Ethnicity and Choice Criteria in Retail Banking : A Malaysian Perspective

This exploratory research empirically investigates variation in bank choice criteria between three distinct cultural groups in Malaysia, namely Malays, Chinese and Indians. The author predicted that, due to differences in values and lifestyles of the three ethnic groups, cultural differences would be observed in the level of importance members of each group attached to specific bank choice criteria. A quantitative methodology, using responses given by 482 Malaysia’s undergraduates, is employed in the analysis. The results of this comparative analysis indicate that there is a difference in bank choice criteria based on ethnic backgrounds, and bank marketers can capitalize on these differences, by identifying them and designing appropriate operations and marketing strategies, leading to increased customer satisfaction and competitive advantage.


Introduction
There have been numerous studies over the years that analyze the issue of choice criteria in retail banking context.For example, Anderson et al. (1976), Javalgi et al. (1989) and Boyd et al. (1994) conducted their studies in the USA; Laroche et al. (1986) and Kaynak (1986) in Canada; while others have provided general studies in the European context: Thwaites and Vere (1995) and Devlin and Gerrard (2004) in the UK; Holstius and Kaynak (1995) in Finland; Mylonakis et al. (1998) in Greece; Kennington et al. (1996) in Poland and Zineldin (1996) in Sweden.A minority of studies has been conducted in the Middle East: Yavas (1988) in Saudi Arabia; Erol et al. (1990) in Jordan and Almossawi (2001) in Bahrain.While two studies have been conducted in the South Asian region: Rehman and Ahmed (2008) in Pakistan and Rashid and Hassan (2009) in Bangladesh.A number of studies at different points in time have been conducted in the Far East; for example, Kaynak and Kucukemiroglu (1992) in Hong Kong; Huu and Kar (2000) and Gerrard and Cunningham (2001) in Singapore and Lang and Chua (1989); Che Wel and Mohd Nor (2003) and Mokhlis et al. (2008) in Malaysia.Other researchers have recently investigated international consumers' choice criteria of banks in a range of cultural and country economic scenario (Blankson et al. 2007(Blankson et al. , 2009)).
A related theme that emerges from the existing literature is that of analyzing bank choice criteria related to demographic factors and analyze differences in the choice criteria across groups.In the first major study of note, Laroche et al. (1986) found that there are some significant differences in choice criteria for retail banks in Canada with respect to gender, language, age, income and educational level groups.Kaynak et al. (1991) reported differences in bank selection criteria based on gender, age, education background of bank customer in Turkey.In Bangladesh, Rashid and Hassan (2009) also found that customers with different demographic background reported different level of importance towards choice criteria for Islamic banks.Omar (2008) investigated gender differences in the relative importance of bank choice criteria in Nigeria whilst Srivatsa and Srinivasan (2008) investigated similar matters in India.Both authors found some differences in choice factors used by male and female customers in selecting a retail bank for patronage.
As indicated above, previous studies into choice in retail baking have investigated choice criteria in a number of countries and in variety of cultural context.In addition, various demographic factors have been employed in attempts to highlight differences between groups of customers.Only a limited consumer research on ethnicity, however, has extended beyond consumption and choice behavior into the domain of retail banking behavior.Even in those instances in which ethnicity have been empirically examined in a retail banking context, the research effort has been somewhat limited to certain aspects of consumer behavior.For example, Joy et al. (1991) investigated ethnicity as a factor influencing the use patterns of financial services in Quebec, Canada.Snow et al. (1996) focused their study on ethnicity influences on expectations of retail financial services in Canada while a study conducted by Burton (1996) studied consumer financial behavior among British Asians in the pension markets.There is no evidence of any previous published research investigating the influence of ethnicity on choice criteria in retail banking.This exploratory research article offers an analysis of choice criteria in retail banking for groups with differing ethnic identity.Rossiter and Chan (1998) highlighted the latter factor as being potentially important for use in consumer research because ethnic microcultures are potent forms of influence in society and may affect decision making process.Notwithstanding many marketers have a strong bias towards demographic segmentation, perhaps, fearful of being labeled racist (Chudry & Pallister, 2002), ethnicity is an obviously relevant causal construct and is so manifestly important in business and consumer behavior that it cannot be neglected as an area of scientific study (Rossiter & Chan, 1998).Studies done in the USA, which has predominantly focused on retailing, have shown that ethnicity is significant in relation to consumers' preference for advertisements, shopping behavior, consumer values, media usage and evaluation of retail store attributes, among others (e.g.Valencia, 1989;Delener & Neelankavil, 1990;Herche & Balasubramanian, 1994;Bristow & Asquith, 1999;Kim & Kang, 2001;Dublish, 2001;Chudry & Pallister, 2002), suggesting that ethnicity may prove to be a desirable bases for market segmentation.However, compared with what retailers know of the relationship between ethnicity and consumer behavior, financial institutions know relatively little (cf.Burton, 1996).As multiculturalism has become a social reality in many countries of the world, impacting every facet of business operations, including marketing, there is a clear need for ethnicity and cultural differences to be fully integrated into the bank marketing literature.
This study is one step in the direction of closing the gap in knowledge of ethnic banking by examining variation in bank choice criteria between three distinct cultural groups in the Malaysian context, namely Malays, Chinese and Indians.Due to differences in values and lifestyles across ethnic groups, it is inferred that consumers from different ethnic groups employ different and embedded cultural dimensions in the selection of banks.Thus the research question posed in this study is: "How do consumers from a range of ethnic cultures evaluate and make bank choice decisions?" The knowledge obtained from this study will provide commercial bank managers with better insights into the salient criteria used by ethnic customers in deciding which bank to patronize.An accurate appreciation of choice criteria may well aid bank marketers as they attempt to develop more precise, targeted marketing strategies to better cater to the needs of different customer segments.If ethnicity is related to customers' bank choice criteria, then this information would allow for an easy and inexpensive segmentation of the market.
The remainder of this paper is organized in four sections.Section 2 reports the method and data collection procedures.This is followed by Section 3, which presents the results of data analysis and its interpretation.Section 4 has a discussion, which includes conclusions and implications of the study.

