THE APPLICATION OF MULTIVARIATE STATISTICAL ANALYSIS TO THE VALUATION OF DURABLE GOODS BRANDS

Dziechciarz-Duda, Marta; Król, Anna Article THE APPLICATION OF MULTIVARIATE STATISTICAL ANALYSIS TO THE VALUATION OF DURABLE GOODS BRANDS Statistics in Transition New Series Provided in Cooperation with: Polish Statistical Association Suggested Citation: Dziechciarz-Duda, Marta; Król, Anna (2017) : THE APPLICATION OF MULTIVARIATE STATISTICAL ANALYSIS TO THE VALUATION OF DURABLE GOODS BRANDS, Statistics in Transition New Series, ISSN 2450-0291, Exeley, New York, NY, Vol. 18, Iss. 1, pp. 75-90, https://doi.org/10.21307/stattrans-2016-058 This Version is available at: https://hdl.handle.net/10419/207846 Standard-Nutzungsbedingungen: Die Dokumente auf EconStor dürfen zu eigenen wissenschaftlichen Zwecken und zum Privatgebrauch gespeichert und kopiert werden. Sie dürfen die Dokumente nicht für öffentliche oder kommerzielle Zwecke vervielfältigen, öffentlich ausstellen, öffentlich zugänglich machen, vertreiben oder anderweitig nutzen. Sofern die Verfasser die Dokumente unter Open-Content-Lizenzen (insbesondere CC-Lizenzen) zur Verfügung gestellt haben sollten, gelten abweichend von diesen Nutzungsbedingungen die in der dort genannten Lizenz gewährten Nutzungsrechte. Terms of use: Documents in EconStor may be saved and copied for your personal and scholarly purposes. You are not to copy documents for public or commercial purposes, to exhibit the documents publicly, to make them publicly available on the internet, or to distribute or otherwise use the documents in public. If the documents have been made available under an Open Content Licence (especially Creative Commons Licences), you may exercise further usage rights as specified in the indicated licence. https://creativecommons.org/licenses/by/4.0/ STATISTICS IN TRANSITION new series, March 2017 75 STATISTICS IN TRANSITION new series, March 2017 Vol. 18, No. 1, pp. 75–90 THE APPLICATION OF MULTIVARIATE STATISTICAL ANALYSIS TO THE VALUATION OF DURABLE GOODS BRANDS Marta Dziechciarz-Duda 1,Anna Król 2 ABSTRACT Nowadays, due to changes in the market and new trends in consumer behaviours, intangible assets, such as brand, have gained fundamental importance. The more frequent conviction that a product with a well-known name is better than other products contributes to the case of replacing the price of a product by its brand name as the predominant factor in the purchase decision process. Thus, for many companies the strengthening of brand equity has become one of the key elements of marketing strategy. The main aim of this study is an attempt to improve the process of analysing the position and value of brands using selected multivariate statistical analysis methods (hedonic regression, multidimensional scaling, classification and linear ordination methods). In the conducted research the direct approach to the evaluation of the position of the brands for a selected ICT good – smartphones – have been applied. The measurement was performed on two levels: the product level, in which the prices of branded products were compared, and the consumer level, where the perception and attitudes of consumers towards the brands were studied. The analyses have been carried out on two sets of data, which enabled fuller and more comprehensible understanding of decision rules that guide consumers in choosing the brand. Key words: brand valuation, multivariate statistical analysis, durable goods. 1. Brand and its value In an era of fast technological development, when durable consumer goods are endowed in number of complex features, become more and more advanced and undergo rapid improvements, the consumers more frequently face the problem of a difficult choice between many variants of complicated commodities. This issue is particularly severe on the highly advanced durable goods markets (such as, for example, ICT goods), which on the one hand are violently expanding thus posing promising prospects for the future, and on the other hand are extremely changeable 1Wrocław University of Economics. E-mail: [email protected] 2Wrocław University of Economics. E-mail: [email protected] 76 Dziechciarz-Duda M., Król A.: The Application of Multivariate ... and, furthermore, constantly experience shortening of the life cycles of products. In such unstable conditions one of the ways to win over the confused customers, induce their purchase decision and gain their trust and future loyalty is to create and strengthen the product brand. Increasing the brand value is particularly important on the markets where it is extremely difficult to fully assess the quality of the product before the purchase, and moreover the frequency of the purchases is low in comparison to the rapid pace of technological development, so that it is very hard for the consumers to build on their own experience of the past. Therefore, the well build brand might become a kind of safety buffer for the customers. It is known that brand names can convey