• No results found

How important consumers perceive online and offline recommendations across different product categories : what are the factors that are associated with differences in evaluation?

N/A
N/A
Protected

Academic year: 2021

Share "How important consumers perceive online and offline recommendations across different product categories : what are the factors that are associated with differences in evaluation?"

Copied!
79
0
0

Bezig met laden.... (Bekijk nu de volledige tekst)

Hele tekst

(1)

MASTER THESIS

How important consumers perceive online and offline recommendations across different product categories.

What are the factors that are associated with differences in evaluation?

University of Amsterdam Faculty of Economics and Business

Master of Science in Business Studies Track: Marketing

Under supervision of: MSc. Umut Konuş Second supervisor: MSc. E. Korkmaz

By:

Student: Milou de Sain Student number: 10420886

(2)

2

ABSTRACT

The Internet and social media have grown intensively (Cheung, Lee, & Rabjohn, 2008). And through electronic word-of-mouth (eWOM), consumers can share their opinions to a wider public than ever before. This sharing of eWOM has become a major information source for consumer purchase decisions (Park, Gu, & Lee, 2012). Recommendations are types of (e)WOM communication and are product-related conversations, which can have either a positive or a negative impact on consumer decision-making, and play a role when considering buying a new product (Arndt, 1967). The increase in Internet use and a change in use of information sources creates new challenges for research in understanding how consumers search for information and what the impact of recommendations is on their purchase behaviour. Current study examined whether, and to what extent, online and offline recommendations might differ in perceived importance across five product categories (cars, home electronics, clothing, holidays and restaurants), and it investigates the association among these recommendations and three demographic variables (gender, age and income) and three personal characteristics (consumer novelty seeking, attention to social comparison and buying impulsiveness). Using survey data from 156 Dutch consumers, it was identified that offline recommendations were perceived as more important for the product categories cars, clothing and restaurants. Online recommendations were perceived as more important for the product category home electronics. Furthermore, demographic and personal characteristics are shown to be associated with the perceived importance of online and offline recommendations. This study concludes with managerial implications of the results and suggestions for directions for future research.

Keywords: Online shopping behaviour, (e)WOM and recommendations, consumers’

(3)

3

ACKNOWLEDGEMENT

I would like to thank several individuals for their guidance and help to support me in writing this thesis. First, I would like to acknowledge Professor Mr. Dr. Umut Konuş for all his support and guidance that he provided during this thesis development process. Furthermore, I would like to thank my friends and family who always supported me and gave me a helping hand when needed. This thesis would not have been possible without their support. At last, my thesis is written.

I hope you enjoy reading this thesis.

(4)

4

TABLE OF CONTENT

1. Introduction ... 7

2. Literature review ... 12

2.1 Online shopping behaviour ... 12

2.2 (e)WOM and recommendations ... 13

2.3 Consumers’ perceived importance ... 22

2.4 Product Categories ... 26

2.5 Associations between of demographic variables and recommendations ... 33

2.6 Associations between personal characteristics and recommendations ... 34

2.7 Conceptual framework ... 36 3. Methodology ... 37 3.1 The sample ... 37 3.2 Research design ... 38 3.2.1 Measures ... 38 3.3 The procedure ... 41 3.3.1 Pilot study ... 41 3.3.2 Main study ... 41 4. Results ... 42 4.1 Reliability ... 42 4.2 Correlation check ... 43 4.3 Model testing ... 45 5. Discussion ... 58 5.1 Academic relevance ... 61 5.2 Managerial implications... 61

(5)

5

5.3 Limitations and further research ... 63

6. Conclusion ... 64

References ... 66

Appendix ... 71

I Questionnaire ... 71

II Results hypotheses testing ... 73

(6)

6

INDEX OF TABLES

Table 1: Recommendation sources ... 22

Table 2: Overview product categories ... 30

Table 4: Reliability of scales ... 43

Table 5: Correlations recommendations ... 44

Table 6: Mean importance of online and offline recommendation sources for purchases ... 45

Table 7: H1a. Perceived importance of recommendations per product category ... 46

Table 8: H1b. Difference in perceived importance of online and offline recommendations .. 47

Table 10: H2. Differences between product categories ... 47

Table 12: H3. Associations between age and recommendations ... 48

Table 14: H4. Associations with income ... 49

Table 15: H5. Associations between gender differences and recommendations ... 51

Table 16: H6. Associations between innovativeness and recommendations ... 53

Table 18: H7. Associations between attention to social comparison and recommendations .. 54

Table 19: H8. Associations between buying impulsiveness and recommendations ... 55

Table 20: Impact of demographic and personal characteristics ... 58

Table 22: Covariates of recommendations ... 61

INDEX OF FIGURES Figure 1: Conceptual framework ... 37

(7)

7

1. INTRODUCTION

For the last two decades, the Internet and social media have grown intensively (Cheung et al., 2008). Through the Internet, consumers can share their opinions to a wider public than ever before through its accessibility, reach and transparency (Hennig-Thurau, Gwinner, Walsh, & Gremler, 2004). This information technology development caused a transformation of the traditional Word of Mouth (WOM) to the electronic Word of Mouth (eWOM) (Hennig-Thurau et al., 2004). Consumers’ options to share content extended from traditional, offline WOM, to an interaction with a geographically dispersed population, by engaging in eWOM. Specifically, through eWOM communication, more consumers share their opinions on, experiences with, and evaluations of brands and products via discussion forums, web-based opinion platforms, news groups, and social media sites, with a multitude of other consumers (Hennig-Thurau et al., 2004). This sharing of eWOM has become a major information source for consumer purchase decisions (Park et al., 2012). The effects of this eWOM development have been studied intensively. This development creates new challenges for research in understanding how consumers search for information and what the impact of recommendations is on their purchase behaviour. Different communication channels have impact on what gets shared, which in turn has impact on purchase decisions (Berger & Iyengar, 2013). Within these communication channels, online and offline recommendations are types of (e)WOM communication sources. Specifically, recommendations are product-related conversations, which can have either a positive or a negative impact on consumer decision-making, and play a role when considering buying a new product (Arndt, 1967).

