The Causes of Impulse Buying Behavior among Iranian Shoppers

The retailers have a tendency to involve in the procedure of buying and to produce an atmosphere to convince the people for buying goods. The overall purpose of the paper is to gain insights into impulse buying behavior among Iranian shoppers and to arrive at a better understanding on how situational, personal, and product related factors influence impulse buying. The cross-sectional survey design is the specific survey method employed in this research. The instrument for collecting data was a questionnaire and the samples were selected based on the cluster sampling method that includes 207 participants. Findings show that situational variables (store environment) have a positive influence on impulse buying through positive mood and the felt urge to buy has a mediator role between impulse buying tendency and impulse buying. The study also found that demographic variables (job status) has not moderation role between the hedonic shopping and felt urge to buy impulsively. The findings of the research contribute to the knowledge, retailers and salesperson training. This study demonstrates empirically which important factors influence on impulse buying behavior among Iranian consumers.


Introduction
According to Agins (2004), in U.S.A impulse buying behavior covers four billion dollars of annual sales, hence impulsive buying behaviors are considered as very common behavior.According to Beatty and Ferrell (1998) individuals who buy impulsively do not go with the particular reason of visiting a certain shop and buying a definite item.The impulse buying behavior happen after understanding an urge to buy and such behaviors are effected by internal states and environmental/external causes (Kim & Kim, 2008).Because of its occurrence, scholars from unlike disciplines, for instance consumer behavior, have shown attention in investigating this behavior (Parboteeah, 2005).According to Hausman (2000), sellers tend to engage in the procedures of buying and to make a situation to persuade individuals to purchase goods.Marketing science has been paid attention on consumer shopping manners to try to anticipate their expectations of spending.The theory that researchers will use in the present study is Stimulus-Organism-Response (S-O-R).They proposed situational stimuli caused primarily an emotional response, and this response in turn caused reactions to the environment; in other words, behaviors.The theory indicates that, the emotional conditions as important mediators between situational stimuli and human behavior.As applied to the present study this theory holds that independent variables: situational, product related, and personal characteristics influence the dependent variable impulse buying.

Impulse Buying (IB)
Impulse buying behavior is defined as" an unplanned purchase" that is distinguished by (1) relatively quick decision-making (Rook & Gardner, 1993).Impulse buying behavior is typically connected with an unplanned a unexpected purchase, which is started on the spot (Verplanken & Herabadi, 2001).It is explained as more exciting, less purposeful, and more on tempting buying behavior compared to the planned shopping behavior (Munusamy et al., 2010).Synthesizing the conceptualizations of previous researchers in this study, the researcher used the definition of impulse purchasing as an accidental purchase that is distinguished by 1) being hedonically compound and more exciting than rational (Rook, 1987), 2) comparatively quick decision-making (Rook, 1987), and 3) not including the shopping as of an easy prompt item such as buying a present for somebody that fulfils a considered duty (Beatty & Ferrell, 1998)

The Influence of Product Related Features on Impulse Buying Behavior
Product characteristics used for self-expression such as clothing, accessories, and shoes were frequently bought on impulse.In previously noted research participants reported that jewellery, bakery products, and women's sportswear/footwear were impulsive-purchased items (Bellenger et al., 1987).Dittmar et al. (1995) asked 40 students at the University of Sussex in Southern England to pick the top five items they bought on impulse from 20 categories of consumer goods.Music and clothing items were identified as the top two items.The disadvantage of their research is a sample size.Mai et al. (2003) collected qualitative data from a focus group with nine participants and conducted personal interviews with 20 participants.They asked participants to list 5 to 10 items purchased on impulse.Clothing and shoes were the top items listed.In this study the researcher divided products categories: hedonic or unhedonic products.Thus, based on this review the following hypotheses formulated: H1: The product characteristics has positive relation with impulse buying behavior Beatty and Ferrell (1998:172) state that: "Felt urge to buy impulsively is a state of desire that is experienced upon encountering an object in the environment.It clearly precedes the actual impulse action and it is spontaneous and sudden''.The final dependent variable in Beatty & Ferrell (1998) model is an impulse purchase; it is the actual buying of the product.Physical proximity helps link browsing to urge and urge to impulse buying and felt urge to buy which will increase the impulse buying.Thus, the first set of hypotheses follows: H2: The felt urges to buy impulsively has a positive relationship with impulse buying behavior

