Journal of Accounting Management and Economics Vol. 20, No. 1, 2018, pp. 37-47 Published by Faculty of Economics and Business, Universitas Jenderal Soedirman Published online on January, 8 2018 in http://jos.unsoed.ac.id/index.php/jame Print ISSN: 1410-9336 OnlineISSN: 2620-8482  Correspondence to : intan.praditya@yahoo.com Received: December 5, 2017 Revised: December 19, 2017 Accepted: January 2, 2018 INTRODUCTION The varied and increasing needs of consumers requires business people to meet the needs and desires of consumers more effectively than their competitors. Consumer behavior becomes very important basis in marketing and advertising. Consumer behavior study focuses on how individual decides to utilize their available resources (time, money, effort) in order to buy things related to consumption. This includes what, why, when, and where they buy, how often they buy, and how often they use them (Schiffman and Kanuk, 2008:6). According to Engel, Blackwell, and Miniard (1995:201) purchase of products or services made by the consumer can be classified into three kinds, namely: (1) fully planned purchases, (2) half planned purchase, (3) impulse buying. Impulse buying is the behavior of a purchase in a store, where purchases are different from what have been planned by the consumer at the time they get into the store. Impulse buying is an act of purchase made without planning, or purchasing decisions are made at the time in the store. Impulse buying may occur when a consumer is not familiar with the store, under the pressure of time, or someone who is reminded of the need to buy a unit when they look at the rack in the store (Utami, 2010:50). Setiadi and Warmika (2015) in their research revealed that the involvement of fashion has a positive influence on impulse buying of fashion consumers in Denpasar, the involvement of fashion has a positive effect toward the positive emotion of fashion consumers in Denpasar, positive emotionhas a positive effect on impulse buying of fashion consumers in Denpasar, and positive emotion positivelymediatesthe effect of fashion involvement on impulse buying of fashion consumers in Denpasar. The results of the research conducted by Dhurup (2014) indicates that the behavior of hedonism, the involvement of fashion, and emotional satisfaction have positive effect on impulsebuying behavior. Dewiet.al (2015) in her research suggests that the trend of hedonism consumption has positive effect on impulse The influence of fashion involvement, hedonic consumption, and visual merchandising on impulse buying with positive emotion as mediation variables SRI MURNI SETYAWATI1, SUMARSONO2, INTAN PRADITYA3 1,2,3 Faculty of Economics and Business, Jenderal Soedirman University, Indonesia Abstract This research aims to analyze the influence of fashion involvement, hedonic consumptions, and virtual merchandising on impulse buying with positive emotion as mediator variable at Popies Girls Shop. The population of this study was the female students of Economics and Business Faculty, Jenderal Soedirman University. The researcher used 145 samples. Sampling technique used in this research was random sampling. Data were analyzed using Structural Equation Modeling (SEM). The result of this research analysis shows that: (1) the involvement of fashion had a positive effect on positive emotion, (2) the trend of hedonic consumption did not have effect on positive emotion, (3) visual merchandising had positive effect on positive emotion, (4) positive emotion had positive effect on impulse buying, (5) positive emotion mediates the relationship between involvement fashion and impulse buying, (6) positive emotion does not mediate the relationship between trends in hedonic consumption and impulse buying, (7) positive emotion mediates the relationship between the visual merchandising and impulse buying. Keywords Fashion involvement; hedonic consumption; trends; visual merchandising; positive emotions; impulse buying Journal of Accounting Management and Economics , Vol. 20, No. 1, 2018, pp. 37-47 38 buying and positive emotion. The higher the hedonic consumption experienced by consumers when shopping directly increase positive emotion of consumers when shopping and increase impulsebuying. In variabel of visual merchandising studied by Sari et al (2015) suggests that visual merchandisinghas positive effect on impulse buying. Meanwhile, the research conducted by Yanthi and Japarianto (2014) suggests that visual merchandising has positive effect on positive emotion, and impulse buying. It means that visual merchandising affects buyer‟s emotional condition that make them perform buying. Emotional condition will create two dominant feelings i.e. pleasure and raising desire. According to Rachmawati (2009: 195), there are two influencing factors in impulse buying, one of them is positive emotion. As an individual, consumers have deep feeling and emotion that affect purchase and ownership of certain products (Utami, 2010:291). According to Donovan and Rossiter