QR Codes: Effective In-Store Marketing Tools or only further Stimuli that
lead to Information Overload?
Master Thesis, Master of Science Marketing Management University of Groningen
Faculty of Economics and Business, Department of Marketing
January 12, 2015
Mona Mieske
Briljantstraat 397, 9743 NP Groningen monamieske@googlemail.com
S2626578
Abstract
The aim of this thesis was to investigate the influence of providing additional product information by using QR codes as a means of in-store marketing to influence buying decisions. Perceived product involvement was expected to moderate this effect. An online based experiment has been conducted to analyze the suggested relationships by either providing participants with the possibility to access additional product related information or by actually exposing them to this information of either a low or a high involvement product. 205 participants took part in this experiment. The main finding of this thesis is that there is a positive influence when customers are also exposed to this additional information as more than twice as many participants chose the product that contained a QR code over the product they intended to buy. However, the additional information did not lead to choice difficulties due to information overload and also the assumed interaction effect of perceived product involvement could not be supported. Thus, the use of QR codes should not be limited to high involvement products as they can also influence customers’ decisions when less effort is made to choose a product. A main implication that was derived from the results is that practitioners should not only focus on implementing QR codes on their product designs but should put even more thoughts in ways to motivate customers to scan QR codes in order to be more effective.
Master Thesis Mona Mieske Table of Contents 1 Introduction ... 1 2 Literature Review ... 4 2.1 Mobile Marketing ... 4 2.2 Information Overload ... 8 2.3 Brand Choice ... 11
2.4 Perceived Product Involvement ... 12
2.5 Conceptual Model ... 15
3 Methodology ... 15
3.1 Study Design ... 16
3.2 Procedure and Participants ... 16
3.3 Manipulations ... 16
3.4 Measurement ... 17
4 Results ... 18
4.1 Descriptive Statistics ... 18
4.2 Preparing for Analysis ... 21
4.3 Manipulation Check ... 21
4.4 Analysis of Main and Moderating Effects ... 22
5 Discussion ... 27
6 Conclusion ... 29
7 Implications ... 30
8 Limitations and Future Research ... 31
References ... 33
List of Figures
Figure 1: Conceptual Model ... 15
Figure 2: Distribution of Age in Years ... 19
Figure 3: Distribution of Occupation ... 19
Figure 4: Distribution of Income... 19
Figure 5: Smartphone Usage in Everyday Life and Shopping Situations ... 20
Figure 6: Distribution of Age in Years per Group ... 21
Figure 7: Results Chi-Square-Test ... 22
List of Tables Table 1: Study Design ... 16
Table 2: Overview Experimental Groups ... 20
Table 3: Overview Results One-Way ANOVA ... 23
Table 4: Results of Variables in the Equation ... 24
1 Introduction
In-store shopper marketing gains increased attention due to new opportunities and challenges created by the advent of the internet and, in line with that, the augmented use of mobile devices and possibilities of mobile marketing activities (Shankar 2011, p. 34; Shankar et al. 2010). In order to successfully implement in-store marketing activities, it is important to understand how the connection of offline shopping situations and the online world affects the behavior and decisions of consumers. In the past, most research focused on how online search behavior and, for example, reading online reviews influences shopper behavior in offline shopping settings (Senecal and Nantel 2004; Zhu and Zhang 2010). However, a field that has not yet been researched in depth is how customers can be motivated to use online devices during their in-store purchase and, furthermore, how this could influence their buying decision. As 82% of consumers in the United States use their smartphone to look for information while shopping in 2010 (Goldberg 2010), mobile devices might be a promising tool to connect in-store marketing and mobile marketing activities.
Possible touch points which marketers can use to connect an offline shopping situation to the online world are Quick Response codes (QR codes). These codes are two-dimensional barcodes that contain information which can be retrieved when they are scanned with a smartphone. QR codes were developed back in 1994 but only recently gained attention by researchers and practitioners in the field of shopper marketing (Shin, Jung, and Chang 2012). In an in-store shopping situation, these codes can be used to offer customers the possibility to access additional product information like, for example, nutritional aspects for groceries or detailed product features for electronic devices. Thus, a QR code itself can be seen as a means of transparency created by companies. However, in order to make use of this transparency and to retrieve the information, an intended action is required from the customer by scanning the code with a smartphone. The goal of this thesis is to examine if providing and confronting customers with additional product information via a QR code can have an influence on buying decisions in an in-store shopping environment.
be the case that also providing additional information via QR codes as an in-store marketing activity will lead to overwhelmed customers as well as to information overload. On the other hand, it could be possible that more specific product information facilitates the decision process as it deepens the customers’ product knowledge and hence reduces confusion (Jacoby, Speller, and Kohn 1974).
Existing research in the field of information overload shows that more information does not necessarily lead to better decisions or higher levels of satisfaction (Iyengar and Lepper 2000; Scheibehenne, Greifeneder, and Todd 2010). In addition to that, there is support for the fact that more information can have negative effects in leading to confusion and uncertainty and even to the customer’s disability to choose a product that best suits his or her needs (Jacoby, Speller, and Kohn 1974; Scheibehenne, Greifeneder, and Todd 2010; Greifeneder, Scheibehenne, and Kleber 2010). However, these findings are based on higher levels of information in terms of a bigger assortment size. Hence, one can differentiate information overload that has been caused by a large product variety to choose from. In addition to that, a high amount of information concerning a specific product can be a further reason for the phenomenon of choice overload, on the other hand (Iyengar and Lepper 2000; Scheibehenne, Greifeneder, and Todd 2010; Greifeneder, Scheibehenne, and Kleber 2010).
Due to the gap in existing knowledge of how additional information provided by QR codes affects in-store shopper behavior, this thesis will deal with the following research questions:
First of all, does the possibility of accessing additional information provided by a QR code influence customers’ in-store buying behavior by changing their brand choice? Can just the fact that more information is available via a QR code on the package change the brand a customer intentionally wanted to buy and lead to a different brand choice?
Second, how does exposure to this information affect the brand choice? In order to retrieve the information, an intended action is necessary to actually scan the code and be able to look at the information. Can it be the case that this additional information leads to information overload? As in the first research question, also the goal of investigating the second one is to find out whether additional product information can influence the brand choice.
Additionally, this thesis will examine if the product category can have an influence on this effect. Does it matter if the planned purchase concerns a low or a high involvement product? Can we gain insights whether customers would prefer personal advice over mobile marketing via their smartphone when buying high involvement products? As involvement is a construct that can hardly be generalized over categories, the perceived product involvement of a customer will be examined.