The Relationship between Impulse Buying Tendency (IBT) and Felt Urge to Buy Impulsively
Impulse buying tendency (IBT) connected to a individuality attribute lack of control" where individuals who buy impulsively without the cognitive manage of not shopping on impulse beside people with high IBT are more likely to affect by advertisements and thus join in in-store browsing (Dawson and Kim, 2010;Youn and Faber, 2000).The individual internal differences with regard to impulsive acts are one of the important factor influences on an impulse purchase and psychologists have been studied impulsivity extensively as a personality trait (Chen, 2008).Therefore impulse buying tendency has been conceptualized as a consumer attribute and defined shopping impulsiveness as shopping impulsively, unreflectively, right away (Chen, 2008).Those with a higher impulsive buying tendency, tend to purchase more on impulse.The high impulse buying tendency individual engages in-store browsing and impulse buying produces positive reinforcement (Beatty and Ferrell, 1998).The general impulse buying tendency refers to the consumer's impulse buying in general (Hansen & Olsen, 2006).H3: The impulse buying tendency (IBT) has positive effect on of urges felt to buy impulsively.H4: The impulse buying tendency (IBT) has positive effect on impulse buying

The Effect Positive Mood on Felt Urge to Buy Impulsively
Some shoppers feel stress and sadness which conceptualize negative affect (Beatty & Ferrell, 1998).In the aspect of psychology, when one person is in a superior mood such as experiencing positive influence he/she is more engaged in approaching behavior than evasion behavior (Beatty & Ferrell, 1998).The positive effect causes individuals to reward themselves more generously and will create behaviors purpose to maintain a positive mood condition (Isen, 2001).Negative effect, experienced simultaneously, may cause someone to urge to buy impulsively which decreases the approach behavior.The mood (positive affect) influences approach buying behavior and the effect of negative mood remains minimal (Beatty & Ferrell, 1998).

H5:
The positive mood has a positive effect on the felt urge to buy impulsively

The Effect of Hedonic Shopping Value (HSV) on Felt Urge to Buy Impulsively
The HSV offers insights into impulse buying behavior and it has different types that effect on impulsive buying, the ones which activates impulse buying the most (Yu & Bastin, 2010).HSV is far more subjective, personal and experiential than its useful complement.A far more exciting practice, the HSV reflects shopping's potential amusement value and the possible symbiotic and/or synergistic association amid the consumers and their shopping reference group (s) (Overby & Lee, 2006).Hedonic consumption tendency influences consumers purchase amount directly and indirectly, by positive emotion.The HSV reflects the value establish in the buying experience itself, lack of task associated activities (Overby & Lee, 2006).H6: The hedonic shopping value (HSV) has positive relationship with urges felt to buy impulsively.

The Effect of Physical Stimuli on Positive Mood
Previous researchers (Lee, 2008;Zhou & Wong, 2004) found impulse buyers who were more likely to be influenced by stimuli such as physical surroundings than non-impulse buyers.Visual stimulation, merchandising effect, product features, pricing stimuli, and sound stimuli were identified as components of the physical surroundings that influenced participants' impulse buying.Thus, investigating the effects of store environments that create pleasurable shopping experiences is important to understanding impulse buying.The pleasant environments encouraged consumers stay longer in the selling environment and to make unplanned purchases.H7: There is a relationship between physical stimuli and positive mood.

The Effect of Product Involvement (PI) on Impulse Buying Tendency (IBT)
In general, products used for self-expression such as shoes; accessories and clothing were frequently bought on impulse, further the marketing literature has suggested that there are a few types of products such as: hedonic products or functional products (Lee, 2008).In addition, product involvement and product categories' are dimensions of product characteristics.Likewise, product involvement is measured a significant factor effecting impulse buying behavior by Jones et al. (2003).Product involvement as: an individual's supposed importance of the purpose based on intrinsic needs, value (Zaichkowsky, 1995).Customer involvement plays an important role in the decision making process.Highly involved customers are willing to make more effort when shopping.They believe that purchasing is one of the significant parts of their life.Thus, they are more likely to spend more time browsing and searching for different products (Arnould et al., 2002).H8: The higher product involvement the greater the frequency of urges felt to buy impulsively

The Moderating Effect of Job Status between HSV and Felt Urge to Buy Impulsively
The findings from a study performed by the market investigate company Taylor Nelson Sofres suggest that emerging consumers, or the "consuming class," account for an average of 37% of the population in major cities and have an average monthly income per household of US$350.Higher income is as one of several factors that can account for the high level of impulse buying in the United States compared with other countries (Mai et al., 2003).
H9: The effect of hedonic shopping value on felt urge to buy moderates by job status.The existing literature on the pleasant appearance of retail stores has considered mainly on the significant effect on store image, the significance of location and size.However, past research (Ergin, 2010) just considered the impulse buying in a shopping mall and did not investigate on stores located in the street.According to the result of previous study's type of store (street store vs. mall store) may also be considered.However in this research the impulse buying will be investigated in two locations (shopping mall & stores located in the street).H10: There are significant differences in location of shopping (shopping mall & stores located in the street) with regards to impulse buying behavior.