Even though much research elaborates on WOM and recommendations, little is known about whether, and to what extent, online or offline recommendations are perceived as more important for different product categories, and which factors might be associated with

(8)

8

this evaluation. Therefore, this research examined whether, and to what extent, consumers’ perceived importance of online and offline recommendations differs across product categories (based on five product categories that differ in their tangibility, level of involvement, and level of consumer experience) and whether demographic and/or personal characteristics are associated with this evaluation. An expensive product category (cars) has been included, because previous research suggests that consumers might find both online as offline information sources important, to gather as much as possible information before buying an expensive product (Cheema & Papatla, 2010). The other product categories that are included in this study are clothing, holidays, home electronics, and restaurants. Furthermore, to test whether there are different factors associated with the perceived importance of recommendations across product categories, this study tested associations between recommendations and personal and demographic characteristics. The demographic characteristics that were used are gender, age, and income. The personal characteristics that were used are innovativeness: consumer novelty seeking (CNS), consumers’ attention to social comparison information, and buying impulsiveness. To be able to conduct this research, a Web survey with an explanatory questionnaire was done.

The reason for investigating this research is that it gives clarity to managers and firms about the factors that might influence the purchase decisions of consumers, for both online as offline purchases, and whether managers should focus more on online or offline recommendations, depending on their product offering and target group. Furthermore, this research is important because the literature has paid little attention to the perceived differences in importance of online and offline recommendations across various product categories, and whether there are other factors associated with this perceived importance. While there is a lot of research into the effects of online and offline WOM on purchase

(9)

9

intentions (Arndt, 1967; Cheung et al., 2008; Hu, Liu, & Zhang, 2008), it is not clear how this effect differs across product categories. Although, Cheema & Papatla (2010) did study what the relative importance of online versus offline information for Internet purchases across hedonic versus utilitarian products is, and how Internet experience moderates this relationship, they did not test associations between other personal and demographic characteristics and the perceived importance. These factors might be associated with this relationship as well. Overall, the current study is aimed at closing the existing gap in the literature about the differences in perceived importance between online and offline recommendations, and which factors are associated with differences in evaluations of online and offline recommendations, with the main research question being: “To what extent does

consumers’ perceived importance of online and offline recommendations differ across

product categories, and are demographic and personal characteristics associated with this evaluation?”

Contribution

It is the purpose of this research to contribute to the existing literature in the area of WOM and recommendations in three primary ways.

First, differences in perceived importance of online and offline recommendations across different product categories were tested. An understanding of the varying perceived importance of the sources is important for several reasons. Firstly, recommendations have the highest trustworthiness, and are therefore the most powerful selling tools for companies (Nielsen, 2007). Since product categories differ in their tangibility, their level of involvement, their perceived risk, their social demonstrance, etc. (Beatty & Smith, 1987), it is important to know for which product categories recommendations have a bigger influence on consumers’ perceived importance and their purchase decisions. These differences in product

(10)

10

characteristics might be associated with the perceived importance of online and offline recommendations. Secondly, electronic commerce (ecommerce) is growing; more firms sell their products both offline as online. Hence, it is important to know how consumers’ purchase intentions for products within different product categories get influenced. Furthermore, consumer knowledge sharing through the Internet has strongly increased in recent years (Cheema & Papatla, 2010). Therefore, it is important to know the impact of this knowledge sharing on consumer buying situations. Moreover, consumers have access to both online as offline recommendations about a diversity of products within different categories. Potential differences in perceived importance of online and offline recommendations across product categories need to be better understood. Finally, people show varying behaviour when searching for product information via online communication channels or via offline communication channels (Berger & Iyengar, 2013), and this search process might be related to consumers’ perceived importance of the recommendation.

Second, when differences in perceived importance of recommendations are shown, this research tries to map how this level of importance differs across product categories. As previous studies already show, consumers tend to find information from other consumers more important than from advertisement (Goldsmith & Horowitz, 2006). Therefore, it is interesting to test if this importance holds across different product categories.

Third, other factors might also be associated with the perceived importance of recommendations across product categories. For example, differences in perceived importance might be associated with consumer characteristics, like demographic and personal characteristics. When differences in demographics and personal characteristics are associated with differences in perceived importance of online and offline recommendations, firms can make use of these recommendations accordingly, depending on their target group.

(11)

11

Managerial Relevance

This research is of practical relevance, because it is important for a firm to know what factors might influence the purchase decisions of consumers, for both online as offline purchases. Hence, if there is consensus about which factors are associated with the perceived importance of recommendations when intending to purchase a specific product, managers can allocate their advertising budget per product category more efficiently, and they can increase their return on advertising investment. To achieve this higher return, focused targeting of potential consumers is essential (Cheema & Papatla, 2010). Also, when managers know which recommendation channel is most suited for a certain type of product category, they can optimize their recommendation channels accordingly. Furthermore, it is important to know for which product categories this perceived importance is higher. When managers know this, they can take into account which benefits and risks may be accompanied with the impact of the recommendations per product category.

Theoretical Relevance

This research is of theoretical relevance, because it adds to the literature by investigating more deeply whether, and to what extent, recommendations through offline and online channels differ across product categories in their perceived importance (more than only the difference between hedonic and utilitarian products), and which other factors (besides Internet usage) might be associated with this evaluation.

Thesis Overview

The research is structured as follows. In the next chapter, a review of the literature on the key concepts about online and offline recommendations, product categories, and consumers’ perceived importance are discussed. Furthermore, a theoretical framework with hypotheses about potential differences in perceived importance of online and offline

(12)

12

recommendations across five different product categories are developed. This framework is examined in the field research, for which the research method is explained in chapter three. The model is presented in chapter four. Finally, there is a discussion of the empirical findings, the managerial implications are given, the limitations and suggestions for further research are discussed and the conclusion is given.

2. LITERATURE REVIEW

This chapter provides a comprehensive review of the literature about the key concepts in this study, in order to analyse what has already been studied about the perceived importance of online and offline recommendations across product categories. First, some basic information about online shopping behaviour is given. Next, previous research about the key variables online and offline recommendations, consumers’ perceived importance, and product categories, are reviewed. Then, a detailed analysis of the possible associations with demographic and personal characteristics is given. Finally, based on this review, hypotheses and a conceptual framework are developed and are tested in this study.

2.1 Online shopping behaviour

Recently, the Internet as a marketing channel has boomed. Still, most companies mainly use the Internet for advertising or to promote corporate images (Kiang, Raghu, & Shang, 2000). When looking at the use of the Internet as a an online shopping channel, Levin, Levin & Weller (2005) show that Internet marketing is developing rapidly as a new transaction and distribution channel, besides using the Internet mainly as a communication channel.

