This article was downloaded by: [Universidad Pública de Navarra Biblioteca] On: 24 September 2014, At: 02:36 Publisher: Routledge Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered office: Mortimer House, 37-41 Mortimer Street, London W1T 3JH, UK
The International Review of Retail, Distribution and Consumer Research Publication details, including instructions for authors and subscription information: http://www.tandfonline.com/loi/rirr20
The use of consumer's price information search behaviour for pricing differentiation in retailing Carmen Berne , Jose M. Mugica , Marta Pedraja & Pilar Rivera Published online: 15 Apr 2011.
To cite this article: Carmen Berne , Jose M. Mugica , Marta Pedraja & Pilar Rivera (1999) The use of consumer's price information search behaviour for pricing differentiation in retailing, The International Review of Retail, Distribution and Consumer Research, 9:2, 127-146, DOI: 10.1080/095939699342606 To link to this article: http://dx.doi.org/10.1080/095939699342606
PLEASE SCROLL DOWN FOR ARTICLE Taylor & Francis makes every effort to ensure the accuracy of all the information (the “Content”) contained in the publications on our platform. However, Taylor & Francis, our agents, and our licensors make no representations or warranties whatsoever as
Downloaded by [Universidad Pública de Navarra - Biblioteca] at 02:36 24 September 2014
to the accuracy, completeness, or suitability for any purpose of the Content. Any opinions and views expressed in this publication are the opinions and views of the authors, and are not the views of or endorsed by Taylor & Francis. The accuracy of the Content should not be relied upon and should be independently verified with primary sources of information. Taylor and Francis shall not be liable for any losses, actions, claims, proceedings, demands, costs, expenses, damages, and other liabilities whatsoever or howsoever caused arising directly or indirectly in connection with, in relation to or arising out of the use of the Content. This article may be used for research, teaching, and private study purposes. Any substantial or systematic reproduction, redistribution, reselling, loan, sub-licensing, systematic supply, or distribution in any form to anyone is expressly forbidden. Terms & Conditions of access and use can be found at http:// www.tandfonline.com/page/terms-and-conditions
Downloaded by [Universidad Pública de Navarra - Biblioteca] at 02:36 24 September 2014
The International Review of Retail, Distribution and Consumer Research 9:2 April 1999 127–146
The use of consumer’s price information search behaviour for pricing differentiation in retailing Carmen Bern´e, Jose M. M´ugica, Marta Pedraja and Pilar Rivera Abstract The purpose of this paper is to analyse the potential of price-information-seeking behaviour (PISB) as a basis for pricing differentiation in the retail grocery market. Based on a model developed by Urbany et al. (1996), we analyse the presence of consumer clusters showing differences in their PISB and the possibility of identifying a particular pro le for each of the clusters. The cluster analysis identi ed two clearly separated clusters: high and low intensity price-information seekers. Based on this clustering, a discriminant analysis revealed that some of the independent variables proved to be good descriptors of the consumer pro le of the clusters. The joint consideration of the clusters and their pro les allows us to provide some managerial implications for retailers.
Keywords Retail, pricing, prices, search, segmentation Introduction Driven by pro t-optimization goals, retail companies frequently adopt pricediscrimination decisions across their stores; this interstore price dispersion Carmen Bern´e, Business Department, Universidad de Zaragoza, C/Pedro Cerbuna 12, 50009 Zaragoza, Spain (e-mail: cberne@posta,unizar.es). Jose M. Mugica, ´ Professor of Marketing, Departamento de Gestion de Empresas, Universidad Publica de Navarra, 31006 Pamplona, Spain (tel: (34) 4816 9400; fax: (34) 4816 9404; e-mail:
[email protected]). Marta Pedraja, Business Department, Universidad de Zaragoza, C/Pedro Cerbuna 12, 50009 Zaragoza, Spain. Pilar Rivera, Business Department, Universidad de Zaragoza, C/Pedro Cerbuna 12, 50009 Zaragoza, Spain. Copyright © Routledge 1999 0959–3969
Downloaded by [Universidad Pública de Navarra - Biblioteca] at 02:36 24 September 2014
128
The International Review of Retail, Distribution and Consumer Research
usually re ects differences among the demand functions of the customer bases (Hoch et al. 1995). Differences in the price elasticity then become a primary cornerstone of their pricing decisions. However, pricing is a more complex decision; retailers need not only to x a certain price but also to communicate it to their target customers. The mix of price xing and communication results in varied policies, such as hi-lo pricing or every-day-low-prices. As shown in a recently published work, price discrimination may be also made effective through the use of merchandising (Dhar and Hoch 1996). One alternative in differential pricing is to segment the customer base according to their information-seeking behaviour about prices. The information that consumers have about retail prices will have a certain impact on the major behavioural dimensions which retailers try to in uence: store selection (where), product/brand choices (what), quantity (how much) and timing (when). Then, if consumers in a particular market show different patterns of price-informationseeking behaviour (PISB), retailers may guide their pricing decisions, whether by focusing on a certain segment or by serving the market with two or more differentiated pricing policies. The retail grocery market is an interesting ground to analyse differences in PISB patterns. Consumers supposedly utilize a very limited amount of information in their grocery shopping process. A rst explanation for this lack of search has been provided by economists (Stigler 1961), who maintain that consumers follow cost-bene t criteria in deciding the amount of information to search for. As the bene t potential of information search is limited primarily by the price level, low-price products such as non-durables would require less search for information. But, from a marketing perspective, the acquisition of information prior to the purchase is a more complex problem: it is not just the amount of search but, also, the information sources utilized, the heuristics developed to manage overload information problems and the inferences made under incomplete information. As noted by Urbany et al. (1991), consumers may have good reasons for varying their propensity to search for prices in different retail stores, but the discipline of marketing has experienced some dif culty in both understanding and explaining the most elemental characteristics of consumer search in markets in which price search is an ongoing activity. In general, there has been an overestimate of the proportion of consumers who actively search for prices and respond to promotions. This observation may be due to the fact that there are no studies which adequately explain PISB in, for example, day-to-day shopping product categories, since the tendency has been to examine relatively small sets of predictors and to use dependent measures that do not directly re ect the search for price information (Lichtenstein et al. 1993). The purpose of this paper is to analyse the potential of PISB as a basis for pricing differentiation in the retail grocery market. To achieve this objective we will adapt the model used by Urbany et al. (1996) to explore a very similar set of activities included in price-information seeking. Based on this model, we will analyse the presence of consumer clusters showing differences in their PISB and the possibility of identifying a particular pro le for each of the clusters. Finally, the discussion of the results will lead us to provide some managerial implications for pricing differentiation in the retail grocery market.