The remainder of this thesis will be structured as the following. First of all, there will be given a literature review on the state of the art knowledge concerning the aspects of mobile marketing in in-store settings, information overload, brand choice as well as perceived product involvement in this context. Furthermore, hypotheses will be derived. In the main section, the methodological approach will be explained. Here, it is the goal to develop, apply and analyze an online survey that will test the expected relationships. Third, the findings will be discussed. Finally, the thesis concludes with limitations as well as implications for theory and practice.
2 Literature Review
In the following section of the thesis, the relevant literature and theoretical background will be presented. First of all, an overview will be given on the developments in the field of mobile marketing as a part of in-store marketing activities. A special focus will lie on the usage of QR codes as a possibility to implement mobile marketing as well as on the phenomenon of information overload which might by created by doing so. Furthermore, the concepts of brand choice and perceived product involvement, which is a potential moderator, will be introduced.
2.1 Mobile Marketing
Increased expenditures for shopper marketing activities and a growing body of literature show that shopper marketing is a field of rising interest in marketing research and practice (Neff 2007). According to Shankar (2011, p. 3), shopper marketing can be seen as “the planning and execution of all marketing activities that influence a shopper along, and beyond, the entire path-to-purchase, from the point at which the motivation to shop first emerges through to purchase, consumption, repurchase, and recommendation”. As the number of online shoppers grows, also the interest in developing and implementing successful internet based marketing activities that, for example, make use of applications or QR codes for mobile devices, increases (Zhang, Prybutok, and Strutton 2007).
intuition than on a profound understanding. Furthermore, companies find it rather challenging to motivate existing customers to use online channels in general (Pousttchi and Wiedemann 2006; Persaud and Azhar 2012; Falk et al. 2007). Therefore, there is a need for more research to understand what makes mobile marketing activities successful, especially with regards to in-store shopping situations, and why customers are motivated to engage in mobile marketing, for example by actively scanning a QR code that is on a product package and use the provided information.
In order to do so, an understanding of what mobile marketing means is needed. There are different ways to define the concept. For example, according to Shankar and Balasubramanian (2009, p. 118), mobile marketing is “the two-way or multi-way communication and promotion of an offer between a firm and its customers using a mobile medium, device or technology”. Aspects which the several different definitions have in common are that mobile marketing describes the process of promoting goods and services via mobile technology in an interactive network by creating a relationship between a customer and a company or a brand. In addition to that, mobile marketing offers the opportunity for one-to-one marketing (Shankar and Balasubramanian 2009; Ryu 2013; Pousttchi and Wiedemann 2006; Kaplan 2012). Also, the objectives of this form of marketing can be seen in influencing the brand image, building and increasing brand awareness and also building a customer database (Pousttchi and Wiedemann 2006). Hence, in the long run, mobile marketing activities are expected to increase sales if they are implemented successfully.
which is transmitted via mobile marketing and can be accessed by the customer if desired, is regarded as useful and could influence customers’ buying decisions.
In line with that, there is some research that concentrates on the impact of information provided by mobile marketing activities. For example, Sultan, Rohm, and Gao (2009) find evidence for the fact that this information has a positive influence on risk acceptance and personal attachment which then positively influence the acceptance of engaging in mobile marketing activities. Furthermore, Pousttchi and Wiedemann (2006) also determine information to be an important aspect as they developed four types of mobile marketing with one of them being information. Next to information concerning the product, also aspects like mobile newsletters or market rates can be regarded as information in this context. The remaining three types of mobile marketing are entertainment, raffle and coupon. The authors point out that further investigation concerning the effectiveness of those four types is necessary. Hence, there is a need for additional research to examine the role provided information by mobile marketing activities plays on the success of those activities.
One possibility to implement mobile marketing in an in-store setting is by adding QR codes on product packages that include product information. As mentioned before, QR codes are two-dimensional quadratic matrixes that are composed of black and white modules and can be transferred into a message when scanned with a mobile device. In the last years, the use of such codes increased rapidly and became a part of product marketing, in-store product labeling but can also be found in print ads and commercials (Dou and Li 2008; Ryu 2013). Customers receive product related information, URL links, addresses or even videos when scanning QR codes. In addition to that, users can also create their own codes by making use of specific apps and thereby share content with other users (Shin, Jung, and Chang 2012; Ryu 2013). Furthermore, the success of QR codes depends strongly on the customers’ interaction as the codes need to be actively scanned in order to access the offered information. Thus, communication via QR codes engages customers more than traditional means of communication so their success depends on the individual’s willingness to scan and use the codes (Dou and Li 2008).
Ashford (2010) conducted a study on how QR codes can successfully be implemented in academic libraries. The main finding was that such codes are only appreciated by customers in situations where they expect to benefit and receive added value as a consequence of scanning the code. In line with that, Okazaki, Navarro, and Lopez-Nicholas (2013) suggest that the amount of information that is needed when making a buying decision determines the acceptance of using a QR code. They argue that customers that purchase a high involvement product are looking for more information and, thus, are more willing to use a QR code. Hence, the willingness to use QR codes can be considered as situation and product specific. Further reasons why consumers might want to use QR codes are that a minimum of timely effort and little to no cost is required to do so (Goldberg 2010).
Additionally, there is a very recent study by Watson, McCarthy, and Rowley (2013) which analyzed consumers’ attitudes towards QR codes. However, the setting of this study was different compared to what this thesis aims at analyzing as the QR codes that were scanned were on print ads or posters so the scanning took place outside or at home. The scanning of a QR code in this thesis happens during the decision process, right in a situation that offers a purchase possibility and also direct competition is present. In the study by Watson, McCarthy, and Rowley (2013), scanning QR codes provided participants with information concerning products they might already have bought in the past or also products they are considering to buy in the future due to the exposure to the ad which contains the QR code. Therefore, it can be assumed that this study led to different results as the scanning of the QR code took place in a different situation and did not affect the customer during an in-store decision making process. The authors conclude the following findings. First of all, the attitude and response towards QR code marketing is more positive than towards mobile marketing activities that were implemented by using short text messages. A reason for that was found in the fact that customers using QR codes feel more in control due to the fact that they have to take the initiative instead of passively receiving a message. Furthermore, the primary motivation to use QR codes was found to be accessing information (Watson, McCarthy, and Rowley 2013).
makes them not generalizable. Thus, further research is needed to examine how information provided by QR codes influences shopping behavior and customers’ buying decisions.
