AN ASSESSMENT OF CUSTOMER PERCEPTION OF THE BRAND PERSONALITY OF GLOBACOM, NIGERIA

Scientific Editor Ilan Avrichir Objective: (i) to adapt Aaker’s (1997) scale to measure brand personality in the Nigerian telecommunications market using the methodology identified by Avis (2012) to be the most stable in literature; (ii) to identify the traits that could be used in marketing communication in the telecommunications industry; (iii) to reveal the brand personality factors of the telecommunications industry. Method: The study used a mixed design and survey method to sample 700 consumers of the Nigerian telecommunications industry. Data was collected using a questionnaire and analyzed using inferential statistics. Main Results: Aaker’s (1997) scale was adapted to measure the brand personality in the Nigerian telecommunications industry and it measured six factors, namely, Sincere, Rugged, Sophistication, Competence, Innovative and Excitement.. Relevance/originality: The study provides local empirical evidence of the brand personality factors of the Nigerian telecommunication industry. The scale will be relevant in measuring the brand personality of other brands within the industry. Theoretical/Methodological contributions: The study supported the use of Aaker’s (1997) methodology in developing brand personality scales. The findings were consistent with terminology that could be used to describe brands in the industry. Social Contributions/For management: Managers of Nigerian telecommunication industry brands can make use of the traits identified in marketing communication in order to better present their brands.


Introduction
Customer's willingness to pay for telecommunications service relies on the value such a customer attaches to that service. Consumption of telecommunication services can either be functional or symbolic (Ekinci, Sirakaya-Turk & Baloglu, 2007). It is not uncommon to see many telecommunication brands focussing on the functional benefits of their service such as boasting of fast internet service or guaranteeing quality voice calls. However, symbolic consumption may be a more useful tool to help to drive increased customer loyalty (Kim, Kim & Holland, 2018). This is because symbolic consumption can represent the values and beliefs of a culture (Aaker et al, 2001) as well as allowing the customer to be selfexpressive (Keller, 1993). This self-expression may take the form of identifying with values the brand has stood for or aspiring towards a lifestyle promoted by the brand and has been found to be a significant predictor of customer loyalty (Ekinci, Sirakaya-Turk & Preciado, 2013).
Brand personality has been identified as a driver of symbolic consumption (Chi, Pan & Del Chiappa, 2018;Usakli & Baloglu, 2011). Thus is can be a useful tool in driving customer loyalty for telecommunication brands as it can help to develop marketing communication that can engage customer self and lifestyle congruity and increase the chances of loyalty. However, Aaker (2001) and Aaker, Benet-Martinez, & Garolera, (2001) explained how brand personality's symbolic and expressive attributes associated with commercial brands are structured and how this structure varies across various countries with seemingly different cultures and discovered that although the utilitarian attributes of commercial brands tend to exhibit limited variability in meaning or importance across cultures, the symbolic or value-expressive functions (the brand personality) associated with a brand tend to vary to some degree because of the variation of individuals' needs and self-views and socialization. This suggests that there in a need to need to empirically investigate Brand Personality across various product categories and cultures as there may be variations from Aaker's original five dimensions which were measured in the American context. This is also supported by Hosany, Ekinci & Uysal (2006) who argued that cultures that are quite different in their values and needs are more likely to exhibit culture-specific differences in brand personality.
Brand Personality has been measured empirically using both quantitative and qualitative research designs (Avis, 2012) with quantitative methods being mostly dominated by Aaker's (1997) methodology (Freling, Crosno & Henard, 2010), and so far, only two qualitative methods being found (Avis, 2012). This study aims to measure Brand Personality within the Nigerian context by adopting the widely used Aaker's (1997) methodology in scale development and measurement. The Nigerian market is the largest economy in Africa (Naidoo, 2020). Which provides justification for carrying out the study in Nigeria. Moreover, such a study can be replicable in any other country.
The Nigerian Communications Commission (NCC) (2020) reported that the Nigerian Global System for Mobile Communications (GSM) market is one of the fastest growing markets in the world contributing about 10% of the National GDP. However, the industry has been lagging other sectors in terms of customer satisfaction and loyalty as argued by Oghojafor, Ladipo, Ighomereho & Odunewu (2014). In their analysis of the sector, Sanusi, Ajilore &Oloyede (2014) identified the biggest players in the industry ('The Big Four") as -MTN Nigeria, Glo Mobile (Globacom), 9Mobile and Airtel Nigeria. Furthermore, researchers have studied the Nigerian telecommunications industry from a variety of marketing viewpoints including Advertising and Brand Loyalty (Sanusi, Ajilore & Oloyede, 2014); Corporate Communication and Brand Image (Onikoyi & Onikoyi, 2013); Customer Satisfaction and Loyalty (Oghojafor et al., 2014); Aggressive Marketing and Product Performance (Obasan, Ariyo & Soyebo, 2013), however, not much seems to have been done in the area of brand personality. This research will fill provide some insight to the perception of consumers within the Telecommunications industry to the Glo Mobile Brand which is owned by Globacom Telecommunications (Nig.) Limited. It will also develop a framework that can be replicated in other cultures. The Glo Mobile Brand has been chosen as it is the only brand out of the Big Four -MTN Networks Nigeria Limited, Airtel Nigeria Limited, Etisalat Networks Nigeria Limited and Globacom Telecommunications Nigeria Limited, which appears to be wholly Nigerian, owned and developed. It is hoped that this fact would have infused it with wholly Nigerian values and needs.
This study contributes to the brand personality literature in three ways, first it adapted Aaker's (1997) scale to measure brand personality in the Nigerian telecommunications market using the methodology identified by Avis (2012) to be the most stable in literature; second, it identified the traits that could be used in marketing communication in the telecommunications industry; and finally, it revealed the brand personality factors of the telecommunications industry. The rest of the paper is structured thus: After reviewing the extant literature, the methodology employed was explained. This was followed by results and findings. Subsequently, discussions were elaborated upon. The study concludes with managerial implication for telecommunications managers and marketing professionals.