Bern´e et al.: The use of consumer’s price information search behaviour
129
Downloaded by [Universidad Pública de Navarra - Biblioteca] at 02:36 24 September 2014
Price-information-seeking behaviour Generally, the information which consumers need in order to assess the alternatives that are available in the market can be obtained at two levels, namely internal and external search (Beales et al. 1981). The former involves the initial search being carried out by the individual, often subconsciously, drawing on past experiences and learning structures. If the result of this search is not suf cient for decision-making purposes, then the consumer moves onto the second level, making use of external information sources. This external information-seeking decision will be based on how the individual envisages the return to be obtained as opposed to the costs (monetary and non-monetary) to be borne in his/her search. Some studies have focused on the antecedents and factors explaining the effort and time devoted to the activities involved in search (Kleimenhagen 1966–7; Newman and Staelin 1971; Kiel and Layton 1981; McLelland and Turner 1983). Other studies have analysed a great variety of aspects of external information seeking. To name a few: the number of establishments visited (Newman and Lockeman 1975; Claxton et al. 1974; Midgley 1983); the number of brands assessed (Dommermuth 1965; Furse et al. 1984); the information resources used (Newman and Staelin 1972), including advertisements (Bucklin 1965; Udell 1966; Thorelli 1971; Kiel and Layton 1981) and word-of-mouth (Udell 1966; Thorelli 1971; Kiel and Layton 1981; Mu´ gica and Yag¨ue 1993); or the individual’s belief on how a particular market behaves (Duncan and Olshavsky 1982). Very recently, Grewal et al. (1998) have provided an additional perspective by exploring this issue in terms of behavioural intentions rather than past or present behaviour. To assess the relative importance of these activities in PISB for non-durables, Table 1 provides the self-reported measures found in a survey for the different information sources used by consumers. About half of the respondents reported no use of external information sources, showing a predictable pattern for this type of low-price item. But more interesting for our research purposes is that, among the different external sources used, consumers relied more on those mediated by retailers. Actually, more than half of the sample said that they looked for price information in mailings from retailers and in the different instore information devices. In this direction, the ability of retailers to in uence price perception by customers is con rmed and the PISB becomes then a potential criterion for effective price differentiation. The model and the hypotheses The model The model applied in this empirical study is an extension of the one used by Urbany et al. (1996), to analyse price search in the retail grocery market. The model speci es that the overall behaviour concerned in seeking price information
130
The International Review of Retail, Distribution and Consumer Research
Table 1 Self-reported measures of the use of information sources n 1 Use external sources
%
n
Yes
%
Total n
Detergent Milk
186 226
(42) (51)
259 217
(58) (49)
445 443
Detergent Milk
288 310
(65) (70)
158 133
(35) (30)
446 443
Detergent Milk
396 389
(89) (88)
50 54
(11) (12)
446 443
Detergent Milk
204 243
(46) (55)
242 200
(54) (45)
446 443
Media ads
Detergent Milk
305 335
(68) (76)
141 108
(32) (24)
446 443
In-store
Detergent Milk
182 224
(41) (51)
264 219
(59) (49)
446 443
Detergent 270 Milk 301 Source: adapted from Mugica ´ and Yagu¨ e (1993).
