulse buying is a purchase impulse made because of an emotional outburst from consumers so that they make purchases of products outside of their buying habits. 3. Reminder Impulse buying 4. Reminder Impulse buying is a purchase that occurs because consumers suddenly remember to purchase the product. Thus, the consumer has made a previous purchase or has seen the product in an advertisement. 5. Suggestion Impulse buying 6. Suggestion Impulse buying is a purchase that occurs when consumers see a product, see how to use it or how to use it, and decide to make a purchase. Suggestion impulse buying is done by consumers even though consumers do not really need it, and its use will still be used in the future. 7. Planned Impulse buying 8. Planned Impulse buying is a purchase that occurs when consumers buy products based on special prices and certain products. It indicates that consumers do not actually buy products spontaneously, but a cognitive element is still used before buying something. Impulse buying is also a purchase that is made without planning and does not need the goods/services immediately.

Hedonic Value
Researchers have recently abandoned the perspective that shopping is only a cognitive activity and have begun to examine hedonic values as controllers for shopping, such as shopping for leisure and recreation or the emotional role of mood and pleasure (Holbrook, 1982). Hedonism is motivated by the desire to have fun and play. Therefore, hedonism reflects the shopping experience's values, including fantasy, arousal, sensory stimulation, enjoyment, pleasure, curiosity, and escape. The hedonic values of spending have been confirmed (Babin et al., 1994;Holbrook & Hirschman, 1982;Scarpi, 2006). Hedonic value is related to fulfilling a product's pleasure or aesthetic aspects (Deelman et al., 2005). Hedonic is a stimulus that selects the quality of the shopping environment in terms of perceived enjoyment, visual appeal, and escapism (Subagio, 2012).
Hedonic consumption includes behavioral aspects of multi-sensory, fantasy, and emotional consumers driven by benefits such as pleasure in using products and aesthetic approaches (Holbrook & Hirschman, 1982). Therefore, the buying experience may be more important than acquiring the product. Determination of impulse buying has an important role in fulfilling hedonic desires related to hedonic consumption (Hausman, 2000;Piron, 1991;Rook, 1987). This role supports the conceptual relationship between hedonic shopping motivation and behavior-induced impulse buying. It suggests consumers are more likely to buy impulsively when motivated by hedonistic desires or economic reasons, such as pleasure, fantasy, or social or emotional satisfaction. According to Yu & Bastin (2010), aspects of hedonic value are divided into: 1. Novelty Shopping activities are a way to provide new experiences such as exploring new worlds.

Fun
Shopping activities are a way to fulfill the need for fun, joy and provide positive emotions directly.

Escape
Shopping activities allow consumers to escape from reality, forget their worries and forget the problems they are facing.

Social Interaction
Shopping is a way to communicate and increase a sense of kinship or friendship.

Utilitarian Value
Consumer behavior oriented towards utilitarian values will choose products efficiently based on rational reasons (Holbrook & Hirschman, 1982). According to Deelman et al. (2005), utilitarian value is a person's opinion about the benefits of a product. Utilitarian values are considered objectively and rationally (Neda & Kambiz, 2011). According to Ferrand et al., (2010) and To et al. (2007), the utilitarian values include: 1. Cost Savings Cost savings are a significant factor for repurchasing, where members will look for the lowest prices with the same product and service quality.

Services
Consumers also consider the services offered to make repurchases. The statement shows that utilitarian value is a form of attitude from consumers when they shop by making purchases or not making purchases of goods that they have determined according to their needs. The perception of utilitarian value can depend on what consumers want to achieve from their shopping activities. Consumers will feel satisfied if they get a product that efficiently suits their needs, especially regarding time spent. Intentional purchases made by consumers characterize it to meet their needs quickly.

Browsing
Browsing is an ongoing search activity without a specific purchase plan (Bloch et al., 1986). It is an undirected, unfocused, and stimulus-driven exploratory search process (Moe, 2003). Browsing often takes time, which allows consumers to experience the impulse to buy something impulsively in the process. Kollat & Willett (1969) argue that exposure to merchandise and in-store stimuli can lead to people's unplanned buying behavior. In online stores, (Verhagen & van Dolen, 2011) also show that online store browsing is positively related to consumers' impulse to buy impulsively. Previous research has shown that people browse not only to gather information but also to have fun (Bloch et al., 1989;Floh & Madlberger, 2013).

