ANTECEDENTS OF STORE LOYALTY: AN INVESTIGATION OF THE KENYAN SUPERMARKET INDUSTRY

Dr. Ogutu Robert Peter 1 and Dr. Busolo Hillary Oundo 2 . 1. Geothermal Development Company, Kenya. 2. Department of Curriculum, Instruction and Educational Media, School of Education, Moi University, Kenya. ...................................................................................................................... Manuscript Info Abstract ......................... ........................................................................ Manuscript History

Store loyalty is considered one of the keys in retail business success and has largely been investigated by academia during the past few decades. The Kenyan market has witnessed growth in the supermarket industry across all major towns. However little is known about the antecedents of consumer store loyalty in Kenya. This study investigated the association between store affect and store loyalty in the Kenyan supermarket industry by exploring antecedents of store affect as indirect determinants of store loyalty through the mediating effect of store affect. Explored too is the relationship of demographic characteristics (age, gender, income and distance) with store loyalty. The study adopted a cross sectional survey design with adult shoppers as the target population. A survey instrument developed around antecedents of store loyalty was used. A representative sample of 400 respondents was determined using both proportional and random sampling techniques. A chi square test was carried out to establish the relationships between the various variables. Results show a significant association between antecedents of store affect (store atmosphere, store uniqueness, store familiarity and merchandise value) and store affect. A significant association was also established between store affect and store loyalty. Of the demographic factors under study only monthly income had a significant relationship with store loyalty. These findings contribute empirically to store loyalty concepts and help retail outlets understand why customers patronize their stores and thus develop and strengthen store loyalty drivers that focus on repurchase.

…………………………………………………………………………………………………….... Introduction:-
Store loyalty is the most important factor in retail business success (Ray, 2009) since it has been shown to generate increased profits through enhanced revenues (Hallowell, 1996). This concept continues to receive a lot of attention among stakeholders with a focus on how to develop and sustain store loyalty (Anic, 2006). To scale up store performance, marketers are interested in store loyalty drivers and their impact (Anic, 2006) since it requires less marketing effort and resources to retain an existing customer than to acquire a new one besides a loyal customer is less price sensitive (Tsai et al., 2010) and is more willing to recommend the store to other people, therefore drawing in new customers.
According to the 2016 Oxford Business Group consultancy report, compared against other African markets, Kenya's formal retail penetration rate, which ranges from 30 per cent to 40 per cent, is the second highest in Sub-Saharan Africa. The report attributes the growth to increased expenditure level for the average Kenyan consumer which has risen by as much as 67 per cent in recent years, making Kenya Africa's fastest-growing retail market. Already the increased demand has also fuelled interest from external players some of who have established a presence in the country. Supermarkets are indeed on an upward trajectory buoyed by an expanding middle class, improved infrastructure and an enduring property boom. They continue to grow their market share and penetration as they become the preferred shopping outlet for many middle and high-income consumers in towns. According to (Sunday, 2012), supermarkets and shopping malls are redefining skylines in towns like Nakuru, Eldoret, Kisii and Kakamega. The competition pits three major supermarket chains Nakumatt, Tuskys and Naivas. With the many players coming in and the existing one expanding, store loyalty determinants are definitely going to be key for retail business success.

Statement of the Problem:-
Store loyalty continues to receive a lot of attention since it is crucial for retail success and survival in today's competitive market (Yap et al., 2010). Research has demonstrated several factors to influence store loyalty in different contexts. This includes perception of store attributes (LeHew et al., 2002;Sirohi et al., 1998), satisfaction (Aniac, et al., 2006;Stoel, et al., 2004), and demographic characteristics (Aniac, et al., 2006). However, (East et al., 1995) notes that demonstration of store loyalty need to take account of differences in countries and the evolution within the retailing industry. Nakuru town has experienced growth in supermarkets with little known about antecedents of consumer store loyalty. This study therefore seeks to address this gap by investigating antecedents of store loyalty with focus on supermarkets in Nakuru town.

Objectives:-
The purpose of this study is to examine significant store loyalty variables marketers need to focus on in order to improve and maintain customer loyalty.

