You only hear what you want to hear:
The moderating influence on the willingness to pay
of having functional and social goals of listening
to word-of-mouth
Amber Kip, s2407671
16-01-2017
Supervisor: dr. Felix Eggers Rijksuniversiteit Groningen
2
Abstract
With this research is investigated if having more functional and social goals of listening to word-of-mouth influences the impact of positive word-of-mouth on the decision-making of consumers. There is also investigated what the impact is of the volume, type of word-of-mouth (functional or social valence) and the tie strength with the sender of the word-of-word-of-mouth on the effect of positive word-of-mouth. Furthermore, it is investigated if this is again
moderated by the amount of functional and social goals of listening to word-of-mouth people have.
This research can provide more insights in the influence of the goals of consumers in how usefull they perceive information and how this affects their decision-making.
This is researched in a choice-based experiment in which consumer had to choose between different smartphones (N=53). The impact on the choice probability is translated in the effect on the incremental willingness to pay.
4
Introduction
The importance of word-of-mouth for consumer decision-making is recognized the recent years by managers, there is a lot invested in influencing and understanding this (Libai, Muller and Peres, 2013; Baker, Donthu & Kumar, 2016). The large role of word-of-mouth in the consumer decison-making is also recognized by researchers (East, Hammond & Lomax, 2008; Liu & Zhang, 2010).
The research on what aspects influence the effect that word-of-mouth has on the decision-making of consumers is growing (Kostyra, Reiner, Natter & Klapper, 2016). In research is found support for that the perceived usefulness of word-of-mouth is one of the most important aspects that influence whether consumers use word-of-mouth in their decision-making (Liu & Zhang, 2010). This can be derived from the Model of Information Adoption (Sussman & Siegal, 2003). If word-of-mouth supports the goals of consumers can be seen as an important determinant for whether word-of-mouth is perceived as usefull by consumers and therefore used in their decision-making (Liu & Zhang, 2010; Zhang, Craciun & Shin, 2010; Relling, Schnittka, Sattler & Johnen, 2016). In this research this theoretical view will be used, by investigating if positive word-of-mouth has more positive influence on the consumer decision-making if it matches the goals of consumers, and therefore is perceived as more usefull. Also findings will be integrated that have offered support that the volume of word-of-mouth (Kostyra et al, 2016),and the tie strength with the sender of word-of-word-of-mouth (Baker et al, 2016) have an influence on the effect that word-of-mouth has on the decision-making of consumers.
In this research I will investigate what the influence is of having more functional and social goals of listening to word-of-mouth on the effect that positive word-of-mouth has on the willingness to pay of consumers. It will also be investigated what the influence is of the volume of word-of-mouth, the type of word-of-mouth (social or functional) and the tie
strength with the sender of word-of-mouth on the impact of the positive word-of-mouth and if this differs through the amount of goals people have.
What distinguishes this research from most previous research is that the research on the influence of having specific consumers goals on the perceived usefulness of positive word-of-mouth (Relling et al, 2010; Zhang et al, 2010) and research on the influence of perceived usefulness of word-of-mouth on the use of word-of-mouth in consumer decision-making (Liu & Zhang, 2010) are integrated in one conceptual model. This has also be done in the research of Liu & Zhang (2010), but in this research the fulfilling of goals of listening to word-of-mouth has not been specified and the research was also not in a setting in which consumers had to choose between different produts. Further research in experimental form has also been recommended by the authors (Liu & Zhang, 2010). By this research more information can become available of what the influence is of having functional and social goals of listening to word-of-mouth on the effect that positive word-of-mouth has on the product choice of consumers. In previous research, the use of word-of-mouth in a functional manner, to lower the insecurity about the quality of a product is mostly highlighted (e.g. Kostyra et al, 2016). There is less emphasize yet that positive word-of-mouth can also have a social function by providing social validation (e.g. Rellling et al, 2016) and therefore increase the willingness to pay of consumers. This represents two different dynamics. It is also
5
strength in this. What is distinghuishing is that this research includes what the effect is of the type of word of mouth, if it has a stronger functional or social valence. In the research of Relling et al (2016) there is also called for research to investigate this.
The practical managerial relevance of this research is that consumers pay a lot of attention to word-of-mouth in their purchase decisions (e.g. Kostyra et al, 2016; Baker et al, 2016), so it is important for managers to understand the dynamics of how they are influenced. Customers could be willing to pay more when in word-of-mouth the insecurity about the quality of a product is lowered. In this research I investigate if they are also willing to pay more when they gain more social validation about a product. Also, what the effect is of the type of word-of-mouth, the volume and tie strength with the sender of word-of-mouth on the consumer decision-making could provide more insights and also if this differs for people with more functional and social goals.
This information can perhaps be used in designing information systems by brands that are adjusted to this by providing information for different goals if these appear to have an impact on the willingess to pay. The information can also be used in knowing how to encourage consumers to share word-of-mouth in the most effective manner.
The following main research questions can be formulated, with after this the hypotheses in which they are adressed:
1. Does having more functional and social goals of listening to word-of-mouth influence the perceived usefulness of positive word-of-mouth and therefore the willingness to pay? (H2, H3).
2. Is their a difference in the impact of having more functional and more social goals? (H4) 3. Is the influence of positive word-of-mouth on the willingness to pay affected by the type of word-of-mouth(H5), the tie strength with the sender of word-of-mouth(H6) and the volume of word-of-mouth(H7)?
4. Is this effect moderated by the amount of social and functional goals people have? (H5a, H6, H7a)
6
Literature summary
The effect of word-of-mouth on consumer decisions
The importance of word-of-mouth for the decision making of consumers is recognized (Liu & Zhang, 2010; Baker, Donthu & Kumar, 2016). A lot of research has been done on what causes people to express word-of-mouth (see Berger, 2014 for an overview) and the research on what the effects are of word-of-mouth on consumers is also growing (Cheung & Thadani, 2012). In this research, word-of-mouth is defined as ‘informal advice passed between consumers’ (East, Hammond & Lomax, 2008, p. 215). Within this broad definition of word-of-mouth falls research that is done on the effects of online customer reviews (Kostyra, Reiner, Natter, & Klapper, 2016) and strictly interactive forms of word-of-mouth (Baker, Donthu, & Kumar, 2016). A distinction that is made in the literature is between positive and negative word-of-mouth. In this research only the effect of positive word-of-mouth is investigated.
