Freemium business model in relation to
purchase intention and product strategy
“A survey and multiple case study on the effects of purchase intention and product
strategies for a new business model in the music industry”
University of Amsterdam
Master Thesis: Final version
MSc Business Administration: Strategy track
Name of institution Amsterdam Business School
Supervisor: dhr. dr. drs. S. von Delft
Student name: Joris Damave
Student ID: 10464360
STATEMENT OF ORIGINALITY
This document is written by Joris Damave, who declares to take full responsibility for the contents of this document.
I declare that the text and the work presented in this document is original and that no sources other than those mentioned in the text and its references have been used in creating it.
The Faculty of Economics and Business is responsible solely for the supervision of completion of the work, not for the contents.
ACKNOWLEDGMENTS
This thesis is written for the Master Business Studies on the University of Amsterdam, Strategy track. Before this Master program I graduated at the University of Applied Science in 2013 on Sport, Management and Entrepreneurship at the Hogeschool of Amsterdam. I had an enormous drive to continue studying and signed up for the pre-master of this master program. Even though I was rejected for this program at first, a detour allowed me to still take part of this master.
Finishing this master study at the UvA therefore feels even more satisfying. I have developed myself immensely in many different areas and while doing so also as a human in total. This development has also been made possible by a number of people. I would like to thank Wim van Riet for his important role in my development as a student and the enormous patience he has had in my younger years. I would like to thank Laurien Drenth, Eva Damave and, Cor Strobos because they often assisted me with support on my grammar. And finally, I want to thank Stephan von Delft for his support and feedback as supervisor for this thesis.
Amsterdam, 25 June 2015
ABSTRACT
More and more online services use the freemium business model. A good example would be the music industry, where new companies offer music streaming as a freemium service. Freemium business models are used to give customers the chance to try the product for free and attract as many customers as possible. Then, these companies push customers to the premium (paid) version of their service to generate revenue. In this process it is important to determine what factors influenced the customer's purchase intention. For companies it is important to know which product strategy they should use to convince customers as well. This study examines the different factors that influence the purchase intention of customers in freemium music services. Besides that, it focuses on the effects of differentiation and low cost product strategies in the process of purchase intention. Although purchase intention is a widely investigated topic, for freemium services this process is perhaps more complicated and works different because customers already use the product. In order to examine the antecedents of purchase intention, a survey was distributed via Facebook and email. A total 124 completed the questionnaire. Brand attitude, moral awareness and subjective norms seem to have no significant impact on purchase intention. Moral awareness however, appears to have a negative effect on subjective norms. The effect of product strategies is examined through a case study about the top three music-streaming services in the questionnaire: Spotify, SoundCloud and Deezer. According to the results a differentiation strategy has a positive effect on the relationships between brand attitude, moral awareness and subjective norms on purchase intention, in contrast to the low cost approach. Although most hypotheses are rejected, this study provides a good starting point of the decision process of customers related to the freemium business model and the effect of different product strategies in this new business model.
Keywords: freemium business model, brand attitude, moral awareness, subjective norms, purchase intention, product strategies, differentiation and low cost approach.