Specific objectives:-(i)
To determine the relationship between store atmosphere and store effect (ii) To determine the relationship between store familiarity and store effect (iii) To determine the relationship between store uniqueness and store effect (iv) To determine the relationship between merchandise value and store effect (v) To establish store loyalty differences between females and males (vi) To determine the effect of age on store loyalty (vii) To establish the effect of distant on store loyalty (viii) To establish the effect of income on store loyalty. (ix) To determine the relationship between store affect and store loyalty Hypotheses:-H1: Store atmosphere has a significant association with store affect. H2: Store familiarity has a significant association with store affect. H3: Store uniqueness has a significant association with store affect. H4: Merchandise value has a significant association with store affect.

184
H5: Gender has a significant association with store loyalty. H6: Distance has a significant association with store loyalty. H7: Age has a significant association with store loyalty. H8: Income has a significant association with store loyalty. H9: Store affect has a significant association with store loyalty.
Justification and significance of the Study:-This study's focus is to examine how Kenyan supermarkets can position their stores better and establish store loyalty variables marketers need to focus on and improve in order to maintain store loyalty.

Scope and Limitation of the Study:-
Investigated is customer store loyalty within supermarkets in Nakuru town, Kenya. Factors considered include store affect driven by store uniqueness, merchandise value, store familiarity and store atmosphere and demographic characteristics as predictors of store loyalty. Generalizability of results should be done with caution since the study only focused on supermarkets in Nakuru town. ). These drivers are not universally applicable due to the diverse factors that influence consumer behavior in different settings, be they markets or countries (Ball et al. 2004;Schofield, 2016). This study will adapt (Ray, 2014) model on store loyalty. Discussed below are antecedents of store loyalty.
Store affect:-Store affect has been conceptualized as affective experience evoked during a store visit which may facilitate attachment to the brand (Orth, et al., 2010). Ray (2009) defines store affect as a stores potential to evoke favourable emotional responses in its customers. He further identifies a significant relationship between store affect and store loyalty and establishes that store atmosphere, store uniqueness, merchandise value and store familiarity can impact store loyalty directly or through store affect. Affective responses as a predictor of re-patronage intentions have also received considerable academic attention (Donovan et al., 2002;Mattila, et al., 2001).
Antecedents of store affect:-Store atmosphere, store uniqueness, merchandise value and store familiarity can impact store loyalty directly or through store affect. In this study these factors will be considered to indirectly impact on store loyalty through store affect.
Store atmosphere:-Store atmosphere forms part of the underlying dimensions of the retail store image (Baker et al., 2002). Indeed it contributes to the physical environment of a retail store that projects a desired image thereby attracting customers (Berman et al., 2013). Store attributes such as music ( Merchandise value:-Zeithaml (1988) describes value as a tradeoff between give and get mechanisms. Such evaluations are mainly derived from an interaction between quality and price (Grewal et al., 1998;Sirohi et al., 1998). Customers make such evaluations based on products price and quality (Grewal, et al., 1998). Merchandise value is an important factor for retail stores seeking to boost sales (Vogel, 2008). This is paramount since customers consider price comparisons in selecting the retailers (Zielke, 2006) besides taking into account competing alternatives as well (Sirohi et al., 1998). There is need for retailers to consider customer variance in the way they view value since there are those who search for low prices, who are willing to pay higher price for convenience and superior service, and those who shop in prestigious stores to get status (Vogel, 2008). Research to determine association between merchandise value and store loyalty has been undertaken by various scholars. ( Gilbert et al. (2002) found that discounts significantly affect customers' purchase behaviors. In this study merchandise value has been operationalizes as high quality merchandise providing good value for money.