2.2 Information Overload
As already mentioned in the first part of the thesis, customers could be overwhelmed by additional information provided by QR codes on product packages. In this case, one would speak of information overload, which is a concept that has already been developed 40 years ago (Jacoby, Speller, and Kohn 1974). The phenomenon of choice overload is recently becoming more relevant in shopping environments due to the increased usage of the internet as a stimulus in online shopping situations (Lee and Lee 2004). As a result, consumers are confronted with a large number of products and their descriptions. A common finding of information overload is that it leads to less satisfaction with buying decisions and that customers are more likely to buy in situations when they are confronted with less attributes, which means that they are provided with information on a smaller number of product features (Fasolo, McClelland, and Todd 2007).
There is a traditional approach to characterize information overload, which was defined by Jacoby, Speller, and Kohn (1974), and a more recent approach, that has been imprinted by Iyengar and Lepper (2000). In the following, both approaches as well as prior research related to these will be presented accordingly. First of all, Jacoby, Speller, and Kohn (1974) examined the effect of information quantity on decision making by altering the amount of product information in that sense that they varied the number of brands participants could choose from and the number of information items they were exposed to. They found out that human beings are able to process only a limited amount of information. Once this limit is exceeded, the decision quality decreases. This means that even though a certain load of product information is necessary to make rational decisions and consumers also do prefer more information, a very high amount of information does not lead to better decisions (Jacoby, Speller, and Kohn 1974; Jacoby, Speller, and Berning 1974).
confronting consumers with more information does not lead to poorer buying decision making when reviewing and re-analyzing earlier studies conducted in this field of research. However, Keller and Staelin (1987) show that consumers’ evaluations and buying decisions can be supported but also hampered by providing additional information depending on the level of information load. Participants in their study were confronted with 17 different attributes and a level of attributes around seven was found to be the optimal number. Furthermore, they claim that attribute information should be differentiated in information quality and information quantity and examine the influence of those two aspects on decision effectiveness. Their findings indicate opposing effects of the two aspects; information quality improves decision effectiveness while information quantity decreases decision effectiveness. Nevertheless, it could be shown that the relationship of information quantity and decision effectiveness can be depicted by an inverted U shape. Thus, this study also finds support for the fact that additional information during the decision making process is only supportive to a certain extend and that a high quantity of information leads to less effective decisions due to the phenomenon of information overload (Keller and Staelin 1987).
Furthermore, the traditional approach by Jacoby, Speller, and Kohn (1974) also was criticized by Helgeson and Ursic (1993) for not being complete. According to these authors, there are not only the two task effects, number of alternatives and attributes, but also two context effects that create information overload, namely the variability of information on the attributes and alternative similarity. Hence, those findings are partly contradicting or demand an extension of the earlier results of Jacoby, Speller, and Kohn (1974) by further aspects which supports the fact that there is an inconsistency in the research concerning information overload.
influenced by the fact whether consumers have prior attribute preferences, meaning that they know what kind of product features they are looking for, or not. A large assortment only leads to more satisfaction if consumers do have prior preferences. In this case, the greater number of alternatives to choose from is not perceived as overwhelming due to the prior preferences and might increase the probability of selecting a product that creates satisfaction.
Furthermore, Greifeneder, Scheibehenne, and Kleber (2010) conducted a study on choice complexity, which is related to a large assortment size as it describes the situation when too many options lead to negative consequences, like less satisfaction with the choice. Additionally, the similarity of alternatives and the amount of information provided are also pointed out as being of great influence. By doing so, the authors (Greifeneder, Scheibehenne, and Kleber 2010) deal with Helgeson and Ursic’s (1993) critique concerning the traditional approach of information overload that has been mentioned above. In line with what has been stated before, also this study found that when alternatives were different with regards to many attributes, there was a high choice complexity which led to information overload.
2.3 Brand Choice
Since this thesis aims at analyzing how additional information provided by QR codes influences brand choice, it is essential to comprehend how in-store buying decisions are made. Additionally, it is important to find out which role branding plays in this process as there is a current trend of increasing influence of private labels in consumer goods markets (Baltas 1997). This raises the question, if and how additional information provided by certain brands is able to affect customers’ decision-making.
A customer’s choice for a certain product could be influenced by strong brands due to the fact that they serve as a form of guidance in in-store settings (“Why did you buy that?” 2011). By implementing a consistent communication and presenting customers what they expect, they can be influenced in choosing a familiar brand. In addition to expectations, factors like price, packaging and quality also influence buying decisions. However, those aspects give an explanation for rational buying decisions. Opposing to that, brand choices can also be created through unconscious triggers which lead to an automatic emotional response. If marketing activities are implemented successfully, they trigger implicit feelings and automatic emotional responses which can then lead to a rational buying decision (“Why did you buy that?” 2011).
According to Erdem et al. (1999), there is no specific definition of consumer decision-making. However, they suggest a multi-staged and dynamic characterization of consumer choice processes. Customers are confronted with product attributes of which parts are encoded and become part of the memory during a learning process. These memories can be retrieved again in the future, for example in a decision-making situation as in a buying decision. This decision is then made depending on the utility that can be ascribed to the attribute representation. Hence, this process can be applied for single products or services but also for products in a category which are part of a customer’s choice set (Erdem et al. 1999).
Besides the price, brand credibility affects consideration and also brand choice. The two main components of credibility are trustworthiness and expertise. More specifically, “brand credibility is defined as the believability of the product information contained in a brand, which requires that consumers perceive that the brand has the ability (expertise) and willingness (trustworthiness) to continuously deliver what has been promised,” (Erdem and Swait 2004, p. 192). Moreover, credibility affects consumer choices through perceived risk, information costs saved and perceived quality (Erdem and Swait 2004). This is in line with Aaker (1991), who argues that higher perceived quality, lower information costs and risk associated with credible brands will increase consumer evaluations of a brand and might also influence the brand choice.
Thus, one can see that the costs to retrieve information concerning a product and also the information itself, which is for example used to communicate the product quality and product features, have an influence on brand choice. However, it is not clear what impact the quantity of information provided has on the buying decision. Hence, is it possible that providing more information might lead to choice difficulty and influence the commitment to a brand or even lead to brand switching? As the research of this thesis focuses on planned purchasing, it is being assumed that additional information concerning the product that a customer intends to buy will support this decision. Hence, additional product information is expected to have a positive effect on brand choice. Besides that, does the possibility to receive additional information provided by a QR code influence customers’ brand choice or can brand choice only be affected if customers actually are able to read this information? In order to address these questions, the following hypothesis can be derived:
H1a: There is a positive influence of providing the possibility to obtain additional product information by using QR codes on brand choice.