information about various aspects of the product, such as reputation, reliability, quality, and also are synonyms of certain prestige and even social status and identity. Thus, they might provide the means for the customers to reduce risk level involved in the buying process and increase information efficiency of the purchase (see Keller and Lehmann (2006)). The contemporary notion of brand name is understood as a much broader concept than originally comprehended. According to the classic definition formulated by the American Marketing Association (1960) brand name is „(...) a name, term, symbol, or design, or logo or a combination of them used to identify and differentiate a product or service from the competitor in the marketplace”. Thus, the main role of brand name in this early approach was to inform the potential customers about the product existence and discriminate it from other similar commodities. Nowadays the brand name notion encompasses much more. The brand represents broadly understood trust, as well as certain connection between the company and the consumer, and is extended to include representing certain quality level, introducing the image of prestige and social status, as well as building customers identity, among other things. Detailed discussion on modern brand definitions may be found, among others, in papers by de Chernatony and Dall’Olmo Riley (1998) and Maurya and Mishra (2012). Maurya and Mishra list 12 different aspects in which the brand have been regarded in the literature: brand as a logo, brand as a legal instrument, brand as a company, brand as a shorthand, brand as a risk reducer, brand as an identity system, brand as an image in consumer’s mind, brand as a value system, brand as a personality, brand as relationship, brand as adding value and brand as an evolving entity. This multitude of various approaches and issues accounted for in brand name analyses shows the complexity of the brand name evaluation problem. The presented approaches to brand definition can be related to the question of the brand equity assessment by assigning them to two major trends present in the literature: cognitive psychology and informational economics (cf. Aaker (1991), Erdem and Swait (1998), Erdem et al. (1999), Baltas and Saridakis (2010)). STATISTICS IN TRANSITION new series, March 2017 77 The authors of the cognitive psychology concept define brand equity as the effect that the knowledge of the brand and its features has on the response and revealed preferences of the consumer (Aaker (1991); Keller (1993)). Crucial elements in the process of brand valuation include brand awareness (the degree to which consumer precisely recognizes the brand and associates it with the specific product), brand associations (the extent to which a particular brand calls to mind the attributes of a general product category), brand loyalty (the extent of the faithfulness of consumer to a particular brand), brand perceived quality (consumer’s opinion of a brand’s ability to fulfill expectations) and other proprietary brand assets (patents, trademarks, and channel relationships that prevent the competitors from eroding a customer base). The stronger and more positive those factors, the higher brand price premium. Informational economics approach somewhat differently presents the issue of brand equity (Erdem and Swait (1998)). This concept emphasizes the incompleteness of signals coming from the market, which forces the consumer to make decisions in a situation of partial misinformation. In this context, the brand is understood as a source of information on the quality of goods, as well as a way to reduce purchase risks and costs of searching for the right product on the market. Brand price premium reflect the return on the investment in creating and strengthening the brand made by the company. The value of the brand name and its strength affect the mechanisms that occur during the process of purchasing goods by the consumer. Cobb-Walgren et al. (1995) showed a significant positive correlation between brand equity and consumer’s brand preference. Moreover, an increase in brand value translates into higher purchase intentions and significantly influences the final purchase decision made by the customer. The more frequent conviction that a product with a wellknown name is better than other products contributes to the case of replacing the price of a product by its brand name as the predominant factor in the purchase decision process. Thus, for many companies the strengthening of brand equity has become one of the key elements of marketing strategy. The relationship and interactions between the brand and the actual purchase is schematically shown in Figure 1. Brand is especially important on durable goods market, because of the specificity of the market and durable goods features. Durables typically cost substantially more than nondurable products and thus entail greater financial risk for consumers. Even though consumers are in