Store familiarity:-
Research has established that people unfamiliar with a place primarily use external sources of information in their navigation, whereas those who are familiar with the environment rely more heavily on information stored in their long term memory (Chebat et al., 2005). When it comes to shopping, (Iyer, 1989;Park et al. 1989) argue that customers without repetitive shopping experiences in a store will attend to a larger number of in store cues in their navigation than customers who are familiar with the store's layout. Indeed research shows familiarity to improve a person's performance on navigational tasks (Chebat et al., 2005;Dogu et al., 2000;Hölscher et al., 2006). Titus et al. (1996) found that individuals with little or no familiarity with a supermarket performed a larger number of product searches than individuals who were familiar with the shopping environment. In this study store familiarity is operationalized as a combination of familiarity with the store and frequency of visits. Given the empirical evidence above and considering that the lack of familiarity contributes to elimination of the brand from consideration set for purchase decisions (Dick et al., 1994) it is proposed that SB familiarity have a positive relationship with store affect.
Store Uniqueness:-According to (Small Business Encyclopedia, 2016) before you can begin to sell your product or service to anyone else, you have to sell yourself on it. This is especially important when your product or service is similar to those around you. They further note that very few businesses are truly unique and that there is need for firms to pinpoint what makes their businesses unique in a world of homogeneous competitors in order to target sales effort successfully. Ultimately, the goal is to differentiate a store from competition and subsequently give customers a reason to purchase goods from the store (Schreiber, 2015). Research shows store's uniqueness helps generate positive affect towards the store and thus likely to drive store loyalty (Ray et al., 2009). It's therefore critical that stores become more creative in determining what sets their businesses apart from others because ultimately this is what will make the business stand out and become known and therefore drive store loyalty. 186

Demographic determinants:-Age:
Age is a demographic characteristic that continues to attract considerable attention. Research suggest that store loyalty is manifest more often within the 25 -44 age group (East et al., 1995) with Wright et al (1999) suggesting higher loyalty in the 35 -44 year old people. Store loyalty increases with increase in age (Anne-Françoise et al., 2016) and that older shoppers often exhibit more routine behaviour, and are less willing to accept new ideas and uncertainty (Straughan et al., 2001). They are less willing to experiment and wish to remain associated with the stores they have been patronizing (Ghosh et al., 2010). Indeed as they age, they begin to patronize fewer shops (Baltas et al., 2010). It is therefore expected that the older the consumer the more loyal they will be to the store.

Gender:
Gender difference regarding loyalties has been an important area for researchers and marketers since significant differences might require different selling approaches (Melnyk, 2009). Indeed, several empirical studies on buying behavior point to a relationship between customer demographics and store loyalty (Stan, 2015). Research has found significant difference between women and men on buying behaviour (Stan, 2015;Jose, 2009). Anne-Françoise et al., (2016) for instance found men to be more loyal than women to the store chain, while women were more loyal than men to individual stores. According to Melnyk et al. (2009) prevailing wisdom assumes that female consumers are more loyal than male consumers. They further note that depending on the object of customer loyalty, female consumers tend to be more loyal than male consumers to individuals, whereas men are more loyal to a group as opposed to female consumers. (Anic, 2006) avers that women tend to be significantly more conative loyal to a store and that they exhibit higher purchase intentions and are more likely to recommend store to other people compared to man. According to Ndubisi (2006) women are emotionally involved in shopping and get detailed information about the products and services and their satisfaction manner is also varying from men, and they are more loyal than men. Stan (2015) too found women to be more loyal than men to stores. Other empirical studies have found females not to show strong customer loyalty as compared to their male counterparts (Melnyk et al., 2009). Robert (2001) for instance found men to exhibit greater loyalty to retailers than women.

Income:
Income has a significant effect on store profits and customers tend to spend more, if they have more income (Anic, 2006). Research suggests a relationship between loyalty and income level (Homburg et al., 2001). Customer's income has a strong effect on choice decision (Zeithaml, 1985). Indeed research shows customers with high incomes to be more involved in location convenience, merchandise quality and pleasant shopping experience and to patronize fewer stores (Baltas et al., 2010). Customers with low incomes need to visit more stores to get value for money, but customers with high income can simplify their lives with one stop shopping (East, 1997;Tate, 1961). Tate (1961) identified high tendency of disloyalty in middle income or upper income families whereas low-income category tie themselves with their previous brand. East (1997) overs that store loyalty is lower among customers who have low income. Thus, higher income customers emerge as switchers of stores.