There is found support in previous empirical research that positive word-of-mouth has a positive influence on the product choice of consumers (East et al, 2008, Kostyra et al, 2016). There is also found support that negative word-of-mouth has a negative impact (East et al, 2008, Kostyra et al, 2016).
In this research the effect of positive word-of-mouth on the willingness to pay of consumers for a certain product is investigated. Central is the question if positive word-of-mouth has a larger influence on the willingness to pay of consumers if consumers have to a greater extent functional and social goals of listening to word-of-mouth. With a functional goal of listening to word-of-mouth is meant that consumers want to reduce their uncertainty about the quality of a product (Relling, Schnittka, Sattler & Johnen, 2016). With a social goal of listening to word-of-mouth is meant that consumers want to gain social validation about a product and/or enjoy a positive interaction about a product (Relling et al, 2016). There will be also
investigated which of the goals has the largest impact.
In this study is also investigated whether it matters that the positive word-of-mouth is expressed in a more functional manner, with the emphasis on quality aspects, or in a more social manner, with the emphasis on social aspects. Besides this there is looked at the effect of the volume of word-of-mouth, the amount of word-of-mouth that is received. There is also investigated what the effect of the tie-strength is on the effect of positive word-of-mouth, whether the word-of-mouth comes from a close tie or a weak tie (Baker et al, 2016).
Research on interaction effects
The area of research on interaction effects on the effects of word-of-mouth on consumer decision-making is growing (Kostyra et al, 2016). An example is the research of East,
Hammond & Lomax (2008) in which they found that the influence of positive word-of-mouth and negative word-of-mouth is larger when people have a lower brand purchase probability. There has also been found that when word-of-mouth comes from strong ties as friends this has a stronger positive impact on the effect of positive and negative word-of-mouth on the
purchase probability then when it comes from weak ties (Baker et al, 2016). In the research of Kostyra et al (2016) is found that for positive word-of-mouth, a higher volume of positive word-of-mouth has a more positive effect on the choice probability. It is also found that a low variance of word-of-mouth has a positive effect for positive word-of-mouth (Kostyra et al, 2016). In research is found support that if word-of-mouth is perceived as more usefull, it will be used more in consumer decision-making (Liu & Zhang, 2010; Sussman & Siegal, 2003). In research is found that the perceived information quality and perceived information
7
Siegal, 2003; Cheung & Thadani, 2012) to explain findings that the quality of word-of-mouth has a stronger impact on the perceived usefulness for people who are more involved in the decision-making.
The effect of usefulness of word-of-mouth on the consumer decision making
As mentioned previously, the idea is that when people perceive word-of-mouth as more usefull, it will have a greater influence on their purchase decision (Liu & Zhang, 2010). This has been explained by Liu & Zhang (2010) by the Model of Information Adoption (Sussman & Siegal, 2003). This is the main theoretical framework of this study. According to this model, the degree to which the information is perceived as usefull is the key factor that determines the intention to use this information. These assumptions are derived from the Theory of Reasoned Action (TRA) (Ajzen and Fishbein 1980 in Sussman & Siegal, 2003) and the Technology Acceptance Model (TAM) (Davis 1989 in Sussman & Siegal, 2003). In the research of Liu & Zhang (2010) is found support that the perceived usefulness of the information has a significant impact on the actual use of the word-of-mouth in the decision-making.
Research on the moderating role of goals of listening to of-mouth on the effect of word-of-mouth
This research focuses on investigating if positive word-of-mouth is perceived as more usefull and therefore has a greater influence on the consumer decision-making if it fulfills the goals of consumers of listening to word-of-mouth. In research on this is used as support the theory and supporting findings that the perception of people towards word-of-mouth is influenced by the degree to which it fulfills their goals (Ferguson & Bargh, 2004; Relling et al, 2016; Zhang, 2010). This is named by Relling et al (2016) the goal dependence of perception theory. When information fulfills the goals of consumers, people are more positive about it (Ferguson & Bargh, 2004). The link can be made that as argumentated by Liu & Zhang (2010), that when word-of-mouth fulfills the goals of consumers, it is perceived as more usefull.
There are two studies found in which there is empirical support found for that word-of-mouth is perceived as more usefull if it matches the goals of listening to word-of-mouth of
8
made: positive reviews are said to be more in line with the goal of wanting to hear positive information, and negative reviews are more in line with the goal of wanting to hear
information to help avoid negative outcomes.
Distinction from previous research and hypotheses
What distinguishes this research from the research of Relling et al (2016) is that there is not only compared what the influence is of having functional and social goals of listening to word-of-mouth on the effect of positive word-of-mouth, but also what the influence is of the extent to which people have functional and social goals of listening to word-of-mouth. Also is investigated what the effect is on the actual decision making of consumers and not only on the perceived usefulness. In the research of Liu & Zhang (2010) there is also investigated what the influence is of a larger perceived importance of word-of-mouth, so also the fulfilling of consumer goals, on the decision making of consumers. But in this research there has not been asked how important the information is for consumers when they read the word-of-mouth but how important it is for them when they shop online. This seems more mixed with how usefull they find it to hear word-of-mouth when they shop online. In this research specific goals of listening to word-of-mouth have not been specified and also is it a survey study and not an experiment so the effect on product choice has not been taken along. What distinguishes this research is also that the influence of whether word-of-mouth has a social or functional valence has been included. In this research there is also investigated what the influence is of having functional and social goals on the effect of tie strength, type of word-of-mouth and volume.