in Peter and Olson (2000:251) consumer‟s positive emotion consists of pleasant feeling, passionate, and dominating feeling that are felt when shopping. In the competition of fashion product, consumers will buy current trend fashion products. Popies Girl Shop Purwokerto provides variouus female fashion especially for students such as clothes, shoes, sandals, bags, and accesories with up to date models and affordable price. Visual merchandising served by Popies Girl Shopfor consumers is attractive enough. Popies Girl Shopis located strategically at Jl. HR. Bunyamin No. 489 and near college. Based on the explanation above, the effect of fashion involvement, trend of hedonic consumption, and visual merchandising on impulse buying with positive emotion as interverning variable is interesting to be studied. PROBLEM FORMULATION 1. Does fashion involevement have positive effect on positive emotion? 2. Does trend of hedonic consumption have positive effect on positive emotion? 3. Does visual merchandising have positive effect on positive emotion? 4. Does positive emotion have positive effect on impulse buying? 5. Does positive emotion mediate the effect of fashion involvement on impulse buying? 6. Does positive emotion mediate the effect of visual merchandising on impulse buying? AIMS OF THE RESEARCH 1. To analyze the effect of fashion involevement on positive emotion. 2. To analyze the effect of trend of hedonic consumption on positive emotion. 3. To analyze the effect of visual merchandising on positive emotion. 4. To analyze the effect of positive emotion on impulse buying. 5. To analyze the effect of mediation of positive emotion on causal relationship between fashion involvement and impulse buying. 6. To analyze the effect of mediation of positive emotion on causal relationship between trend of hedonic consumption and impulse buying. 7. To analyze the effect of mediation of positive emotion on causal relationship between visual merchandising and impulse buying. RESEARCH MODEL Figure 1. Research Model HYPOTHESIS H1 : Fashion involvement has positive effect on positive emotion. H2 : Trend of hedonic consumption has positive effect on positive emotion. H3 : Visual merchandisinghas positive effect on positive emotion. H4 : positive emotion has positive effect on impulse buying. H5 : Positive emotion mediates the effect of fashion involvement on impulse buying. H6 : positive emotion mediates the effect of trend of hedonic consumption on impulse buying. Journal of Accounting Management and Economics , Vol. 20, No. 1, 2018, pp. 37-47 39 H7 : Positive emotion mediates the effect of visual merchandisingon impulse buying. METHODS Population and Sample of the Research This is descriptive research wiht observation and survey method. The population of this research was female students of S1 Program, Business and Economy Faculty Jenderal Soedirman University that have performed impulse buying at Popies Girl Shop Purwokerto as many as 721 people. The samples of this research were female students of S1 Program, Business and Economy Faculty Jenderal Soedirman that were still active for the class of 2012, 2013, 2014, and 2015 that have performed impulse buying at Popies Girl Shop Purwokerto. The appropriate sample size for SEM is 100-200, and using comparison with five observations for each parameter estimation (Hair et.al, 2010). In this research, there were 22 indicators and 7 line coefficients so that total estimated parameterwere 29 parameters. Thus, the total samples of this research were 145 respondents. The samples were taken using probability samplingtechnique with simple random samplingtechnique i.e. a sampling technique that gives equal opportunity to the population to be a sample (Suliyanto, 2006:113). The sources of data were primary taken from observation and survey, and secondary data taken from theoretical review, scientific articles, scientific journals, and other sources related to research taken from internet. Conceptual and Operational Definition Dependent Variable Dependent variable used in this research was impulse buying. Conceptual definition: impulse buying is buying behavior based on impulse or sudden desire without considering the consuequence, or being emotionally encouraged (Schiffman and Kanuk, 2008:489) Operational definition: buying behavior that is not planned due to impulse or sudden desire without considering consequence when customers are at Popies Girl Shop. Indicators: spontanous buying; not considering consequence; and emotional impulse (Schiffman and Kanuk, 2008:489) Intervening Variable The dependent variable used in this research was positive emotion. Conceptual definition: consumer‟s positive emotion consists of pleasant feeling, passionate, and dominating feeling that are felt when shopping. (Donovan and Rossiter in Peter and Olson, 2000:251) Operational definition: individual‟s moodfelt as positive response against environment and situation at Popies Girl Shop. Indicators: pleasant feeling; passionate feeling; and dominating