Conceptual Framework
This study consists of four variables. There are two independent variables such as hedonic value and utilitarian value, impulse buying as dependent variable and browsing as mediating variable. The conceptual framework of this study can be viewed below.

Figure 1. Conceptual Framework
On the basis of the conceptual framework above, the proposed hypotheses in this study are as follow: 1. Hedonic value has a significant effect on impulse buying. 2. Utilitarian value has a significant effect on impulse buying. 3. Hedonic value significantly affects browsing. 4. Hedonic value significantly affects browsing. 5. Browsing has a significant effect on impulse buying. 6. Hedonic value significantly affects impulse buying mediated by browsing. 7. Utilitarian value significantly affects impulse buying mediated by browsing.

Materials and Methods
This study is designed using a quantitative approach through a survey questionnaire. The population in this study is the people who do shopping online at Bengkulu city. The sampling technique used is non-probability sampling. The sample used in this study was 201 respondents in the city of Bengkulu. After collecting data, primary data obtained from research respondents was processed using SEM (Structural Equation Modeling) to determine the relationships between latent variables. According to Hair Jr. et al. (2017), the analytical technique used is Structural Equation Modeling (SEM) which is operated through the AMOS program. The data analysis technique using SEM (Structural Equation Modeling) was carried out to thoroughly explain the relationship between the variables in the study. The method of analysis using SEM was chosen because the research model that was analyzed was quite complicated when using multiple linear regressions.

Demography
The demography analysis indicates that gender of respondents are dominated by female as much as 54.2% and male 45.8% from total 201 respondents. In term of age, it is dominated from the age 41-as many as 41.3% along with age from 36-40 is 21.4%. Respondent's income is dominated income from Rp4.100.000-Rp5.000.000 as much as 41.3% and over Rp5.000.000 as many as 28.9%. In terms of the latest education level, respondents are dominated by the community with an undergraduate education level as big as 50.7% and followed by a postgraduate education level as much as 35.8%. In terms of frequently used e-commerce platform, the study is dominated by people who use Shopee as many as 80.1% followed by Tokopedia with percentage 17.4%. For product category, this study is dominated by respondents who shop for Fashion/Apparel products as much as 76.52%. The rest are health product of 35.82%, Technology of 21.39%, Sport of 18.41% and daily need of 23.38%.

Goodness of Fit
According to Ghozali (2016), the goodness of fit test was carried out to measure the accuracy of the sample regression function in estimating the actual value statistically.  Table 1 displays the Chi-square value shows a value of more than 0.0001, which is 661.258. However, the CMIN/DF measurement index shows good information with a value of 2.926 < 3. Other measurement indices also show good information such as RMSEA with a value of 0.069 < 0.08, PGFI of 0.626 (ranging between 0 and 1), CFI which shows a number of 0.910 > 0.90, and IFI with a value of 0.910 > 0.9. In an empirical study, researchers are not required to meet all the criteria goodness-of-fit, but it depends on the justification of each researcher. In this study, the chi-square shows the number 661,258 > 0.0001. (Katsikatsou et al., 2012) explained the reason why chi-square cannot be the only measure for the overall fit of the model, namely because chi-square is sensitive to sample size. The larger the sample, the greater the chi-square value and will lead to the rejection of the research model even though the value of the difference between the sample covariance matrix and the model covariance matrix is minimal or small.

Result of AMOS Structural Model
The structural model is the path model, which relates the independent variable to the dependent variable (Hair Jr. et al., 2017). The measurement model allows researchers to use several variables for one independent or dependent variable.  Table 2 captures that the coefficient value is regression weights 1.036 and the CR is 4.070. This shows a positive relationship between variables hedonic value and impulse buying. P value shows the number *** < 0.01 which means this number is below 0.01 which is better than the significant value at the level 0.05 or 0.01 so that there is a significant influence between the hedonic value variable and the impulse buying variable H1 is accepted. It can be concluded that hedonic value has a positive and significant effect on impulse buying. Also, the coefficient of regression weights is -.595 and CR is -2.372 (< -1.96). This indicates a negative influence between the variables utilitarian value and impulse buying. P value which shows .018 (< 0.05) which means that there is a significant influence between the utilitarian values variables on impulse buying.H2 is accepted. It can be concluded that the utilitarian value has a negative and significant effect on impulse buying.

