Determinants of Purchase Intention: An Interpretive Structural Modelling Approach

In the present paper, the factors which impact consumer behavior are explored through extensive literature review. Also, common determinants of online and offline consumer behavior are identified. Subsequently, ISM approach is used to find the inter-relationship among these factors. Finally, ISM model depicting this interrelationship is developed and MICMAC analysis is performed to categorize factors into clusters depending on their drive and dependence power. Identified factors common to both offline and online buying behavior include convenience, promotion, availability, brand image, demography, safety, quality, word of mouth and price. ISM reveals that promotion, brand image, demography, quality and price are the linking factors with strong dependence and driving power. Additionally, convenience, availability and safety have emerged as the independent factors with strong driving power and weak dependence power. Keywords: Offline buying. online buying. Determinants. ISM (Interpretive structural modelling). MICMAC DOI: 10.7176/JMCR/68-02 Publication date: May 31 st 2020


Introduction
Increase in the number of e-commerce websites has led to a steep rise in global online shopping. At the same time, the significance of offline shopping cannot be overlooked. This paper attempts to find factors that affect consumer behavior while buying products and services offline as well as online. Key objectives of this paper are (a.) to find factors which are common to both offline and online shopping and (b.) to develop an interrelationship among these factors. Eight factors are considered for study. These include convenience, promotion, availability, brand image, demography, safety, quality and price. Inter-relationship among these factors is analyzed through ISM approach. Further, the factors are classified on the basis of driving and dependence power through MICMAC analysis.
In the remainder paper, factors affecting offline and online buying behavior from literature are given in sections 2 and 3 respectively. Section 4 depicts the diagrammatic representation of offline and online factors with the clubbing of common factors and their references/sources. Section 5 depicts the ISM model generation and in Section 6, MICMAC analysis is performed. The conclusion and limitations are provided in sections 7 and 8 respectively.

Offline buying behavior
Offline purchase is a conventional way of purchasing products and undertaking services by directly visiting the store/shop or vendor. Sethi, Inderjeet (2018) have explored that the most significant factors affecting consumer behavior in telecom industry includes promotion, social class, perception, religion. As per Legeza & Brunner (2019) the major social factors for buying eco-intelligent products are employment status and store location. Rana, Jyoti & Paul, Justin (2017) found that the factors which influence a consumer to buy organic products include health consciousness , expectation of wellbeing, quality and safety, willingness to pay, fashion trend, social consciousness and found that lack of promotion and distribution affects the presence of green products in developing countries. Shende, Vikram (2014) indicated that disposable income is the most influential factor for an automobile passenger car customer. Chopra (2014) conducted a study on factors influencing consumer behavior for cosmetic products and found that promotional strategy, festive offer discount, availability, billing speed and ambience to be most significant. Diallo et al. (2013) concluded that value consciousness to be major factor impacting consumer behavior towards store brand in French market. Lakshmi, Niharika, Lahari (2017) conducted a study to find the impact of gender on consumer purchasing behavior and the findings indicated that gender significantly influence consumer purchasing behavior. Kumar, John, Senith (2014) revealed that there existed significant differences among income level and behavior dimensions like social and cultural factors but no significant differences existed of income level with personal and psychological factors. Rengarajan, et al. (2014) indicated that income level is the most influential factor affecting consumer behavior for selected brands of milk products.
from Vishal mega mart.