feeling (Donovan and Rossiter in Peter and Olson, 2000:251) Independent Variable a. Fashion involvement (X1) Conceptual definition: fashioninvolvement is level of interest in fashion product (for example: clothes) (Park, 2006:436) Operational definition: is level of consumer‟s interest in fashion products at Popies Girl Shop. Indicators: having one fashion product or more with up to date model; Fashion is an important thing that supports activities; and more interested in shopping at fashion store than in di department store to meet need for fashion (Park, 2006:439) b. Trend of hedonic consumption (X2) Conceptual definition: hedonic consumption is shopping behavior that is related to feeling, imagination, and other aspects related to feeling of someone experience against a product (Utami, 2010:49) Operational definition: is shopping behavior related to feeling, imagination, and pleasure felt by consumer when shopping at Popies Girl Shop Indicators: shopping is an experience; spending time to shop with family or friends is a pleasing thing; shopping gives positive feeling; shopping makes someone know the latet fashion trend; shopping is special pleasure when buying something for other people; and looking for cheap price or discount when shopping (Utami, 2010:49-50) c. Visual Merchandising Conceptual definition:Visual Merchandisingis a basic element in creating shopping atmosphere in the store since such store performancesas view, voice, touching, taste, and smell are components that can create representative form of merchandise in the store (Dale M. Lewison& M. Wayne DeLozier, 1989:647) Journal of Accounting Management and Economics , Vol. 20, No. 1, 2018, pp. 37-47 40 Operational definition: Visual Merchandising is an element that creates shopping atmosphere at Popies Girl Shop, such as appearance, display, color selection, the music played, lighting, perfume, and assortment of the products. Indicators: Display is attractive product layout that enables consumer to find easily; Color is selection, combination, and display of color that creates identity and attraction for consumers; lighting is appropriate beam of light against a product so that it is clealy seen and can create buying interest of consumers; Hear is relaxing music that gives exciting shopping experience to consumers; Smell is aroma that makes shopping comfortable; assortment is variety of product that offers more options to consumers to choose. Data Analysis Technique 1. Measurement of Research Variables Variable measurement is done using scale measure. Measurement scale is used as the reference standard to determine the length of the interval in order to produce quantitative data. With this measurement scale, the value of variables is measured by certain instrument that can be stated in numbers, so that it will be more accurate, efficient, and communicative (Sugiyono, 2003:127). This research used Likert scale i.e. scale to measure individual‟s response to social object, there are five levels of Likert scale (Suliyanto, 2011:10): a. SA: Strongly Agree with score = 5 b. A : Agree with score = 4 c. N : Neutral with score = 3 d. D: Disagree with score = 2 e. SD : Strongly Disagree with score = 1 2. Structural Equational Modelling (SEM) Model and hypothesis are tested using SEM analysis through the following steps (Ferdinand in Suliyanto (2011:273-275)): 1) Development of a model based theory What should be done is perform a set of scientific exploration through theoretical review in order to obtain theoretical justification on the model developed. 2) Development of path diagram. Theoretical model that has been built will be presented in a path diagram to be estimated. 3) Conversion of path diagram into equation. In this step, model is presented in path diagram which then is converted into series. 4) Assess identification problem. Identification problem is basically an ignorance problem from the model developed in order to generate unique estimation. 5) Evaluation of Goodness-of-fitcriteria. Suitability of model is evaluated through review of various goodness-of-fit criteria with three kinds of evaluation, they are: a) Evaluation of SEM assumption - Normality, using critical value criteris ±2.58 at significance level 0,01. If Z- value is higher than the critical value, it can be assumed that the data are not normally distributed. - Outliers, is an observation or data that has unique characteristic which seems to be very much different from observations, either for single variable or combination of variables. Using critical value criteria ±3, data will be stated outlierif it has Z-score value >3 or <-3 - Multicollinearity and Singularity, what to observe is the determinant of covariant matrix of its sample. Lower determinant or close to zero indicates the existence of multicollinearity and singularity, thus the data can be used for research. b) Compliance test and statistic test Here are several compliance indexes and cut-off valueused to test whether a model is accepted or rejected: - X2 Chi-Square statistic, the lower the value of X2, the better the model and is accepted based on the probability with cut-off value at p > 0,05 or p > 0,010. - RMSEA (The Root Mean Square Error of Approximation), is an index used to compensate chi-squarein big sample. The lower value of RMSEA or equal to 0,08 is an index in order for the model to be accepted based on degree of freedom. - GFI (Goodness of fit Index), is non- statistical measure that has interval from 0 to 1. Higher value in this index shows better fit. - AGFI (Adjusted Goodness of Fit Index), is a criterion that estimates weighted proportion from variant in a covariant matrix of samples. The recommended level is when AGFI is equal to or higher than 0,90. Journal of Accounting Management and Economics , Vol. 20, No. 1, 2018, pp. 37-47 41 - CMIN/DF (The Minimum Sample Discrepancy Function Devided with degree of Freedom), is a statistic of chi-square X2 divided its degree of freedomso that it is called X2 relative. Value of X2 relative < 2,0 or 3,0 is an indication of acceptable fitbetween model and data - TLI (Tucker Lewis Index), is incremental index that compares a tested model against a baseline model, where the recommended value as a reference of the acceptance of a model is ≥0,95 and close to 1 value shows a very good fit. - CFI (Comparative Fit Index), interval between 0-1, where the closer to 1 indicates the highest fit level. Table 1: Goodness of Fit Index and Cut of Value Goodness of Fit Index Cut Of Value -Chi Square Expected to be lower Significance Probability ≥ 0,05 RMSEA ≤ 0,08 GFI ≥ 0,90 AGFI ≥ 0,90 CMIN/DF ≤ 2,00 TLI ≥ 0,95 CFI ≥ 0,95 c) Validity test in SEM (Ghozali, 2008:134) consists of: - Convergent Validity Items or indicators of a latent construct should harus convergeor sharethe highest variant proportion and it is called convergent validity. Therefore, standardized loading estimateshould be equal to 0,50 or more. - Discriminant validity Discriminant validitymeasures how far a construct is really different from other constructs. It is done by comparing the square root of AVE ( ) and the correlation value among constructs. d) Reliability test and Variance extract - Reliability test, where the reliability value that is accepted is ≥ 0.70.Reliability test in SEM can be obtained from the following formula Construct reliability = ( ) ( ) Where: Standard loading is obtained from standardized loadingfor each indicator that is obtained from the result of computer calculation. ε.j is measurement error of each indicator - - Variance extract, where the value that can be accepted is ≥0,50. Here is the formula: Variance Extract = Where: Standar loadingis obtained from standardized loading for each indicator that can be obtained from the result of computer calculation. ε.j is measurement errorof each indicators (1- Std.Loading) 2 6) Interpretation and modification of model. The final step is, interpret model and modify models for those that do not meet the testing requirement. Cut-offat 2,58 can be used to assess the significance of residual generated by model. Residual value that is higher than or equal to 2.58 is interpreted to be statistically significant at 5%. 7) Criteria of hypothesis testing Hypothesis is accepted if CR (Critical Ratio): CR ≥ t table Hypothesis is rejected if CR (Critical Ratio) : CR < t table Journal of Accounting Management and Economics , Vol. 20, No. 1, 2018, pp. 37-47 42 RESULTS AND DISCUSSION Validity Test Convergent Validity Summary of Convergent Validity Test Result Variable Indicator Loading Factor Cut of Value KF x1 0,720 0,500 x2 0,747 0,500 x3 0,525 0,500 KKH x4 0,756 0,500 x5 0,763 0,500 x6 0,653 0,500 x7 0,668 0,500 x8 0,802 0,500 x9 0,601 0,500 VM x10 0,673 0,500 x11 0,849 0,500 x12 0,662 0,500 x13 0,777 0,500 x14 0,846 0,500 x15 0,596 0,500 x16 0,591 0,500 EP x17 0,804 0,500 x18 0,699 0,500 x19 0,585 0,500 PTT x20 0,545 0,500 x21 0,506 0,500 x22 0,748 0,500 The value of loading factorfor the indicator of fashion involvement, trend of hedonic consumption, visual merchandising, positive emotion,and impulse buying was each higher than 0,500. Thus, all indicators of each variable were proven to meet convergent validity so it could be used as data collection instrument of this research. Discriminant Validity Table 2. Value of Square Root of AVE Variable AVE KF 0,786 0,887 KKH 0,848 0,920 AT 0,853 0,924 EP 0,706 0,841 PTT 0,683 0,826 Table 3. Correlation among Constructs and Square Root of AVE Construct KF KKH VM EP PTT KF 0,887 KKH 0,213 0,920 VM 0,066 0,335 0,924 EP 0,397 0,250 0,409 0,841 PTT 0,473 0,494 0,462 0,628 0,826 It can be clearly seen from the table that each latent construct has good discriminant validity. It can be seen from the value of square root of AVE of each latent construct is higher than the correlation coefficient among its constructs (Ghozali, 2008:235) Journal of Accounting Management and Economics , Vol. 20, No. 1, 2018, pp. 37-47 43 Reliability Test Based on the calculation result of reliability test, the