Aaker's (1997) Dimensions of Brand Personality
Consumers associate with brands that reinforces their social identity and add values to their reputation. People tend to have social connections with groups and organisations, and in many instances, their relationship or association with brands is a function of personal identity, self-concept and quite importantly, social identification with groups. Consumers identify with brands that ensures a sense of belongingness especially among groups they belong to or aspire to belong (Chua, Lee, Kim & Han, 2017;Kim, Kim, & Holland, 2018). Consumers decisively choose brands and products that they perceive are consistent with their own image of selfidentity and congruent with their own personality (Kressman, Sirgy, Herrmann, Huber, Huber, & Lee, Lu & Xu, 2015). Understanding personalities of brands has become more crucial due to increase in customers' awareness about available alternatives (Chi, Pan, & Del Chiappa, 2018;Ekinci & Hosany, 2006). This was buttressed by Sheena and Naresh's (2012) assertion that a brand without personality has trouble gaining awareness and developing relationship with customers. Brand personality has been examined from various points of view since Aaker's seminal work on dimensions of brand personality. It was investigated as an independent variable by Wang, Yang, and Liu (2009) who studied the effects of brand personality on purchase intention and found that strong and positive productbrand personality can exert significant influence on consumer purchase intention and also as a dependent variable by Seimiene and Kamarauskaite (2014) who used a qualitative interview to study the effects of brand elements on brand personality. It has also been investigated on a variety of product categories. For instance, Merabet and Benhabib (2012) examined the brand personalities of cars and soft drinks while Pantin-Sohier, Decrop and Brée (2005) investigated the effects of changes in product packaging of mineral water and tinned coffee on customer perception of their brand personality. Brand personality has also been assessed in terms of places, states or countries such as Geuens, Weijters and De Wulf's (2009) study within the Belgian context and Hultman, Skarmeas, Oghazi and Beheshti's (2015) study of Las Vegas where they found that brand personality is a determinant not only of customer satisfaction and promoting, but also of repurchase intentions. Brand personality can also be used for differentiation within industries as shown by Carlson, Donovan and Cumisky (2009) in their study on the sports industry and Ekinci, Sirakaya-Turk and Preciado (2013) in their assessment of the tourism industry.
This suggests evidence that brand personality may be a useful tool in providing differentiation in the telecommunications industry. Avis (2012) presented a few models of brand personality developed using various methodology (qualitative and qualitative) noting that it is surprising to see the degree of variability amongst the models, when each model was developed with similar methods and concluded that Aaker's (1997) methodology appeared to be the most dominant and stable method of all existing methods. This is supported by several other researchers such as Usakli and Baloglu (2011);Geuens, Weijters & De Wulf (2009);Muniz (2012) who have developed scales of their own using Aaker's Methodology.