H1b: There is a positive influence of being exposed to additional product information by using QR codes on brand choice.
2.4 Perceived Product Involvement
According to Zaichkowsky (1986), three types of involvement can be classified, namely involvement with advertisements, with products and with purchase decisions. In addition to that, personal factors, object or stimulus factors and situational factors can be identified as antecedents of involvement. In the following, the focus will lie on product involvement which can be defined as “A person’s perceived relevance of the object based on inherent needs, values, and interests.” (Zaichkowsky 1985, p. 343). Thus, if a product is of low or high involvement, depends mainly on the relevance or importance that is attributed to it by an individual customer (Hupfer and Gardner 1971; Traylor 1981; McQuarrie and Munson 1992).Due to this, the involvement with certain products or even categories cannot be defined since individual differences and perceptions are decisive in determining product involvement (Richins and Bloch 1986). As a consequence, the variable that will be examined in the course of this thesis will be the perceived product involvement.
The most common approach that is used to explain the role of involvement is the so called Elaboration Likelihood Model of persuasion which has been developed by Petty, Cacioppo, and Schuhmann (1983). The authors argue that there are two possibilities to process information or stimuli in order to develop an attitude. On the one hand, information can be processed via the central route by thorough consideration and high effort. On the other hand, peripheral route processing takes place on the bases of heuristics as well as superficial cues under low effort (Petty, Cacioppo, and Schuhmann 1983). Hence, the individual amount of effort or involvement plays an important role in the way information is processed and attitudes are developed.
Further existing research showed that product involvement can affect customers’ commitment to a brand (Traylor 1981) so one might also argue that it can have an influence on brand choice. When the involvement is low, consumers are more likely to choose a different brand than they initially intended to buy. In contrast to that, when the involvement is high, customers have a higher motivation to look for and also process product related information. Additionally, they show a higher brand loyalty which makes brand switching less likely but also strengthens commitment to the brand (Traylor 1981; Warrington and Shim 2000). Product involvement seems to have a moderating effect on consumer behavior. As mentioned before, this effect will be examined by using two levels of product information: first of all, the influence of the possibility to receive additional product information and second of all, the influence of actually being opposed to this additional information. In line with that, the following hypotheses can be derived:
H2b: When the perceived product involvement is high, the influence of having the possibility to obtain additional product information provided by a QR code on brand choice is stronger than when the perceived product involvement is low.
H3a: When the perceived product involvement is low, the influence of being exposed to additional product information provided by a QR code on brand choice is weaker than when the perceived product involvement is high.
H3b: When the perceived product involvement is high, the influence of being exposed to additional product information provided by a QR code on brand choice is stronger than when the perceived product involvement is low.
2.5 Conceptual Model
The relationships which have been explained in the literature review above and which will be tested in the course of this thesis are depicted in the conceptual model below.
Figure 1: Conceptual Model
First of all, the goal of this thesis is to analyze whether additional product information that is provided to customers via QR codes as a form of mobile marketing will have an influence on a customer’s brand choice. It is being assumed that there is a positive relationship; providing more information will facilitate the decision-making process. More specifically, it will be examined whether only the fact that a QR code is present on a package will have an influence on brand choice, on the one hand. On the other hand, it will be analyzed if exposing participants to this additional information also has a positive influence on their brand choice.
Furthermore, the perceived product involvement is expected to moderate this effect. If the perceived involvement is low, the effect will be weaker as additional information is not of great importance for these customers. However, the effect will be stronger for high perceived involvement due to the higher need for information. Also here, the two conditions of only offering the QR code or also being able to read the additional information will be examined.
3 Methodology
3.1 Study Design
In order to examine the assumed effects and relationships, the independent variable consists of three levels and the moderator, perceived involvement, has two levels. Therefore, the study’s design can be described as a 3x2 full factorial design (see table 1 below). As each level of the independent variable will be tested on both levels of the moderator, there will be six experimental groups in this study which will be analyzed and compared in chapter 4 of this thesis.
Perceived Product Involvement low high Additional Information
via Mobile Marketing
No additional information Group 1 Group 4
Possibility to obtain additional information Group 2 Group 5 Exposure to additional information Group 3 Group 6 Table 1: Study Design
3.2 Procedure and Participants
The study described above was conducted as an online experiment using the survey tool Qualtrics. Participants were invited to complete the survey by personal Emails and messages as well as by sharing the survey link via social media like Facebook. In addition to that, participants were kindly asked to share and distribute the survey link in their personal networks to achieve a snowball effect and reach a greater variety of respondents.
In the introduction part of the survey, participants were told that the goal of the survey is to examine the influence of mobile marketing on in-store shopping behavior without providing further information on the content in order to not bias the responses. Additionally, participants were told that their participation is anonymous and that the data will be exclusively used for this master thesis. With regards to the experimental groups mentioned above, participants have been randomly assigned to one of these groups.
Before the survey link was distributed, a pretest has been conducted to make sure that the tool works and that the settings are selected in the best way. This pretest included 15 participants, which were almost all students and the majority indicated to be between 20 and 34 years old.
3.3 Manipulations
depending on the experimental group they have been assigned to. In the control group, participants were exposed to a picture of a product and were told that they usually buy this product (product a). In addition to that, they were shown a second brand offering the same product (product b) and were asked which of the products they would buy.
In the second group, participants were also exposed to a picture of a product and were told that they usually buy this brand (product a). In addition to that, they saw a second brand that offered a QR code on the product (product b) but participants in this group were told that they do not scan the code. So they knew that there was the possibility to obtain additional information but did not actually see this information. Afterwards, they were also asked which of the two products they would buy. In contrast to that, participants in the third group have been told the same as the second group except for the fact that that they were actually exposed to the additional information behind the QR code. In the following, they were asked to indicate their choice as well.
In order to manipulate the moderator, which is the perceived involvement with the product, participants were shown different product categories, depending on the experimental group they were randomly assigned to. In the low perceived involvement condition, the product that has been used is frozen vegetables. For high perceived involvement, participants were asked to choose between different laptops as the interest and relevance in choosing such product types is considered to be high. To check if the manipulation works, an independent samples t-test was performed with the results of the pretest. As the test was significant (p-value = 0.011 < 0.05) and the average perceived involvement of participants in the low involvement groups (M = 3.17, SD = 1.130) was significantly different compared to the average of participants in the high involvement groups (M = 4.72, SD = 0.443), it can be concluded that the manipulation was successful.