Hypotheses
The influence of positive word-of-mouth on the willingness to pay
It is hypothesized that positive word-of-mouth has a positive influence on the choice
probability and therefore on the willingness to pay. For this, empirical support has been found (Kostyra et al., 2016; East et al, 2008). As mentioned previously, positive word-of-mouth can fulfill consumer goals of listening to word-of-mouh and therefore be perceived as usefull for the decision making (Liu & Zhang, 2010; Zhang, 2010). As argumentated before, it is thought that positive word-of-mouth has a greater influence on the decision making of consumers if it is perceived as more usefull (Liu & Zhang, 2010; Sussman & Siegal, 2003). Although with this hypothesis is not tested the influence of having specific consumer goals of listening to word-of-mouth, it is thought that positive word-of-mouth has a positive influence on the choice probability when perceived as usefull. Therefore the following hypothesis is expected:
H1: Positive word-of-mouth has a positive influence on the willingness to pay.
The matching of PWOM and social and functional goals
A distinction between functional and social goals of consumers of listening to word-of-mouth is made in the article of Relling et al. (2016). The goals that word-of-mouth can fulfill are in this research connected to the product, because I am interested in how positive word-of-mouth can heighten the willingness to pay for products.
In this research is assumed that having social goals of listening to word-of-mouth entails that consumers want to gain confirmation that a product can fulfill certain social consumption goals of consumers and the same holds for functional goals of listening to word-of-mouth.
9 goals (Touré-Tillery & Fishbach, 2014; Xiao, 2016).
In the functional goals that someone can have there can mostly be goals related to the product itself: ‘the utility, performance and quality’ (Xiao, 2016, p. 7) (Xiao, 2016; Relling et al, 2016). These are mostly outcome-focused. It is hypothesized that when people have more functional goals of listening to word-of-mouth, the positive word-of-mouth is perceived as more usefull because it reduces their uncertainty about the quality of the product (Relling et al, 2016; Kostyra et al, 2016, Ho-Dac, Carson & Moore, 2013).
As social goals concerning the use of products and word-of-mouth can be outcome-focused motivations distinghuised as wanting to obtain social validation with a product (Na Xiao, 2016; Relling et al, 2016; Mandel, 2003) and also process-focused motivations as wanting to share excitement about a product and feel connected to others (Relling et al, 2016; Dholakia et al, 2004), benefits that lie in the interaction with others (Carlson et al, 2008). It is found in research that consumers take these social goals into account when making purchase decisions, for example if with their purchase they want to reduce the chance that they won’t gain
approval of the purchase from the people around them (Xiao, 2016; Mandel, 2003). It is therefore hypothesized that when people have more social goals of listening to word-of-mouth, the positive word-of-mouth is perceived as more usefull.
The hypotheses that follow from this are:
H2: When people have more functional goals of listening to mouth, positive word-of-mouth has a stronger positive effect on the willingness to pay.
H3: When people have more social goals of listening to mouth, positive word-of-mouth has a stronger positive effect on the willingness to pay.
The strongest effect on the willingness to pay
There is hypothesized if having more functional goals of listening to word-of-mouth or having more social goals of listening to word-of-mouth has the largest influence on the effect of positive word-of-mouth on the willingness to pay.
It can be argued that for people with more functional goals of listening to word-of-mouth, the positive word-of-mouth is most usefull in choosing a product. This line of reasoning is based on the prior purchase probability that should influence the size of the effect of positive word-of-mouth on the purchase decision (East et al, 2008). In the research of East et al (2008) is found that for people with a lower prior purchase probability, positive word-of-mouth has a larger influence on the purchase probability because there is more to gain still in purchase probability. It is expected that having more functional goals of listening to word-of-mouth relates to a lower prior purchase probability because their is still uncertainty about the quality of products in a product category. It is expected that when people have more social goals of listening to word-of-mouth, they are already quite positive about certain products in a product category. It can be argued that therefore compared to a situation in which there is no word-of-mouth present, positive word-of-word-of-mouth will heighten the willingness to pay more for people with more functional goals of listening to word-of-mouth han for people with more social goals.
10 usefull and heighten the willingness to pay.
A larger positive influence on the perceived usefulness of positive word-of-mouth for people with social goals compared to people with functional goals of listening to word-of-mouth is found in the research of Relling et al (2016), although the found difference in effect was not large. As an explanation is given by Relling et al (2016) that people with functional goals find negative word-of-mouth more usefull because it is more difficult to obtain. This mechanism can be theoretically supported with the accessibility–diagnosticity (A–D) theory of
judgmental response as argumentated by East et al (2008) (Feldman & Lynch, 1988 and Lynch, Marmorstein, & Weigold, 1988 in East et al, 2008) in which is stated that information that is less available is perceived as more usefull for forming an opinion.
Because in this research there is researched what the effect is of positive word-of-mouth on the purchase decision of people, there is expected that having more functional goals of listening to word-of-mouth has a larger effect on the purchase decision due to a greater usefulness for making a purchase decision. Therefore it is expected that having more functional goals of listening to word-of-mouth has a larger positive effect than having more social goals of listening to word-of-mouth on the effect of word-of-mouth on the willingness to pay.
H4: Having more functional goals of listening to word-of-mouth has a stronger positive effect on the effect of positive word-of-mouth on the willingness to pay than having more social goals.
Type of word-of-mouth
A distinction between more functional word-of mouth and more social word-of-mouth can be made. It is thought that because there is expected that having more functional goals has a stronger impact, there is also expected that word-of-mouth with a functional valence has a stronger positive impact on the willingess to pay than word-of-mouth with a social valence.
It can be thoought that if the word-of-mouth has a stronger functional valence, it enhances the effect of positive word-of-mouth if people have more functional goals of listening to word of mouth. The same can be thought for social goals. It is expected that this works again through a greater perceived utility of the word-of-mouth (Susmann & Siegal, 2013; Liu & Zhang, 2010).
The following hypotheses follows from this:
H5: The positive effect of word-of-mouth on the willingness to pay will be stronger when it has a functional valence than when it has a social valence.