TABLE OF CONTENTS
1 Introduction ... 7
1.1 Practical contribution ... 8
1.2 Theoretical contribution ... 9
2 Literature review ... 10
2.1 Freemium as a business model ... 10
2.2 Purchase intention ... 12
2.3 Brand attitude ... 13
2.4 Subjective norms ... 14
2.5 Moral awareness ... 15
2.5.1 Moral awareness on purchase intention ... 16
2.5.2 Moral awareness on brand attitude ... 17
2.5.3 Moral awareness on subjective norms ... 18
2.6 Product positioning strategy ... 20
2.6.1 Brand attitude ... 21 2.6.2 Subjective norms ... 22 2.6.3 Moral awareness ... 23 3 Research method ... 24 3.1 Procedure ... 24 3.2 Data collection ... 25 3.3 Operation variables ... 26 3.3.1 Brand attitude ... 26 3.3.2 Subjective norms ... 27 3.3.3 Moral awareness ... 27 3.3.4 Purchase intention ... 27 3.4 Data analysis ... 28 3.5 Case selection ... 28
4 Results ... 30
4.1 Model evaluation ... 30
4.2 Measurement model ... 30
4.2.1 Reliability and Validity ... 31
4.2.2 Composite reliability and Convergent validity ... 32
4.2.3 Conclusion reliability and validity ... 33
4.3 Structural model ... 33 4.4 Moderating effects ... 36 4.4.1 Occupation ... 36 4.4.2 Gender ... 37 4.5 Summary ... 39 5 Case studies ... 41 5.1 Spotify ... 41 5.2 SoundCloud ... 43 5.3 Deezer ... 45
6 Effect of product strategies ... 47
6.1 Brand attitude and purchase intention ... 47
6.2 Subjective norms and purchase intention ... 49
6.3 Moral awareness and purchase intention ... 50
7 Discussion and Conclusion ... 51
7.1 Theoretical implications ... 55
7.2 Managerial implications ... 56
7.3 Limitations ... 57
8 Literature ... 59
1 INTRODUCTION
Over the past decade the business model freemium has become a dominant business model in different industries. The idea behind this business model is that the core service or product is free, in order to generate a large user base, while additional features or functionalities can be purchased by users in a premium account (Wilson, 2006). Although the popularity and clear benefits, the business model works different in different industries and is difficult to understand (Cheng & Tang, 2010; Kumar, 2014). This research focuses on the music industry, because the number of companies using this business model is rising but the results are disappointing for these companies (Dredge, 2015).
If companies will succeed with a freemium business model, it is hence important to convince customers to purchase the premium version in order to generate revenue. Customers’ intention to purchase a service is an important aspect because it has a
significant prediction on someone’s actual buying behavior (Chang & Wildt, 1994; Fishbein & Ajzen, 1975; Wang, Yeh, & Liao, 2013). The purchase intention may be more complicated in this business model because customers already use the free service, and may not
understand why they have to switch to the premium version. In order to begin this exploratory study of the purchase intention, this study starts with the theory of reasoned action (TRA). The TRA explains the process of purchase intention based on two drivers, namely attitude and subjective norms (Ajzen, 2005; Fishbein & Ajzen, 1975; Hartmann & Apaolaza-Ibáñez, 2012). Then, within this model moral awareness is added to assess which effect moral awareness has on brand attitude and subjective norms, as well as the effect on purchase intention. Moral awareness is included in this study because it plays an important role in today's society and many companies are dealing with it (Savitz, 2012).
The research field of freemium business models is not yet explored and these three drivers are proven to have influence on the purchase intention in other business models. This study will investigate which effect brand attitude, subjective norms, and moral awareness has on the purchase intention of the premium version. Also it investigates which effect moral
awareness has on brand attitude and subjective norms of the free version of the music services. The following research question have been formulated:
RQ: Which effect do brand attitude, subjective norms and moral awareness have on purchase intention of the premium version of music streaming services, and which effect does moral awareness have on brand attitude and subjective norms in this process?
Further, the effect of product strategies on these drivers will be investigated. Product strategies refers to how firms allocate their resources and realize its goals, and how firms compete in the market with their products (Wei, Samiee, & Lee, 2013). Successful product strategies will result in a competitive advantage. Porter (1980) identified three generic strategies for companies to create competitive advantage, namely low cost, differentiation and focus strategy. These generic strategies are widely used in the business literature and will be used in this thesis (Ha & Janda, 2012; Yi & Yoo, 2011). However, the focus approach is more a sub-strategy and in this research is therefore focused only on low cost and
differentiation. In this study the following research question is formulated:
RQ: Which effect has differentiation and low cost strategies in freemium music services on the relationship between brand attitude, subjective norms and moral awareness on the purchase intention?
1.1 Practical contribution
Findings of this study support organizations that are considering to implement a freemium business model, because companies attain a better understanding whether they should implement the freemium model or not. In addition to other studies, companies get a broader and more complete view of the conversion process in freemium models. This study in particular gives organizations with insights on the importance of brand attitude, subjective
norms and moral awareness in the purchase intention. Also, organizations have insights into the effects of the differentiation and low cost strategies into the purchase intention.