reliability coefficient of fashion involvement, trend of hedonic consumption, visual merchandising, positive emotion,and impulse buying was each higher than 0,700.Thus, the construct of each variable in this research was reliable and could be used as data collection instrument. Variance Extract Test The result of variance extract test shows the value of variance extractfor the indicators of fashion involvement, trend of hedonic consumption, visual merchandising, positive emotion,and impulse buyingwas higher than 0,500. Thus, construct of each variable in this research was proper to test causal relationship in this research. Hypothesis Testing The significance of causal relationship in SEM analysis was tested through null hypothesis which showed that the coefficient of causal relationship among variables was equal to zero using t test commonly used in regression.Here is the result summary of causal relationship among variables: Table 4. Standarized Regression Weight for Hypothesis Testing Variable C.R. t table EP <-- KKH 0,954 1,660 EP <-- VM 3,748 1,660 EP <-- KF 3,275 1,660 PTT <-- EP 2,773 1,660 The value of critical ratio (CR) for the effect of fashion involvement variable on positive emotion was 3,275 higher than that of t table (1,660). Thus, it can be concluded that the first hypothesis which states that fashion involvement had positive effect on positive emotion, was accepted. The value of critical ratio (CR) for the effect of trend of hedonic consumption variable on positive emotion was 0,954 smaller than that of t table (1,660). Thus, it can be concluded that the second hypothesis which states that trend of hedonic consumption had positive effect on positive emotion, was rejected. The value of critical ratio (CR) for the effect of visual merchandising variable on positive emotion was 3,748 higher than that of t table (1,660). Thus, it can be concluded that the third hypothesis which states that visual merchandising had positive effect on positive emotion, was accepted. The value of critical ratio (CR) for the effect of positive emotion variable on impulse buying was 2,773 higher than that of t table (1,660). Thus, it can be concluded that the fourth hypothesis which states that positive emotion had positive effect on impulse buying, was accepted. For hypothesis testing with intervening variable, this research used sobel test; a method of significance testing of mediation effect. The result is as follows: Table 5. The result of Sobel test for Hypothesis Testing of Mediation Effect Variable tsobel ttable PTT  EP  KF 2,068 1,984 PTT  EP  KKH 0,850 1,984 PTT  EP  VM 2,187 1,984 The result of Sobeltest for fashion involvement variable on impulse buying through positive emotion was 2,068 higher than that of ttable (1,984). Thus, it can be concluded that the fifth hypothesis which states that positive emotion mediated the effect of fashion involvement on impulse buying, was accepted. Trend of hedonic consumption variable on impulse buying through positive emotion was 0,850 smaller than that of t table (1,984). Thus, it can be concluded that the sixth hypothesis which states that positive emotion mediated the effect of trend of hedonic consumption on impulse buying, was rejected. Visual merchandising variable on impulse buying through positive emotion was Journal of Accounting Management and Economics , Vol. 20, No. 1, 2018, pp. 37-47 44 2,187 higher than that of t table (1,984). Thus, it can be concluded that the seventh hypothesis which states that positive emotion mediated the effect of visual merchandising on impulse buying, was accepted. DISCUSSION The Direct Effect of Fashion Involvement on Positive Emotion Based on the hypothesis testing, it has been proven that there was positive effect fashion involvement variable. It means the higher the fashion involvement felt by consumers, the more increase the positive emotion felt by consumers when shopping. From the result of survey based on respondents‟ perception conducted using three indicators, it can be assumed that most of respondents feel that fashionis an important thing that supports their activities. Therefore, respondents are very happy and enthusiastic when they are shopping at Popies Girl Shopbecause the products offered at Popies Girl Shopare up to date. Wearing up to datefashion products can increase respondents‟ self-confidence during their activities. This result is relevant with the previous research conducted by Setiadi and Warmika (2015) which states that fashioninvolvement had positive and significant effect on positive emotion. The result is also consistent with the research conducted by Pattipeilohy et al., (2013) which also proves that fashion involvement has positive effect on positive emotion. The Direct Effect of Trend of Hedonic Consumption on Positive Emotion The result of hypothesis testing shows that trend of hedonic consumption does have any effect on positive emotion. The causal effect shows that the higher the level of trend of hedonic consumption is not always