Advantages of the Internet as an online shopping channel are that it can serve as both a transaction channel and a physical distribution channel, it allows companies to respond more quickly to market changes and customer preferences, and it allows access to a broader

(13)

13

customer base (Kiang et al., 2000). Moreover, online and offline shopping channels have three major channel functions in common, which are availability, product customization, and logistics (Kiang et al., 2000). A disadvantage of online shopping channels compared to traditional offline shopping is the transaction complexity. This complexity increases the perceived risk of online transactions, and therefore, online shopping preference is less at the purchase stage than at the search stage (Levin et al., 2005).

The growth of the Internet as a new sales channel creates important challenges for marketers. One of the challenges is to know the differences between online and offline marketing across product categories, and their impact on consumer decision-making. Benefits for a company for using Internet as an online shopping channel greatly depend on the product characteristics. A key question is “what drives a consumer to shop online or offline for a particular product? (Levin et al., 2005).”

Besides knowing the preference of consumers to shop online or offline for a specific product, it is interesting to know “what drives a consumer to look for online or offline recommendations for a certain product and how does this influence its decision-making process?” Because WOM and recommendations are important factors impacting consumers’ shopping behaviour, the next chapter will elaborate on WOM and recommendations and it will study differences between online and offline recommendations.

2.2 (e)WOM and recommendations

Consumers use various communication channels to interact and communicate with others about products. Before the Internet age, this consumer-to-consumer communication, like WOM, mostly took place via face-to-face communication. WOM can be defined as “oral, person-to-person communication between a receiver and a communicator, whom the receiver perceives as non-commercial, regarding a brand, product or service” (Arndt, 1967). WOM is

(14)

14

a consumer-dominated marketing communication channel. The sender is independent of the market. Hence, consumers perceive this channel as more reliable, credible and trustworthy than firm-hosted communication channels (Arndt, 1967). Furthermore, WOM is recognized as one of the most influential resources for information transmission (Mayzlin & Godes, 2002). WOM is mainly driven by personal relevance, and personal experience has the biggest influence on WOM, and almost everybody participates in a social network (Allsop, Bassett, & Hoskins, 2007). Allsop et al. (2007) show that 81% of respondents’ perceptions is positively influenced by WOM and 34% finds WOM a very credible source. WOM mainly takes place via social networks, which operates to gather information from others, incorporate them in consumers’ own knowledge, and pass this on to others (Allsop et al., 2007).

It is interesting to investigate how people respond to things that other people say. People can respond differently, depending on how important they perceive the information. This perception, on the other hand, can depend on the type of source or on the context where the source talks about, like differences in responses between and within a product or service.

In this study, the focus is on online and offline recommendations, which are types of (e)WOM communication. Recommendations are product-related conversations, which can have either a positive or a negative impact on consumer decision-making, and can play a role when considering buying a new product (Arndt, 1967). Specifically, Nielsen (2007) shows that recommendations have the highest trustworthiness and are therefore the most powerful selling tools for companies. 78% of the respondents in the study trust the recommendation of other consumers that they do not know. Furthermore, nine of every ten Internet consumers worldwide trust recommendations from people they know, and seven of every ten Internet consumers trust consumers’ opinions posted online by people from outside their direct social network (Nielsen, 2009). As shown by Arndt (1967), this high level of trust may be due to the

(15)

15

fact that in recommendation channels, the sender is independent of the market. Moreover, when looking at the duration of the impact of WOM referrals compared to traditional marketing, Trusov, Bucklin & Pauwels (2008) show that word-of-mouth referrals have a substantially longer carryover effect than traditional marketing actions.

In this study, a distinction is made between online and offline recommendations, because of the increasing interest in how and when consumers use online or offline communication sources for purchase behaviour (Cheema & Papatla, 2010). The next two paragraphs will discuss the different types of recommendations more in depth.

WOM and offline recommendations

The main sources of recommendations before the Internet age were face-to-face conversations. Arndt (1967) finds that in an offline context, exposure to favourable comments about a new product stimulates purchase behaviour, while unfavourable comments hinder it. Besides the effect of positive recommendations, the perceived risk of buying that product impacts the likeliness of considering buying a new product; high-risk perceivers are less likely to buy a new product.

EWOM and online recommendations

Sales over the Internet are increasing (Cheema & Papatla, 2010) and Internet has changed the structure of communication by reducing search time and costs for consumers (Kulkarni, Ratchford, & Kannan, 2012). Through the Internet, more recommendations sources are available to consumers, and online recommendations start to play a bigger role in consumers’ purchase decisions (Hu et al., 2008; Lee & Kwon, 2008; Cheema & Papatla, 2010). A reason for this increase in online recommendation sources is that they reduce information overload for consumers, they can give advice in finding the right product, and they support online consumer decision-making (Lee & Kwon, 2008).

(16)

16

When looking at how consumers use new media, it is interesting to know which activities consumers perform most frequently in this channel. Harris Interactive (used in Allsop et al., 2007) shows that 59% of the respondents use new media mainly to forward information found on the Internet to colleagues, peers, friends or family, and 48% of the respondents read newspapers online.

Going more in depth into the impact of online recommendations on consumer decision-making, Lee & Kwon (2008) show that online recommendations can lead to enhancing consumers' attitude towards the recommended product and can positively influence consumer decision-making. Also, the amount of views and the reputation of the reviewers impact consumer decision-making. Specifically, Hu et al. (2008) show that consumers respond more favourably to online reviews that have a higher exposure rate and are written by reviewers with a better reputation, but that this impact diminishes over time.

Online retailers also picked up the increasing importance of online recommendations. Many online retailers encourage consumers to share online recommendations about their products, by allowing consumers to write online reviews about the products they offer (Gupta & Harris, 2010). These recommendations are important, because in an online shopping environment, 92% of the online shoppers reads product reviews before they make a purchase decision (Li, Huang, Tan, & Wei, 2013). There are differences in characteristics between online and offline recommendation sources due to several factors, and these differences can impact consumers’ perceived importance of the recommendation. Therefore, these differences are discussed in the next paragraph.

Differences in characteristics of online and offline recommendation sources

When looking at the differences in characteristics between offline and online recommendation sources, online sources are more likely to be written, undirected,

(17)

17

anonymous, involving a larger audience, and reduce social presence (Berger, 2013). Moreover, offline recommendations have a higher trustworthiness than online recommendations (Nielsen, 2009). Another difference in characteristics is the amount of search. Specifically, Rose & Samouel (2009) show that in contrast to the offline context, in an online context, the amount of search effort is reduced, because listing information can be done more efficiently.