Online Buying Behavior
Online shopping is a form of buying goods and availing services through internet without actually visiting the stores.
Chincholkar &Sonwaney (2017) found that the most significant factors which consumer consider while selecting a website for shopping are availability and quality of products/services and also there is no significant difference existed between men's and women's attitude towards website. Shanthi, Kannaiah (2015) revealed that price is the most significant factor impacting consumer to buy online followed by security, guarantees and warrantees. According to Vikash & Kumar (2017) quality, convenience, satisfaction, availability of products, security and privacy, quickness, attractive, flexibility, spatial convenience and awareness are the most significant factors influencing consumer to buy online. Vadivu (2015) found that variety, quick service and reduced price are the most influential factors for consumer to buy online. Islam, Md. Shariful (2015) proposed a conceptual model consisting of independent variable as financial risks, convenience risks, non-delivery risk, infrastructural variables, return policy, attitude and subjective norms whereas dependent variables being attitude, online shopping behavior. The findings are such that financial risks and non-delivery risks have a negative effect on consumer attitude for online buying behavior. Nagra, Gopal (2013) concluded that age, gender, income, family size, marital status has a positive impact on online shopping behavior of consumers. According to Abdullah et al. (2016) the most influential factors for consumer buying stimuli are perceived ease of use, vendors characteristics, perceived usefulness and website design. Vaghela, Pratiksinh (2016) revealed that factors such as perceived ease of use, perceived usefulness, website design are the most influential factors affecting online shopping behavior of consumer. Further it is found out that there is no significant difference existed between male and female purchase behavior. Raunaque, Zeeshan, Imam (2016) concluded that easy return and refund policy, money back guarantee are among the factors to improve customer trust and hence the perception towards online shopping. Deshmukh & Sanskrity (2016) revealed that demographic factors, product features, website design has the most significant influence on consumer to buy online. Singh et al. (2016) concluded that Gen Y has a positive bend towards online shopping and they prefer to shop anytime they feel like. Dange & Kumar (2012) proposed a model factor, filtering element and filtered buying behavior (FFF). Factor (internal and external) make consumer to get into filtering elements (security concerns, privacy concerns, trust) which leads to filtered buying. Rao et al. (2018) found the factors which women consider important while purchasing online are ease of use, convenience, security, utility, time effectiveness, outbound logistics and feedback. Bauboniene & Guleviciute (2015) revealed that the major factors impacting consumers to do online shopping are convenience, simple to access and relatively better prices. Mahalaxmi & Ranjith (2016) concluded that the consumers are aware of digital marketing and give preference to digital channels to purchase any types of products. Rai (2018) revealed that consumer prefer to use digital channels to buy any sought of products irrespective of their monthly income.
Prasath & Yaganathen (2018) concluded that SMM (social media marketing) as an independent variable has a positive and linear relationship with CBDM (consumer buying decision making) the dependent variable. Yapa (2017) conducted a study to find the effect of independent variables i.e. (users generated communication and firms generated communication) on dependent variables i.e. Brand awareness. Results stated that independent variables have positive correlation with dependent variables. Ioanas & Stoica (2014) concluded that social media has a positive influence on online buying behavior of consumers. Emir et al.(2016) with the aid of stimulus-Organism-Response model concluded that while booking hotel online consumer consider stimuli as independent variables (information quality, perceived interactivity , safety and privacy, price and promotion and e-word of mouth ) expected to influence perceived value (organism) which will lead them to book hotel online(response). Ahmed et al. (2018) concluded that perceived benefits, domain specific innovativeness and shopping orientation had significant and positive relation with consumer buying behavior. Whereas Jadhav & Khanna (2016) conducted a study among college students to know the online buying behavior and found that availability, low price, promotions, comparison, convenience, customer service, ease of use, attitude, time consciousness, trust and variety seeking to be the most influential factors considered for online purchase. Akbar et al. (2017) revealed that quality is the most significant factor followed by convenience and trust while promotion did not have any significant relationship with consumer behavior. Hong & Deng (2018) conducted a study to know the factors affecting trust in online healthcare services and concluded that among three kinds of online healthcare services information search requires less amount of trust using online appointment requires moderate level of trust whereas online consultation requires the highest level of trust. Santaso, Bidyati, Hendar (2019) revealed that hedonic motivation does not plays a significant role in online purchase behavior whereas trust, website quality and design play a significant role in purchase intention of customers. Adnan, Hooria (2014) revealed that perceived advantages and psychological factors (trust and security concerns) have a positive influence on consumer buying behavior whereas perceived risk impact consumer buying negatively and website design and hedonic motivation were found to be insignificant.Ofori, Boakye & Narteh (2016) conducted a study to find factors which influence consumer loyalty towards 3G mobile data service provider and found corporate image, service quality, trust, satisfaction and loyalty are the major factors. Soomro et al. (2012) concluded that ease of booking, e-booking and clearing time to be most significant factors amongst service quality, ease of online booking, boarding and clearing time for preferences in airline industry.

Factors affecting offline and online buying behavior
In the above fig. the factors affecting offline buying behavior of consumers are shown on the left-hand side while factors affecting online buying behavior are shown on the right-hand side. Factors in the middle are common to both offline and online buying behavior . These factors are convenience, promotion, availability, word of mouth, demography, safety, quality and price.