followed by the increase in positive emotion. From the result of survey by research that has been done using six indicators, respondents think that shopping is an experience and when they are shopping they tend to look for products with cheap price or discount. However, it is not the reason why they feel happy, passionate, or have dominating feeling when shopping at Popies Girl Shop. Trend of hedonic consumption does not have positive effect on positive emotion because not every person who has high hedonic personality will increase their positive emotion. The respondents of this researchare female students with limited pocket money especially for those who come from other towns. Thus, when they are shopping with high hedonic consumption personality, it is not always followed by pleasure, passionate nor dominating feeling at Popies Girl Shop. When they see the products they like and compare the price and the money they have, they might not buy and it makes them sad. Thus, it does not always increase consumers‟ positive feeling. It may happen because some products have higher price than the other products sold by Popies Girl Shop such as dress, high heels,and bags. This result is not relevant with the research conducted by Dewi et al., (2015) that trend of hedonic consumption has positive effect on positive emotion. This research does not support the previous research conducted by Pattipeilohy et.al (2013) which suggests that trend of hedonic consumption has positive effect on positive emotion. Direct Effect of Visual Merchandising on Positive Emotion Based on the result of hypothesis testing, visual merchandisinghas positive effect on positive emotion. The causal relationship shows that the stronger the effect of visual merchandising, the stronger the positive emotion of consumers. Visual merchandisingfelt by consumers at Popies Girl Shopmakes them excited and gives pleasure shopping experience so that it increases consumer‟s positive emotion. Visual merchandising variable has the biggest effect on consumers‟ positive emotion seen from the result of hypothesis testing. From the result of survey by research that was done using six indicators to evaluate visual merchandising based on respondents‟ perception, it can be assumed that most of respondents think that visual merchandising at Popies Girl Shopis good because the products are set very well so that it makes them easy to choose. This result is consistent with the research conducted by Sari et al (2015). Direct Effect of Positive Emotion on Impulse Buying Based on the result of hypothesis testing, positive emotion has positive effect on impulse buying. The stronger the positive emotion is, the bigger the possibility of impulse buying done by consumers will be. Journal of Accounting Management and Economics , Vol. 20, No. 1, 2018, pp. 37-47 45 This result supports Mowen and Minor (2002) who states that impulse buying can be explained as a choice made at that moment due to strong positive feeling against a product. From the result of survey by research conducted using three indicators, it can be known that most of respondents will buy products that attract them, without considering the consequence due to being motivated by their passionate feeling when shopping at Popies Girl Shop. It is supported by Donovan and Rossiter (in Peter and Olson, 2000:251) which states that enthusiasm caused by shop can lengthen the stay in the shop and willingness to interact with the shopkeepers that will influence consumers to shop more and more beyond what is previously planned. This result is consistent with the study conducted by Naentiana and Setiawan (2014) that positive emotion has significant positive effect on impulse buying and the research conducted by Amiri et.al (2012) which states that establishing consumer‟s positive emotion against products or shop environment can increase consumers‟ motivation to perform impulse buying. Indirect Effect of Fashion Involvement on Impulse Buying through Positive Emotion The result of hypothesis testing shows that positive emotion mediates the effect of fashion involvement on impulse buying. The causal relationship shows that the higher the fashion involvement encourage by strong positive emotion will make consumers perform impulse buying “Positive emotion mediates” means that there is indirect relationship between fashion involvement and impulse buying. The higher the fashion involvement will increase positive emotion, and consumers with positive emotion when shopping at Popies Girl Shopfinally will trigger impulse buying. Based on the survey, respondents believe and are confident that fashion is an important thing that supports their activities. By wearing up to datefashion product, they will feel more confident in their activities. They are also more interested in shopping at such fashion store as Popies Girl Shopthat at department store to meet their fashion need. Through positive emotion variable, many respondents fairly agree that they feel happy and enthusiastic