the market for a short period and after purchase stay away for a long time, they undoubtedly spend a substantial amount of money in that period. The individual consumer is present on the market intermittently so it is difficult to evaluate the quality of the product before the purchase. Thus, customers 78 Dziechciarz-Duda M., Król A.: The Application of Multivariate ... Figure 1: Brand equity in the purchase process (Source: based on Cobb-Walgren et al. (1995)) often rely on the brand name reputation in order to reduce the perceived risk and search costs. Since in many cases brand is one of the most valuable intangible assets of the company, there is a necessity of developing accurate measures of brand equity. Brand equity measurement approaches include three main areas - one financial, and two more closely related to the marketing concepts (Keller and Lehmann (2006)): •financial-based approach, which focuses on the monetary or financial value of the brand in the marketplace; •consumer-based brand equity, which involves the value added to a product or service by consumers’ associations and perceptions of a brand name; •product-market-based approach. It is worth noticing that brand in financial terms, as an asset of the company, is dependent on such factors as brand loyalty, brand awareness, perception of brand quality, brand associations (cf. Aaker (1991), which in fact are closely linked to the area of marketing in the company. Thus, financial and marketing approaches to STATISTICS IN TRANSITION new series, March 2017 79 brand valuation are closely related and the measurement of brand value is a complex multi-level issue. This paper focuses mainly on methods of brand valuation and brand equity measurement from marketing perspectives, and thus consumer-based and product-market-based measurement approaches are analysed. Measuring brand as a financial resource - analyses of the company’s assets and all related aspects in the accounting areas of the company’s operations are not the subject of this study. Further aspects of the presented research concerning consumer preferences may be found in a paper by Dziechciarz-Duda and Król (2016). 2. Marketing approaches to brand equity measurement From the marketing point of view both the product-market-based and consumerbased approaches are applied to brand valuation. In the product-market-based approach a strong brand is one that increases the effectiveness of advertising, differentiates the product from the competition and facilitates prospective expansion into new market segments. In this context, the brand value can be understood as an additional bonus due to the ownership of a particular brand, which is absent in the case of an equivalent, comparable product that does not have a strong brand. A common tool for brand valuation from the product-market perspective is the analysis of the price premium as an indicator of brand equity (Erdem et al. (1999); Netemeyer et al. (2004)). The price premium is defined as the amount of money the consumer is willing to pay for the product of preferred brand in comparison with other products having similar characteristics but different brand names (Kamakura and Russell (1993)). The price premium is considered as a valuable and comprehensive tool for brand equity measurement, which is in addition relatively easy to calculate and straightforward in interpretation. However, there are some critical views in the literature (Ailawadi et al. (2002)) indicating that the models used to calculate the price premium (e.g. hedonic models) do not directly capture the essential elements of the marketing mix (e.g. advertising). Nevertheless, it is assumed that the forces of supply and demand, as well as other market mechanisms take into account all these aspects indirectly. From that point of view hedonic modeling is a useful statistical tool in the analysis of brand value at the product level. Numerous studies devoted to issues of brand valuation focus on the consumerbased brand equity. This approach to brand value analysis is considered especially valuable because it directly examines the consumer behavior, which provides the basis for formulating marketing strategies (Keller (1993)). According to this approach the subject of analysis is the type of associations with the brand that comes to the mind of the customer. Moreover, it is assumed that the unique, positive and strong associations influence purchasing decisions of potential customers. Brand 80 Dziechciarz-Duda M., Król A.: The Application of Multivariate ... value at the consumer level is connected with the knowledge of the brand and types of reactions toward the brand, which can be expressed in five areas (Keller and Lehmann (2006)): awareness (ranging from recognition to recall), associations (encompassing tangible and intangible product considerations), attitude (ranging from acceptability to attraction), attachment (ranging from loyalty to addiction) and activity (including purchase and consumption