3.4 Measurement
In the following section, the scales that have been used to measure the relevant constructs will be briefly presented.
services. They consist of seven-point Likert-scales with three items measuring perceived usefulness and four items measuring ease of use. In order to get comparable results, a five-point Likert scale was used which is in accordance with the scales for the other tested constructs.
Before the participants were exposed to the different levels of product information, they were asked to rate statements concerning their involvement with the respective product. The scale that has been used was developed by Chandrashekaran (2004) and consists of three items which measure interest and relevance by using a five-point Likert-scale (ranging from strongly agree to strongly disagree).
After the manipulation, brand choice was tested by simply asking participants to choose product a (product which they had intended to buy) or b (manipulation) as there is no scale in existing research for this variable. Additionally, the survey included questions concerning choice difficulty, commitment to a brand and brand switching in order to gain more insights in consumer’s decision making when choosing a product. For the construct choice difficulty, a scale that has been used in previous research by Laroche et al. (2005), who adapted it from a scale by Breivik, Troye, and Olsson (1999), was applied. The scale measures the extent to which choosing between two brands is seen to be demanding by using four five-point semantic differentials. To measure commitment to a brand, a scale composed of three items and a 5-point Likert scale that has previously been applied by Coulter, Price, and Feick (2003) has been used. In order to measure the likeliness of brand switching, which is also closely related to brand choice, three items have been used on a 5-point Likert scale as well. Also in this case, the scale has been derived and adapted from previous studies, namely research by Raju (1980) as well as Moore-Shay and Lutz (1989).
4 Results
In the next section, the overall sample as well as the six experiment groups will be described before the procedure and the results of the data analysis will be presented with regards to the proposed hypotheses.
4.1 Descriptive Statistics
34 years old as it can be seen in figure 2 below. The overall sample consists of participants from 29 different countries, the biggest part is from Germany (49.8%), followed by the Netherlands (20.5%) and the USA (7.3%).
Figure 2: Distribution of Age in Years
Furthermore, most of the participants were students and indicated to dispose over a rather low income. Nevertheless, the sample also included participants representing the other categories of occupation and income (see figures 3 and 4 below).
Figure 3: Distribution of Occupation Figure 4: Distribution of Income
Figure 5: Smartphone Usage in Everyday Life and Shopping Situations
Moreover, participants were asked about their acceptance of using QR codes before the manipulations. In this case, descriptive analysis shows that the acceptance is rather positive, as 42% agree or even strongly agree with the corresponding statements. Nevertheless, it is also important to mention that 47.8% stated to neither disagree nor agree which could be an indicator for the fact that they do not have previous experience with using QR codes.
In the following, the six experimental groups will be described. First of all, the groups are of comparable size with between 31 and 37 participants per group. As for the overall sample, also the number of female participants is slightly higher than the number of male participants. Besides in group six, which only has one more female than male participants, the relative distribution of gender is equal among the groups (see table 2).
Table 2: Overview Experimental Groups
In addition to that, the majority of all groups indicated to be between 20 and 34 years old (see figure 6 below). As only a small proportion of the overall sample indicated to be in the groups of 50 to 64 and over 64 years old, it might be the case that age is a confounding variable. However, as the average age of the respective groups (see table 2) are close to the average age of the overall sample (1.99), it can be concluded that this is not the case.
Group 1 Group 2 Group 3 Group 4 Group 5 Group 6
Group Size 37 31 37 35 32 33 Gender Female I Male 20 17 18 13 22 15 20 15 19 13 17 16 Average Age 2.22 1.83 1.86 1.94 1.94 2.09 6.3% 1.5% 9.3% 26.3% 56.6% 23.4% 21.5% 31.7% 20.0% 3.4% 0% 10% 20% 30% 40% 50% 60%
very seldom seldom sometimes often very often
Figure 6: Distribution of Age in Years per Group
4.2 Preparing for Analysis
Before the hypothesized effects can be analyzed, it should be ensured whether the used questions measure the respective constructs. In order to do so, correlations, reliability tests and factor analyses have been performed for the constructs acceptance (perceived usefulness and ease of use), involvement, choice difficulty, commitment as well as brand switching.
The correlation analyses showed that all items correlate significantly with each other (p-values < 0.05) for the respective constructs. In addition to that, the reliability analysis resulted in Cronbach’s alphas above 0.7 for all used items, which speaks for a high internal consistency. As appendix 1 shows, the Cronbach’s alphas range from 0.737 to 0.916, so they are all above the critical value of 0.7. The alpha increases if one item is deleted only in one case, namely for item three of involvement. Besides this exception, the alpha does not increase when deleting items. These results indicate that the used items can be used as a sum variable of the respective constructs. Furthermore, factor analyses resulted in Kaiser-Meyer-Olkin Measures above 0.5, which confirms the appropriateness of this analysis. As the items loaded high on the respective constructs, they have been combined in order to reduce complexity for the further analysis.
4.3 Manipulation Check
In order to see if the manipulation was successful, meaning that frozen vegetables were rated as lower perceived involvement than a laptop, an independent samples t-test with the manipulation of the experimental groups and the perceived involvement was performed. The results show that this test was significant with a p-value of 0.000 which is lower than 0.05. The average perceived involvement of participants in the low involvement experimental groups (M=2.92, SD=0.769) does significantly differ from the average perceived involvement of participants in the high involvement
10.8% 18.9% 22.9% 18.8% 15.2% 73.0% 71.0% 78.4% 68.6% 75.0% 69.7% 6.1% 16.2% 2.7% 8.6% 6.3% 9.1% 3.2% 0% 10% 20% 30% 40% 50% 60% 70% 80% 90%
Group 1 Group 2 Group 3 Group 4 Group 5 Group 6
experimental groups (M=4.13, SD=0.660). Moreover, as the mean was significantly higher in the high involvement groups, participants in these groups did actually perceive the involvement with the respective products to be higher (see appendix 2). Thus, it can be concluded that the manipulation was successful.