(61) (68)
176 142
(39) (32)
446 443
2 Type of source: 2.1 Interpersonal sources
Downloaded by [Universidad Pública de Navarra - Biblioteca] at 02:36 24 September 2014
No
2.2 Neutral sources 2.3 Retailer sources Retailer mailing
3 Visits to stores
is explained by ve different sets of factors: PISB 5
f ((Economic Returns:1 PPD, 1 BC); (Search Costs: 2 TIME, 2 MOB, DIFF, 2 YC); (Human Capital: 1 K, 2 IS, 1 TM); (Demographics: 1 AGE, 1 EDU); (Psychosocial Returns: 1 MM, 1 SE)) 2
where, Economic Returns : Search Costs :
PPD 5 perceived price dispersion; TIME 5 time constraints;
DIFF 5 dif culty of store comparison Human Capital : K 5 market knowledge; TM 5 time management skills Demographics: AGE 5 age; Psychosocial Returns: MM 5 market maven motivation;
BC 5 budget constraints MOB 5 mobility constraints; YC 5 presence of young children IS 5 investment search; EDU 5 education SE 5 shopping enjoyment
Moreover, Urbany et al. distinguish two large components of the PISB: the activities involved in price comparisons and those targeting the search for price specials. PISB: Price-Comparison-Seeking Behaviour (PCSB) 1 Price-Specials-Seeking Behaviour (PSSB)
Downloaded by [Universidad Pública de Navarra - Biblioteca] at 02:36 24 September 2014
Bern´e et al.: The use of consumer’s price information search behaviour
131
As far as the components of the model were concerned, i.e. the collection of items included in the dependent variable and in the different sets of factors, our study included only some minor modi cations required to adapt the questionnaire to the Spanish grocery market (Table 2). The inclusion of these modi cations does not affect the initial premises of the model concerning the overall price-information-seeking behaviour. These minor modi cations might be summarized as follows; rst, given the importance of specialized foodstores (butchers, sh and sea-food, and frozen products), three more items were included in the PCSB to re ect price comparisons at these types of stores. Second, the PSSB was slightly extended to cover not only price specials but prices in general and, by adding another information source, price enquiry in the store. The second set of activities concern information sources mostly under the control of retailers. When consumers rely on these it is because they are conceding them some credibility or because they cannot afford to get involved in more costly price-search activities. If the pro les of consumers using one or another type of PISB more intensively are different, then there would be a good ground for effective price differentiation policy between segments. Model hypotheses The model is based on the premises of some basic hypotheses which relate PISB to ve kinds of potentially determinant factors: economic returns, search costs, human capital, demographic characteristics and psychosocial returns (Figure 1). The role of each of these ve factors is discussed in the following text including the hypotheses describing the expected relationships included in Figure 1. Economic Returns Within the purchasing process, and considering the price variable, one of the factors faced by consumers is that a particular product might be sold at different prices, depending on the point of sale,1 the time and the terms of payment. In fact, price dispersion across stores and along time is the origin of most of the activities included in PISB. Economic returns deriving from PISB are identi ed with the acquisition of the product at a lower price, so it is expected that consumers will increase their search for prices when perceiving larger ranges in the distribution of prices (Urbany 1986; Bucklin 1969). But, also, PISB is a function of the impact that the cost of a product may have on the budget of one individual or household (Stigler 1961). That is, the economic returns have to be measured in relative terms, framed within the budget constraints of the consumer. H1:
The greater the economic returns perceived in purchasing by a consumer, the greater the probability of being an intensive priceinformation seeker.
H1.1: The greater the price dispersion perceived among stores, the greater the probability of being an intensive price information–seeker. H1.2: The greater the budgetary constraints, the greater the probability of being an intensive price-information seeker.
132
The International Review of Retail, Distribution and Consumer Research
Table 2 Independent and dependent variable Factor Economic returns
Subfactor
Items
Perceived price c dispersion c
A cart full of the same groceries bought from each of my local grocery stores will cost about the same (PPD1) Some grocery stores in Zaragoza have a lot lower prices than others (PPD2) The price of meats and produce varies a lot between Zaragoza grocery stores (PPD3) The price of individual items often varies a lot between stores (PRICEDIF) I frequently have problems making ends meet (PBC1) My budgeting is always tight (BUDGET) I often have to spend more money than I have available (PBC2) Intervals of income (IPF)
Downloaded by [Universidad Pública de Navarra - Biblioteca] at 02:36 24 September 2014
c c Perceived budget constraints c c
c
Search costs
Income per family c
Perceived time constraints c c
I am time poor, I never have enough time (PTC1) Everything I do is rushed (TIME) Hours per week paid employment (PTC2) Number of children less than 8 years of age (CHILDREN) It is very dif cult to compare the prices of grocery stores (PDSC) It is very dif cult to compare the quality of products between grocery stores (DIFFICULTY) I have a lot of energy to do things (MC1) My health restricts my activities (HEALTH) I do have reliable transportation to get out and about (MOBILITY) To have to use transportation to shopping is not a problem for me (MC2) c c
Perceived dif culty of store comparison Mobility constraints c
c
c
c
c
c Human capital
Market knowledge c
c
c c c c
Investment search
c
c
c
I know a lot about Zaragoza grocery stores I know which stores have the best prices I know which stores have the best price specials I know which grocery stores have the best meat department I know which grocery stores have the best sh department I know which grocery stores have the best greengrocery department I shop back and forth between several different stores before choosing where I now do most of my grocery shopping I compare the prices of different stores before nally deciding where to do most of my grocery shopping I made an extra effort in the beginning to learn about different stores so as to simplify the grocery shopping I do now
Bern´e et al.: The use of consumer’s price information search behaviour
133
Table 2 Continued
Downloaded by [Universidad Pública de Navarra - Biblioteca] at 02:36 24 September 2014
Factor
Subfactor Perceived time management skills c
Demographic characteristics
Age Education c
Psychosocial returns
Market mavens c
Items
c
I am very good at organizing and scheduling acitivities I am very good at time management Age range Levels of education c
c
c c c c c
Shopping enjoyment c
c
c c c
Dependent variables PCSB
I like introducing new brands and products I like helping people by providing them with information about many kinds of products I like it when people ask me for information about products, places to shop or sales I like it when someone asks me where to get the best buy on several types of products I think of myself as a good source of information for other people when it comes to new products or sales I know a lot of different products, stores and sales and I like sharing this information I like giving people information about prices I view grocery shopping in a positive way I enjoy grocery shopping Shopping is amusing Shopping is funny Shopping is a pleasure
I compare the prices of different stores (PI) I often compare the prices of fruit and vegetables at two or more grocery stores (P2) I often PCSB the prices of meat at two or more grocery stores (P3) I often PCSB the prices of sh at two or more grocery stores (P4) I often PCSB the prices of frozen food at two or more grocery stores (P5) How often do you PCSB the speci c prices of grocery stores? (P6)
c c c c c c
INFOR c
c
c c c c
I take into account the information about prices of substitute products or specials showed in the store, before shopping (P7) Regularly read ads or iers to PCSB prices and/or check price specials (P8) I decide to visit some stores before shopping (P9) Decide where to shop based upon ads/ iers I receive at home (P10) Often talk to friends about price specials before shopping (P11) Regularly shop for the price specials at one store and then the price specials at another store (P12)
Search costs PISB activities involve confronting a series of costs or time constraints, and undergoing a decision-making process based on personal perceptions of the complexity of the search activity, as well as possible physical
134
The International Review of Retail, Distribution and Consumer Research
constraints on the consumer. These costs and restrictions are unevenly distributed and perceived among consumers, so it is expected that some of the variance in PISB could be explained by differences in search costs. The general hypothesis and the subhypotheses proposed are as follows:
Downloaded by [Universidad Pública de Navarra - Biblioteca] at 02:36 24 September 2014
H2:
The higher the costs perceived in information seeking, the less the probability of being an intensive price-information seeker.