Distance:
The location of a store affects a customers' frequency to shop from the store. Anic (2006) argues that customers who live near to a store tend to be more conative loyal than distant customers. Rhee et al. (2002) determined stores easily reached to be the most preferred and therefore significant to store loyalty. Also East (1997) found that among those customers who can assess store quickly, tend to score in store loyalty higher, compared to other customers. If a particular store is more accessible than others, customer tends to spend more money in there, which can explain part of loyalty (Anic, 2006), however some studies have found that there is no significant relationship between store location or time access and loyalty (East et al., 1995).

Theoretical Model:
This study is hinged on the Mehrabian et al., (1974) Stimulus-Organism-Response Theory which outlines physical stimuli act as antecedents that directly shape emotional states (such as pleasure, arousal, and dominance) which in turn, influences an individual's behavior . Research has found affective states experienced while shopping in retail stores to influence customers' store loyalty (Ray, 2009;Orth, et al., 2010;Donovan et al., 2002). When applied to the retail context, it is proposed that consumers will utilize the stimulus cues (store atmosphere, uniqueness, familiarity and merchandise value) to determine both hedonic and utilitarian appraisals of the store which, in turn will lead to commitment and/or re-patronage intentions. In this study store atmosphere, uniqueness, 187 familiarity and merchandise value are the stimulus cues which in turn leads to store affect and subsequently store loyalty.

Conceptual Model on Store Loyalty:
The model posits that store affect and demographic characteristics have an association with store loyalty. Store affect is driven by store uniqueness, merchandise value, store familiarity and store atmosphere. The framework is presented in Figure 1.

Sample Design:-
The sampling frame consisted of shoppers in Nakuru town. Krejcie et al. (1970) approach on determination of sample size was used to determine the number of shoppers who formed the sample. Both proportional and random sampling techniques were used to select a total of 400 respondents.

Data Collection Instruments:-
A closed ended survey questionnaire was used to collect primary data on antecedents of store affect (store familiarity, store uniqueness, merchandise value and store atmosphere), store affect, demographic characteristics (Age, gender, income and distance) and store loyalty. Scale items were adopted from (Ray, 2014). Respondents were drawn from adult residents of Nakuru town, Kenya. All Likert scale items were measured on a 5-point Likert scale (from 1=strongly disagree to 5=strongly agree).

Reliability Test:-
Cronbach's (1951) alpha coefficient was used as a quality indicator of the scale items. The reliability coefficient for the 17 items in the data collection instrument was found to be 0.718 which is within the minimum acceptable threshold of 0.70 as recommended (Hair et al., 2006).

Data Analysis:-Inferential Analysis:-
A chi square test of independence was undertaken to determine the relationship between; (i) Store Atmosphere, Store Uniqueness, Merchandise Value, Store familiarity with Store Affect (ii) Demographic factors of (gender, age, income and distance) with store affect (iii) Store affect with store loyalty Store Affect Store Loyalty Store familiarity Store Atmosphere

Merchandise Value
Store Uniqueness

Demographic Factors
With a pvalue < 0.05 was considered significant. Cohen (1998) guideline on the strength of association between variables was employed. An effect size between r = 0.1 − 0.23 was considered small; r = 0.24 − 0.36 medium; and r = 0.37 and above was considered large.