4.4 Analysis of Main and Moderating Effects
In the following paragraphs, the results of analyzing the main effect of providing additional information via mobile marketing on brand choice as well as on choice difficulty, commitment and brand switching will be presented before the results of the moderation analyses will be explained. With regards to the nominal dependent variable brand choice, a Chi-square test has been performed to examine if it is affected by the different levels of the independent variable. As a result, the test showed that there is a statistically significant relationship between the information provided via mobile marketing and brand choice (p Pearson-Chi-Sqaure = 0.000). Figure 7 below shows that the difference between participants who chose product a and product b is rather low in the control group (no additional information). In the second group, where participants saw a further product which had a QR code but were not exposed to the additional information, the majority of participants chose the product they intended to buy (product a). In contrast to that, the majority of the participants in the third group, who were also exposed to the additional information, chose the product that contained the QR code (product b) over the product which they intended to buy (product a).
Figure 7: Results Chi-Square-Test
As a consequence, it could be shown that there is a direct effect of the information provided via QR codes on brand choice only if participants were exposed to the additional information. Thus, hypotheses 1a cannot be supported as the majority of participants in the second group (45 out of 73) stuck to the product they intended to buy so their brand choice was not influenced by the QR code.
H1a: There is a positive influence of providing the possibility to obtain additional product information by using QR codes on brand choice.
Accordingly, hypotheses 1b can be supported. In the experimental groups that were also exposed to the additional information that was provided by a QR code, the decision of 44 out of 70 participants was influenced by this information as they chose the product with the QR code over the one without it. Hence, there is a positive effect on brand choice that in this case.
H1b: There is a positive influence of being exposed to additional product information by using QR codes on brand choice.
Furthermore, three one-way analyses of variance (ANOVA) were performed for the dependent variables choice difficulty, commitment and brand switching. Also here, the purpose is to examine if the amount of information the participants were exposed to has an influence on their decision making. Table 3 below shows that the p-values for these three tests were above 0.05 which means that there are no significant differences between the means of the experimental groups (also see appendix 3). Thus, it can be stated that there is no significant effect of providing additional information via mobile marketing on the aspects choice difficulty, commitment and brand switching of customers in in-store situations.
Dependent Variable Sig. Choice Difficulty 0.773
Commitment 0.608
Brand Switching 0.520
Table 3: Overview Results One-Way ANOVA
can be seen when looking at the odds ratios. For both group 2 and group 3. this ratio is below 1. This means that the probability of choosing product a when the participants either have the possibility to obtain more information or are exposed to this information is lower than when there is no additional information since this is the reference category. As the odds ratio is even lower for group 3 (0.229) than for group 2 (0.483), this shows that the effect is even stronger; the probability of participants choosing product a over b is lower here. This is also in line with the findings of the Chi-square test presented before which has been performed to examine the direct effect.
B Sig. Exp(B) Group 1 0.000 Group 2 -0.728 0.036 0.483 Group 3 -1.473 0.000 0.229 Involvement 0.280 0.265 0.755 Group 2 * Involvement 0.510 0.141 1.666 Group 3 * Involvement 0.459 0.223 1.582 Constant 0.552 0.028 1.737
Table 4: Results of Variables in the Equation
However, the direct and the interaction effect of involvement are not significant as the p-values are above the critical value of 0.05 for these variables (see table 4). As a conclusion, it can be noted that perceived product involvement neither has a direct effect on brand choice nor moderates the effect of additional information on brand choice. Consequently, hypotheses 2a, 2b, 3a and 3b cannot be supported.
H2a: When the perceived product involvement is low, the influence of having the possibility to obtain additional product information provided by a QR code on brand choice is weaker than when the perceived product involvement is high.
H2b: When the perceived product involvement is high, the influence of having the possibility to obtain additional product information provided by a QR code on brand choice is stronger than when the perceived product involvement is low.
H3a: When the perceived product involvement is low, the influence of being exposed to additional product information provided by a QR code on brand choice is weaker than when the perceived product involvement is high.
Thus, when trying to influence customers’ brand choice by attracting them to a different product they intended to buy with a QR code, it does not make a difference whether this product or even product category is being perceived as low or high involvement.
In order to examine the influence of perceived involvement as a potential moderator for the dependent variables choice difficulty and commitment, two factorial ANOVAs have been performed. With regards to the dependent variable choice difficulty, the model explains 15.6% of the variance (see appendix 5). Furthermore, the Levene’s test for equality of variances is significant (p = 0.002). This means that the null hypotheses, stating that the variances in the groups are equal, can be rejected. Hence, the variances of the groups are all close to the mean. However, this is in line with the further results of the ANOVA, which will be presented in the following, showing that the groups do not significantly influence the dependent variable. The results of the between-subjects effects for the dependent variable choice difficulty show that the overall model is significant (F = 7.361, p = 0.000). However, there is no significant direct effect of the independent variable, providing additional information via mobile marketing, on choice difficulty (F = 0.213, p = 0.809). Also, the overall test results of the custom hypotheses tests, which were performed to compare the groups, indicate that this test is not significant (p = 0.809). In line with that, the contrasts between group 1 and 2 (p = 0.748) and between group 1 and 3 (p = 0.515) are not significant. These results are in line with the outcome of the ANOVAs explained above. Furthermore, the interaction effect is also not significant (F = 0.301, p = 0.740). Yet, the influence of perceived involvement on choice difficulty is statistically significant with a p-value of 0.000. Consequently, it can be stated that there is a direct effect of perceived involvement on choice difficulty but no moderating effect on the relationship of groups on choice difficulty. Due to a B coefficient of 0.845 for low involvement, there is a positive relationship: if the involvement increases by one unit, choice difficulty is predicted to change by 0.845. Thus, it can be stated that if the level of involvement is increased from low to high, the choice difficulty will also increase.
19.299, p = 0.000). However, the coefficient for low involvement is negative (-0.790) for commitment so there is a negative relation: if the involvement increases from low to high, the level of commitment is predicted to decrease.
To conclude, it can be noted that the analyses of the moderation effect of choice difficulty and commitment also cannot find support for hypotheses 2a, 2b, 3a and 3b so there is no interaction effect of involvement on any of the examined dependent variables. An overview of the outcome of all tested hypotheses can be found in table 5 below.
H1a: There is a positive influence of providing the possibility to obtain additional
product information by using QR codes on brand choice. x
H1b: There is a positive influence of being exposed to additional product information
by using QR codes on brand choice.