Scale Development and Brand Personality
Furthermore, Ivens and Valta (2012) prepared a taxonomy of Brand Personality research, and identified various groups of selected research built on Aaker's Methodology. These groups varied in the way brand personality featured in the research. They identified research on cross-country similarities and differences in consumer perception of Brand Personality, research on cross-industry similarities and differences in consumer perception of Brand Personality, as well as research linking the brand personality construct to various outcome variables. This suggests that the Brand Personality construct can be measured in various ways and for varying reasons according to Ivens and Valta (2012). Their findings also provide a basis for the development of a scale within the Nigerian context for a wholly Nigerian brand within the telecommunications industry as a search of three databases (EBSCOHost, Jstor and ScienceDirect) which are available to the researcher as at the time of the study do not reveal any previous research carried out within the Nigerian context. Even though literature search revealed that only two (ie. Opoku, Abratt & Pitt, 2006;Pitt, Opoku, Hultman, Abratt, & Spyropoulou, 2007) carried out within the African Context in general.
This study, therefore, attempts to bridge the lacunae of knowledge in the perception of Brand Personality in the Nigerian consumer market.

Method
The study employed a Quantitative Research design in line with Avis's (2012) assertion that quantitative methods are more stable when measuring brand personality.

Population and Sampling
The population for this study was consumers of the Nigerian telecommunication industry resident in Lagos State, which is a melting pot of all the Nigerian cultures due to its designation as Nigeria's economic focal point (About Lagos, n.d.). Klabi and Debabi (2011) as well as Čáslavová and Petráčková (2011) supported this category of subjects. In line with Aaker's (1997) use of non-probabilistic volunteer sampling 700 volunteers who were all students of the University of Lagos Distance Learning Institute completed copies of the questionnaire. Students were used as they are customers of the telecommunications industry which were immediately accessible to the researchers. in line with Aaker's (1997).

Instrumentation and Validity
The data was collected using a self-administered survey questionnaire, which was distributed to all sample subjects. The questionnaire was developed in line with Aaker's (1997) methodology using 6 academic scholars in the field of Business Administration to undergo free association tasks to develop 5 traits each peculiar to the Nigerian Cultural Context, which were added to Aaker's (1997) Brand Personality Scale. Redundancies and close synonyms were eliminated using two separate online thesauruses. The resulting list was then given to two senior lecturers teaching postgraduate courses in the Marketing Program of the University of Lagos for face and content validation and possible elimination of traits not relevant to the telecommunications industry. This process generated a list of 48 items, which were considered valid for the study. 39 of the original Aaker's 42-item scale) (small town, western and good-looking were not adjudged to describe the telecommunications industry) were retained and 9 new traits -Innovative, Indigenous, Efficient, Effective, Distinct, Competitive, Diligent, Responsive and Accessible were generated. A 5-point descriptive rating scale (5 = Very Descriptive, 1 = Not at all descriptive) was used to collect responses about the generated traits; this is in line with Aaker (1997) use of such a scale.

Data Analysis
The data was analysed using Exploratory Factor Analysis to extract the major dimensions while traits in each dimension were further Factor Analysed to determine the facets. This is in line with Aaker's (1997) methodology. Confirmatory Factor Analysis was used to validate findings, this method of analysis has also been used by Chi, Pan and Del Chiappa (2018) as well as Geuens, Weijters and De Wulf, (2009).

Demography of Respondents
The sex of respondents was evenly distributed with 50.1% of respondents being male and 49.9% being female. Age range of respondents also appeared normally distributed with 80% of the respondents being between 26 and 40 years old. Marital Status was also evenly spread with up to 51.3% being single and 46.9% being married.

Exploratory Factor Analysis
The Exploratory Factor Analysis was carried out using Principal Component Extraction with a VARIMAX Rotation and latent root criterion (eigenvalues >1) (Usakli & Baloglu, 2011), on SPSS 20 Software. All loadings less than 0.5 were suppressed as recommended by Hair, Anderson, Tatham and Black (2005), since factor loadings greater than .50 are considered necessary for practical significance. Some traits not loading up to 0.50 were dropped.

Source: SPSS Output
The results initially showed a 6-factor model, however, one of the factors only had one item and was subsequently dropped in line with Raubenheimer's (2004) recommendation of a minimum of 3 items per factor. Thus, a 5-factor model (See Figure 1), which explained about 66% of the Total Variance, which is within accepted levels was adopted. The Kaiser-Meyer-Olkin Measure of Sampling Adequacy was sufficient (0.969) suggesting that the Principal Component Analysis was appropriate to use on the data (Usakli & Baloglu, 2011). The Bartlett's Test of Sphericity (Table 2, p value < 0.01, Chi Square = 13676.488, df = 703) suggested that there existed enough correlations between the variables to run a factor analysis and that there was a good model fit (Aaker, 1997).

Reliability
The reliability test carried out using Cronbach's Alpha showed high reliability of each factor (Aaker, 1997). The researchers further carried out an ANOVA test using the F-Statistic, which was also found to be significant (p < 0.01) which corroborates the goodness of model fit. Individual item reliability also tested highly as the corrected item-total correlation showed high correlations all above 0.50. The findings of these tests suggest that the scale is reliable and also suggests good validity. All reliability results are presented in Table 3.