Other differences in characteristics are that offline recommendations are face-to-face, have a narrow reach, and for which the identity is disclosed. This is in contrast to characteristics of online recommendation sources, which have an indirect interaction, have a broad reach, and for which the identity is not disclosed (Tong & Xuecheng, 2010).

Furthermore, the increase in online communication changed the structure of WOM interactions by letting consumers expose to online recommendations from virtual strangers (Steffes & Burgee, 2009). Differences in characteristics can be related to differences in how recommendations affect shopping behaviour. Therefore, the next paragraph will discuss the effects of these recommendations on shopping behaviour.

Effects of online and offline recommendations on shopping behaviour

This paragraph focuses on the effects and importance of offline and online recommendations on the consideration-set and choice of products by consumers. When looking at the importance of online recommendations, Goldsmith & Horowitz (2010) show that consumers seek opinions of others online and find this information more important than advertising. Online recommendations also impact the time consumers spend considering the recommended product (Gupta & Harris, 2010). Specifically, online shoppers are overwhelmed with many information sources available on the Internet, and therefore, pay more attention to some sources than others (Cheema & Papatla, 2010). Moreover, when

(18)

18

consumers have a higher level of involvement, they spend more time on the choice of a product in general (not only considering the recommended product), which is in contrast to consumers who are less motivated to process information. Consumers who are less motivated to process information have a lower level of involvement and are more likely to focus only on the recommended product. They are more likely to make irrational, suboptimal decisions (Gupta & Harris, 2010).

When looking at the effects of offline recommendations on purchase decisions, Cheema & Papatla (2010) show that for consumers with a high level of Internet experience, offline information sources become relatively more important than online sources. This holds for both utilitarian as hedonic product types. The next paragraph will discuss how and why consumers might respond differently to recommendations.

Differences in consumer responses to recommendations

Consumers respond differently to things that others say and to recommendations. There are many factors impacting consumers’ response. For example, differences in which channels have been used (online or offline) can cause a distinction in consumers’ response. A first factor that impacts this is the level of trust. Specifically, recommendations through offline channels have a bigger impact on shopping behaviour than recommendations through online channels, because they are more persuasive (Nielsen, 2007). This credibility gap even exists when communicating with a best friend (Nielsen, 2007). This difference is further tested in this research.

Differences in consumers’ response also depend on the type of product or service where the recommendation is about. Specifically, the perceived risk is greater for products that are intangible, higher priced, more visible, more complex, and more difficult to evaluate (Beatty & Smith, 1987; Laroche, Nepomuceno, & Richard, 2010). Accordingly, this

(19)

19

perceived risk increases the importance of recommendations and influences how consumers respond to these recommendations. Bearden & Etzel (1982) show that reference groups have a strong influence on which products and brands they buy when consumers intend to buy public consumed luxury products, like a sailboat. Furthermore, the reference group has the weakest influence on consumers’ choice when they intend to buy private consumed necessity products, like a mattress. Moreover, reference groups have a weak influence on the product choice and a strong influence on brand choice for public consumed necessity products and reference groups have a strong influence on product choice and a weak influence on brand choice for private consumed luxury products.

Besides classifying products based on these two axes, products can be classified as experience and search products. Online recommendations are shown to have a bigger influence on consumers’ decision-making for experience products than for search products (Senecal & Nantel, 2004).

When looking more in depth to the impact of recommendations in a services purchase decision context, Bansal & Voyer (2000) show that there are three relations between WOM and consumer decision-making. In the first relation, the interpersonal influence relation, actively seeking WOM and tie strength positively influence the impact of WOM on receiver’s decision-making. In the second relation, the non-interpersonal influence relation, perceived risk and sender’s expertise positively influence actively seeking WOM. Sender’s expertise also positively influences the impact of WOM on receiver’s decision-making. In the third relation, non-interpersonal forces influence interpersonal forces. Receiver’s expertise has a negative impact on actively seeking WOM and the influence of the sender’s WOM on the receiver’s purchase decision. This shows that actively seeking WOM, tie strength,

(20)

20

perceived risk, sender’s and receiver’s expertise influence the impact of WOM on receiver’s decision-making in a services context.

Moreover, the type of source plays an important role in how consumers respond to the recommendation. Consumer reviews are perceived as more trustful than company-generated reviews (Bronner & de Hoog, 2010). Also, when the consumer knows the source, the recommendation is perceived as more trustful than when the source is a stranger (Nielsen, 2009). Varies types of sources and their perceived importance are discussed in the next paragraph.

Sources of online and offline recommendations

Online and offline recommendation sources can be classified in different ways. In this part, different categorizations of recommendation sources and factors that influence the likelihood of choosing a particular recommendation source are discussed. For an overview of the different types of recommendation sources and the perceived importance, see Table 1.

Product knowledge is a factor that impacts the type of channel a consumer uses to search for information (Westerman, Van Der Heide, Klein, & Walther, 2008). Specifically, when consumers search for information about less-known topics, they use channels where their identity is not disclosed, while for well-known topics, consumers use channels where their identity is disclosed.

Online recommendations can be exposed to consumers inter alia through websites, blogs, chat-rooms, or e-mails (Hennig-Thurau et al., 2004). Examples of consumer-generated media through which online recommendations can be exposed to consumers are blogs, online communities and online review sites. Online communities are online channels where consumers come together to interact with others that share the same interest (Granitz & Ward, 1996). Online reviews can be divided into expert-written and customer-written reviews

(21)

21

(Li et al., 2013). Specifically, Li et al. (2013) show that customer-written reviews with a low level of content abstractness have the highest perceived review helpfulness. Furthermore, social networking websites are useful for information search about both well-known and unknown topics (Westerman et al., 2008).

Offline recommendations mostly take place through face-to-face (FtF), oral conversations, inter alia through recommendations of friends, vendors, or through editorial recommendations (Cheema & Papatla, 2010). A distinction in offline recommendations can also be made between oral and written conversations. According to Berger & Iyengar (2013), consumers are more likely to share interesting brands and products through written conversations than through oral conversations, because of the higher level of asynchrony.

