Consumers ’ attitudes towards different product category of private labels

This research attempted to investigate the consumer perceptions on product quality, price, brand leadership and brand personality towards convenience goods and shopping goods based on international private labels (IPLs) and local private labels (LPLs). Data were collected outside the entrances of the main rail station of Taipei, Taiwan. A systematic sampling was adopted and 254 questionnaires were eventually collected. The findings revealed that the consumers in Taiwan believe that international and local hypermarkets both produce better convenience goods than shopping goods that have high quality and deliver more value. This research represents one of the few pioneer works that empirically investigate the aforementioned issues.


INTRODUCTION
In recent years, large retailers have exercised private label (PL) strategies to reap the ever-increasing market share in most consumers' product categories when competing with national brands (Cheng et al., 2007;Ishibashi and Matsushima, 2009).For example, Wal-Mart, a leading retailer in the USA, the sales volume of PL has accounted for as much as 25% of the operating revenues.In grocery outlets, they form over 15% of US supermarket sales and over 44% grocery shoppers regularly buy PL (Chen, 2007).In Switzerland, where the top 5retailers capture up to 88% of the market, the number of PL (38% of business is private labeled) makes Switzerland the most PL country in the world.In Great Britain, 31% of total business is composed of PL, which owns an 83% share of sales in the retailing sector (Anonymous, 2004).
PL, also store brand, is the brand which retailers are responsible for promoting, shelf placing, pricing, and positioning its own brand in the product space.In this study, we group the PL into two categories that is, the international private label (IPL) for example, American Costco and British Tesco, and the local private label (LPL) for example, Taiwanese RT-Mart.Taiwanese consumers perceive that these two categories of PL are different in some ways.Meanwhile, product classification, for example, convenience goods versus shopping goods, can also affect consumers' attitudes toward these brands.
The notion of national brands (NBs) and PL have been heavily studied in the marketing literature (Sanjoy and Oded, 2001;Karray and Martín-Herrán, 2009).Most of them attempt to compare consumer perceptions of product quality and price among different brands.However, there are few articles that distinguish the different effects on IPL and LPL and there are few researches work that introducing or show the effect of different product classification on such in this matter.This research, therefore, investigates the attitudes of Taiwanese consumers towards convenience goods and shopping goods based on two types of brands: IPL and LPL.This research is organized as follows: literature review and research hypotheses are presented in the next section.In Section 3, we will discuss the method used in the paper.The data analysis and results are shown in Section 4. This paper is then concluded in the final section.

REVIEW AND RESEARCH HYPOTHESES
Some companies license their PLs and then launch new businesses.They sometimes search for manufacturers with excess capacity to produce their PL at a lower cost, and then earn a higher profit margin (Parker and Kim, 1997;Gómez and Benito, 2008).The past two decades have seen manufacturers' brands becoming less important.Meanwhile, retailers have been growing in influence and gaining power in the marketing channel (Farris and Kusum, 1992;Shocker et al., 1994;Frederick, 2000;Suárez, 2005).The various arenas of NBs and PLs take place and the PLs possess unique competitive means to control the market power of NBs, including price, shelf space, quality, innovation and brand advertising (Steiner, 2004).
This research distinguishes two types of PLs namely IPL and LPL.The purpose of this paper is to investigate consumers' perceptions towards convenience goods and shopping goods based on IPL and LPL qualities.First, the study defines the constructs used in this paper and their application.It then proposes the viewpoints by raising a number of hypotheses.There are four constructs to be discussed namely, perceived quality, brand leadership, price perception, and brand personality.These constructs will be discussed in relation to two product categories which are the convenience goods and shopping goods.

Perceived quality
According to Aaker and Joachimsthaler (2000), perceived quality is "a special type of association, because it influences brand associations in many contexts and has been empirically shown to affect profitability".It is also defined as "the consumer's judgment about a product's overall excellence or superiority" (Chueh and Kao, 2004).

Brand leadership
According to Aaker and Joachimsthaler (2000), "brand leadership is an excellent, interesting, insightful and thought-provoking draw that significantly extends the boundaries of what we know about managing brands."Aaker (1996) develops three dimensions to interpret leadership.These are syndrome, innovation and dynamics of customer acceptance.Steiner (2004) regards innovation as the most important of all.Stefan and Don (2003) also indicate that innovation is one of the National Wang and Chen 7617 Brand American Core Values.

Price perception
The price of an item plays a powerful role in marketing.
Consumers see price more than just the money that the buyer hands over to the seller.The broader view is that the price is the sum of all the values that the buyer exchanges for obtaining the product.Thus price is a key variable in communicating to the customer about the value of the product.The shoppers' perceptions of price are regarded to influence their purchase behavior (Miranda and Joshi, 2003).Thus, the lower the price, the more favorable the brand perception.