H5a: When the valence of the positive word-of-mouth (functional or social) alignes with the goal of listening to word-of-mouth, the postive effect on the willingness to pay will be stronger.
Volume
A greater volume of word-of-mouth could make the effect of the positive word-of-mouth stronger, because it adds reliablity (Kostyra et al, 2016; Ho-Dac et al, 2013). It is found in the research of Kostyra et al (2016) that a greater volume of reviews increases the impact of positive word-of-mouth. It is expected that the volume of word-of-mouth has a stronger effect for people with functional goals of listening to word-of-mouth because they could be more interested in the accuracy of the information.
11
word-of-mouth on the willingness to pay will be stronger.
H6a: This effect will be the largest for people who have more functional goals of listening to word-of-mouth.
Tie strength
It is expected that word-of-mouth between strong ties has a greater influence on the purchase probability than word-of-mouth between weak ties (e.g. Baker et al, 2016).
In the research of Baker et al (2016) is found that positive word-of-mouth between strong ties, online as well as offline, compared to weak ties, has the largest positive effect on purchase intentions. This is because there is often more trust (McPherson, Smith-Lovin and Cook; 2001, Baker et al, 2016) between strong ties and more personalized word-of-mouth (Baker et al, 2016).
It is found in research that people are motivated to gain social validation from friends and family and to avoid situations when they get a negative response and that this affects their purchasing decisions (Mandel, 2003; Na Xiao, 2016). It is therefore expected that the opinion of friends will be perceived as more usefull for people who have more social goals than functional goals because the positive word-of-mouth can signal to them that they can gain social validation by the word-of-mouth.
H7: When the tie strength increases, the positive effect of the positive word-of-mouth on the willingness to pay will be stronger.
H7a: This effect will be the largest for people who have more social goals of listening to word-of-mouth.
12
Methodology
Choice of product
The product category that is chosen to research are smartphones. This product is chosen because there can be functional as well as social goals by listening to word-of-mouth about the product. It is a product for which the quality can play an important role, this relates tot he functional goals. Also, is it a visible product which could make social validation more important what relates to the social goals.
Measuring the goals of listening to word-of-mouth
For the creation of the two variables that reflect the extent to which consumers have functional goals and social goals of listening to word-of-mouth, there are several items
included in the questionnaire to measure these goals. By taking the average score on the items of respondents that measure social and functional goals of listening to word-of-mouth, two scale variables are created, the Cronbach’s alpha for the scale to measure the functional goals is 0,767 and for social goals 0,797.
A way to measure how strong the underlying goals are for people to listen to word-of-mouth is to research how positive they are about doing what could fulfill certain goals (Touré-Tillery & Fishbach, 2014). In previous research on which needs are important for consumers in the use of communication and technological communication platforms value perception measures are used by asking to consumers to rate how usefull they rate the media in satisfying functional goals, as ‘getting information’, as well as social goals, as ‘to impress others’ and ‘to have something to do with others’ (Flanagin & Metzger, 2001, p. 166) and how much they use social groups to fulfill this goals (Dholakia, Bagozzi & Klein Pearo, 2004). The functional and social goals can be divided in wanting to obtain certain outcome-focused goals and process-focused goals, as specified in the theory.
To assess to what degree the respondents have functional and social goals of listening to word-of-mouth, these benefits are translated in statements that highlight these goals as is did in Flanagin & Metzger (2001) and Dholakia et al (2004). To assess how positive they are about the listening to word-of-mouth when the word-of-mouth has these aspects, the activity that could fulfill the goal (Touré-Tillery & Fisbach, 2014), it is asked how important this is for them. From this could be derived how strongly the respondents possess functional and social goals of listening to word-of-mouth (Touré-Tillery & Fisbach, 2014).
Items (own development based on Flanagin & Metzger (2001), Dholakia et al (2004) and Touré-Tillery & Fishbach (2014)):
Question asked:
Imagine you are considering buying a new smartphone.
You read and/or hear things about smartphones from other consumers, this is further called word-of-mouth.
You are asked to assess to what degree you find it important to hear certain aspects when you listen to and/or read word-of-mouth about smartphones. You are asked to assess this on a 5 point scale, with 1 indicating not important at all to me and 5 indicating very important to me.
13 (Functional goals)
- people communicate that the quality of the phone is high - people communicate that the phone performs well - people communicate that I can use the phone as I wish to
(Social goals)
- people communicate that they approve of the choice for the phone - people communicate that they agree with my feelings about the phone - people communicate that they are enthusiasic about the phone
Procedure
First there was an introduction in which the structure of the survey and the central product was explained. There was also confidentiality and responsible handling of the information promised and contact information was shared. First there was asked in the survey for a few demographics. Then the questions to measure the functional and social goals were asked. The questions about functional and social goals were not asked on the same page to prevent anchoring. Subsequently a question was asked to consumers to rate their brand attachment to the smartphone brands included in the experiment. Then the choice sets were presented to the respondents.
Choice-based conjoint experiment: Measuring the impact of positive word-of-mouth on the willingness to pay
For measuring the impact of positive word-of-mouth on the willingness to pay, a choice-based conjoint experiment will be performed with a fractional factorial design. An example of a choice set can be found in figure 2. There will be made use of 12 choice sets, including one holdout set. There are three alternatives presented in each choice set. Above each choice set a text is placed in which is explained that all phones are available in comparable sizes in a good quality store. There is made use of a dual response option: after the presented choice set respondents are asked to indicate if they would buy their preferred phone if this were the alternatives. The first choice set is not used to achieve a higher reliability. There are attributes included about the product as well as the positive word-of-mouth. As attributes are included brand and price because these are seen as important in the decision-making of consumers (e.g. Kostyra et al, 2016). For brand the levels Samsung, Apple and Huawei are chosen. It is
expected that there are preferences of consumers between the brands, but the brands all have an image of delivering phones in the premium segment, so the impact of brand is not expected to be too large. For the price there are chosen the levels of 375, 450 and 525. These are
14
tie friends. This because it is difficult to specifiy these things because then there is a difference between electronical word-of-mouth and interpersonal word-of-mouth. In the research of Baker et al. (2016) there is found support that the effect of positive word-of-mouth is equal in an online and offline setting, so this distinction has not been taken into account.