H2.1: The less time available, the less the probability of being an intensive price-information seeker. H2.2: The greater the perception of dif culty in comparison, the less the probability of being an intensive price-information seeker. H2.3: The greater the constraint on mobility, the less the probability of being an intensive price information seeker. Human capital The human capital dimension refers basically to the knowledge accumulated by the individual over time. In this study, human capital is measured in the same way as in Urbany et al. (1996): a composite of (1) the degree of present knowledge of the competing stores, (2) the previous investment made in price search and (3) the time-management skills. The impact of knowledge on price search is not clear as there are two implicit con icting in uences: knowledgeable consumers are less dependent on external search but are also more motivated to seek information. However, a stronger association with the latter is expected, so a positive relationship is hypothesized. Previous investments in price search are expected to have a negative impact as consumers with an a priori knowledge of prices and their distribution tend to limit their future searching, on the supposition that prices are correlated over time (Stigler 1961). Finally, consumers self-reporting better time-management skills will be more determined to get involved in price-information seeking as they are more con dent of getting positive results. H3.1: The greater the knowledge of local stores, the greater the probability of being an intensive price-information seeker. H3.2: The greater the previous investment in search, the less the probability of being an intensive price information seeker. H3.3: The greater the time-management skills, the greater the probability of being an intensive price-information seeker. Demographics The determining role of demographic characteristics is usually explored in consumer behaviour since they are an excellent basis for effective differentiation. However, very often their estimated effects in consumer behaviour models are not signi cant or are contradictory. Demographics, such as age, sex or education, include many implications operating in opposite directions. The role of these variables when modelling price-information seeking is not an exception: older consumers are more experienced and would need less external price search but they also have fewer competing activities; educated consumers might be more skilful and ef cient in information seeking, resulting in a reduction of the search effort intensity, but they also have the potential of being
Bern´e et al.: The use of consumer’s price information search behaviour
135
more search extensive using a wider variety of information sources. As noted by Urbany et al. (1996), the impact of demographics on price-search activities is uncertain, but they observed signi cant correlation of age and education with other independent variables. In a further analysis of the age variable they found that older consumers reported more extensive price search. In this direction, our initial premise is that both age and education have a positive effect.
Downloaded by [Universidad Pública de Navarra - Biblioteca] at 02:36 24 September 2014
H4:
The probability of being an intensive price-information seeker is higher for older and more educated consumers.
Psychosocial returns Psychosocial returns are the enjoyment and self-esteem bene ts associated respectively with the activities and outcomes of the priceinformation-seeking effort. In this context they are associated, rst, with how much enjoyment consumers get when seeking for price information. This attitude, which may be found in those consumers who enjoy shopping in general, re ects an incremental utility that is primarily associated with the possibility of nding lower prices. H5.1: The greater the self-reported enjoyment when shopping, the greater the probability of being an intensive price-information seeker. Second, some consumers might take some interest in taking a leading role in the gathering of marketplace information and sharing it with others, with the intention of proving their worth as experts on the subject among their circle of friends or acquaintances, a concept which has come to be known as being a ‘market maven’.2 This behaviour will in uence price search, especially through its importance as an information transmission mechanism in the retail grocery market (Urbany et al. 1996). H5.2: The greater the self-reported market mavenism, the greater the probability of being an intensive price-information seeker. Research method Data-gathering procedure A survey was conducted in 1997 in a large town in the northeast of Spain (Zaragoza), an urban setting with a large supply of alternatives for grocery shopping. After a pretest, it was decided to use the personal interview mode for the gathering of information, using a questionnaire based on that employed by Urbany et al. (1996). The respondents were the primary grocery shoppers in their respective households. A geographic criterion was used in the allocation of the sample so as to preserve the distribution of the population across the different districts. In the eldwork 231 grocery shoppers were interviewed, although only 196 of the questionnaires were nally considered valid. Characteristics of the sample Given the requirement of being the main grocery shopper in the household, most of respondents in the sample were women (85.3 per cent). Since sex
136
The International Review of Retail, Distribution and Consumer Research
Downloaded by [Universidad Pública de Navarra - Biblioteca] at 02:36 24 September 2014
differentiation was deemed to be non-existent, this variable has not been included in the determinants of PISB in this study. Other descriptive elements of the sample are: 50 per cent of the sample was aged between 36 and 50; 60.5 per cent self-reported a low–medium or medium gross annual income. Most of
Figure 1 Model of the determinants of price-information-seeking behaviour
Bern´e et al.: The use of consumer’s price information search behaviour
137
the respondents (90.4 per cent) had some degree of education, with an even distribution between those with primary, secondary and further education (31 per cent, 28 per cent and 31 per cent respectively).