Recommendation sources can also be categorized according to the tie strength, which is the closeness of the relationship between the decision-maker and the recommendation source (Duhan, Johnson, Wilcox, & Harrell, 1997). This tie strength is strong when the decision-maker knows the source personally (family or friends), and is weak when the decision-maker does not know the source personally (acquaintances or strangers). There are different factors that influence the likelihood that consumers either choose a strong-tie or a weak-tie recommendation source. These factors are product knowledge level, perceived level of task difficulty, and the type of evaluative cues sought by the consumer (Duhan et al., 1997). Task difficulty and prior knowledge influence the likelihood of choosing strong-tie sources (Duhan et al., 1997). The importance of instrumental cues and subjective prior knowledge influence the likelihood of choosing weak-tie sources (i.a. professional reference services). Even though there are many different recommendation sources and many different distinctions, in this study, there will only be a distinction between online and offline

(22)

22

recommendations. The next section will discuss consumers’ perceived importance, and how this can differ across consumers, communication sources and product categories.

Table 1: Recommendation sources

Sources Difference in perceived importance

Source perceived as more important Source perceived as less important Online vs. offline

recommendations

Offline recommendations (oral or written conversations) are face-to-face, have a more narrow reach, and for which the identity is disclosed.

Order of importance of offline sources: 1. Friend recommendation 2. Editorial recommendation 3. Seeing the product 4. Vendor recommendation 5. Advertisement

Online recommendations (via social media, review websites, company websites)

are more written, undirected, anonymous, involve, larger audiences, and reduces social presence (compared to offline).

Order of importance of online sources: 1. Friend recommendation 2. Editorial recommendation 3. Seeing the product 4. Vendor recommendation 5. Advertisement

Oral vs. written Written channels (asynchrony) Oral channels

Online reviews Customer-written reviews Expert written reviews

Online versus offline information sources

Offline information sources for hedonic products

Online information sources for utilitarian products

Tie strength Strong-tie sources Weak-tie sources

Product knowledge: less-known topics

Non-identity disclosed channels Identity disclosed channels

Product knowledge: well-known topics

Identity disclosed channels Non-identity disclosed channels

New media/online channels Consumer-generated media Company-generated media 2.3 Consumers’ perceived importance

To study whether consumers value online and offline recommendations differently across product categories, the perceived importance of these recommendations across product categories will be measured. There can be a big difference in how important consumers perceive products and services to be. Perceived importance can be seen as a component of the perceived value that consumers receive for a product or service. Retail companies see delivering customer value as the most important aspect in order to deliver quality (Sweeney & Soutar, 2001). Perceived value is defined as “consumer’s overall assessment of the utility of a product based on perceptions of what is received and what is given” (Zeithaml, 1988, p.14). Perceived value has four components that are interrelated: emotional value, social value, functional value (based on price) and functional value (based on performance and quality). These components are perceived differently across consumers, based on the

(23)

23

perception of what consumers receive and what they give (Sweeney & Soutar, 2001). Perceptions are mainly influenced by the emotional and functional components. For example, when considering buying a product in a durable product category, emotional value is the most important component affecting consumers’ willingness to buy (Sweeney & Soutar, 2001).

When looking more in depth at the perceived importance of product information, which is the value component, a difference can be derived from the level of involvement that consumers have when they buy a product within a specific product category. This level of involvement differs across consumers and consists of five antecedents. These antecedents are the perceived importance of the product or the situation (personal meaning), the perceived risk (two facets: perceived importance of negative consequences of poor choice and the perceived probability to make a poor choice), the perceived symbolic value, and the perceived pleasure value (hedonic value) (Laurent & Kapferer, 1985). Moreover, based on the level of involvement of the consumer, there are two routes to persuasion for buying a product or service, which are the central and the peripheral route to persuasion. These routes are explained through the Elaboration Likelihood Model (ELM) (Petty & Cacioppo, 1986). According to this model, if a consumer is highly involved when intending to buy a product, a consumer is more likely to take the central route. In this route, the quality of arguments is more important. If a consumer has a lower level of involvement when intending to buy a product, the consumer is more likely to take the peripheral route. When taking this route, the number of arguments and source expertise are more convincing. Which route to persuasion the consumer will take, is influenced by the motivation and ability to process persuasive information. Furthermore, Gupta & Harris (2010) show that consumers who are motivated to process information spent more time on the purchase decision overall, which is driven by an online recommendation. Consumers with a lower level of involvement and motivation base

(24)

24

their purchase decisions easier solely on online recommendations. Inherently, they make irrational decisions based on these online recommendations.

Hence, consumers’ perceived importance of online and offline recommendations differs across consumers and across product categories, depending on many different factors that are interrelated. One of the factors is the product category. Therefore, differences in perceived importance across product categories are discussed below.

How perceived importance changes across product categories

Consumers perceive the importance of online and offline recommendations across products or services differently, which can be influenced by several factors. Examples of factors that can impact this difference are perceived risk and cost for search of information. Specifically, the factor cost of search is important for offline information search effort, but not for online information search effort, since there are (almost) no costs involved when searching for information online.

Cheema & Papatla (2010) show that for both hedonic and utilitarian products, the relative importance of offline information sources is greater than the relative importance of online sources. Also, the relative importance of online sources is greater for utilitarian products than for hedonic products. Furthermore, the level of Internet experience affects this relative importance: for Internet purchases, an increase in Internet experience decreases the relative importance of and trust in online sources. Moreover, perceived importance of recommendations can depend on the knowledge that a consumer has about a product category. Specifically, when consumers know more about the product category, they will put less effort into searching for information about a product in this product category (Beatty & Smith, 1987).

(25)

25

As explained, the level of involvement impacts the perceived importance of a product or service. This level of involvement can vary across product categories. When a consumer is more involved in buying a particular product, this product is perceived as more important. In this situation, the consumer is more likely to take the central route to persuasion, for which argument quality is more important (Petty & Cacioppo, 1986). It is expected that this route increases the perceived importance of trustworthy recommendation sources. For example, when a consumer is intending to buy an expensive product, the perceived risk for making this purchase is higher, and in this case, the consumer is more likely to be more involved in this purchase decision-making process. The level of involvement is expected to increase the perceived importance of information about this product (Petty & Cacioppo, 1986).

Another distinction in perceived importance can be made between products and services. Because services are intangible, consumers perceive a higher level of risk when buying a service. This perceived risk increases the evaluation difficulty (Laroche et al., 2010), which can impact the level of involvement and the perceived importance when buying a service. Hence, this perceived importance changes across products and services categories.