Context
The study draws its data from a multi-ethnic country located in the Southeast Asia -Malaysia.Among examples of plural societies, contemporary Malaysia shows an unusually balanced ethnic structure of two dominant groups, the Malays who make up 53.4 per cent and the Chinese who make up 26 per cent.As well there are 7.7 per cent Malaysians of Indian ethnic origin (Department of Statistics Malaysia, 2005).Due to the cultural differences that exist in the origins of different communities, there is a noticeable absence of homogeneity in the behavior of consumers in Malaysia where the nature of its domestic market is highly characterized by the "ethnically segmented consumer markets" (Mohd.Salleh, Teo & Pecotich 1998, p. 481).Such unique characteristic provides a particularly appropriate context for this study from which a sample representing consumers from different ethnic groups can be drawn from its population.

Instrument
A structured questionnaire was prepared for use in the survey based on literature review and objectives of the study.The questions were organized into three sections according to the following topics: bank choice criteria, banking behavior and personal background.
The first section of the questionnaire asked respondents to rate the relative importance of 28 bank attributes when choosing which banks to patronize.They were measured on a five point Likert-type scale of importance ranging from 1 (not important at all) to 5 (very important).The list was based on previous similar studies (Thwaites & Vere, 1995;Almossawi, 2001;Gerrard & Cunningham, 2001).
The second section of the questionnaire sought to obtain information on the banking behavior of respondents.The respondents were asked for the name of banks at which savings accounts were maintained.The length of time that customers have been with their banks was also measured.Finally, to obtain personal background of the respondents, questions regarding their gender, age, ethnicity, faculties and course studied included in the last part of the questionnaire.
To determine the potential effectiveness of the questionnaire and whether further revision is needed prior to conducting the survey, the questionnaire was pilot tested.The questionnaire was distributed to twenty students as a sample group.The subjects were asked if they had any problems understanding the questionnaire or have specific comments regarding the questionnaire.The subjects were encouraged to be very free with their responses, make suggestions for improvement and delineate any difficulties they found.No serious problems were found and minor amendments were made to the survey questions based on the feedback received.