Brand personality
Aaker (1997) defines brand personality as "the set of human characteristics or traits that consumers attribute to a brand".Aaker (1996) describes brand personality as the linkage to the emotional and self-expressive benefits of a brand.Boyle (2003) advocates that, "a key aspect of a brand's personality is its values and therefore, one of the tasks of brand builders is to find a way of imbuing the brand with these values".Though it is no easy work to imbue a brand with the values, nevertheless, the brand personality must be distinctive, robust, desirable, and consistent enough to successfully differentiate it from another brand.

Research hypotheses
Consumer products are those that are purchased by the final consumer for his/her consumption.These products are further classified into convenience, shopping, specialty.This research will focus on convenience goods and shopping goods.Convenience goods are those that are purchased frequently with little planning or shopping effort.They are usually at low prices and widely available.
Shopping goods are those which are purchased less frequently, such as furniture and major appliances, and are compared on the bases of suitability, quality, price and style (Kotler et al., 1998).Empirical evidence indicates that retail concentration has increased dramatically throughout Europe as supermarkets take an increasing share of the convenience goods business (Cullen, 1997).Retailers have offered convenience goods as their PLs for a long time now in many countries.Sometimes these products represent a prime product type for brand imitation.Consumers will buy a brand imitator instead of an original if they do not perceive significant differently on product quality given the price difference.Astous and Gargouri (2001) indicate "the goodness of the imitation may be of less importance as long as the quality appears to be there".
In an empirical study done by Leblanc and Turley (1994), only an average of 2.19 brands can be remembered from a consideration set sizes of 23 shopping goods.In the research conducted by LeBlanc and Turkey (1994) and LeBlanc and Herndon (2001), they record that "consumers seem to have initial evoked sets with only a single brand in them."Baker and Wilkie (1992) have the same results reporting that consumers have small-evoked set sizes for shopping goods.
When they compared convenience goods and shopping goods, Murphy and Ben (1986) found out that the former has lower risk and effort than the latter.On a concept of intergenerational (IG) influence research, Heckler et al. (1989) discovered that consumers have stronger preference effects for convenience goods than for shopping goods (Moore et al., 2002).Therefore, the study proposes the following hypotheses.
Consumers perceive that the quality of convenience goods is superior to that of shopping goods for private label.H 2.
Consumers perceive that brand leadership of convenience goods is superior to that of shopping goods for private label.H 3.
Consumers perceive that the price of convenience goods is higher than that of shopping goods for private label.H 4.
Consumers are more conscious of the brand personality for convenience goods than for shopping goods for private label.

RESEARCH METHODOLOGY
In this section, we discuss the research setting, scale development, as well as the sampling framework.

Research setting
As mentioned earlier on, this research seeks to investigate consumer perception of two product categories among two different labels.We therefore compare consumer's attitude toward IPL, and LPL between convenience goods and shopping goods.
The study considers a liquid detergent as the convenience good in the research based on Cullen's (1997) work.There are many hypermarkets in Taiwan.However, this research chose two of them to be the research subjects.Not only are these hypermarkets well known, but they also have their own store brands in many product categories.TESCO (merged with Carrefour in Taiwan) liquid detergent (one of the store-brand products of British Tesco Co.) and FP liquid detergent (a PL by a local Taiwanese RT-Mart) are our IPL and LPL, respectively.We choose the electronic appliance as the item representing shopping goods based on the paper by Varinder and Krish (2001).As pointed out earlier on TESCO and FP are our IPL and LPL, respectively.

Scale development
The questionnaire was developed mostly according to the scales of Aaker (1996) and Miranda and Joshi (2003), but with some minor modifications that fit our research purpose.The measurement items of price perception are cited from Miranda and Joshi (2003).In addition, the study takes into consideration the classification by Aaker (1996) on measuring brand equity across products and markets.All items were measured as perceptions on a 7-point Likert scale: 1=strongly disagree, 4=neutral and 7=strongly agree.The interviewees were asked about their perceptions toward IPL, and LPL products.Finally, the study adopted four sets of items that are discussed below as follows.
Each scale of perceived quality, brand leadership, and brand personality contain three items.The Item-total value of the price construct suggested the elimination of item 2 and the amount of the final measurement items for price perception is only one.The items of perceived quality are "high quality", "the worst brand", and "consistent quality".Leadership scale contains "innovative", "growing in popularity", and "the leading brand".Price perception scale is comprised of "lower"; "has a personality", "interesting", and "clear image of the type of users".These items are related to the personality scale.

and data collection
Taipei, the metropolis of Taiwan, is selected as the target population in this research.The survey was administered to the Taiwanese residents at the public location (Taipei railway station) with the systematic sampling method (one out of ten passing entrance).After discarding some invalid respondents, we collected a total sample of 254 for the final data analysis.

ANALYSIS AND RESULTS
A series of statistical procedures are used to analyze the research questions.The respondents' profile will be reported first, followed by an examination of the reliability of the scales.Testing of the research hypotheses will then be placed at the end of this section.