Taking these findings into account, in this study, perceived importance is defined as: “How important consumers perceive recommendations about a product category depends on their level of involvement with the product and the level of trust in the recommendation source”. Because it is shown that offline recommendations have a higher level of trust compared to online recommendations (Cheema & Papatla, 2010; Nielsen, 2009; Nielsen, 2007), it is expected that the perceived importance of offline recommendations is higher than the perceived importance of online recommendations across all product categories. This difference is tested across all product categories independently, and by comparing the combined mean of the perceived importance of online recommendations with the combined

(26)

26

mean of the perceived importance of offline recommendations Therefore, the following hypotheses are tested:

H1a. The perceived importance of offline recommendations will be greater than the perceived importance of online recommendations across all product categories.

H1b. The perceived importance of offline recommendations will be greater than the perceived importance of online recommendations.

2.4 Product Categories

Besides studying differences in perceived importance between online and offline recommendations across all product categories, differences between product categories are studied. Previous research about consumer behaviour shows that the type of product and consumers’ decisions influence how consumers perceive the value of these products, in both purchase-intention stage and post-purchase stage (Levin et al., 2005). This difference between product types can influence consumers’ preference for online versus offline purchases. Levin et al. (2005) show that this difference in preference for shopping online or offline is driven by the difference in importance of product attributes that are perceived to be better delivered online or offline. Specifically, product attributes related to the experience or the delivery process are perceived as being better delivered offline (Levin et al., 2005). Product attributes related to the search process are perceived as being better delivered online (Levin et al., 2005). Furthermore, online shopping preference is bigger at the search stage than at the purchase stage, due to the perceived risk of online transactions (Levin et al., 2005).

Since there is a difference in preference for shopping online or offline, it is interesting to see if there is also a difference in preference for using online or offline recommendation sources before purchasing a product. Therefore, in the next paragraph, characteristics of the

(27)

27

five product categories and the perceived importance of recommendation sources across product categories are discussed.

Product categories used in this research

In this study, five product categories are selected to represent contrasting profiles on the dimensions level of involvement, tangibility, and level of consumer experience. The perceived importance of online and offline recommendations is measured for the five product categories home electronics, cars, holidays, restaurants, and clothing. ortance of recommendations.

In the next paragraphs, differences in perceived importance of recommendations across product categories are discussed.

Table 2 gives an overview of the differences across product categories based on these different dimensions. This paragraph reviews the dimensions for all product categories individually.

When looking at the product category holidays, Bronner and de Hoog (2010) show that when searching for information about booking a holiday, consumers use both user-generated as company-user-generated websites, and they have a complementary role. Many consumers use the Internet to search for information before booking a holiday. Moreover, consumers see consumer-generated websites as more relevant than company-generated websites, through scoring higher on unknown and useful information. When looking at the amount of search for, and providing recommendations about holidays, Allsop et al. (2007) show that 67% of consumers search for recommendations where to go, and 89% of the consumers provide recommendations where to go, which shows that customer experience is an important factor for providing recommendations. Furthermore, Susskind, Bonn & Dev (2003) show that the Internet is mostly used to look for information to book a holiday and not

(28)

28

to actually book a holiday. A holiday is a service, and is therefore intangible. Also, holidays are a relatively high expenditure. Because of holidays’ intangibility and relatively high expenditure, it is expected that holidays have a high-perceived risk, which increases the level of involvement when booking a holiday.

For the product category restaurants, online recommendations are shown to be more persuasive when individuals are highly involved and have a low ambiguity tolerance (Wang & Wang, 2010). Yang (2013) shows that predictors of online recommendation intentions about restaurants are the experience-based factor satisfaction, the knowledge sharing-based factors egoistic and altruistic needs, and the technology-based factors perceived usefulness and ease-of-use. Furthermore, when looking at the amount of search for, and providing recommendations about restaurants, Allsop et al. (2007) show that 89% of online consumers seek information and advice provided by other consumers, and 85% provide information and advice to other consumers. Compared to other product categories, the product category restaurants scores the highest on providing information to and seeking information from other consumers. This shows that customer experience is very important in this product category (Allsop et al., 2007). Restaurants are a service and therefore intangible, which increases the perceived risk. Restaurants expense is rather medium, which reduces this perceived risk. Therefore, a medium level of involvement for the product category restaurants is expected.

Cars are major durable (tangible) goods, which have a high level of involvement and

a higher perceived risk when intending to buy one. Many consumers use Internet sources for information about cars (Kulkarni et al., 2012). Yang, Hu, Winer, Assael & Chen (2012) show that first-time buyers are more likely to consume both online as offline recommendations, which is mainly because of consumers’ limited product knowledge. Interestingly, a person’s driving experience increases the consumption of both online and offline recommendations.

(29)

29

This shows that more product experience makes it easier to process new information (Yang et al., 2012). 60% of car buyers use the Internet for information, and see the manufacturer source as the most important information source. Consumer reports and buying services are also seen as important information sources (Kulkarni et al., 2012). Furthermore, Allsop et al. (2007) show that 74% of the respondents seeks for information about cars from others, and that 88% of the respondents provides information to others. Moreover, Kulkarni et al. (2012) show that Internet-users and non-Internet users differ in whether they rely on recommendation sources when making automobile choices. Non-Internet users rely more on recommendations, while Internet users rely more on ratings. This difference impacts the choice of automobile, which stresses the importance for implications of this knowledge for organizations.

Home electronics are experience products and are seen as medium-level risk

purchases, with a medium level of involvement (Gupta & Harris, 2010). Consumers show a medium-level of offline information search effort for risk reduction (Maity, Hsu, & Pelton, 2012). Furthermore, Gupta & Harris (2010) show that for laptops, online recommendations increase the time that consumers take to analyse information about the recommended product, but only when consumers are motivated to process information. When looking more in depth to the use of recommendations for computers, Allsop et al. (2007) show that 84% of the respondents seek information provided by others, and 78% provides information to others. Search effort for home electronic products affects the use of recommendations and is positively related with purchase involvement; attitude towards shopping and time availability; and it is negatively related to product class knowledge (Beatty & Smith, 1987).