H2a: When the perceived product involvement is low, the influence of having the possibility to obtain additional product information provided by a QR code on brand choice is weaker than when the perceived product involvement is high.
x
H2b: When the perceived product involvement is high, the influence of having the possibility to obtain additional product information provided by a QR code on brand choice is stronger than when the perceived product involvement is low.
x
H3a: When the perceived product involvement is low, the influence of being exposed to additional product information provided by a QR code on brand choice is weaker than when the perceived product involvement is high.
x
H3b: When the perceived product involvement is high, the influence of being exposed to additional product information provided by a QR code on brand choice is stronger than when the perceived product involvement is low.
x
Table 5: Results of Tested Hypotheses
5 Discussion
After having presented the results of analyzing the main and interaction effect, they will be discussed in the following section with regards to the research questions.
The overall goal of this thesis was to examine whether additional product information, which is provided via a QR code, can influence customers’ in-store buying behavior. More specifically, it has been differentiated between the influence of having the possibility to access additional information by providing a QR code and the exposure to this information offered by the QR code. First of all, it could be shown that the brand choice was influenced when customers were actually being exposed to the additional product information. More specifically, it could be shown that only 18 out of 63 participants that only had the possibility to obtain additional information chose this product whereas 45 participants, which is the majority, stuck to the product they intended to buy. On the other hand, 44 out of 72 participants in the experimental group that was also exposed to this information chose the product that offered a QR code over the one they intended to buy. Hence, there is an opposing effect between the participants decision behavior in experimental groups 2 and 3 due to the amount of information that was provided via the QR code. With regards to the proposed research questions, it can be noted that QR codes are only effective in influencing brand choice if customers also read the additional information that is captured by them. Due to this, it is especially important that brands or companies do not only put effort in the process of adding QR codes to their product package designs, but even more promote the usage of such codes. This aspect will be further explained in the previous section of the thesis.
experiment was additional product related information, like technical specifications, concerning a particular product. As the participants intended to buy the respective product anyways, it could be assumed that this additional information facilitated the decision making process in this case due to the fact that it provided supportive arguments for it. Nevertheless, the options to choose from, which are referred to as the assortment size, were not changed by the information provided in the experiment. Another possibility that might have led to different results would have been confronting participants with information that provides additional product alternatives. For example, participants in the high involvement groups could have been exposed to different design options the laptop brand offers. By promoting a product in different colors or warranty service options, QR codes could also be used to increase alternatives to choose from. In this case, the increased assortment size might lead to an inertia to choose. Thus, it is important to distinguish between different kinds of information when examining possible effects on consumer behavior and decision making.
In addition to the influence of QR codes on brand choice, also the direct effect on choice difficulty, commitment and brand switching have been analyzed. However, as these direct effects could not be confirmed, these aspects cannot be used to support the findings concerning the influence on customers’ decision making when choosing a product. A possible reason for this might be that the concepts of choice difficulty, commitment and brand switching are not influenced by the amount of information a customer is exposed to during the decision process of a planned purchase in in-store situations but is more affected by other aspects. For example, commitment to a brand could be more affected by brand trust and the level of satisfaction a consumer has due to these aspects (Delgado‐Ballester and Munuera-Alemán 2001; Gustafsson, Johnson, and Ross 2005). Brand switching could be essentially influenced by stimuli at a different point in the buying cycle other than at the point of sales, for example by advertisements of competitive brands which can lead to a different decision making, thus to brand switching. However, especially for choice difficulty, this result is other than expected as the additional information could also confuse consumers while making decisions as it has been mentioned before. Since this aspect could not be approved with this research, this might be a further indicator for the fact that additional information concerning a specific product might actually be helpful instead of hindering when it comes to the decision making process.
not only be considered for high involvement products but can also influence customers decision making with low involvement. This aspect can be further supported with the findings as a direct effect of involvement on the variables choice difficulty and commitment was found. Also here, the effects have opposing directions as the relationship of involvement and choice difficulty is positive, whereas the effect on commitment was found to be negative. Hence, when the involvement increases from low to high, the choice difficulty is also expected to increase due to the positive relationship of the variables. This means that choosing a high involvement product can be more challenging than selecting a low involvement product. Yet, when the involvement increases from low to high, the commitment is expected to decrease which, in turn means that the level of commitment is rather high in the case of low involvement. This might be due to the fact that high involvement products are purchased with a lower frequency than low involvement products. Regarding the latter, it might be the case that customers buy certain low involvement products on a regular base and tend to choose the same brand due to habitual purchases. So in this case, the commitment would be high even though the involvement with the product is rather low. When considering high involvement products, it can be assumed that these are bought less often and that customers make decisions more thorough fully by, for example, taking a comparatively higher number of alternatives into consideration as when deciding on a low involvement product. This might lead to the fact that they decide to purchase a product from another brand than the last time they bought such a product and, thus, a lower level of commitment is prevalent. Hence, one could assume that customers with high involvement are more likely to be influenced in their purchasing decision.
6 Conclusion
7 Implications
One of the main findings of this study is that there is an influence of additional product information on brand choice when customers are also exposed to this information. Due to this, it is an important implication that customers should not only notice the QR code but should also be aware of how to use those codes and be willing to do so in order to be able to influence the brand choice. Moreover, as this additional information can change customers’ decisions, QR codes can also be used as a tool to attract new customers that bought competitors’ products in the past. Therefore, companies that implement such codes should not only make the effort to integrate them in the product design, but even more also create additional value for customers to scan the QR codes and communicate the advantages that come with it. An example of how QR codes were successfully implemented is the American consumer electronics retailer Best Buy who added the codes on price tags next to their product. When scanning the codes, customers are forwarded to price comparisons with other retailers. By offering this service, the retailer creates transparency and directly faces competition in order to promote a low price guarantee and increase sales (Schottmuller 2011). As almost half of the American smartphone owners make use of their mobile devices to compare prices in shopping situations, this might be a promising strategy (Kagan 2011). Beyond that, this strategy does not focus on manufactures, who usually add QR codes to their product packages, but addresses retailers. It is a possibility for marketing managers of retail stores to implement the relatively new technology, to build loyalty by communicating in a very transparent way and thereby increasing sales.
A further strategy to motivate customers to use QR codes is by offering price promotions. First of all, this will create attention as customers are price sensitive and, second of all, serves as a means to introduce QR codes to customers. One can assume that customers are more willing to make use of such codes if they are motivated to overcome the barrier of new technology and gain first experience. For example, customers can be invited to like a store on Facebook via the code and will be offered a special coupon as well as information about upcoming discounts when doing so.