Sampling and data collection
A self-administered questionnaire was distributed to a non-probability sample of 600 undergraduate students of a public higher learning institution in the state of Terengganu.The survey was taken in a controlled classroom environment; allowing for a stronger research design.Although the sample is selected on the basis of convenience and ease, data were gathered at different locations (classrooms and faculties), on different days of the week, and at different times of the day, thus reducing location and timing biases.
Although there has been concern regarding the use of students as surrogate consumers and also about the validity and generalizability of student samples (Beltramini, 1983), college students were deemed appropriate for this study, for they are actual customers of banks (i.e.owned bank accounts), conversant with bank products and represent a critically important target markets for banking services (Mokhlis et al. 2008).From a statistical standpoint, using a relatively more homogeneous group such as college students is particularly helpful to minimize random error that might occur by using a heterogeneous sample such as the general public (Calder et al. 1981).This is because the likelihood of error within the measurement model being inflated by situational factors inherent in diverse samples (e.g.age, income, education and other potential confounds) is reduced when respondents are homogeneous across demographic and behavioral characteristics, thus resulting in less "extraneous variation" (Assael & Keon, 1982;Peterson, 2001).The applicability of using students as a unit of analysis evidence in previous studies of retail bank choice (see for example, Huu & Kar, 2000;Mokhlis et al. 2008;Blankson et al. 2007Blankson et al. , 2009)).

Analysis and Results
From a total of 600 questionnaires distributed, 520 were returned, out of which 482 were deemed usable (valid and completed), thereby yielding a response rate of about 80.3 per cent.Such a response rate was considered sufficient for statistical reliability and generalizability.

Sample profile
Table 1 shows the key characteristics of the respondents.Of the 482 respondents, 61 per cent were females.The over representation of female respondents is expected since the population of female students in universities in Malaysia is 60 per cent females and 40 per cent males (Ministry of Higher Education Malaysia, 2009).The median age of respondents is 21.8 years (+ s.d. 1.32).In terms of ethnic group, 55.8 per cent of the respondents were Malay, 26.8 per cent were Chinese and 17.4 per cent were Indian.

Bank choice factors
In the first phase of the data analysis, exploratory factor analysis (EFA) was carried out to explore the underlying structure of the data.Varimax-rotated factor analysis was performed on the constituent items representing the different constructs to validate empirically the theoretical structure of the scale.The Bartlett test of Sphericity (Approx.Chi-square = 3557.7;df = 351; sig.0.000) and the Kaiser-Meyer-Olkin (KMO) measure of Sampling Adequacy index (value of 0.773) confirmed the appropriateness of the data for EFA.
Principal component analysis was selected to reduce the number of factors where the eigenvalue greater than 1 and a cumulative percentage of variance explained being greater than 50 percent were the criteria used in determining the number of factors.An accepted method of interpretation of factor loadings is to regard as significant any variable with a loading of 0.4 or greater as associated with the appropriate factor (Hair et al. 1998).In addition, to ensure that each factor identified by the factor analysis would have only one dimension, any item loading on more than one factor with a loading score equal to or greater than 0.4 on each factor was eliminated from the analysis.Based on these criteria, nine factors were extracted (see Table 2).Together the nine factors explained more than half of the variance observed in the variables (63.74 per cent), which satisfies the percentage of variance criterion for social science research (Hair et al. 1998).The nine factors identified were (1) attractiveness, (2) people influences, (3) service provision, (4) ATM service, (5) secure feeling, (6) marketing promotion, (7) proximity, (8) branch location, and ( 9) financial benefits.
To examine the internal consistency of the factors obtained, Cronbach of 0.5 was applied as a general guideline throughout the computation (Kerlinger & Lee, 2000).As depicted by Table 2, the Cronbach's value for all factors range from 0.508 to 0.762, exceeding the adopted criteria, thus the constructs measures are deemed reliable.