Product attributes Levels Positive word-of-mouth attributes Levels Brand Samsung Apple Huawei Presence of word-of-mouth Not present Present Camera 8 megapixels 14 megapixels 20 megapixels Type of word-of-mouth
Social: ‘the phone is cool to have’
Functional: ‘the quality of the phone is high’
Price 375
450 525
Tie strength Strong tie: from friends
Weak tie: from unknown people
Volume Low volume:
a few
High volume: many
15 Figure 2: Example of choice set
As can be seen in figure 2, the different attributes and levels of word-of-mouth specified in table 1 are combined as a sentence that is included in each alternative in the choice set. There is also a sentence included to represent that no word-of-mouth is included. These items will be recoded back again into the variables tie strength, volume of word-of-mouth and type of word-of-mouth to analyze their impact.
Items that are included as word-of-mouth in the choice sets : No positive word-of-mouth heard or read yet (3x)
WOM from a few friends that the phone is cool to have WOM from many friends that the phone is cool to have
WOM from a few unknown people that the phone is cool to have WOM from many unknown people that the phone is cool to have WOM from a few friends that the quality of the phone is high WOM from many friends that the quality of the phone is high
16 Model
The model that is investigated in this research, presented as a utility function and logit model:
Utility function: 𝑉𝑖 = 𝛽𝐵𝐵𝑟𝑎𝑛𝑑𝑖 + 𝛽𝐶𝐶𝑎𝑚𝑒𝑟𝑎_𝑚𝑝𝑖+ 𝛽𝑃𝑃𝑟𝑖𝑐𝑒𝑖+ 𝛽𝑁𝑊𝑁𝑜_𝑤𝑜𝑚𝑖 + 𝛽𝑇𝑦𝑇𝑦𝑝𝑒𝑖 + 𝛽𝑉𝑉𝑜𝑙𝑢𝑚𝑒𝑖 + 𝛽𝑇𝑇𝑖𝑒𝑖 + 𝛽𝑁𝑁𝑜𝑛𝑒𝑖 + 𝛽𝐹𝑁𝑊 𝐹𝑢𝑛𝑐𝑡𝑖𝑜𝑛𝑎𝑙𝑔𝑜𝑎𝑙𝑠𝑖 ∙ 𝑁𝑜𝑤𝑜𝑚𝑖 + 𝛽𝑆𝑁𝑊 𝑆𝑜𝑐𝑖𝑎𝑙𝑔𝑜𝑎𝑙𝑠𝑖? ∙ 𝑁𝑜𝑤𝑜𝑚𝑖 𝛽𝐹𝑇𝑦 𝐹𝑢𝑛𝑐𝑡𝑖𝑜𝑛𝑎𝑙𝑔𝑜𝑎𝑙𝑠𝑖 ∙ 𝑇𝑦𝑝𝑒𝑖 + 𝛽𝐹𝑉 𝐹𝑢𝑛𝑐𝑡𝑖𝑜𝑛𝑎𝑙𝑔𝑜𝑎𝑙𝑠𝑖 ∙ 𝑉𝑜𝑙𝑢𝑚𝑒𝑖 + 𝛽𝐹𝑇 𝐹𝑢𝑛𝑐𝑡𝑖𝑜𝑛𝑎𝑙𝑔𝑜𝑎𝑙𝑠𝑖 ∙ 𝑇𝑖𝑒𝑖 + 𝛽𝑆𝑇𝑦 𝑆𝑜𝑐𝑖𝑎𝑙𝑔𝑜𝑎𝑙𝑠𝑖 ∙ 𝑇𝑦𝑝𝑒𝑖 + 𝛽𝑆𝑉 𝑆𝑜𝑐𝑖𝑎𝑙𝑔𝑜𝑎𝑙𝑠𝑖 ∙ 𝑉𝑜𝑙𝑢𝑚𝑒𝑖 + 𝛽𝑆𝑇 𝑆𝑜𝑐𝑖𝑎𝑙𝑔𝑜𝑎𝑙𝑠𝑖 ∙ 𝑇𝑖𝑒𝑖 Logit model: 𝑝𝑟𝑜𝑏(𝑖 | 𝐽) = exp(𝑉𝑖) ∑3𝑗=1exp (𝑉𝑗)
First, there will be investigated what the influence is of the presence of positive word-of-mouth on the choice probability of the respondents. Then there will be looked if there is a moderation of this effect by the amount of functional and social goals people have. This will be done by adding an interaction term between the attribute that indicates the presence of word-of-mouth and the scale variables of functional and social goals. It will also be
investigated whether there is a moderation effect of the goals of listening to word-of-mouth on the effect of the type of word-of-mouth the volume and the tie strength. There will also be interaction terms added between these attributes and the goal variables to test this. The willingness to pay is calculated based on the choices made.
Data collection
The data will be collected by gathering a sample through personally asking possible
respondents to fill in the questionnaire online, via Facebook and e-mail. The largest part of the respondents are expected to be students.
Descriptives of sample
Descriptives of the used sample can be found in table 2.
Number of respondents Gender Age
53 30,2% male, 69,8% female Average age: 25,3
Functional goals Social goals
Min: 2, max: 5, mean: 4,09 Min: 1, max: 5, mean: 3,02
17
Results
The results of the choice-based conjoint analysis can be found in table 3.