The product category clothing is a visibly consumed product category with a symbolic meaning. Clothing is an experience product category, for which consumers like to be able to

(30)

30

see, touch and handle the product before buying it, and these attributes are therefore better delivered offline (Levin et al., 2005). Also, personal service and speedy delivery are important clothing attributes better delivered offline (Levin et al., 2005). In this product category, personal characteristics that influence social behaviour in the social environment can play an important role (Bertrandias & Goldsmith, 2006). An example of such a characteristic is consumers’ attention to social comparison information, which is positively related to fashion opinion seeking (Bertrandias & Goldsmith, 2006). This relationship can be associated with the perceived importance of recommendations.

In the next paragraphs, differences in perceived importance of recommendations across product categories are discussed.

Table 2: Overview product categories Product category Differences

Level of customer

involvement Tangibility

Level of customer experience level

Holidays High Intangible High

Restaurants Medium Intangible High

Cars High Tangible High

Home electronics Medium Tangible Medium

Clothing Medium - High Tangible High

Perceived importance of recommendation sources across product categories

How consumers perceive the value of online and offline recommendation sources can differ across product categories. This difference can be driven through several factors. As shown in previous research, the level of perceived risk, the level of involvement, the product consumption visibility, and prior product knowledge, are important factors impacting the perceived importance of recommendations across product categories (Beatty & Smith, 1987). When looking at the impact of perceived risk on the perceived importance of recommendations, Beatty & Smith (1987) show that offline search effort increases when the level of perceived risk is higher, as a risk-reduction process, but this level of perceived risk does not increase online search effort. This is in line with the findings of Maity et al. (2012),

(31)

31

who show that consumers tend to search for more information when intending to buy a product that creates a greater intrinsic perceived risk. There are many factors that can influence this difference in perceived risk across product categories. A few examples are products’ tangibility (as explained in the previous section), product price, visibility of the product, complexity of the product, consumers’ experience with the product, consumers’ knowledge about the product, the gender of the consumer, and consumers’ level of involvement with the product (Beatty & Smith, 1987; Rose & Samouel, 2009; Yang, 2013). The perceived risk is greater for products that are intangible, higher priced, more visible, more complex, and more difficult to evaluate (Laroche et al., 2010).

The level of involvement can impact the relationship between evaluation difficulty and the perceived risk for buying a product in a certain product category (Rose & Samouel, 2009; Racherla & Friske, 2012). Specifically, a higher level of involvement generates a stronger relationship between evaluation difficulty and the perceived risk in a product category (Laroche et al., 2010). Also, a higher level of involvement with the product increases the level of motivation to search for information about this product (Rose & Samouel, 2009). The next paragraph discusses the perceived importance per product category.

Differences in perceived importance per product category

This paragraph reviews existing literature about the differences in perceived importance between online and offline recommendations that are related to the product category. Cheema and Papatla (2010) show that the relative importance of online information is higher for utilitarian products, like computer software and hardware, than for hedonic products, like books and music. They suggest including an expensive product category, like

(32)

32

cars, in future research, and expect that consumers search for more information via both online and offline channels before buying a car.

Furthermore, the factors perceived risk when intending to buy a certain product and the perceived costs for searching for information about that product play a bigger role in offline information search than in online information search (Maity et al., 2012).

When looking at what information consumers search for and provide to other consumers via recommendations, Allsop et al. (2007) show that the amount of consumers involved in this varies across product categories. Interestingly, most people that search for recommendations also provide recommendations. Furthermore, they show that the product category restaurants is the most popular product category to talk about. From the twelve products that Allsop et al. (2007) study, athletic shoes score the lowest when looking at how many consumers search for information from others (37%) and how many consumers provide information to others (30%).

Home electronics have a medium-level of risk (Rose & Samouel, 2009) and cars have a high-level of risk (Cheema & Papatla, 2010). Therefore, it is expected that the perceived importance of offline information is higher for cars than for the other product categories, based on the relatively high-perceived level of risk.

Furthermore, consumers perceive a higher level of risk when evaluating intangible products or services, like holidays and restaurants, than when evaluating tangible products and services, like clothing and home electronics.

Also, consumers have a higher level of involvement when they purchase a product with a higher perceived risk due to the complexity of a product or when purchasing a higher priced product (Maity et al., 2012). When the level of involvement is higher, consumers tend to search for more information before buying a product (Maity et al., 2012). Hence, it is

(33)

33

expected that consumers have a higher level of involvement for the product categories cars and holidays, because these categories are higher priced and are more complex. Based on the factors intangibility, perceived risk, and level of involvement, the following hypothesis is developed:

H2. The perceived importance of offline and online recommendations will be greater for the product categories holidays, restaurants, and cars than for the product categories home electronics and clothing.

The following two sections will discuss the possible associations between the perceived importance of online and offline recommendations and demographic and personal characteristics.

2.5 Associations between of demographic variables and recommendations

To test whether demographic characteristics might be associated with the perceived importance of both online as offline recommendations across product categories, the characteristics age, gender, and income are included in this study (Cleveland, Papadopoulos, & Laroche, 2011). These characteristics might be related to how consumers search for information and which products they buy. For example, consumers who search more for product information and who tend to use the Internet are mostly younger, more educated, and have a higher income (Kulkarni et al., 2012). Other possible associations with these demographic factors and the perceived importance of online versus offline recommendations are discussed below.

Younger consumers are more open to innovative products (Cleveland et al., 2011). Also, younger consumers tend to have more Internet experience (Cheema & Papatla, 2010). Therefore, it is expected that age is negatively associated with the perceived importance of

(34)

34

both online as offline recommendations across all product categories. Therefore, the following hypothesis is developed:

H3. Age is negatively associated with the perceived importance of online and offline recommendations across all product categories.

Consumers with a higher income are more likely to buy luxury products (Cleveland et al., 2011), and are more likely to consume and generate more WOM (Yang et al., 2012). Furthermore, the perceived risk is higher for more expensive products, like luxury cars. Hence, it is expected that this perceived risk is associated with an increase in the amount of offline search before buying an expensive product to reduce this risk. Because cars are expensive products, the perceived risk for buying a car is high. Therefore, the following hypothesis is developed:

H4. Income is positively associated with the perceived importance of offline recommendations for the product category cars.

Gender can also be related with the perceived importance of both online as offline recommendations across product categories. Gender differences impact information processing. Specifically, women tend to search more for information and make use of more information sources than men do. For example, women are more likely to read online reviews and involve the assistant agent when searching for a product (Park, Yoon, & Lee, 2009). Therefore, the following hypothesis is developed:

H5. Women perceive the importance of both online and offline recommendations across all product categories as more important than men.