Validity of the Scale
Analysis to test the stability of the scale using confirmatory factor analysis suggested a there might not be good model fit with a significant chi square value of 2 = 2201.95, df = 688, p = 0.000 2/df = 3.20051163. However, according to Hooper, Coughlan and Mullen (2008) chi square can be very sensitive to large sample sizes, this is also supported by Kline (2005). This suggests the need for a look at other indices. RMSEA (RMSEA = 0.59) value was within the acceptable range of 0.50 and 0.80 (Hair et al, 2005). CFI (.909) was above Kline's (2005) (>.90) suggestion and SRMR (0.065) values were also less to 0.08, which was adjudged to be acceptable by Hu andBentler (1998, 1999). Thus in accordance with the criteria set by Hair et al (2005) and supported by Hooper, Coughlan and Mullen (2008), the convergent validity of the scale is confirmed. Figure 1 presents a graphical representation of the 5-Factor model. The 5-factor model had three factors that show significant similarity to three of Aaker's (1997) Dimensions. For instance, Aaker's Competence Dimension was well represented in the 6-Factor model with many of its original traits loading well for under one factor. Interestingly however, the reliable trait was removed while independent and upper-class appeared to load higher under this factor. This may be explained by the Nigerian belief of upscale labels as a sign of competence over lower scale brands. The Sincerity Dimension also loaded reasonably the same with only the down-to-earth trait being removed completely due to low loading. All other traits from the original Sincerity Dimension loaded highly in our 6-Factor model. Another similar factor was the Ruggedness Dimension, which loaded all original traits but also added the contemporary trait, which had previously loaded on the Excitement Dimension on Aaker's Model. Aaker's Excitement Dimension was seen to drop quite a lot of its traits due to low loading with traits like daring, trendy and up-to-date being dropped entirely while the independence trait loaded higher for the Competence Dimension and the contemporary trait loaded higher under the Ruggedness Dimension. An entirely new factor however emerged with all its traits loading highly (0.514 -0.700). This new factor we have named Innovative. This is because it is made up of traits that can connote innovation such as -distinct, effective, innovative, competitive, diligent, and indigenous. All these traits can be said to fit rather well with a brand for an organization that is supposed to have a high dependency on technological advancement such as one in the telecommunications industry in which our study belongs. This also lends credence to our claim of a good model fit.  This may prove valuable to marketers managing the brands of such organisations. Some of the traits of the Innovation factor such as diligent or effective can be used to create the impression of good quality service. Additionally, the scale can be used to measure how well the individual organisations are managing their brands in line with the factors revealed. Such measurement can guide the managers of such brands modifying the marketing process.
In Nigeria specifically, exploring the performance of Globacom Nigeria's brand personality can yield avenues for the organisation to modify it's marketing communication to better position it in the minds of the customers.

Conclusion
The aim of this study was to explore brand personality dimensions in the Nigerian context, by adapting Aaker's(1997) scale to the Nigerian telecommunications industry.
It also provides support for the use of Aaker's methodology in the measurement of brand personality. The study also revealed some of the traits that can be used in marketing communication by marketing professionals in the telecommunications industry such as Family-Oriented to connote that the brand has family values or Successful to show that the brand is acceptable to high income earners.
Furthermore, this study employed a confirmatory approach, making it possible to achieve a glimpse of the perception that consumers have of the Globacom Brand as a wholly Nigerian Brand, enabling the identification of the dimensions of brand personality which were labelled as Competence, Sincerity, Ruggedness, Innovative, Sophistication and Excitement.
Findings of this study made it possible to compare the Nigerian dimensions to those found in other contexts, which underlines the proposition that consumer symbols such as brands can carry different and sometimes varying meanings between cultures thus reconfirming culturally significant meanings of a given culture (Aaker et al., 2001). The findings support that of authors studying across various cultures (Glińska & Kilon, 2014;Hosany, Ekinci & Uysal, 2006;Sung and Tinkham, 2005).

Limitations and Further Research
This study was limited to only one brand within the Nigerian Telecommunications Industry due to the need to assess a brand that was wholly owned by Nigerians in order to see if such a brand perceived as wholly Nigerian could exhibit cultural inferences that would indicate a dimensional shift from Aaker's (1997) dimensions. Future research can focus on more brands within the industry to get a more generalized view of Brand Personality perceptions of the industry as a whole.
This study also concentrated on only one product category in accordance with Glińska & Kilon (2014) while researchers such as Aaker (1997) and Muniz (2012) measured Brand Personality across various product categories. Thus, future research can include other product categories within the Nigerian context.