Respondents' profiles
The respondents included more females (60.63%) than males (39.37%).Most of them were less than 30 years of age.About 91.74% of the respondents held a university/college degree or higher.As for their occupation, white collar workers, blue collar workers, and students all occupied around one third of the respondents.
In terms of income, about 45% earned total monthly incomes of NT$15,000 or less.Finally, most of the respondents were single.Detailed descriptive statistics relating to the respondents' profiles are shown in Table 1.

Measurement accuracy analysis
A confirmatory factor analysis (CFA) was conducted on the sample data in order to get evidence on the robustness and reliability of the scales in a brand equity context.The CFA model had an overall Chi-square of 225.10, a category for Inference (CFI) of 0.97, an IFI of 0.97, a NNFI of 0.96 and a root mean square error of approximation (RMSEA) of 0.065.The CFI, IFI, and NNFI measures fall above the recommended value of 0.9 and RMSEA is below the recommended value of 0.08.All of the item loadings were satisfactory and the t-values were significant.Thus, the fit of the model is good.Related reliability and validity assessment followed.Coefficient alpha values, composite reliability (CR) indexes and an average-variance-explained (AVE) value of the 6 dimensions were computed for reliability tests.All the alpha, CR and AVE values shown in Table 2 exceeded generally recommended levels of 0.6, 0.6 and 0.5, respectively (Shook et al., 2004).Thus, the results provided evidence of reliability.As for the test of validity, examining significant t-value factor loadings checked for convergent scale validity.Table 2 shows significant t-values, ranging from 20.68 to 41.32.The AVE score, achieved for the entire model constructs, for each factor being over the minimum threshold of 0.5 (Fornell and Larcker, 1981) and ranged from 0.51 to 0.90.Considering the evidence of reliability and validity, the scales should be considered generally reliable and valid overall.Individual scale items and test summary related to research construct accuracy are shown in Table 2.

Hypotheses testing
The statistical analysis software (SAS) package is used in this study.The t-tests are used to compare the consumers' perception on both convenience goods and shopping goods within a company.As can be seen from Table 3, in terms of the perceptions of IPL (British Tesco) and LPL (Taiwanese RT-Mart) across the 4 factors, there was a great degree of homogeneity in the convenience and shopping goods except quality perceptions.It's credible that consumers perceived the quality for convenience goods to be better than shopping goods.Therefore, H 1 is accepted.This denoted that customers in Taiwan believed the hypermarkets with more effort and experience on convenience goods are better than shopping goods.

DISCUSSION AND CONCLUSION
This research investigated the consumer perceptions of IPL and LPL on convenience goods and shopping goods categories in Taiwan.Most of previous works have focused on NBs and private brands only, but we distinguish PL into IPL and LPL.Since Taiwan is getting more internationalized, more and more international businesses come to the island.Therefore, it is important to understand how consumers perceive these private branders.
The mean value of consumers' perception on convenience goods is significantly higher than that of shopping goods.That is to say the consumers in Taiwan believe that these two hypermarkets produce better convenience products than shopping products in quality and deliver more values to the customers.
IPL has the advantage of economies of scale and the image of commodities of foreign make, which manifests itself as exotic lifestyle and culture.LPL received the lowest score in almost all aspects, which means that consumers do not regard highly this label.But convenience goods branders of this label can make use of the advantage of their low price strategy.For shopping goods, the study suggests the branders emphasize the origin of those products.
Findings from the comparison of convenience and shopping goods for IPL/LPL, show that customers perceived shopping goods inferior to convenience goods.Retailers should make more effort to create high value shopping goods, especially on quality conception enhancement.

LIMITATIONS AND DIRECTION TO FURTHER RESEARCH
This paper tests the hypotheses by conducting the survey method.The study adopts part of the constructs from the work of Aaker (1996) and Miranda and Joshi (2003).Their objectives are to measure brand equity without subject to any specific product categories.Though the study adopt only four constructs to measure Taiwanese consumer perceptions on two different brand types, most of the results support the viewpoint.That means, the study was able to measure consumer perceptions using these constructs.The study does not use the whole original scale, but this could be an interesting issue for further study.
There is little research distinguishing IPL and LPL from PLs.The findings indicate that consumers have different perceptions on these two brand types.This calls for marketers to exploit the opportunities in these segments.
Finally, the research approach and findings of this study can provide directions for future research.Future study could follow the research approach to get more empirical evidence by adopting different categories, brands, as well as focusing on different countries other than Taiwan.

Table 2 .
Measurement accuracy analysis statistics.
* Not applicable

Table 3 .
Comparison between convenience and shopping goods for IPL/LPL.Mean values for convenience goods on IPLs, based on a 7 Likert-type scale with strongly disagree; 7, to strongly agree; 1, mean values for shopping goods on IPLs, based on a 7 Likert-type scale with strongly disagree; 7, to strongly agree; 1, mean values for convenience goods on LPLs based on a 7 Likert-type scale with strongly disagree; 7, to strongly agree; 1, mean values for shopping goods on LPLs based on a 7 Likert-type scale with strongly disagree; 7, to strongly agree; 5, significant level; *, significant different (at 0.05 level) for convenience goods and shopping goods.