1.2 Theoretical contribution
This thesis offers insights in the drivers (brand attitude, subjective norms and moral
awareness) of purchase intention and influence of product strategy. The freemium business model still has a lot of undefined areas, purchase intention being one of the most important ones (Cheng & Tang, 2010). This study contributes to the research field of freemium by investigating three variables of purchase intention; brand attitude, subjective norms and moral awareness. Previous studies concluded that the effect of these drivers is significant on the purchase intention in other business models (Ajzen, 2005; Benjamin Tan, 2002;
Hartmann & Apaolaza-Ibáñez, 2012). However, the freemium model works differently than other business models because customers already use the service before they decide whether to buy the product. This study examine whether these drivers also affect the purchase intention in the music industry for the freemium business model.
In the following section of this research is examined which effect the product strategies has in the process of purchasing intention. The effect of product strategies on purchase intention has been investigated (Collins-Dodd & Lindley, 2003; Li & Li, 2008; Yi & Yoo, 2011), but this study will identify what effect product strategies has on the freemium business model. In addition, it is one of the first studies that investigate the variables of the purchase intention linked with the effect of product strategies.
2 LITERATURE REVIEW
In this section the literature regarding this study is outlined and hypotheses are formulated. The research exists of two different sections. First, freemium is discussed and the process of purchase intention is defined. Second, the effects of product strategies on the relationships of purchase intention are outlined.
2.1 Freemium as a business model
Business models describe the way an organization creates, delivers and captures value (Teece, 2010). With the introduction of the digital economy, firms had the opportunity to experiment with different business models in order to create value (McGrath, 2010). A new concept is the freemium business model. Freemium is made possible because of the digital world, where companies can increase the number of users without much extra cost (Kumar, 2014). In the freemium business model companies give their service away for free, might use adds to generate revenue and acquire a lot of customers efficiently through word-of-mouth, referral networks, organic search marketing. When customers become a member, premium priced value added services or an enhanced version of the service are offered. Companies then hope customers are convinced of the free version and switch to the premium (paid) version (Anderson, 2009; Wilson, 2006).
Industry
While freemium was first only used in the software industry, more and more other industries like newspapers, music, publishing and education got familiar with the freemium model (Kumar, 2014). Although freemium, in the basis, is about generating lots of users, the use of the concept can differ per industry. In the gaming industry they focus on short-term revenues, for example customers make a one-time payment in exchange for unlockable items, upgrade of characters or to enter the next levels. In the music industry they attract customers on the basis of extra features in exchange for a monthly payment. Within the different industries
companies can differentiate freemium business models based on quantity, feature or
distribution (Pujol, 2010). Thus, factors that influence the success of the freemium model are different, depending on the industry and how freemium is used. This study will focus only on the music industries where the music provider Spotify, Pandora and SoundCloud are well known examples that use freemium. The music industry is chosen because several studies have already investigated freemium within this industry, but still the process of why
customers decide to purchase the premium version is not fully understood (Cheng & Tang, 2010; Laran & Tsiros, 2013; Thomes, 2013).
Value creation and capturing
Value creation and capturing is often used to explain how business models function within an industry or market (Brandenburger & Stuart, 1996; Zott, Amit, & Massa, 2011).
Brandenburger and Stuart (1996) explain how firms create and capture value. Value creation derives from the vertical chain of players, suppliers, firms, and buyers and is the willingness-to-pay of buyers minus the opportunity costs. Value capturing is how value is divided among the different players in the chain.
According to Zott et al., (2011) value creation is the relative amount of value that is subjectively realized by the organization for the customer. This subjective value realization must at least translate into the user’s willingness to pay for the value received. Firms develop core competencies, capabilities and positional advantages that focus on the creation of value for customers. With the introduction of the freemium business model in the music industry, customers were given the option to stream music on a legally and low budget way.
Customers could unlimited stream music, listen to almost every song and use it on multiple devices. The music streaming services had added value in comparison with the expensive CD’s and illegal music downloading.
Value capturing is the process by which the value is divided among all players (Zott et al., 2011). According to Brandenburger and Stuart (1996) bargaining power among the players determines how the value is divided. The amount of bargaining power a consumer,
firm or supplier has, determine how much value they can capture. In relation with the freemium business model customers have the strongest bargaining power because they make the decision to purchase or download free music.
