Moderation Effect of Perceived Benefits of Loyalty Programs on the Link between Store Attributes and Customer Loyalty

The objective of this paper is to examine the moderation effect of perceived benefits of loyalty programs on the link between store attributes and customer loyalty. This study uses data from loyalty program members covering different apparel retailers of Northern districts of Karnataka. In the first step variables were identified through literature review and interview method. In the next step, exploratory factor analysis was performed to identify the critical factors which measure customer loyalty in the apparel retailer’s context. Fi-nally, moderation analysis was employed using SPSS and results indicate that perceived benefits of loyalty programs significantly moderate the relationship between stores attributes and customer loyalty.


Introduction
In retailing, relationship marketing has gained importance in both academics and industrial fields as a strategy to manage, develop, and evaluate the relationship between retailers and customers [1] [2] [3] [4] [5]. Studies have shown that relationship-marketing outcomes are influenced by the relational constructs of trust, commitment, relationship satisfaction, and relationship quality influences buying behaviour [6], increases loyalty [7] [8] [9], and the expectation of continuing interaction and willingness to advice [10]. However, often retail business executives are disappointed with the performance of relationship marketing efforts [11] and customers may not want a relationship with a retailer because they consider it to be a burden [12].

Store Attributes
Store attributes can be defined as retail environments which arouse consumer's desire to purchase, and these include product, service, and store qualities [13].
The retail store attributes have been found to be one of the significant inputs into the consumer's perceptions of global store image and overall attitudes toward a store [14]. Different stores differ in the combinations of store attributes, but previous studies have identified nine store attribute factors, i.e., store atmosphere (an in-store environment consisting of layout, visual display, colours, lightings, and sounds which create an image of the store in consumer's mind), convenience (accessibility to a shopping mall or a store), merchandise (characteristics of products carried by a store or shopping mall such as quality, selection or assortment, and styles of products), promotions (activities a store undertakes to inform customers of its offerings and images), post-transaction service (physical services relating to products after transactions such as merchandise delivery and installation, product warranties, and exchanges or refunds), direct mail (activities performed by a store or a shopping mall to communicate directly with customers through mail brochures and pamphlets), interpersonal communication (employees interactions with customers in a friendly manner), preferential treatment (a customer's perception that he/she receives special recognition from a store or a mall by means of better service which is unavailable to non-regular customers), and tangible rewards (such as discounts and rewards offered to customers in return for their business with a store or a shopping mall). But, previous studies have shown that a quantitative study is needed to identify the suitable variables in the chosen retailing context.

Customer Loyalty and Perceived Benefits of Loyalty Program
The marketing literature present a broad array of loyalty measures [15], and their effectiveness depends on the particular market and study objective. Conventionally, customer loyalty has been defined as a behavioural measure such as amount of purchase [16], possibility of purchase [17], and purchase behaviour [18]. Shoemaker and Lewis 1999 [19] asserted that customers become loyal toward a retailer when their important needs are taken care by its products and services.
"True" customer loyalty is difficult to build and sustain without considering basic attitudinal aspects of customer behaviour into account [20]. In other words, both behavioural and attitudinal aspects of loyalty should be considered simultaneously [21].
The apparent benefits of loyalty program refer to the values that the program provides for members [22]. Therefore, the success of a loyalty program depends on how member observe the program benefits [23]. The perceived benefit produced by loyalty program rewards is the association between the consumer's perceived benefits in relation to the perceived costs of receiving these benefits, and represents a affirmative touching reaction such as prejudiced belief of satisfaction or pleasure, a source of satisfaction and motivation, because the rewards  [27]. Firstly, economical rewards and financial savings which match to an economic purchase motivation-e.g. price reductions, purchase vouchers [28]; secondly, convenience in which they assure commodity motivations e.g. ease of purchase, decrease in shopping time [29]; and lastly informational rewards, which are related to examination [30] [31] [32].

Research Gap
Many retailers fail to identify the appropriate variables and factors leading to customer loyalty intentions in the context of retailers. And off late companies have realised the importance of store attributes in the context of retail, but fail to emphasize the important aspects to get the desired results in the long run. Also from the review of literature it was found that, loyalty programs are the most frequently used marketing activity to induce loyalty among customers in the context of organized apparel retailers. It was intended that, knowing the moderating role of perceived benefits of loyalty programs on the link between store attributes-customer loyalty enables the retailers to formulate strategies to achieve desired results in the long run.

Objective of the Study
To test whether perceived benefits of loyalty programs are moderating the relationship between store attributes and customer loyalty.