shopping at Popies Girl Shopsince the products offered by Popies Girl Shopare up to date. It encourages respondents to perform buying even though they do not plan and consider the consequence. This result is consistent with the findings of the research conducted by Beatty et al., (in Park et al., 2006) that positive emotion during shopping can be a significant mediator in encouraging impulse buying. This result is also consistent with the research conducted by Amiri et al., (2012) that the higher the fashion involvement of consumers will show more positive emotion during shopping and can increase involvement in impulsive buying. Indirect Effect of Trend of Hedonic Consumption on Impulse Buying Through Positive Emotion The result of hypothesis testing shows that positive emotion does not mediate the effect of trend of hedonic consumption on impulse buying. The causal relationship shows that in impulse buying performed by consumers with high trend of hedonic consumption is not always caused by strong positive emotion. It means that trend of hedonic consumption has direct effect on impulse buying. This result is supported by Dewi et.al (2015) who suggests that trend of hedonic consumption has positive direct effect on impulse buying. Such behaviors of hedonic consumption as willingness to satisfy curiosity, willingness to obtain new experience, and looking for cheap product or discount can influence their behavior to perform impulse buying. According to the result of survey, respondents think that shopping is an experience because they can compare prices and products sold in other shops. They also admit that when they are shopping, they will look for product with cheap price or discount especially for those who come from other towns. It can make consumers perform buying for products they want, even though it is not previously planned. This result is supported by the research by Kusuma (2014) which states that trend for hedonic consumption has positive effect on impulse buying. The higher the trend of hedonic consumption felt by consumers, the higher the impulse buying will be. It is because when someone shops hedonically, s/he will not consider usefulness of the products, so that the possibility of impulse buying will get higher Journal of Accounting Management and Economics , Vol. 20, No. 1, 2018, pp. 37-47 46 Indirect Effect of Visual merchandising on Impulse Buying through Positive Emotion The result of hypothesis testing shows that positive emotion mediates the relationship between visual merchandising on impulse buying. The causal relationship shows that the stronger the visual merchandising supported by positive emotion will make consumers perform impulse buying at Popies Girl Shop. „Positive emotion is mediating‟ means that there is indirect relationship between visual merchandising and impulse buying. The increase in impulse buying is caused by a good visual merchandising through positive emotion. Respondents who feel a good visual merchandising will have increasing positive emotion. Consumers with high positive emotion when shopping at Popies Girl Shop will finally trigger impulse buying. This result is consistent with the research conducted by Amiri et al., (2012) which suggests that the establishment of consumers‟ positive emotion on products or shop environment can increase consumers‟ motivation to perform impulse buying.This result is also consistent with the research by Donovan and Rossiter (in Peter dan Olson, 2000) that environment stimulus affects the status of consumer‟s emotion, where in turn it will affect consumer‟s behavior. In addition, this result also supports the previous research performed by Kurniawan and Kunto (2012) that visual merchandising has positive effect on consumer‟s positive emotion, where visual merchandising affects buyer‟s emotion which finally influences buying. CONCLUSION Based on the result of the research, it can be concluded that: 1. Fashion involvement has positive effect on positive emotion, it means that the higher the fashion involvement, the higher the consumer‟s positive emotion during shopping. 2. Trend of hedonic consumption does not have effect on positive emotion. It means that the higher the trend of hedonic consumption is not always followed by the higher the positive emotion. 3. Visual Merchandising has positive effect on positive emotion, meaning that the higher the effect of PopiesGirl Shop, the higher the positive emotion felt by consumers during shopping. 4. Positive emotion has positive effect on impulse buying. It means that the higher the positive emotion felt by consumers during shopping will increase impulse buying at Popies Girl Shop. 5. Positive emotion mediates the effect of fashion involvement on impulse buying. It means that there is indirect correlation between fashion involvement and impulse buying through positive emotion. 