2.6 Associations between personal characteristics and recommendations

To test whether personal characteristics are related to the perceived importance of recommendations across product categories, the following personal characteristics were

(35)

35

tested: innovativeness: consumer novelty seeking (CNS), consumers’ attention to social comparison information, and buying impulsiveness.

Innovativeness can be conceptualized as consumer novelty seeking (CNS) and can be related to the new product adoption process. Manning, Bearden & Madden (1995, p. 330) define CNS as “the desire to seek out new product information". They show that in the adoption process of new products, CNS is positively related to early stages in this process. CNS consumers discover new products by seeking information (Manning et al., 1995). Because innovative consumers maximize their search for information (Konuş, Verhoef, & Neslin, 2008), it is expected that CNS is positively associated with the perceived importance of online and offline recommendations. Therefore, the following hypothesis is developed:

H6. Consumer novelty seeking is positively associated with the perceived importance of both online and offline recommendations across all product categories.

Bearden, Netemeyer & Teel (1989, p. 478) define attention to social comparison information as "the need to identify with or enhance one's image in the opinion of significant others through the acquisition and use of products and brands, the willingness to conform to the expectations of others regarding purchase decisions, and/or the tendency to learn about products and services by observing others or seeking information from others". Bearden et al. (1989) show that attention to social comparison information is positively related to the perceived importance of friends’ approval. Because clothing is a visibly consumed product, it is expected that attention to social comparison is positively associated with the perceived importance of both online and offline recommendations for the product category clothing. Therefore, the following hypothesis is developed:

H7. Consumers’ attention to social comparison information is positively associated with the perceived importance of both online and offline recommendations for the product

(36)

36

category clothing.

Buying impulsiveness refers to "a consumer's tendency to buy spontaneously,

unreflectively, immediately, and kinetically" (Rook & Fisher, 1995, p. 306). Rook & Fisher (1995) show that the relationship between buying impulsiveness and buying behaviour is moderated by consumers’ normative evaluations. Specifically, they show that this relationship is only significant when consumers believe that the impulsive behaviour is appropriate. It is expected that impulsive buyers tend to put less effort in searching for recommendations about a product. Especially for products for which impulsive buyers are not highly involved when intending to buy a product, like for home electronics and clothing. Therefore, the following hypothesis is developed:

H8. Impulsive buying is negatively associated with the perceived importance of both online and offline recommendations for the product categories home electronics and clothing.

2.7 Conceptual framework

To test the hypotheses, to analyse to what extent consumers’ perceived importance of online and offline recommendations differs across product categories, and to test the influence of demographic factors and personal characteristics on this relationship, the following conceptual framework in Figure 1 is developed. How data is collected is explained in the next chapter. How the hypotheses are tested is explained in the results section in chapter 4. Results.

(37)

37

Figure 1: Conceptual framework

3. METHODOLOGY

To be able to test the proposed hypotheses, data were collected. This chapter describes how data have been collected. In the first part of this chapter, the sample is described. In the second part of this chapter, the research design is described. In the third part of this chapter, the measures, the measure items and variables are clarified. In the last part of this chapter, the procedure about how data were collected is described.

3.1 The sample

The population for this study consists of the shopping consumers in the Netherlands, who have Internet access, which consists of approximately 11.8 million people (CBS, 2014). Non-probability sampling was used, because no sampling frame could be retrieved for this large research population. The main goal for data collection was to achieve a sample as large as possible, to increase the chance of having a representative sample and to be able to generalize conclusions over the population. The minimum sample size consisted of data from at least one hundred shopping consumers in the Netherlands.

On the 6th of May 2014, 160 questionnaires had been filled in. Four questionnaires were deleted because of missing too many data. After deleting these four, the remaining 156

(38)

38

were used for analysis. 76 respondents were male (49%) and 79 respondents were female (51%). The mean age of the respondents was 30. The youngest respondent was 18 and the oldest respondent was 70. The mode age of the sample was 26. The standard deviation of age was 11.62.

3.2 Research design

To be able to identify and explain differences in consumers’ perceived importance of online and offline recommendations that might be related to five product categories, and to test whether demographic and personal characteristics are associated with differences in evaluation, a quantitative research through a survey was most appropriate. The reason for this is that statistical analysis can be done to compare a diversified and large sample of respondents (Saunders & Lewis, 2012). Surveys provide formal outcomes to use in quantitative research to test hypotheses statistically. This survey contains a structured, explanatory Web questionnaire for collecting and analysing the data systematically. The study is cross-sectional, in which respondents fill out the questionnaire at one moment. 3.2.1 Measures

Questions about the perceived importance of online and offline recommendations across product categories and questions about personal characteristic statements were answered with a five point Likert-scale. All questions are shown in appendix I Questionnaire.

The questionnaire consisted of three parts: part one consisted of two variables, which were the perceived importance of online recommendations and offline recommendations across five product categories. Respondents were asked to rank this perceived importance. To be able to test whether personal characteristics are associated with the perceived importance of the recommendations, the second part of the questionnaire consisted of seventeen items about six personal characteristics. To be able to test whether demographic characteristics are

Referenties

GERELATEERDE DOCUMENTEN

Support is sought for the relationship between the main effect variables, CEO stock ownership and CEO option ownership on the development of the M-score in the period of low

Om de inschatting van de politieke oriëntatie van aanzienlijk afwijkende vrienden te berekenen is binnen de ego-netwerken van een student een onderscheid gemaakt tussen

The questions are aimed to measure the dependent variables: general satisfaction of customers in online and offline, and independent variables such as the common factors (frequency

Ook wegens gebrek aan goede alterna- tieven voor gasfiltratie bij verhoogde druk en temperatuur blijkt een magne- tisch gestabiliseerd wervelbed precies bij die condities

Ten einde die aandag van die publiek pertinent hierop te vestig en steun te verwerf vir die hermeuhilering — in die vorm van skenkings van voorwerpe of geld — het die

Next, the total number of repurchased brands, repurchased sizes, promoted, non- food, non-sensory, private label brand, premium- & national-brand and high market share

In the case of apparel product pages, it is hypothesized that including more features on the product page would decrease the feeling of risks (and increase benefits) which

When type of recommendations systems like described in this research will be introduced it is important for managers to realize that weekly reminders is the most effective