Attributes Β z-value SE Wald Importance
Brand - Samsung - Apple - Huawei ,2271 ,2252 -,4523 3,9048 3,8475 -6,8168 ,0582 ,0587 ,0664 46,4760*** 20,57% Camera megapixels - 8 mp - 14 mp - 20 mp -,3603 ,0733 ,2871 -5,5226 1,2416 4,9610 ,0652 ,0590 ,0579 35,8901*** 19,60% Price (linear) -,0066 -9,1280 ,0007 83,3196*** 29,97% Tie strength weak -,1638 -3,0057 ,0545 9,0339** 4,96% Volume many -,0128 -0,2330 ,0551 ,0543 0,39% Type of word-of-mouth functional ,1969 3,6447 ,0540 13,2838*** 5,96% No word-of-mouth -,6130 -5,4060 ,1134 29,2246*** 18,56% None option -3,3662 -10,3333 ,3258 106,7769***
*** = p-value <.001; ** = p-value <.01; * = p-value <.05.
18 Log Likelihood B McFadden’s R2
adjusted
Goodness of fit: Predictive validity
-827,2266 p(Chisq)<0,001
0,116 Mean absolute error
= 5%
Table 4: Model fit statistics
As can be seen from table 3, there is found support for hypotheses 1: Positive word-of-mouth has a positive influence on the willingness to pay. This can be seen from the negative béta of no word-of-mouth (β= -,6130 , p=<0,001). When there is positive word-of-mouth presented, the probability of choosing the alternative is 64,862%. The willingness to pay increases when they are confronted with positive word-of-mouth, consumers are willing to spend €92,88 more compared to a situation when there is no word-of-mouth present.
There is found support for the hypothesis 5 that it matters for the effect of positive word-of-mouth on the willingness to pay whether the word-of-word-of-mouth has a functional or a social valence. Functional valenced word-of-mouth has a significantly positive effect on the choice probability compared to social valenced word-of-mouth (β= ,1969, p=< 0,001) (table 3). This means that compared to a choice situation in which there is positive word-of-mouth present, the probability of choosing the alternative when the positive word-of-mouth has a social valence increases with 30.79%. The probability of choosing the alternative when the positive word-of-mouth has a functional valence increases with 39,75% compared to the average when word-of-mouth is present, so 8,96 % more than word-of-mouth with a social valence. When the word-of-mouth has a functional valence, the willingness to pay increases with €29,83 compared to the average increase in willingness to pay when there is positive word-of-mouth presented compared to no positive word-of-mouth. When the word-of-mouth has a social valence the willingness to pay decreases with €29,83 compared to the average increase in willingness to pay when there is positive word-of-mouth presented compared to no positive word-of-mouth.
From table 3 can be seen that there is found support for hypothesis 6 that when the tie strength increases with the sender of the word-of-mouth, this has a positive effect on the willingness to pay. This can be seen from the negative bèta of tie strength weak (β= -0,1638, p=<0,01). This means that compared to a choice situation in which there is positive word-of-mouth present, the probability of choosing the alternative when the positive word-of-word-of-mouth is from weak ties increases with 31,5%. The probability of choosing the alternative when the positive word-of-mouth is from strong ties increases with 39% compared to the average when of-mouth is present, so 7,5 % more than of-mouth for weak ties. When the word-of-mouth is from strong ties the willingness to pay increases with €24,82 compared to the average increase in willingness to pay when there is positive word-of-mouth presented compared to no positive word-of-mouth. When the word-of-mouth is from weak ties the willingness to pay decreases with €24,82 compared to the average increase in willingness to pay when there is positive word-of-mouth presented compared to no positive word-of-mouth.
19
willingness to pay increases as the volume of word-of-mouth increases. The effect is not significant (β= -,0128, p=> ,05).
When looking at the importance (table 3), the importance of word-of-mouth is almost as high as that of brand.
The total model predicts signficantly better than the null model (table 4) (Chisq=244,813, p<0,001). Also when estimated on the holdout sample, Mean Absolute Error is only 5% (see table 4). The McFadden’s R2 adjusted seems allright but could be heightened, for example by using segmentation.
Table 5 moderation 1 with mean centering
*** = p-value <.001; ** = p-value <.01; * = p-value <.05.
Attributes Β z-value SE Wald
20 None option -3,3713 -10,2925 ,3276 105,9348*** Funtional goals1 x no wom -,1333 -,7172 ,1859 ,5144 (p=,47) Social goals1x no wom -,0880 -,7512 ,1172 ,5643 (p=,45) Functional1x tie strength weak -,0732 -,8048 ,0910 ,6476(p=,42) Functional1x volume many ,0779 ,8316 ,0937 ,6915(p=,41) Functional1x type wom functional -,0962 -1,0860 ,0886 1,1793(p=0,28) Social1x tie strength weak -,0533 -,8750 ,0609 ,7656(p=,38) Social1x volume many ,0716 1,1538 ,0621 1,3313 (p=,25) Social1x type wom functional ,0608 1,0206 ,0595 1,0417 (p=,31)
*** = p-value <.001; ** = p-value <.01; * = p-value <.05. Table 5: moderation 1 with mean centering
Log Likelihood B McFadden’s R2 adjusted
-823,5949 p(Chisq)<0,001
0,112
21
As can be seen from table 5 in which the predicted moderations are tested, none of the tested moderations are significant. There is no support found for hypotheses 2 and 3, that having more functional and social goals of listening to word-of-mouth leads to a stronger positive effect of positive word-of-mouth on the willingness to pay. The interaction between the presence of positive word-of-mouth and functional goals (β= -0,1333, p= 0,47) and the interaction between the presence of positive word-of-mouth and social goals (β= -0,0880, p= 0,45) both are non-significant. The negative sign would otherwise have indicated that as people have more functional and social goals, they have a more negative effect of having no word-of-mouth present on their choice probability. Because of the non-significant results there can also not be made inferences about which goal of listening to word-of-mouth has the strongest influence, as hypothesized in hypothesis 4.
Also there is no support found for hypotheses 6a, 7a and 8a, that the effect of the type of word-of-mouth, volume and tie strength differs for people with more functional and social goals. The interaction of type of word-of-mouth and functional goals (β= -0,0962, p= 0,28), volume and functional goals (β= 0,0779, p= 0,41) and tie strength and functional goals (β= -0,0732, p= 0,42) are all non-significant. The interaction of type of word-of-mouth and social goals (β= 0,0608, p= 0,31), volume and social goals (β= 0,0716, p= 0,25) and tie strength and social goals (β= -0,0533, p= 0,38) are also not significant.