6. Positive emotion mediates the effect of trend of hedonic consumption on impulse buying. It means that trend of hedonic consumption has direct effect on impulse buying. The higher the trend of hedonic consumption felt by consumers does not affect the positive emotion level of consumers to perform impulse buying at Popies Girl Shop. 7. Positive emotion mediate the effect of Visual Merchandising on impulse buying. It means that there is indirect relationship between Visual Merchandising and impulse buying through positive emotion. IMPLICATION Managerial Implication Management of Popies Girl Shop should encourage impulse buying and consumer‟s positive emotion by selling up to date fashion product. To establish trend of consumer‟s hedonic consumption in order to make them perform impulse buying is by using such marketing strategies as giving discount and buy 1 get 1 free. Management should increase visual merchandising by giving attention to room design, layout, information boards, lighting, coloring, music, and aroma of Popies Girl Shop in order to give comfort to consumers that can encourage positive emotion and stimulate impulse buying at Popies Girl Shop. LIMITATIONS Meanwhile, the limitatons of this research is in data collection such as respondents answer dishonestly and carelessly. Limited time and power were problems for the researcher to perform deeper study. Journal of Accounting Management and Economics , Vol. 20, No. 1, 2018, pp. 37-47 47 REFERENCES Amiri, Farhad, Jalal Jasour, Mohsein Shirpour, and Tohid Alizadeh. 2012. Evaluation of Effective Fashionism Involvement Factors on Impulse Buying of Costumers and Condition of Interrelation between These Factor. Journal of Basic and Applied Scientific Research. 2 (9), pp 9413-9419. Dhurup, Manilall. 2014. Impulsive Fashion Apparel Consumption: The Role of Hedonism, Fashion Involvement and Emotional Gratification in Fashion Apparel Impulsive Buying Behaviourin a Developing Country. Mediterranean Journal of Social Sciences Vol 5 (8). Engel, J.F., Blackwell, R.D.,dan Miniard, P.W. 1995. Perilaku Konsumen. Edisi 6. Jilid 2. Jakarta: Binarupa Aksara Ghozali, Imam. 2008. Konsep Dan Aplikasi Dengan Program Amos. Badan Penerbit Universitas Diponegoro, Semarang. Hair JF.,Anderson RE, Tatham RL., & Black WC. 2010. Multivariate Data Analysis : Perspective. New Jersey: Pearson Education Int. Kotler, Philip. 2005. ManajemenPemasaran. Edisi kesebelas. Jilid 2. Jakarta: Indeks. Kurniawan, Denny dan Yohanes Sundang Kunto. 2013. Pengaruh Promosi Dan Store Atmosphere Terhadap Impulse Buying Dengan Shopping Emotion Sebagai Variabel Intervening. Jurnal Manajemen Pemasaran. Vol 1, No 2. Kusuma, Ardian. 2014. Pengaruh Fashion Involvement, Hedonic Consumption Tendency, Dan Positive Emotion Terhadap Fashion-Oriented Impulse Buying Kalangan Remaja Di Surabaya. Jurnal Ilmiah Mahasiswa Universitas Surabaya. Vol 3, No 2. Lewison, Dale M. & DeLozier, M. Wayne. 1989. Retailling,3rd ed. Ohio: Merrill Publishing Company Mowen, J.C., dan Minor, M. 2002. Perilaku Konsumen. Edisi kelima. Jilid 1 & 2. Jakarta: Indeks. Park, EunJoo, Eun Young Kim, and Judith Cardona Forney. 2006. A Structural Model of Fashion-Oriented Impulse Buying Behavior. Journal of Fashion Marketing and Management. Vol. 10, No. 4, pp 433-446. Pattipeilohy, Victor Ringhard, Rofiaty, and M.S. Idrus. 2013. The Influence of the Availability of Money and Time, Fashion Involvement, Hedonic Consumption Tendency and Positive Emotions towards Impulse Buying Behavior in Ambon. International Journal of Business and Behavioral Sciences. Vol 3, No 8, pp 36-49 Peter, J.Paul, dan Olson, Jerry C. 1999. Perilaku Konsumen dan Strategi Pemasaran. Edisi 4. Jilid 1. Jakarta: Erlangga Peter, J.Paul, dan Olson, Jerry C. 2000. Perilaku Konsumen dan Strategi Pemasaran. Edisi 4. Jilid 2. Jakarta: Erlangga Rachmawati, Veronika. 2009. Hubungan Antara Hedonic Shopping Value, Positive Emotion dan Perilaku Impulse Buying pada Konsumen Ritel, Jurnal Majalah Ekonomi. Agustus, pp 192-208. Sari, Taman Ayu, Dewa dan Alit Suryani, 2015. Pengaruh Merchandising, Promosi dan Atmosfir Toko Terhadap Impulse Buying. Schiffman, Leon dan Kanuk, Lezlie Lazar. 2008. Perilaku Konsumen. Jakarta: Indeks Setiadi, I Made Willy dan I Gde Ketut Warmika. 2015. Pengaruh Fashion Involvement Terhadap Impulse Buying Konsumen Fashion yang Dimediasi Positive Emotion di Kota Denpasar. Jurnal Manajemen. Vol 4, No 6, pp 1684-1700. Setiadi, Nugroho J. 2003. Perilaku Konsumen Konsep dan Implikasi untuk Strategi dan Penelitian Pemasaran. Jakarta: Prenada Media. Sugiyono. 2003. Statistika Untuk Penelitian. Cetakan Kelima. Bandung: Alfabetta. Sugiyono. 2012. Metode Penelitian Kuantitatif Kualitatif dan R&D. Jakarta:Alfabeta Utami, Christina Whidya. 2010. Manajemen Ritel :Strategi dan Implementasi Operasional Bisnis Ritel Modern di Indonesia.Jakarta : Salemba Empat Yanthi, Devi dan Edwin Japarianto. 2014. Analisis Pengaruh Hedonic Shopping Tendency dan Visual Merchandising Terhadap Impulse Buying dengan Positive Emotion Sebagai Variabel Intervening pada Area Ladies Matahari Department Store Tunjungan Plaza Surabaya. Jurnal Manajemen Pemasaran Petra Vol. 2, No. 2, (2014).