Downloaded by [Universidad Pública de Navarra - Biblioteca] at 02:36 24 September 2014
Measures As described in Table 2, according to the previous discussion, the measurement of the dependent variable (PISB) is a composite of twelve items about the priceinformation-seeking activities in which consumers may eventually engage. The total information-seeking effort has been separated into two components. The rst one, PCSB (6 items: P1 through P6), is equivalent to the one used in Urbany et al. and re ects the consumer’s tendency to make effective price comparisons across stores. These initiatives by the consumer in acquiring market information through price comparison across stores imply high degrees of involvement and physical effort. The second, PSSB (6 items: P7 through P12) is a different concept as it includes the rest of information sources available to consumers.3 It is important to note that most of the activities included in PSSB are related to information sources which require less physical effort and time than the activities included in PCSB. In order to check reliability, the alpha coef cients for the two measurements of the price-search dependent variable were calculated, which gave values of 0.82 for PCSB and 0.74 for PSSB. The measurement of the independent variables involved the self-reported answers about forty-three statements (Table 2) using a Likert-type scale with 5 points, ranging from ‘strongly disagree’ to ‘strongly agree’. The forty-three items make up the ve initial factors of the mode: eight items for economic returns, ten for search costs, eleven for human capital, two for demographics and twelve for psychosocial returns. However, this initial structure of the independent variables was altered after a factor analysis was made. For our model, in which some of the variables were a composite measurement of different items, we calculated the alpha coef cients for those with more than two items, and the correlation coef cients for those with only two items (Table 3).4 The calculated coef cients, or correlation, for the variables economic returns (price dispersion, budgetary constraints and household income) and search costs (time constraints, dif culty of comparison and mobility constraints) are not shown in Table 3 as they did not reach acceptable values. To uncover whether there was indeed a common structure between the items forming the variables, both dependent and independent, an exploratory factor analysis was carried out. It was noticeable that the independent variable ‘human capital: market knowledge’, which was originally unidimensional, gave rise to a two-factor structure5 which held 80 per cent of the variance. The rst factor (HCMK1), was formed by the items which indicated an ‘overall knowledge of store prices’ and the second factor (HCMK2) by those which measured the knowledge of prices in ‘special sections in the stores ( sh department, greengrocery and meat department)’.
138
The International Review of Retail, Distribution and Consumer Research
Table 3 Alpha coef cients, independent variables Variable
Downloaded by [Universidad Pública de Navarra - Biblioteca] at 02:36 24 September 2014
Human capital: market knowledge (HCMK) Human capital: investment in information (HCSI) Human capital: time management (HCTM) Psychosocial returns: market maven (PRMM) Psychosocial returns: shopping enjoyment (PRSE)
Alpha/corr. 0.84 0.8 0.75 0.91 0.87
The nal structure of the independent variables is: Economic returns (perceived price dispersion)
Economic returns (perceived budget constraints) Economic returns (income per family) Search costs (perceived time constraints)
Search costs (perceived dif culty of store comparison) Search costs (mobility constraints)
Human capital (market knowledge) Human capital (investment in information) Human capital (time management) Demographics (Age) Demographics (education) Psychosocial returns (market mavens) Psychosocial returns (shopping enjoyment)
PPD1 PPD2 PPD3 PRICEDIF PBC1 BUDGET PBC2 IPF PTC1 TIME PTC2 CHILDREN PDSC DIFFICULTY MC1 HEALTH MOBILITY MC2 HCMK1 HCMK2 HCSI HCTM AGE EDU PRMM PRSE
Results To achieve the purpose of this paper, i.e. to provide some basis for the identi cation of market segments and suggest subsequent managerial implications for price differentiation, the method should focus on analysing the discriminatory power of the independent variables in a market segmentation based on PISB. In this direction, we rst applied a cluster of cases for the dependent variables, and, second, we applied a discriminant analysis to explore the capacity of the independent variables in this way.
Bern´e et al.: The use of consumer’s price information search behaviour
139
Cluster analysis
Downloaded by [Universidad Pública de Navarra - Biblioteca] at 02:36 24 September 2014
The cluster analysis was made for three different speci cations of the dependent variable: 1 the total effort in price-information seeking (PISB) 2 the effort in information seeking through price comparisons (PCSB) 3 the effort in information seeking through other sources (PSSB). A relevant market partition for price differentiation purposes should not result in too many segments; pricing is a mix of prices and communication which is not easily delivered differently to groups of consumers. In our analysis6 we restricted the number of clusters to two. The results for the three alternative speci cations of the dependent variable indicate the existence of two well-differentiated clusters in each of them: a group of intensive price-information seekers and a group of light information seekers (Table 4). For the PISB variable, the intensuve seekers group has signi cantly higher measurements for all the thirteen components of the variable. The same circumstance is true for the two partitions of PISB: PCSB and PSSB.7 The size of each cluster and their crossed relationships are provided in Table 5. For the PISB variable, the size of the heavy seekers group is forty-four consumers; one-third (33.3 per cent) of the consumers in the sample ascribed to any cluster might qualify as intensive price-information seekers while two-thirds (66.6 per cent) show a low-intensity pro le in price-information seeking. Observing the clusters obtained with the PCSB and PSSB criteria (Table 6), we may get a more de ned picture of the distribution of price-information seeking. For PCSB, the segment of intensive seekers is fty-eight; this is the 44 per cent of all consumers ascribed to any cluster. For