Method
The unit of analysis and target population of this study is Iranian shoppers.Probability sampling (cluster sampling) was used in this study.In this study, at first step Tehran was divided into five parts (North, West, South, East and Center) based on geographic area on map of Tehran to avoid possible local bias.In the next step, the list of shopping malls has identified then the researcher chose seven shopping malls randomly based on costefficiency approach.

Data Collection
In this study, the questionnaire has divided in two sections: first part, the researchers asked questions about demographic characteristics the second part, we asked respondents about their impulse buying behavior and factors effect that behavior.The questionnaire of this research was developed in English and then translated into the Persian language by two official bilingual translators.Back translation was also done to check for any inconsistencies or translation errors.Of the 600 initial contacts, 246 have accepted to interview after data screening 207 questionnaires.Around 21 days the researchers spent time for procedures of delivery and collection of questionnaires.

Measures
In this study all items used, were taken from valid scales in the literature and to gather data for each construct seven-point Likert scale was used.To measure impulse buying behavior questions (six items) adopt by Beatty and Ferrell (1998); Hedonic shopping value (HSV) five items from (Hausman, 2000;Yu & Bastin, 2010), felt urge to buy impulsively (four items) from Jeon (Jeon, 1990) and Rook (1987); Weun et al (1998), positive mood (five items) from Lee (2008), IBT (eight items) from (Rook and Fisher, 1995;Weun and Sharon, 1998); physical stimuli (seven items) from Lee (2008).For testing goodness of measures, the validity of questions assessed during convergent and discriminant validity.Table 1 to assess factor loading and based on Hair et al. ( 2010), a cutoff assessment for loadings at 0.5 as significant has been used by researcher.Furthermore, with this table can observe that all the items evaluating a specific variable loaded highly on that variable and loaded lower on the other constructs thus confirming construct validity.For addressing convergent validity, we used the composite reliability, factor loadings, and average variance extracted (AVE) recommended by Hair et al. (2010).

Goodness of Measures
The results of the measurement model have summarized in

Relevance (Assessment) of Structural Model
In this research R 2 (Non-parametric tests) for dependent variables, the Q 2 test is used.The finding shows that all values were above the threshold level (zero).However, regarding CV-redundancy index Q 2 impulse buying tendency had high value.In sum, results indicated that model had an acceptable predictive relevance.

Hypotheses Testing and Discussion
For each relation in the research model (path model), one can assess the effect size.The f 2 is assess as add to in R 2 relative to the amount of variance of the endogenous latent construct that remains sun explained.Based on table 1.6 the effect size of the IBT to Impulse buying (0.99) is large and other effect sizes are small.The result shows, there is a positive and direct relation between the IBT and impulse buying (ß=0.685,std=0.069,f 2 =0.99, t-statistics= 9.91, P-value < 0.05).Table 6 indicated that, there is a positive and direct relation between physical stimuli and positive mood.This relationship with 95% confidence is significant at the 0.05 level (ß=0.193,std =0.108, t statistics=1.79,f 2 =0.038,P-value < 0.05).Also, there is a relation between the positive mood and FUBI.This relationship with 95% confidence is significant at the 0.05 level (ß= 0.205, std= 0.122, t-statistics = 1.677, f 2 =0.038P-value < 0.05).In the present study, there is a positive and direct relation between the product involvement and FUBI.This relationship with 95% confidence is significant at the 0.05 level (ß=0.171, std=0.096, f 2 =0.03, t-statistics=1.67, P-value < 0.05).Also, the positive and direct relation between the HSV and FUBI (ß=0.206, std = 0.114, f 2 =0.036, t-statistic =1.81, P-value< 0.05).Furthermore, there is positive and direct relation between IBT and FUBI (ß=0.270,std =0.113, f 2 =0.066, t-statistics= 2.38, P-value < 0.05).Also, there is a positive and direct relation between the FUBI and impulse buying (ß=0.181,std= 0.088 f 2 = 0.057,T.statistics=2.05,P-value < 0.05).The result revealed that there is not any relation between location and impulse buying, in other words there is no difference in impulse buying behavior in stores located in streets and shopping mall among Iranian shoppers (ß= -0.005, t-statistic = 0.192 , p-value= 0.424).Beside, there is negative relation between product characteristic and impulse buying but this result does not support by previous research due to past researches indicated that the relation between these two variables are positive.It means that the impulse buying behavior is different in products categories such as hedonic or unhedonic products (ß= -0.079, t-statistic = 1.94, p_value= 0.026).In PLS to test moderating effect "When the moderator variable is categorical in nature (as e. g., sex, race, class) it can be used as grouping variable without further refinement (Henseler and Fassott, 2010).For the analysis of the differences job status, the difference between the standardized coefficients of each sample calculated.The result shows job status is not moderating effect (refer to table 6).