Figure 2. AMOS Structural Model
Besides that, the coefficient of regression weights is .817 and CR is 5.188. This shows that there is a positive relationship between variables hedonic value and browsing. The P value shows the number *** < 0.01, which means that there is a significant effect between the hedonic value variable and the variable browsing. H3 is accepted. It can be concluded that the hedonic value has a positive and significant effect on browsing. In addition, the coefficient value is regression weights .193, CR is 1.096 (<1.96) and the P value shows the number .273 > 0.01 which means there is no effect between the hedonic value variable and the variable browsing. H4 is rejected. It can be concluded that the utilitarian value has no effect on the variable browsing. Moreover, the coefficient of regression weights is -.687 and the CR is 4.437. This indicates a positive influence between the variables browsing and impulse buying. The P value shows the number *** < 0.01 which means that there is a significant influence between the browsing variable and the variable impulse buying. H5 is accepted. It can be concluded that browsing has a positive and significant effect on impulse buying.

Mediating Analysis
On the basis of the rules of Baron et al. (1986), there are three kinds of variables in testing mediation. The three variables consist of a predictor, a mediator, and a criterion. In this study, the predictor variables include hedonic value and utilitarian value. The mediator variable is browsing, while the criterion variable is impulse buying. The mediation effect test is carried out through four stages, the mediation effect test can be carried out if the main effect (direct relationship of the independent variable to the dependent) is significant.

Figure 3. Result of Mediating Analysis
In the first stage, the hedonic value was proven to significantly affect the variable criterion (impulse buying) with evidence of a CR value of 4,070 (> 1.96) and a P value of 0.000 (< 0.05). In the second stage, the hedonic value can also affect the variable mediator (browsing) significantly with a CR value of 5.188 (> 1.96) and a P value of 0.000 (< 0.05). In the third stage, the variable mediator (browsing) can also significantly affect the variable criterion (impulse buying) with a CR value of 4.070 (> 1.96) and P 0.000 (<0.05). Based on the theory of ( Baron et al., 1986), the test results above show that the variable browsing partially mediates the variables hedonic value and impulse buying. H6 is accepted. It can be concluded that the variable hedonic value can affect the variable impulse buying mediated by browsing. In addition, the utilitarian value was proven to be able to significantly affect the variable criterion (impulse buying) because it had a P value of 0.018 (< 0.05) with a CR value of -2.372 (< -1.96) which indicated a negative influence. However, in the second stage the utilitarian value also can not affect the variable mediator (browsing) significantly because the CR value is 1.096 (<1.96) and the P value is 0.000 (<0.05). Based on the theory of ( Baron et al., 1986), the test results above show that the variable browsing cannot mediate between the variables utilitarian value and impulse buying. H7 is rejected. It can be concluded that utilitarian value does not affect impulse buying mediated by browsing.