PSSB, the segment of intense seekers goes up to seventy-three; that is, 55 per cent of the consumers. Price comparison across stores is a much more time-and-energy-consuming activity than any of those included in the PSSB measurement. Very likely, this is the reason why the cluster size of intensive seekers is larger in PSSB than in PCSB. Discriminant analysis The following step in our procedure is to explore the discriminant capacity of the independent variables of our model; that is, their in uence in allocating one consumer to any of the previously identi ed clusters. With this step we are trying to provide a picture of the consumers in each cluster so to enable the retail companies to design an effective pricing-differentiation policy. Using the enter method, the ltered results of the discriminant analysis are shown in Table 7. The procedure used to select the variables was based on two criteria: (1) the correlation between variables (variables with low correlation coef cients were excluded) and (2) a joint consideration of the variable coef cient and its correlation with the discriminant function; the variables selected had high or medium coef cients and high or medium correlation with the
ANOVA P1 P2 P3 P4 P5 P6 P7 P8 P9 P10 P11 P12
ME-L 3.35 3.12 2.58 2.10 2.02 2.32 3.94 2.16 2.43 2.60 1.83 3.61
PCSB-PSSB ME-2 F 4.66 3.84 4.27 4.27 4.18 3.59 4.64 2.86 3.14 4.09 2.80 4.66
43.43 17.53 56.89 135.26 142.78 33.62 14.22 7.14 8.27 44.78 20.49 24.41
p-value
ME-1
ME-2
0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.008 0.005 0.000 0.000 0.000
3.26 3.08 2.45 1.84 1.82 2.08
4.47 3.72 4.03 4.09 3.91 3.59
PCSB
F
p-value
40.48 15.45 54.92 200.41 154.71 60.77
0.000 0.000 0.000 0.000 0.000 0.000
ME-1
ME-2
3.80 1.88 1.86 1.90 1.56 3.07
4.48 2.81 3.31 4.07 2.63 4.68
PSSB
F
p-value
15.48 14.49 51.10 199.68 29.74 94.42
0.000 0.000 0.000 0.000 0.000 0.000
MANOVA Value Exact F Hypoth. DF Error DF p-value Value Exact F Hypoth. DF Error DF p-value Value Exact F Hypoth. DF Error DF p-value Test Wilks 0.33
20.45
12
119
0.000
0.30
49.74
6
125
0.000
0.30
48.94
6
125
0.0000
The International Review of Retail, Distribution and Consumer Research
Downloaded by [Universidad Pública de Navarra - Biblioteca] at 02:36 24 September 2014
140
Table 4 Characterization of cluster variables
Bern´e et al.: The use of consumer’s price information search behaviour
141
Table 5 Dependent variables: percentages in the clusters
PCSB
Downloaded by [Universidad Pública de Navarra - Biblioteca] at 02:36 24 September 2014
Low High Total
74 (84.1%) 14 (15.9%) 88 (100.0%)
Low
PISB PSSB
PCSB
55 (62.5%) 33 (37.5%) 88 (100.0%)
0 44 (100.0%) 44 (100.0%)
High
PSSB 4 (9.1%) 40 (90.9%) 44 (100.0%)
discriminant function (variables in bold characters in Table 7 have high coef cients and high correlation). The variables which discriminate jointly are not the same for the different speci cations of the dependent variable. When analysing information seeking in general (PISB), the pro le of an intensive price-information seeker would be: knowledgeable consumer (market knowledge 1 investment in information) perceiving a wide range of price variations across stores, and with a high degree of education, and, although with less intensity, they are people facing a tight budget but neither mobility nor time restrictions. When the analysed behaviour is intensive PSSB, the pro le is detailed with fewer variables: knowledgeable consumers who enjoy sharing and delivering price information (market mavens) and with young children at home. The pro le for intensive PCSB behaviour is: knowledgeable consumer, perceiving a wide range of price variations across stores, with health limitations, but no mobility restrictions and a high degree of education and perceiving dif culties in differentiating the quality of products. Most of these results are consistent with the hypotheses embedded in our model about the individual relationships. However, the roles of two components of the human capital factor need to be discussed. First, although the selfTable 6 Cluster crossing with PCSB and PSSB Other information sources High Low Total
Comparison across stores High Low 44 (33.3%) 14 (10.6%) 58 (44.0%)
29 (22.0%) 45 (34.0%) 74 (56.0%)
Total 73 (55.3%) 59 (44.6%) 132 (100.0%)
Table 7 Discriminant variables in price-information seeking behaviour PISB Variable PRICEDIF HCMK1 HCSI EDU MOBILITY BUDGET TIME
Coef c. 0.55 0.48 0.33 0.54 0.38 0.33 2 0.29
PSSB Variable HCMK1 PRMM HCSI CHILDREN
Coef c. 0.68 0.40 0.20 0.41
PCSB Variable PRICEDIF HCMK1 HCSI EDU MOBILITY DIFFICULTY HEALTH
Coef c. 0.52 0.52 0.49 0.47 0.37 0.32 0.26
Downloaded by [Universidad Pública de Navarra - Biblioteca] at 02:36 24 September 2014
142
The International Review of Retail, Distribution and Consumer Research
reported measurement of market knowledge was initially considered to have an uncertain effect on PISB, it has shown to be probably the most discriminatory characteristic. Second, contrary to the initial argumentation, previous investment in seeking has a positive relationship with the probability of being an intensive price-information seeker. A very plausible explanation for these con icting ndings lies in the nature of the retail grocery market. In these markets information becomes obsolete within very short periods of time. In a simulation in which consumers had to select a neighbourhood store for shopping for a basket of grocery products, the use of previous-week price information resulted in inef cient choices. In an eight-week period, random choices of stores resulted in a lower bill at the end of the simulation than the one resulting from the use of previous-week price information (M´ugica and Deike 1991). This misleading use of information about past prices poses the need to update the information constantly. In this direction, market knowledge and investment in information are just part of PISB, there seems to be no causality, they are not an antecedent of PISB. These different pro les suggest that there are some relevant differences between intensive PCSB seekers and intensive PSSB seekers. Basically, consumers with an intensive use of in-store and at-home information sources are driven in their information-mode choice by their maven orientation and their time restrictions. This result is coherent with previous ndings in the study about store loyalty in the grocery market (Bern´e et al. 1996) in which time restrictions were identi ed as a major barrier to inter-store visits or to multistore shopping.