Research Findings and Analysis:-Sample Characteristics:
The study targeted shoppers in Nakuru town. Of the 400 questionnaires administered, 372 were collected accounting for 93 percent response rate. Of the total respondents who completed the survey (N=372), 187 (50.3%) were male and 185 (49.7%) were female. 171 (46.0%) of respondents were between 18 and 29 years old, 188 (50.5%) were between 30 and 49 years old, 11 (3.0%) were between 50 and 64 years old and 2 (0.5%) were 65 years or older.
Of the total respondents, 113 (30.4) had a monthly income less than 24,999, 199 (53.5) had monthly income between 25,000 and 49,999, 36 (9.7) had a monthly income between 50,000 and 99,999, 24 (6.5) had a monthly income of 100,000 and above.
Looking at how far respondents homes were from the supermarket they shopped from, 33 (8.9%) of respondents lived within less than 1 km from the supermarket, 72 (19.4%) lived between 1 -2 km from the supermarket, 133 (35.8%) lived between 2 -3 km from the supermarket, 56 (15.1%) lived between 3 -4 km from the supermarket and 78 (21.0%) lived at a distance of more than 4 km from the supermarket they shop from. Results are presented below in Table 1 below. Antecedents of store affect:-Each of the variables (Store Atmosphere; χ 2 (9) = 79.683, p = .001, Store Uniqueness; χ 2 (12) = 76.995, p = .001, Merchandise Value; χ 2 (12) = 25.804, p = .001, and Store Familiarity; χ 2 (9) = 63.805, p = .001) had its p-value < 189 0.05. The relationships were therefore statistically significant. We therefore reject the null hypotheses, and conclude that they all have a relationship with store affect. Results are presented below in Table 2. The magnitude of association between antecedents of store affect and store affect:-With the exception of merchandise value (phi = .263, p = 0.011) which had a small, positive and significant effect size with store affect, all the other variables; Store atmosphere (phi = .463, p = 0.001); Store uniqueness; (phi = .455, p = 0.001); and Store familiarity (phi = .414, p = 0.001) had a large, positive and significant effect size with store affect. Results are presented below in Table 3. The relationship between demographic factors with store loyalty:-Gender; χ 2 (4) = 6.259, p = .181, Age; χ 2 (12) = 20.550, p = .057 and Distance; χ 2 (16) = 20.826, p = .185 had their pvalue > 0.05 and therefore not significant. We therefore fail to reject the null hypotheses, and conclude that the three variables are independent of store loyalty. Income; χ 2 (12) = 25.454, p = .011, had its p-value < 0.05 and therefore significant. We therefore reject the null hypothesis, and conclude that monthly income has a relationship with store loyalty. Results are presented below in Table 4. The magnitude of association between demographic factors and store loyalty:-Income (phi = .262, p = 0.013) had a small, positive and significant effect size with Store affect. Gender (phi = .130, p = 0.181); Age; (phi = .235, p = 0.057); and Distance (phi = .237, p = 0.185) had their p-values > 0.05 and thus not significant. Results are presented below in Table 5. The relationship between Store Affect and Store Loyalty:-The p-value < 0.05, we therefore reject the null hypothesis, and conclude that there is a significant association between store affect and store loyalty χ 2 (12) = 105.906, p < .001. Results are presented below in Table 6. The magnitude of association between store affect and store loyalty:-Store affect (phi = .534, p = 0.001) has a large, positive but significant effect size with Store loyalty. Results are presented below in Table 7.

Summary of Findings:-
The study sought to examine the relationships between the antecedents (store atmosphere, store uniqueness, merchandise value, and store familiarity) with store affect and store loyalty. This study finds all the antecedent variables to have an association with store affect. There exists also a significant relationship between store affect and store loyalty. Monthly income had a significant association with store loyalty.

Conclusion:-
Store affect has been conceptualized as affective experience evoked during a store visit which may facilitate attachment. Indeed it is has the potential to evoke favourable emotional responses in its customers. To develop a customer base that will then guarantee good returns as a result of store loyalty, retailers will need to pay special attention to store atmosphere, store uniqueness, merchandise value and store familiarity. These factors can be manipulated to create the desired effect on store affect and therefore store loyalty given they are controllable. Besides, affective responses have been determined as a predictor of re-patronage intentions. The other factor that is 191 significant is the consumer income. There is a great need to understand consumer income levels so as to target them appropriately.
Limitations and Recommendations:-First, assessment of store affect in this study is based upon antecedents identified from the extant literature. This is not exhaustive and future research should explore other industry specific factors that have not been considered in this study. The study employed the static cross-sectional correlational design. Future research will need to use longitudinal design to better understand constructs such as loyalty. This is because store loyalty, takes time to develop and would accurately be determined if measured over time. This study has limited generalizability because of the regional sample used. The study was conducted in Nakuru Town, Kenya, which limits its generalizability to other countries as cultural and geographical differences create significant distinctions in the expectations and evaluations of consumers. Observations based on more representative samples therefore remain an important goal for future research.