ISM Approach
ISM (Interpretive structural modelling) is defined as a system which aims at providing assistance to humans by transforming unclear vague mental models to a well-defined model. The model so formed transforms the complex problem or issue to a well-designed pattern. ISM is a technique in which a cluster of variables or factors which are directly or indirectly related are structured into a systematic model. ISM approach is originally propounded by Warfield (1974). The steps involved in ISM are given below: Step 1: Identification of factors The first step in ISM is to identify the factors which pertains to a situation or issue. In the present paper the various factors which are common in online as well as offline shopping are extracted from the literature. Eight factors taken for study are convenience, promotion, availability, brand image, demography, safety, quality and price.
Step 2: Development of SSIM (Structural self-interaction matrix) The next step in ISM is to do a pairwise comparison by developing a VAXO table. for this a group of experts is consulted from industry or academia. The VAXO table denotes the inter-relationship between factors (i and j) as shown in table 2 where:  V depicts that factor i influences factor j  A depicts that factor j influences factor i  X depicts that both i and j influences each other  O depicts that both i and j are unrelated Pairwise comparison of all the eight factor is done with the help of group of experts. About half of the table is left blank as the pairwise comparison of these factors is already done in other half part.
Step 3: Development of Initial Reachability Matrix SSIM matrix in this step is transformed into reachability matrix by converting the values of SSIM into 1's and 0's as shown in table 3  If (i, j) entry in SSIM is V then it becomes 1 and if (j, i) is V then it becomes 0  If (i, j) entry in SSIM is A then it becomes 0 and if (j, i) is A then it becomes 1  If (i, j) entry in SSIM is X then it becomes 1 and if (j, i) is X then also it becomes 1  If (i, j) entry in SSIM is O then it becomes 0 and if (j, i) is O then also it becomes 0 The factors of SSIM in this step are numbered from 1 to 8 and assigned 0's and 1's in both direction (i ,j) and (j, i). and now the entire table is filled.
Step 4: Development of final Reachability Matrix After preparing the initial reachability matrix it's transitivity is checked and 1* is used in place of 0's where there is error. Then a final reachability matrix is prepared as shown in table 4. Transitivity indicates if factor 1 is related to 2 and factor 2 to 3 then it means 1 is related to 3.
Step 5: Partitioning the final reachability matrix In this step the partitioning of final reachability matrix is done by assessing reachability sets and antecedents sets for each factor. Reachability sets includes all the factors which are present in the row of the given factor, it depicts all the factors on which that particular factor is dependent. The antecedent sets include all the factors which are there in the column of that particular factor, it depicts all the factors which that particular factor drives. Intersection sets depicts all the common factors of the reachability and antecedents sets. If the reachability set and intersection set are same then those factors occupy the first level and then all those factors are removed to find the next subsequent levels. In our case there are 5 factors in the first level of partitioning present in iteration 1 ,which are 2,4 ,5 ,7,8 they occupy first level in the hierarchy and then these are removed from all three sets that is reachability set, antecedent set and intersection set in the next level partitioning ,now the factors left are 1 ,3 and 6 in which reachability set is same as intersection set and all three occupy 2 nd level in the hierarchy. The level of partitioning comes out to be two only as the numbers of factors are limited.

MICMAC Analysis
Matrice d' impact Croises-multiplication applique (cross -impact matrix multiplication to classification) is abbreviated as MICMAC analysis. Based on driving power and dependent power factors can be classified into 4 categories : 1. Autonomous -the factors which have weak drive power as well as weak dependence power are classified as autonomous factors. 2. Dependent -the factors with strong dependence power but weak driving power are classified as dependent factors. 3. Linkage-The factors with strong dependence as well as driving power are classified under linkage factors. 4. Independent factors-the factors with weak dependence power but strong driving power are classified as independent factors. Driving power 8 7 6 1,6 5 3 7 2,4,5,8 4 3 2 1 1 2 3 4 5 6 7 8 Dependence power All the eight factors in MICMAC analysis are placed on the basis of their dependence and driving power in the four quadrants. Factor 2,4,5,7 and 8 occupies place in third quadrant and hence are categorized as linking factors they are dependent on each other and also drive each other. Factor 1,3 and 6 occupies place in 4 th quadrant and are categorized as independent factors, they are not dependent on any other factors but all of them drive factor 2,4 ,5 ,7 and 8.

Conclusion
The major objectives of this paper are to find out the factors which are common to both online as well as offline buying behavior and to establish an interrelationship among these factors. To cater to this requirement, an ISM