Discussion and managerial implications The rst main purpose of this paper was to identify different clusters of consumers according to their behaviour in price-information seeking. The empirical study allowed us to identify two clearly separated clusters for each of the three alternative speci cations of the dependent variable: high and low intensive price-information seekers. Based on this clustering, some of the independent variables proved to be good descriptive devices allowing us to give some hints about the consumer pro le of the clusters. Market knowledge and investment in information, ‘human capital’ following Urbany et al.’s (1996) terminology, seems to have the greatest discriminant power in ascribing consumers to the intense information seekers cluster, both for price comparison activities and for other information sources. However, they are not relevant discriminatory variables in the retail grocery market since, rst, they are among the dependent variables and, second, they are not effective indicators for any differentiation policy as they do not provide external signals to retailers. The need to establish a sound pricing policy is reinforced by the ndings of our cluster analysis in which only forty- ve consumers (34 per cent of our sample) self-reported a low intensity pro le in price-information seeking. The rest of consumers are not just concerned about prices but, more than that, they
Downloaded by [Universidad Pública de Navarra - Biblioteca] at 02:36 24 September 2014
Bern´e et al.: The use of consumer’s price information search behaviour
143
show some explicit initiatives in gathering information from the various sources in the market-place. Retailers use a wide variety of ways to communicate their prices: in-store communication – promotions and all varieties of merchandising devices such as shelf-tags (regular or special) and electronic panels; local media – newspaper ads, magazines, local TV and radio; direct marketing – mail, home delivery, phone. Should retail companies use these communication means disregarding the PISB patterns of their market target? There follow two examples of proposals for pricing differentiation. 1 Pricing policy for intense PCSB seekers (44.0 per cent of the market). The external signals of these consumers are degree of education and transport availability, and they are associated with limited budgets for shopping for groceries. If the store manager believes that this is the market segment served by the retail outlet, the pricing policy should incorporate: competitive price xing for meat and produce; these consumers PCSB prices across stores based on the belief that there are signi cant differences in these type of products. Maintaining a competitive advantage on these prices might be a decisive factor in attracting these customers. the competitive price xing, with claims such as matching the lowest price in the market or 10 per cent below our competition, should be communicated according to the pro le of the cluster. These consumers are educated and ef cient in their information effort; the communication should emphasize the price advantages with subtle and far from deceptive guidelines.
c
c
2 Pricing policy orientated to intense PSSB seekers (55.3 per cent of the market). The only external signal of these consumers is that they will have young children, something basically associated to time restrictions. If the target market has a relevant share of these consumers, the pricing policy should incorporate: c
c
c
consistent price specials in children products: very likely, the cluster will be very sensitive to prices for these products so that price xing will have to consider a promotion planning with the continuous presence of price specials for some of these products. communication orientated to market mavens: these consumers enjoy communicating to other consumers their knowledge about good prices. The communication from the store to these individuals should include appealing arguments which they could replicate in their communication with friends or neighbours. communication orientated towards establishing an identi cation of the customer with the store: while knowledgeable about market prices, these consumers face some restrictions which will result in some behavioural loyalty to the store. The store should include in its communication policy some relationship marketing considerations: frequency programmes or some kind of price advantages based on loyalty.
144
The International Review of Retail, Distribution and Consumer Research
Downloaded by [Universidad Pública de Navarra - Biblioteca] at 02:36 24 September 2014
Finally, the authors feel that it would be of interest to extend the study in various directions. On the one hand, it is necessary to be aware that this present study has been undertaken without differentiating price dispersion types. Thus, it contains information on inter-type and intra-type differences which will be the subject of examination in future research. In addition, other types of variables could well be used as explanatory variables in search behaviour, such as intrinsic consumer variety seeking or the division of the different search costs (perception, prospecting and evaluation) which are borne by consumers. Notes 1 With respect to the point of sale, price differences may be inter-type and intra-type, inter-channel and intra-channel, or inter-chain and intra-chain. Inter-type and intratype price differences take the level of service offered as a classi cation criterion. For the former, price differences arise from the different types of establishment (supermarkets, hypermarkets, etc.). Intra-type differences refer to the fact that no two establishments of the same type are ever exactly the same: the mere siting of an establishment is in itself a service. Inter-channel and intra-channel divisions refer to price differences arising from the type of relationship the establishments have with their suppliers, which is re ected in their purchasing costs, and the form in which these are passed on to the consumer. Finally, the inter-chain and intra-chain division is due to the differences existing in the rms’ marketing policies, as well as the various levels of price control exercised by individual establishments (M´ugica and Yagu¨ e 1993). 2 ‘Market maven’ refers to ‘those who have information on many types of product, places to buy and other facets of the marketplace, and who share such information with other consumers’ (Feick and Price 1987). 3 In Urbany et al. (1996), this second component is named INDEX and is a measurement of the consumer effort only in the search for price-specials. 4 In the variables included in psychosocial returns, two items were eliminated. Speci cally, it was deemed expedient to leave out the item: ‘I like introducing new brands and products’, a component of the market maven independent variable, and the item: ‘I view shopping in a positive way’, a component of shopping enjoyment. This decision gave rise to a slight increase in alpha values by 0.04 and 0.06. Although, at rst, the elimination of the items did not appear relevant, basically because of our intention not to lose information, it was nally decided to exclude them since an exploratory factor analysis con rmed an increase of the explained variance with one factor, from 64 per cent to 73.3 per cent (with the elimination of item ‘I like introducing new brands and products’) and an increase in the retained variance from 70.1 per cent to 82 per cent (with the elimination of item ‘I view shopping in a positive way’, whose structure was unidimensional). In addition, the communality of these variables was 0.11 and 0.28 respectively, while the other variables, in their respective factors, showed communalities higher than 0.6 and 0.7, respectively. 5 This result coincides with that obtained by Urbany et al. (1996). 6 The methodology used was the agglomerative hierarchical cluster analysis applied to the matrix of the squared euclidean distances between cases. 7 For the PISB analysis, the criterion used for case combinations in each step was the average linkage between groups. For PCSB and INFORM the criterion used was the average linkage within groups.