22
Summary
In this research a choice-based conjoint experiment is performed to investigate if having more functional and social goals of listening to word-of-mouth leads people to have a greater positive influence of positive word-of-mouth on their willingness to pay. The product for which alternatives were presented in the choice sets were smartphones.
There is found support for that the presence of positive word-of-mouth about the products increases the choice probability for smartphones and therefore also the willingness to pay. There is not found support that this is influenced by the amount of functional and social goals that people have. There is found support that the presence of functional word-of-mouth increases the choice probability for people more than social word-of-mouth. There has also been found support that word-of-mouth from strong ties, friends in this experiment, increases the choice probability more than word-of-mouth from weak ties, unknown people in this experiment. There has not been found support that the volume of word-of-mouth has an impact on the choice of consumers. The influence of the tie strength with the sender of word-of-mouth and the type of word-word-of-mouth could not be explained by the amount of social and functional goals people have.
Implications
In the results there is found that the effect of having more functional and social goals of listening to word-of-mouth on the effect of word-of-mouth on the willingness to pay is very small compared to the other effects. Because this effect has found to be insignificant, there can’t be said with certainty something about the direction of the effects, but in the results having more goals seems to increase the effect of word-of-mouth on the willingness to pay. This is what theoretically was expected. There are a few possible explanations why the effect is smaller than expected.
It could be that the effect of having more functional and social goals already has an effect from a certain treshold, that the effect doesn’t become larger when people have the goals stronger. To investigate this more with the data, the tests were performed on data split into datasets for people with lower functional goals, higher functional goals, lower social goals, higher social goals and lower social and lower functional goals and higher social and lower functional goals (see appendix 2 for a description and results). This can point to interesting interaction effects between having social and functional goals. Only for having both low functional and low social goals there is not found an effect of word-of-mouth. When only one of the two goals is higher, there still seems to be an effect, but the combination of having both of the two goals lower seems to make the usefulness of word-of-mouth indeed lower and therefore not having a significant impact on the choice. This could point to that having more social and functional goals can increase the usefulness of word-of-mouth, but that the utility is already high enough from having one of the goals, that more goals doesn’t addd to the utility. The tests could be less reliable because the number of cases included is sometimes low due to a lower number of people with lower functional goals and some correlation between the two goals.
A research in which the impact of perceived utility of positive word-of-mouth on the intention to use it in decison-making is investigated is the research of Liu & Zhang (2010). In this survey study the impact was significantly positive and quite large.
23
impact is of the kind of goals that someone has on the perceived usefulness of positive word-of-mouth. Here is found that the kind of goals does have a significant impact on how usefull the word-of-mouth is perceived. In the studies of Relling et al (2016) and Zhang (2010) the fulfillment of consumer goals is not operationalized as fulfilling more goals but the goals are manipulated. It could be that that explains why there is found there a significant impact of the goals on the perceived usefulness of word-of-mouth. In both of the studies of Liu & Zhang (2010) and Relling et al (2016), the impact of positive word-of-mouth is investigated either on the usefulness or the decision to use it. What distinguishes this study from both these studies is that in this study not a survey method is used but a choice-based conjoint experiment in which the word-of-mouth is placed in an environment where consumers have to make a decision between other products and also take into account features as the price and brand when making a decision. That could make the fulfillment of more goals of listening to word-of-mouth have a smaller impact on the perceived usefulness of word-word-of-mouth for the decision-making.
Performing a segmentation analysis (see output appendix 3) shows that when segmenting into two and three classes, there can only be found significant differences between the classes in brand. People who are more loyal to Apple, Samsung or a mix of the two. The difference in functional and social goals is not significant. This could show that other factors in the decison-making as the brand attachment could be very important in explaining the decision-making of consumers. Also, when looking at the correlations between the reported brand attachment and the score on the two goals, the brand involvement of people with social goals seems higher than for people with functional goals. When investigating the impact of having social and functional goals of listening to word-of-mouth further in research it can be
interesting to also take into account the role of brand attachment.
This research has found support that word-of-mouth has a large impact on the decision-making of consumers. This supports previous research that managers should have attention for this aspect. Because there has not been found support with the moderation analyses that the impact of positive word-of-mouth varies for people with more functional and social goals of listening to word-of-mouth, this can show that it is more easy for managers to target effectivley people by word-of-mouth in information systems to influence their decision-making. More research into this is recommended that takes along the possible explanations for the findings in this research.
There has been found support that positive word-of-mouth has a more postive impact when it comes from friends than from unknown people, this supports previous research (Baker et al, 2016). There has not been found a difference for people with more functional and social goals as was expected. A practical recommendation for brands could be to use more positive word-of-mouth from friends, for example to show recommendations from friends on Facebook. There has been found support that functional positive word-of-mouth is more influential than social positive word-of-mouth. The testing of this in this way is new. For both of these results there is not found a moderating impact of the amount of social and functional goals of
listening to word-of-mouth someone has. This is not as expected. It seems that for both goals positive word-of-mouth from strong ties with a functional valence has the strongest impact on the choice. This could be usefull information for brands in using word-of-mouth in
information systems.
24
positive, the postive effect for volume is only found here for very postive word-of-mouth. It can be that when there is also negative of-mouth included, the volume of positive word-of-mouth becomes more important because people are also presented with higher volumes of more negative word-of-mouth.
Limitations
A limitation of the study is that there are not many people included who have a low amount of functional goals, this could make it more difficult to find an interaction effect. It is also a limitation that there is only one item for social word-of-mouth included and one item for functional word-of-mouth, including more different items can give more certainty about the found results for the effect of the type of word-of-mouth.