Discussion
Consumer hedonic values. It can be in the form of advertising products that can spoil the eyes of consumers, various features, and others that can make consumers happy. This study is in line with research conducted by Zhang et al. (2018)) which found an influence between variables of hedonic value on impulse buying for students in China when shopping online.
In the results of this study, utilitarian value has a significant negative effect on impulse buying. The negative influence shows an opposite relationship, which means that the higher a person's utilitarian value, the less spontaneous (impulse) a person is to spend. Conversely, the lower the utilitarian value, the higher the spontaneity of people will occur in shopping. The results of this study are also in line with research conducted by Park et al. (2015), who found a negative influence from the influence of utilitarian value on impulse buying. This study supports previous studies conducted by Zhang et al. (2018), Rook (1987), Piron (1991), Bae et al. (2021), Rezaei et al. (2016), and Babin et al. (1994), which state that there is a significant influence between the variables utilitarian value and impulse buying.
The negative effect shows an opposite relationship, which means that the higher a utilitarian value person's, the less spontaneous (impulse) a person is to spend. Conversely, the lower the utilitarian value, the higher the spontaneity of people will occur in shopping. On the influence of these two variables, the P value shows the number 0.018 (> 0.05), which means that the two variables have a significant effect. Ha & Shawn Jang (2010) said that consumers who have utilitarian values could influence impulse buying. For example, the higher the consideration of the function and value of a utilitarian product in a store online, the higher the consumer's pressure to shop in the store environment online. It shows that the cognitive aspect of a person can still make people do impulse buying, and impulse buying itself is not only limited to the definition of shopping based on emotion alone but a person's rationality and cognition can still influence it.
On the basis of type of impulse buying, namely suggestion impulse buying (shopping after seeing the procedures and uses of the product) and planned impulse buying (shopping based on special prices & discounts) (Beatty & Elizabeth Ferrell, 1998), it shows that people do impulse buying not really automatically. Spontaneously without thinking or considering anything but still using his cognitive abilities, such as considering its usefulness, special prices, and product promotions. Characteristics of respondents who are dominated by an undergraduate education level (50.7%) show that respondents have a high enough level of intellectuality so that impulse buying will not only be based on emotional (hedonic value) but also cognitive levels that take into account factors such as usability, special prices, discounts, and promos that make people make impulse buying in the form of suggested impulse buying and planned impulse buying even though the intensity is not very impulsive. Conversely, the lower a person's level of rationality, the higher the impulsivity. It can create pure impulse buying because only the shopping process is dominated by emotion. Therefore, the hypothesis is accepted because the utilitarian value can affect impulse buying.
In this study, the result shows that hedonic value can affect browsing. Meanwhile, utilitarian value cannot. Hedonistic people seeking joy and happiness tend to browse because it could produce those things. Many features provided by the e-commerce platform make browsing more enjoyable. For example, in Shopee, the app provides interactive design and even users can play games on the app while browsing. It shows that those who have hedonic value are more likely to do browsing. The difference comes from people who have utilitarian values. In this study, utilitarian value cannot affect browsing. People with utilitarian values tend to do their activities effectively and efficiently. That is why in the shopping process, they only browse for products that they have planned and need in the very first place. These characteristics justify that utilitarian value cannot affect browsing.
This study shows a mediating role that connects the independent (variable hedonic value) and dependent (impulse buying). It is because the mediating variable can affect the independent variable (hedonic value) and the dependent variable (impulse buying). The mediating variable (browsing) is a partial mediation variable, which means that the independent variable can directly affect the dependent variable (impulse buying) without having to go through mediation. It indicates that when shopping online, the hedonic value of consumers can still lead to impulse buying even without having to browse because consumers with high hedonistic value shop for happiness, which can lead to impulsive shopping behavior. In other words, consumers with their hedonic value can stimulate impulse buying, with the role of product information search (browsing) as a mediating factor. In fact, without searching for product information (browsing), the hedonic value owned by consumers can still stimulate impulse buying directly.
Based on the results of the study, it was found that browsing could not be a mediating factor between utilitarian value and impulse buying. It is because the utilitarian value is not included in the criteria of the mediation test. Baron et al. (1986) state that the role of mediation can occur if the independent variable (utilitarian value) can affect the dependent variable (impulse buying) and the mediating variable (browsing). It is evidenced by the first stage, namely the effect of utilitarian value on impulse buying, which is negative and significant, but in the second stage utilitarian value has no effect on browsing, which means it cannot be continued to the next stage to be tested for mediation. So, the hypothesis is rejected because browsing cannot mediate the effect of utilitarian value on impulse buying.
It contradicts research findings by Zhang et al. (2018) and Gültekin (2012). The results of this study are different from previous studies because the characteristics of the respondents used are different, namely, people who already have their income and are dominated by the upper middle class, while in the research of Ha & Shawn Jang (2010) and Gültekin (2012) using a sample of students who generally do not have their income. In addition, this study also has respondents with a high level of education, which also represents a high level of utilitarian values. That is, browsing cannot mediate the relationship between the influence of utilitarian value on impulse buying because browsing carried out by people with utilitarian values is only browsing rational where they only browse for the products they want or plan, so it does not bridge the occurrence of impulse buying

Conclusions
In conclusion, this study indicated that hedonic value significantly positively affects impulse buying. The positive influence demonstrates that the higher the people's hedonistic values, the more spontaneous (impulse) people have in their shopping behavior. People will shop to look for fun or entertainment without thinking about price, consequences, etc. Practically, the results of this study contribute to businesspeople online to prepare marketing strategies that can stimulate value.