frequency). As mentioned before, the purchase process is a complex and multi-faceted operation (Figure 1), and the preferences of consumers (including the preference towards the brand), purchase intentions and the actual purchasing decision form composite relations. This raises the question whether declarative behaviours of the consumers always translate directly into actual purchasing decisions. As numerous research on the process of buying suggests, the declared preference towards the brand is highly correlated with customers’ choices, but there are markets where this relationship is weaker. Undoubtedly, consumer durables market falls into that category. On durable goods markets, only few percent of the declared intention of buying is actually implemented (cf. Dziechciarz (2008), Cobb-Walgren et al. (1995), Morwitz and Schmittlein (1992)), which pose a particular challenge for analyses of consumer future reactions to the elements of marketing mix. Given the above concerns, it seems reasonable to confront the results of studies assessing the brand at the consumer level with the objective study of the current market offer (product level approach). The combined application of these approaches to brand valuation and juxtaposition of the results of the brand analyses from the consumer and the market perspectives provides added value in the form of additional information and areas open to interpretation. On the basis of the consumer preferences research one can obtain information about subjective associations with the brand. Such a study is extremely valuable because it provides insights into individual attitudes of potential customers. The obvious disadvantage of this approach lies in its subjectivity. In addition, there are many signs saying that declarative behaviours do not always translate into actual purchasing decisions. Therefore, the authors believe that there is a need to confront the subjective brand valuations provided by the consumers with the information from the market (i.e. the calculated price premium based on existing market offer). In the following part of the paper an empirical example illustrating the possible applications of multivariate statistical methods for the brand equity measurement for selected durable good - smartphone - will be presented. The estimated parameters from an econometric model based on the data showing the market offer will be compared with assessments of brands from a study of consumer preferences towards brands. Thus, on the product-market level of brand valuation, hedonic regression STATISTICS IN TRANSITION new series, March 2017 81 model will be used. In turn, on the consumer-base level multidimensional scaling, linear ordering and hierarchical classification methods will be applied. 3. Data sets The presented analyses have been carried out on two separate sets of data. On the product level the prices and significant characteristics of smartphones (including brands) were collected from price lists available on an Internet website of one of the biggest in Poland price comparison service provider, whereas on the consumer level the perception and attitudes of consumers towards the smartphone brands were measured using a specifically designed on-line survey. This two-sided approach enabled fuller and more comprehensible understanding of decision rules that guide consumers in choosing the brand of a smartphone. 3.1. Data set for hedonic analysis Database used in this part of the study have been created using a tool for data collection from web pages created by the authors. The data originate from Polish price comparison service providers. The data set comprises 910 smartphones of 27 different brands offered in Internet shops in Poland in February 2015. Each offer is described by price (PRICE [PLN]) and the following smartphones’ characteristics: SCREEN – screen size [inch], STORAGE – internal storage [GB] and CAMERA – camera resolution [Mpix]. Moreover, the following dummy variables (take value 1 if the feature is present and 0 otherwise) were used: LTE, GPS, ANDROID, as well as dummies representing the brands. 3.2. On–line survey data set The data from on-line survey was gathered in February 2015 among the students of Wrocław University of Economics. The questionnaire was focused on measuring consumers’ preferences towards smartphone characteristics and possible applications of the device. The sample consisted of 451 respondents selected based on their accessibility and proximity (convenience sampling). The respondents were expected to asses popular brands of a smartphone, its important characteristics, as well as the common usage patterns of the device. Thus, in order to evaluate the analyzed criteria, each respondent have created his individual rankings of the brand names, the criteria, that would be taken into account while purchasing a smartphone and the common usage patterns of a smartphone. Moreover, the respondents have assessed the brand names of smartphones by assigning 82 Dziechciarz-Duda M., Król A.: The Application of Multivariate ... the rate (on 5-point scale: 5 - highest rate, 1 - lowest rate) to five brand attributes: prestige, design, modernity, support and reliability. 