Also is it hypothesized that the effect of the goals of listening to word-of-mouth have an effect through the perceived usefulness of word-of-mouth, but the perceived usefulness has not been measured because the effect is researched through choices in the experiment. It is not sure with this research if the impact of the goals on the word-of-mouth goes through the
25
References
Baker, A., Donthu, N., & Kumar, V. (2016). Investigating How Word-of-Mouth Conversations About Brands Influence Purchase and Retransmission Intentions. Journal Of Marketing Research, 53(2), 225-239.
http://dx.doi.org/10.1509/jmr.14.0099
Berger, J. (2014). Word of mouth and interpersonal communication: A review and directions for future research. Journal Of Consumer Psychology, 24(4), 586-607. http://dx.doi.org/10.1016/j.jcps.2014.05.002
Carlson, K., Janiszewski, C., Keeney, R., Krantz, D., Kunreuther, H., & Luce, M. et al.
(2008). A theoretical framework for goal-based choice and for prescriptive
analysis. Marketing Letters, 19(3-4), 241-254. http://dx.doi.org/10.1007/s11002-008
9043-4
Cheung, C. & Thadani, D. (2012). The impact of electronic word-of-mouth communication: A literature analysis and integrative model. Decision Support Systems, 54(1), 461-470. http://dx.doi.org/10.1016/j.dss.2012.06.008
Dholakia, U., Bagozzi, R., & Pearo, L. (2004). A social influence model of consumer participation in network- and small-group-based virtual communities. International Journal Of Research In Marketing, 21(3), 241-263.
http://dx.doi.org/10.1016/j.ijresmar.2003.12.004
East, R., Hammond, K., & Lomax, W. (2008). Measuring the impact of positive and negative word of mouth on brand purchase probability. International Journal Of Research In Marketing, 25(3), 215-224. http://dx.doi.org/10.1016/j.ijresmar.2008.04.001
Ferguson, M. & Bargh, J. (2004). Liking Is for Doing: The Effects of Goal Pursuit on
Automatic Evaluation. Journal Of Personality And Social Psychology, 87(5), 557-572. http://dx.doi.org/10.1037/0022-3514.87.5.557
Flanagin, A. & Metzger, M. (2006). Internet use in the contemporary media environment. Human Communication Research, 27(1), 153-181. http://dx.doi.org/10.1111/j.1468-2958.2001.tb00779.x
26
Psychology, 72(3), 515-525. http://dx.doi.org/10.1037//0022-3514.72.3.515
Ho-Dac, N. N., Carson, S. J., & Moore, W. L. (2013). The effects of positive and negative online customer reviews: Do brand strength and category maturity matter? Journal of Marketing, 77(6), 37–53.
Kostyra, D., Reiner, J., Natter, M., & Klapper, D. (2016). Decomposing the effects of online customer reviews on brand, price, and product attributes. International Journal Of Research In Marketing,33(1), 11-26. http://dx.doi.org/10.1016/j.ijresmar.2014.12.004
Libai, B., Muller, E., & Peres, R. (2013). Decomposing the Value of Word-of-Mouth Seeding Programs: Acceleration Versus Expansion. Journal Of Marketing Research, 50(2), 161-176. http://dx.doi.org/10.1509/jmr.11.0305
Liu, R. & Zhang, W. (2010). Informational influence of online customer feedback: An empirical study. Journal Of Database Marketing & Customer Strategy
Management, 17(2), 120-131. http://dx.doi.org/10.1057/dbm.2010.11
Mandel, N. (2003). Shifting Selves and Decision Making: The Effects of Self-Construal Priming on Consumer Risk-Taking. Journal Of Consumer Research, 30(1), 30-40. http://dx.doi.org/10.1086/374700
McPherson, M., Smith-Lovin, L., & Cook, J. (2001). Birds of a Feather: Homophily in Social Networks. Annual Review Of Sociology, 27(1), 415-444.
http://dx.doi.org/10.1146/annurev.soc.27.1.415
Relling, M., Schnittka, O., Sattler, H., & Johnen, M. (2016). Each can help or hurt: Negative and positive word of mouth in social network brand communities. International Journal Of Research In Marketing, 33(1), 42-58.
http://dx.doi.org/10.1016/j.ijresmar.2015.11.001
Sussman, S. & Siegal, W. (2003). Informational Influence in Organizations: An Integrated Approach to Knowledge Adoption. Information Systems Research, 14(1), 47-65. http://dx.doi.org/10.1287/isre.14.1.47.14767
Touré-Tillery, M. & Fishbach, A. (2014). How to Measure Motivation: A Guide for the Experimental Social Psychologist. Social And Personality Psychology Compass, 8(7), 328-341. http://dx.doi.org/10.1111/spc3.12110
Xiao, N. (2016). How non-consumption goals (elicited by competitive setting or social risk) and self-confidence influence the importance of trivial attributes in product
27
28
Appendix
1. List of figures and tables
Figure 1: Conceptual model, p. 11 Figure 2: Example of choice set, p. 15
Table 1: Overview of attributes included in the choice sets, p. 14
Table 2: Descriptives of sample, p. 16
Table 3: Results choice based conjoint, p. 17
Table 4: Model fit statistics, p. 18
Table 5: Moderation, p. 19 - 20
29 Appendix 2: additional investigation interaction effect
There is investigated what the effect is of positive word-of-mouth for smaller datasets, only including people with a higher amount of functional and social goals and a lower amount of social goals. Because the remaining effect for people with lower social goals could also be because of a higher amount of functional goals of these people and vice versa, there is investigated what the effects are for people with higher social goals and lower functional goals and people with higher social goals and higher functional goals, higher functional goals and lower social goals etc. This has been done by splitting the data into these datasets, in order to try to make the correlation between the goals less impactfull. The correlation between the two goals is 0,253, so that is also not very high. There is found that the effect for people with higher social goals and lower functional goals is positive and almost significant
30 Low Social goals:
N = 24
High social goals:
31 High functional goals:
N = 40.
Low functional goals:
N= 13.
Social low functional low:
32 Social low functional high:
Social high, functional low
N=7.