8 Limitations and Future Research
There are some limiting aspects in this study that can be improved in the future and serve as indicators for further research. First of all, the overall sample size could have been bigger and could also include more variety in the participants’ age. Even though there were participants representing every age group in each of the six experimental groups, the majority of each group indicated to be between 20 and 34 years old. Thus, the results are mainly limited to students and young people and do not deliver a representation of the overall population.
In addition to that, product involvement played a major role in this research. However, participants have only been exposed to one product during the online experiment which was perceived either as low or high involvement. In order to be able to make holistic statements about the influence of product involvement on customer choices, it might be helpful to conduct this experiment with several different products in the future. Furthermore, future research could also focus on gaining more insights concerning the broader influence of product categories and differing preferences of target segments. For example, it could be tested if customers are in general more willing to use QR codes when buying cosmetics products or groceries to derive further implications and recommendations for practitioners.
emotional outcomes. Hence, one could test whether providing additional information via mobile marketing can lead to more satisfied customers and therefore also increase loyalty.
References
Aaker, David A. (1991), Managing Brand Equity. New York: Free Press.
Andrews, Lynda, Rebekah Bennett, and Judy Drennan (2005), "Mobile Phone Consumption and implications for SMS Marketing," Rejuvenating Marketing: Contamination, Innovation,
Integration: Proceedings of the 34th European Marketing Academy Conference.
Ashford, Robin (2010), "QR Codes and Academic Libraries Reaching Mobile Users," College &
Research Libraries News, 71 (Nov.), 526-30.
Baltas, George (1997), "Determinants of Store Brand Choice: A Behavioral Analysis," Journal of
Product & Brand Management, 6 (5), 315-24.
Barnes, Stuart J. and Eusebio Scornavacca (2004), "Mobile Marketing: The Role of Permission and acceptance," International Journal of Mobile Communications, 2 (2), 128-39.
Barutçu, Süleyman (2007), "Attitudes Towards Mobile Marketing Tools: A Study of Turkish Consumers," Journal of Targeting, Measurement and Analysis for Marketing, 16 (Dec.), 26-38.
Breivik, Einar, Sigurd V. Troye, and Ulf H. Olsson (1999), "Dimensions of Product Intangibility and their Impact on Product Evaluation," Advances in Consumer research, 26, 264.
Celsi, Richard L. and Jerry C. Olson (1988), "The Role of Involvement in Attention and Comprehension Processes," Journal of Consumer Research, 15 (2), 210-24.
Chandrashekaran, Rajesh (2004), "The Influence of Redundant Comparison Prices and Other Price Presentation Formats on Consumers’ Evaluations and Purchase Intentions," Journal of Retailing, 80 (1), 53-66.
Chernev, Alexander (2003), "Product Assortment and Individual Decision Processes," Journal of
Coulter, Robin A., Linda L. Price, and Lawrence Feick (2003), "Rethinking the Origins of Involvement and Brand Commitment: Insights from Postsocialist Central Europe," Journal of Consumer Research, 30 (2), 151-69.
Davis, Fred D. (1989), "Perceived Usefulness, Perceived Ease of use, and User Acceptance of Information Technology," MIS Quarterly, 13 (3), 319-40.
---, Richard P. Bagozzi, and Paul R. Warshaw (1989), "User Acceptance of Computer Technology: A Comparison of Two Theoretical Models," Management Science, 35 (8), 982-1003.
Delgado‐Ballester, Elena and José L. Munuera-Alemán (2001), "Brand Trust in the Context of Consumer Loyalty," European Journal of Marketing, 35 (11), 1238-58.
Dholakia, Utpal M. (2001), "A Motivational Process Model of Product Involvement and Consumer Risk Perception," European Journal of Marketing, 35 (11), 1340-62.
Dou, Xue and Hairong Li (2008), "Creative Use of QR Codes in Consumer Communication,"
International Journal of Mobile Marketing, 3 (2), 61-67.
Eppler, Martin J. and Jeanne Mengis (2004), “The Concept of Information Overload: A Review of Literature,” Organization Science, Accounting, Marketing, MIS, and Related Disciplines, The
Information Society: An International Journal, 20 (5), 325-44.
Erdem, Tülin, Susan Broniarczyk, Dipankar Charavarti, Jean-Noel Kapferer, Michael Keane, John Roberts, Jan-Benedict Steenkamp, Joffre Swait, and Florian Zettelmeyer (1999), "Brand Equity, Consumer Learning and Choice," Marketing Letters, 10 (3), 301-18.
Erdem, Tülin and Joffre Swait (2004), "Brand Credibility, Brand Consideration, and Choice," Journal of
Falk, Tomas, Jeroen Schepers, Maik Hammerschmidt, and Hans H. Bauer (2007), "Identifying Cross-Channel Dissynergies for Multichannel Service Providers," Journal of Service Research, 10 (2), 143-60.
Fasolo, Barbara, Gary H. McClelland, and Peter M. Todd (2007), "Escaping the Tyranny of Choice: When Fewer Attributes make Choice Easier," Marketing Theory, 7 (1), 13-26.
Gao, Tao, Fareena Sultan, and Andrew J. Rohm (2010), "Factors Influencing Chinese Youth Consumers' Acceptance of Mobile Marketing," Journal of Consumer Marketing, 27 (7), 574-83.
Goldberg, Jason (2010), “Best Buy Adds QR Codes to Product Fact Tags,” (accessed October 13. 2014), [available at http://www.retailcustomerexperience.com/articles/best-buy-adds-qr-codes-to-product-fact-tags/].
Grant, Ian and Stephanie O'Donohoe (2007), "Why Young Consumers are Not Open to Mobile Marketing Communications," International Journal of Advertising, 26 (2), 223-46.
Greifeneder, Rainer, Benjamin Scheibehenne, and Nina Kleber (2010), "Less may be More when Choosing is Difficult: Choice Complexity and Too Much Choice," Acta Psychologica, 133 (1), 45-50.
Gustafsson, Anders, Michael D. Johnson, and Inger Roos (2005), "The Effects of Customer Satisfaction, Relationship Commitment Dimensions, and Triggers on Customer Retention," Journal of
Marketing, 69 (2), 210-8.
Hardie, Bruce G., Eric J. Johnson, and Peter S. Fader (1993), "Modeling Loss Aversion and Reference Dependence Effects on Brand Choice," Marketing Science, 12 (4), 378-94.
Helgeson, James G. and Michael L. Ursic (1993), "Information Load, cost/benefit Assessment and Decision Strategy Variability," Journal of the Academy of Marketing Science, 21 (1), 13-20.