Conclusion
The research findings indicated that this study considerably contributed towards an improved understanding of impulse buying among Iranian customers.Impulse buying is an unexpected and instantaneous purchase without any pre-shopping objectives either to buy the certain product or to accomplish a particular purchasing chore (Rook, 1987).For a long time, researchers have been attempting to find out if customers, who frequently engage in impulse buying behavior have some mutual personality traits or not.This research study has further investigated some factors that influenced impulse buying behavior.In an effort to inspect this relationship, this research study principally attempted to elaborate the relationship between impulse buying behavior and some factors influencing it.This research study also served as a bridge to comprehend impulse buying behavior among Iranian customers.In the current study, the research model was theoretically justified and a comprehensive evaluation of this model was performed.Accordingly, for the more specific phenomenon, the principal theory established in this model can be useful in the investigation, detecting important variables with new perspectives.The result revealed that the IBT and felt urged to buy impulsively has a direct relationship with impulse buying, whereas, hedonic shopping, physical stimuli and product involvement have an indirect relationship with impulse buying among Iranian customers.Nevertheless, the relationship between felt the urge to buy and hedonic shopping was not increased by demographic variables.But the relationship between hedonic shopping value and felt urged to buy impulsively has not a significant relationship with respect to employee and unemployed.Furthermore, the effect of shopping location (mall and stores located in the streets) on impulse buying is not significant.In principle these findings will assist Iranian retailers in the areas to concentrate on, in order to improve the level of impulse buying for both employee and unemployed people in stores (in the shopping mall or located in the streets).With refer our results; the generated knowledge can be utilized to improve retail practices.The clear guidelines were provided to retailers and dealers that could help them in developing sales skills and store designs in order to attract consumers and engage them in impulse buying behavior.Furthermore, this study can provide a foundation to educate retailers and inform Iranian customers about the implications of this behavior (impulse buying).In conclusion, the model in the present study conceptualized certain environments (mall and store), location of shopping (shopping mall and stores located in the streets), product categories and responses to this construct that contribute to impulse buying.Due to the increasing number of customers, impulse purchases, the significant strategic importance of the mall and store environment, a mall or store environment that stimulates consumer purchases is likely to receive increasing managerial and academic attention.

Limitation
One of the constraints is the time; it can only be conducted on a limited geographic area confined to the Tehran.
If the research can be extended to cover all the states in Iran, it would give a better picture and increase the research finding generalizability.Also due to limited time, only three kinds of products were being studied, however, there are many other products that could be added to it to make it more meaningful for the body of knowledge.Another limitation is the sample size.

Figure
Figure 1.Research model

Figure
Figure 2. Final model

Table 1 .
Factor loading Table2and indicated that the AVE, were through 0.547 and 0.648.It measures the variance got by the indicators relative to measurement error.Composite reliability values from 0.80 to 0.88 which exceeded recommended value of 0.7.This coefficient depicts the degree to which the construct indicators indicate the latent.In this study the Cronbach's alpha coefficient has used to test reliability of items.The loadings factors and coefficient of alpha Cronbach values have summarized in Table2and as seen from there, all values are above 0.6.Therefore, the measurements are reliable and the other criterion for assessment of measurement model is discriminant validity.Table3describe that diagonal elements are larger than off-diagonal elements in the same row and column.The result describes that the questionnaire had discriminant validity.

Table 6 .
Path coefficient for HSV and FUBI