4. Product level approach 4.1. Hedonic modelling The foundations of hedonic methods are formed by the so-called hedonic hypothesis, which states that heterogeneous commodities are characterized by a set of relatively homogeneous attributes (characteristics) relevant both from the point of view of the customer and the producer (Brachinger (2002); Dziechciarz (2004)). The relationship between the price of commodity (PRICE) and the set of its characteristics (X) described by certain function fis called hedonic regression and may be described in the following general notation: PRICE =f(X;β;ε),(1) where εis the error term of the model. The estimate of the vector of parameters, obtained by estimation of the correctly specified hedonic regression model using data set, allows to calculate the prices of individual characteristics of the given good (so-called hedonic prices or implicit prices). It is assumed that the consumers derive utility from goods attributes, and therefore the hedonic prices reflect the willingness to pay for certain levels of attributes. In that context the hedonic model may be used to measure the brand price premium. 4.2. Estimation results The results of estimation of the hedonic model for smartphone prices are presented in Table 1. The best functional form turned out to be the model with dependent variable transformed to logarithm (lnPRICE), and some of the independent variables in logarithmic transformation (SCREEN, CAMERA), and quadratic transformation (STORAGE). Due to heteroskedasticity of the error term weighted least squares method proposed by White (1980) was applied for model estimation. Out of 27 brand names present in the dataset 18 were statistically significant (in the parentheses number of models representing given brand is given): ACER (11), ALCATEL (24), ALIGATOR (6), APPLE (40), ARCHOS (7), ASUS (6), BLACKBERRY (16), GIGABYTE (22), HTC (62), HUAWEI (37), LG (91), MOTOROLA (17), NOKIA (94), PRESTIGIO (41), SAMSUNG (217), SONY (87), ZOPO (15), ZTE (7). The remaining 9 brands formed the reference group: BE (9), GOCLEVER (25), KRUGER&MATZ (15), MANTA (6), MEDIA-TECH (10), MYPHONE (14), STATISTICS IN TRANSITION new series, March 2017 89 DZIECHCIARZ, J., (2004). Regresja hedoniczna. Próba wskazania obszarów stosowalno´ sci. In Przestrzenno-czasowe modelowanie i prognozowanie zjawisk gospodarczych. A. Zelia´ s (ed.), Kraków: Wyd. AE, 163–175. DZIECHCIARZ, M., (2008). Deklaracje intencji zakupu w analizie wyposa˙ zenia gospodarstw domowych w innowacyjne dobra trwałego u˙ zytku. In Zastosowania bada´n marketingowych w procesie tworzenia nowych produktów (cena, opakowanie, znak towarowy), S. Kaczmarczyk, M. Schulz (eds.), Toru´ n: TNOIK, 157–166. DZIECHCIARZ-DUDA, M., KRÓL, A., (2016). The Analysis of Consumers’ Preferences with the Application of Multivariate Models: Hedonic Regression and Multidimensional Scaling. Archives of Data Science, under review. ERDEM, T., SWAIT, J., (1998). Brand Equity as a Signalling Phenomenon. Journal of Consumer Psychology, 7(2), 131–157. ERDEM, T., SWAIT, J., BRONIARCZYK, S., CHAKRAVARTI, D., KAPFERER, J. N., KEANE, M., ROBERTS, J., STEENKAMP, J-B. E. M., ZETTELMEYER, F., (1999). Brand Equity, Consumer Learning and Choice. Marketing Letters, 10(3), 301–318. HELLWIG, Z., (1968), Zastosowanie metody taksonomicznej do typologicznego podziału krajów ze wzgl˛edu na poziom ich rozwoju oraz zasoby i struktur˛e wykwalifikowanych kadr, Przegl ˛ad Statystyczny, 4, 307–326. KAMAKURA, W. A., Russell, G. J., (1993). Measuring Brand Value with Scanner Data. International Journal of Research in Marketing, 10(1), 9–22. KELLER, K. L., (1993). Conceptualizing, Measuring, and Managing Customerbased Brand Equity. Journal of Marketing, 57(1); 1–22. KELLER, K. L., LEHMANN, D. R., (2006). Brands and Branding: Research Findings and Future Priorities. Marketing Science, 25(6), 740–759. MAURYA, U. K., MISHRA, P., (2012). What Is a Brand? A Perspective on Brand Meaning. European Journal of Business and Management, 4(3), 122–134. MORWITZ, V., SCHMITTLEIN, D., (1992). Using segmentation to improve sales forecasts based on purchase intent: which „intenders” actually buy? Journal of Marketing Research, 29(4), 391–405. 90 Dziechciarz-Duda M., Król A.: The Application of Multivariate ... NETEMEYER, R. G., KRISHNAN, B., PULLING, C., WANG, G., YAGCI, M., DEAN, D., RICKS, J., WIRTH, F., (2004). Developing and Validating Measures of Facets of Customer-based Brand Equity. Journal of Business Research, 57 (2), 209–224. WALESIAK, M., GATNAR, E. (2004). Metody statystycznej analizy wielowymiarowej w badaniach marketingowych, Wrocław: Wyd. AE. WHITE, H., (1980). A Heteroskedasticity-Consistent Covariance Matrix Estimator and a Direct Test for Heteroskedasticity. Econometrica, 48(4), 817–838.