Research Methodology
In this study, firstly a review of literature was carried out by focussing on store attributes, loyalty program and satisfaction by emphasising on customer loyalty in the context of retailers. This literature review has helped in identifying variables and also in knowing the research gap. Considering the variables identified in this study a structured questionnaire was developed. The questionnaire was administered in collecting data from members of loyalty programs in select apparel retail stores of Northern districts of Karnataka. Factor analysis and Moderation analysis was used for analysing the data.
Unstructured interview with members of loyalty program and store managers helped to explore their perception about store attributes, loyalty program benefits and loyalty. Unstructured interview was conducted with 20 loyalty program members and two store managers, which helped in gathering diverse viewpoints on the concepts i.e., store attributes, loyalty program and customer loyalty inten- The store managers were asked to rate each variable as "clearly representative", "somewhat representative", and "not representative". Only items rated as clearly representative and somewhat representative were retained. After elimination the draft questionnaire is divided in to two parts: A and B. The part A contains 55 questions related to loyalty, loyalty program and store attributes. All the items are measured on a five point Likert scale. Section B contains seven questions related to demographic details about the respondents such as Name, Age, Education, Gender, Occupation, Income and Period of membership.
The questionnaire consists of two sections. Section A contains 48 variables (questions) covering both store attribute and loyalty program concepts. A total of seventeen items measures store attributes which consists of store environment, merchandise, interpersonal communication, preferential treatment and post transaction service. Fifteen items are measuring perceived benefits of loyalty program which includes economy, convenience, recognition-social relationship and information. The concept of perceived relationship investment, perceived relationship quality was measured using three items each. Five items measuring attitudinal and behavioural loyalty covering the concept of store attributes, the remaining five items are measuring loyalty (purchase intensity & resistance to counter persuasion) covering loyalty program concept. All the items are measured on five point Likert scale ranges from 1 = strongly disagree to 5 = strongly agree except two questions, measuring behavioural loyalty. The questions are: Of the ten times you select a store at which to buy products, how many times do you select this store? a) 1 -2, b) 3 -4, c) 5 -6, d) 7 -8, e) 9 -10; How often would you buy products in this store compared with other stores where you buy products? a) Very less often, b) Less often, c) Neutral, d) Frequently, e) Very frequently.
Section B includes seven questions intended to collect demographic details about the respondents. The demographic details cover name, age, gender, educa-

Data Collection
For descriptive part of the study survey was conducted among six retailers (three retail companies) which include Levi's, Basics Life store's at Gulbarga and Peter England store's at Gulbarga, Bidar, Raichur and Bellary northern cities of Karnataka region. The only retailers which were running loyalty/reward program more than five years were selected since they are well versed in running reward program. A total of 540 samples were collected using self-administered survey method in eight month duration. This was sufficient period to collect data. Sincere efforts were made to get the list of members of loyalty programs of various identified retailers. But in order to maintain business confidentiality, the retailers were averse to provide the list. Hence, in absence of the sample frame Non probability method of sampling i.e. purposive sampling was used. Purposive sampling is a non-probability sampling technique and it occurs when "elements selected for the sample are chosen by the judgment of the researcher. Researchers often believe that they can obtain a representative sample by using a sound judgment, which will result in saving time and money" [34].
"Five subjects for one variable" was used for deciding the total number of subjects for the sample of the exploratory factor analysis according to [35]. Since proportionately as the size of the population was known.
Questionnaire was administered on all week days to ensure the heterogeneity of samples. Initially store managers had suggested to collect data from respondents after purchase process is over to avoid inconvenience to members. So, only those members who are willing to fill the questionnaire were given the questionnaire to get the valid response. Before administering the questionnaire, each member was briefed about the variables and purpose of this study to avoid biasness. A total of 540 valid questionnaires were obtained out of 615 questionnaires administered to collect data. The factor analysis performed to reduce the information contained in a number of variables into a few set of factors with a minimum loss of information, because this research consists of large number of variables, some of which are correlated and need to be reduced to a manageable level [36]. The study used exploratory factor analysis to develop a theoretical factor structure for concepts i.e., store attributes and loyalty program.
Multiple extraction methods are available in SPSS to extract the factors. Principal component factor analysis method is the most commonly used extraction method in social science research. In principal component analysis, the total variance in the data is considered. It is recommended when the primary concern is to determine the minimum number of factors that will account for maximum variance in the data for use in multivariate analysis.