Ranking importance of bank choice criteria
The second area of interest relating to the ratings of the factors, is which ethnic group rated the factors most important.The factors that were more important to one ethnic group were less important to other groups, resulting in a difference in how the factors were ranked.Table 3 shows the choice criteria for the three ethnic groups ranked in order of mean importance of each factor.
Ranking of the factors by ethnic group indicate three areas of interest.The top factor, secure feeling, is the most important factor for all ethnic groups.The least important factor, people influence, was ranked the lowest for all ethnic groups.The factors between the top and the bottom -the middle group of factors -are different between the ethnic groups.This middle group included factors such as 'ATM service', that ranked as the second most important factor to the Chinese group but the sixth to the Malay group.Similarly, 'financial benefit' was ranked as second most important to the Malay group but fifth to the Indian group.

Differences in bank choice criteria by ethnicity
Analysis of variance (ANOVA) was used to test the statistical differences between the means for each of the selection factors among ethnic groups.In such an approach, the null hypothesis that there is no difference in means between groups is rejected if the F-statistic is sufficiently large to be significant.The probability level accepted for statistical significance of F-statistic in the present study was set at p < 0.1, showing there was 10 per cent probability that the result occurred by chance.The results of this analysis are shown in Table 4.At the 0.01 level, factors of which the three samples had variation were 'people influence' (F = 4.622, p = 0.01), 'marketing promotion' (F = 7.997, p = 0.000) and 'branch location' (F = 8.424, p = 0.000).At the 0.05 significant level, there was a variation for 'financial benefits' factor (F = 2.827, p = 0.06).
To determine where the significant difference(s) lie after the hypothesis has been rejected in ANOVA, pairwise multiple comparison tests were conducted.In keeping with customary practice, the significance level was set at p < 0.05 for all post-hoc comparisons.
With respect to 'people influence' factor, a significant contrast existed between Malay and Indian groups.The Indian group perceived people influence as more important than did the Malay group.For 'marketing promotion' factor, a significant difference was found between Malay and Chinese groups.Malays were more likely to perceive marketing promotion as more important than did their Chinese counterparts.
In relation to 'branch location' factor, significant differences were found between Malay and Chinese and between Malay and Indian groups.An examination of the mean scores showed that Malays viewed branch location as being greater of importance in selecting a retail bank than did their Chinese and Indian counterparts.As for 'financial benefits' factor, a significant difference was found between Malay and Indian groups.The cell means indicate that Malays placed relatively more importance on financial benefit than did Indians.The Chinese sample was intermediate in this regard and not significantly different from either Malay or Indian groups.

Conclusion
This study adds to the limited body of knowledge in the bank marketing literature on similarities and differences between ethnic groups in Malaysia in terms of their choice behavior in retail banking context.Albeit tentative, the preceding analyses demonstrate that differences in bank choice criteria do exist between the three ethnic groups, thus indicating where bank marketers should place attention on salient attributes to attract and retain members of different ethnic segments.In the saturated banking markets of countries and cities with a diverse ethnic mix, ethnic banking may prove a competitive advantage.Banks that can properly identify the selection factor where ethnic differences occur can adapt their marketing strategies to capitalize on these differences.
This study had certain constrains that must be considered when assessing the outcomes of its findings and implications.First, this study poses generalizability questions because the sample frame used is the college student cohort and thus the results do not represent the banking behavior of the general public.Secondly, the findings of this study are obviously exploratory in nature and should be regarded as suggestive evidence, rather than a conclusive demonstration that such variation exist.
Future research are warranted to provide consistent evidence for verification of the differences found in this study and continue this stream of research in order to achieve a fair understanding of variation in consumer choice behavior in different ethnic cultures.Finally, it must be acknowledged, that ethnic background is just one of the demographic factors that might affect the selection criteria employed in selecting which banks to patronize.Future studies should explore explanations for variations that go beyond assigning all differences to ethnic origins.

Table 1 .
The profile of respondents.Due to space limitations, only a summary of the result is presented.Complete result of the factor analysis can be obtained directly from the corresponding author upon request.

Table 3 .
Means test of the relative importance of bank choice factors * Significant at p < 0.1; ** significant at p < 0.05; *** significant at p < 0.01