Bern´e et al.: The use of consumer’s price information search behaviour
145
Downloaded by [Universidad Pública de Navarra - Biblioteca] at 02:36 24 September 2014
References Beales, H., Mazis, B., Salop, S.C. and Staelin, R. (1981) ‘Consumer search and public policy’, Journal of Consumer Research, 8: 11–22. Bern´e, C., Mugica, J.C. and Yague, ¨ M.J. (1996) ‘La gesti´on estrat´egica y los conceptos de calidad percibida, satisfacci´on del cliente y lealtad’, Econom´õ a Industrial, 307: 63–74. Bucklin, L.P. (1965) ‘The informative role of advertising’, Journal of Advertising Research, 5(3): 11–15. Bucklin, L.P. (1969) ‘Consumer search, role enactment and market ef ciency’, Journal of Business, 42: 416–38. Claxton, J., Fry, N. and Portis, B. (1974) ‘A taxonomy of purchase information gathering patterns’, Journal of Consumer Research, 1: 35–42. Dhar, S.K. and Hoch, S.J. (1996) ‘Price discrimination using in-store merchandising’, Journal of Marketing, 60(1): 17–30. Dommermuth, W.P. (1965) ‘The shopping matrix and marketing strategy’, Journal of Marketing Research, 2(May): 128–152. Duncan, C.P. and Olshavsky, R.W. (1982) ‘External search: the role of consumer beliefs’, Journal of Marketing Research, 19(February): 32–43. Feick, L.F. and Price, L.L. (1987) ‘The market maven: a diffuser of marketplace information’, Journal of Marketing, 51(January): 83–97. Furse, D.H., Punj, G.N. and Stewart, D.W. (1984) ‘A typology of individual search strategies among purchasers of new automobiles’, Journal of Consumer Research, 19(4): 7–31. Grewal, D., Monroe, K.B. and Krishnan, R. (1998) ‘The effects of price-comparison advertising on buyers’ perceptions of acquisition value, transaction value, and behavioral intentions’, Journal of Marketing, 62(April): 46–59. Hoch, S.J., Kim, B.D., Montgomery, A.L. and Rossi, P.E. (1995) ‘Determinants of store-level price
elasticity’, Journal of Marketing Research, 32(February): 17–29. Kiel, G.C. and Layton, R.A. (1981) ‘Dimensions of consumer information seeking behavior’, Journal of Marketing Research, 18(May): 233–9. Kleimenhagen, A.K. (1966–7) ‘Shopping speciality or convenience goods’, Journal of Retailing, 42(Winter): 32–9. Lichtenstein, D.R., Ridgway, N.M. and Netemeyer, R.G. (1993) ‘Price perceptions and consumer shopping behavior’, Journal of Marketing Research, 30(May): 234–45. McLelland, C. and Turner, J. (1983) ‘Consumer prepurchase information seeking: a comparison of methodologies’, Marketing Intelligence and Planning, 1: 15–27. Midgley, D.F. (1983) ‘Patterns of interpersonal information seeking for the purchase of a symbolic product’, Journal of Marketing Research, 20(February): 74–83. Mugica, ´ J.M. and Deike, V. (1991) ‘La variabilidad de precios: problemas en la eeneraci´o n y difusi´on de Informaci´on a los consumidores’, Proceedings of the 3rd Encuentro de Profesores Universitarios de Marketing, Salamanca. Mugica, ´ J.M. and Yague, ¨ M.J. (1993) ‘The search for price information by consumers: a study on nondurables’, Proceedings of the 7th International Conference on Research in the Distributive Trades, Stirling, Scotland. September: 40–9. Newman, J.W. and Lockeman, B.D. (1975) ‘Measuring prepurchase information seeking’, Journal of Consumer Research, 2: 216–22. Newman, J.W. and Staelin, R. (1971) ‘Multivariate analysis of differences in buyer decision time’, Journal of Consumer Research, 8(May): 192–8. Newman, J.W. and Staelin, R. (1972) ‘Prepurchase information seeking for new car and major household
Downloaded by [Universidad Pública de Navarra - Biblioteca] at 02:36 24 September 2014
146
The International Review of Retail, Distribution and Consumer Research
appliances’, Journal of Marketing Research, 9(August): 249–57. Stigler, G. (1961) ‘The economics of information’, Journal of Political Economy, 69(June): 213–25. Thorelli, H.B. (1971) ‘Concentration of information power among consumers’, Journal of Marketing Research, 8(November): 427–32. Udell, J.G. (1966) ‘Prepurchase behavior of buyers of small electrical appliances’, Journal of Marketing, 30(October): 50–2.
Urbany, J.E. (1986) ‘An experimental examination of the economics of information’, Journal of Consumer Research, 13(September): 257–71. Urbany, J.E., Dickson, P.R. and Key, R. (1991) ‘Actual and perceived consumer vigilance in the retail grocery industry’, Marketing Letters, 2(1): 15–25. Urbany, J.E., Dickson, P.R. and Kalapurakal, R. (1996) ‘Price search in the retail grocery market’, Journal of Marketing, 60(April): 91–104.