Data Analysis
A moderator is a variable that specifies conditions under which a given predictor is related to an outcome. The moderator explains "when" a Moderation Hierarchical multiple regression is used to assess the effects of a moderating variable. To test moderation, we will be particularly looking at the interaction effect between X and M and whether or not such an effect is significant in predicting Y. Figure 1 shows the schematic representation of moderation analysis.
In this study, X is store attributes, Y is customer loyalty and M is Perceived benefits of loyalty programs.   The "rotated sums of squared loadings" gives the information on the extracted factors after rotation. The value under the "cumulative %" indicates that five extracted factors explain 72.93% of the variance.

Factor Analysis
Similarly Table 3 shows the total variance explained in case of perceived benefits of loyalty program which is 75.36%. Three criteria were used to determine the factor structure: 1) Factor loadings more than 0.40 remained for further analysis, 2) items which cross load more than one factor were excluded, and 3) the difference between the cross loadings should be minimum 0.20 [35] [36].
A total of 16 items grouped under five factors of store attributes emerged from the factor analysis as shown in Table 4. And similarly 12 items were grouped under four factors of perceived benefits of loyalty program as shown in Table 5.   In general there was an attractive range of products The quality of the products that I bought at this store met my expectations I can find variety of products that satisfies my need The price of the products sold by this store was acceptable Signs and decoration in the store were pleasant and tasteful The store spaces were visually appealing It is spacious and easy to move about The store was clean and fresh The store has appropriate return policy and easy to return products The store offered easy exchange services for products I purchased     Above Tables 6-8 shows the SPSS output of moderation analysis. Further, to test whether store attributes positively influences customer loyalty and more specifically whether perceived benefits of loyalty program moderates the relationship between store attributes and customer loyalty intention, a hierarchial multiple regression was conducted. In the first step, two variables were included: store attributes and perceived benefits of loyalty program. These variables accounted for significant amount of variance in loyalty, R square = 0.338, F (2, 537) = 138.82 (P = 0.000). Next, the interaction term between store attribute and perceived benefits of loyalty program was added to the regression model, which accounted for significant amount of variance in loyalty, ΔR 2 = 0.025, F (1, 536) = 21.20 (P = 0.000).
From Table 9 it was evident that both the models 1 and 2 are significant. Model 2 shows the more variance (with interaction) than model 1 (without interaction).
From Table 10 it was noticed that R 2 change = 0.025, P = 0.000, indicating that

Discussion of Research Findings
Store attribute items reduced to the following factors: Store Environment-This factor refers to spaciousness, cleanliness, visual aspect, signs & decoration of the store.
Merchandise-The items that describe this factor are variety, quality, attractive ranges of products, and price of the products available with the retailer.
Interpersonal  [42] and the findings of this study authenticate the same. It is clear that, three store attributes less common in earlier studies appear to reveal store attributes distinctive to organized apparel retailers: merchandise, post-transaction service, and store atmosphere [42].
Items pertaining to Loyalty program reduced to the following factors: Economy-This factor consists of items related to financial benefit, substantial economies (by being part of loyalty program), and reduction in the purchase amount by being part of loyalty program.
Convenience-This factor consists of items such as allowing member to find usual bought products, making purchases easier and granting additional services.
Recognition and Social Relationship-It refers to items such as treating customer as a privileged customer, making customer feel as if the store is paying more attention than others and helping the customer to adhere to a group of privileged customer.
Information-The items that explain this factor refers to providing general information; helping the customers to choose new products and discover good bargains.
Moderation Analysis-The results of moderation analysis performed using SPSS indicate that, perceived benefits of loyalty programs significantly moderate the relationship between stores attributes and customer loyalty. Hence, retailers need to manage their membership/loyalty program meticulously along with store attributes to ensure better customer loyalty.

Conclusion
This research has filled gaps in customer loyalty domain by identifying the important variables and factors measuring loyalty in the context of apparel retailers. This study covers wide variety of the variables and dimensions measuring loyalty. The research found that, perceived benefits of loyalty programs are significantly moderating the relationship between store attributes and customer loyalty.
And hence, the store to stress more in capitalizing the returns from loyalty programs and emphasizing the important store attribute factors identified in this study, to get the desired results in the long run. The non-probability method of sampling i.e., purposive sampling method however, due to unwillingness of retailers to disclose the list of loyalty members is the reason which limits the generalization of the findings. In future, the study may be conducted by considering other type of retail stores. In this study, the customers who are having a membership period of at least a year are considered. In future, researchers can take-up a study considering the customers having longer period of membership, so that deeper understanding about customer loyalty is possible.