In relation with freemium business model the value capturing process is complicated. Customers want to use a music streaming service at a low price and companies want customers switching to the paid version in order to increase their sales. Since the introduction of the freemium model in the music industry, companies have problems convincing their customer to purchase the premium version (Cheng & Tang, 2010; Kumar, 2014). If companies want to succeed using the freemium business model, they need to understand the consumer’s intention to purchase the premium version.
2.2 Purchase intention
Purchase intention is an important aspect in the marketing literature because it represents what customers want to purchase in the future, and therefore is one of the most important predictors of purchase behavior (Fishbein & Ajzen, 1975; Wang, Yeh & Liao, 2013). In the freemium business model the purchase intention is related to customers who use the free version of the product and would like to purchase the premium version. This study will focus on the different drivers for customers to purchase the premium version.
Online purchase intention has widely been investigated over the last few years (Hartmann & Apaolaza-Ibáñez, 2012; Pan, Kuo, Pan, & Tu, 2013; Wang et al., 2013). The different factors that can affect the online purchase intention are immense. A recent study about purchase intention for online shopping already mentioned seven different drivers, usefulness, ease of use, compatibility, privacy, security, normative beliefs and attitude (Pan et al., 2013). In order to understand the different drivers of purchase intention the theory of reasoned action will be used (TRA). TRA suggests that consumer behavior to purchase can be predicted by their intentions (Fishbein & Ajzen, 1975). According to this theory behavioral intention is influenced by someone’s attitude and subjective norms. This study will focus on
brand attitude because product attitude is already investigated in the freemium business model. These will be discussed in the next two paragraphs. Not part of the TRA is the moral awareness, but the importance of this driver is increasing (Bateman, Valentine, & Rittenburg, 2012; K.-H. Lee & Shin, 2010), and therefore this is the last variable. In this study the effect of moral awareness is also examined on brand attitude and subjective norms, because moral awareness may have an indirect impact on the purchase intention. The effects of these three drivers will be examined in this study.
2.3 Brand attitude
Brand attitude is an expression of favor or disfavor towards a brand. Attitudes are formed of a consumer past and present (Petty & Krosnick, 2014). Brand attitude is an important aspect in the research literature because this psychological construct is tightly correlated with purchase intention (Hartmann & Apaolaza-Ibáñez, 2012; MacKenzie & Lutz, 1989; Spears & Singh, 2004). According to the theory of reasoned action, attitudes correlate positively with purchase intention (Fishbein & Ajzen, 1975). The more someone likes a product or brand, the higher is their intention to purchase the product.
In the modern business world companies focus more and more on brand image and their product benefits. A strong brand image can be a powerful asset for companies. A recognized and trusted brand image makes people more confident in buying products of these companies. Thus, how people think about brands is both in the theoretical as in the practical field important. The model of Laroche and Sadokierski (1994) explained that
customers create a certain brand attitude based on the factors of brand familiarity and brand evaluation. For marketers to convince customers of their brand attitude, they need to get customers confident with the brand. The free version makes the customers confident with the service before having to decide to purchase the service.
In the freemium services organizations see more and more that customers use the free product until it is not possible or not satisfying anymore and go on (Zhu & Chang, 2014).
Although this is a common finding in this industry, this issue is more related to product
attitude then brand attitude. In this study is suggested that customers with high brand attitude have a high purchase intention. Therefore, the following hypotheses is formulated:
Hypotheses 1: Brand attitude has a positive relation with purchase intention of the premium version
2.4 Subjective norms
A subjective norm is the influence of people in one’s social environment on his behavioral intentions (Fishbein & Ajzen, 2011). The beliefs of people, weighted by the importance one attributes to each of their opinions, will influence one’s behavioral intention. Customers who buy a product have certain expectations about how others think of their purchase (Ajzen, 2005). Although most of the customers have thoughts about the expectations of others, the degree of influence of these thoughts differs among customers (Shiau & Luo, 2012).
Customers with different types of self-concepts may have different motivations and
experiences. People who classify themselves as independent are more favoured to act as unique and are able to express themselves. On the other hand, if people want to belong to a group, they will use opinions of others in relation with their actions. Customers’ personal factors influence their purchase intention.
According to Ajzen (2005) customers are stimulated with the thoughts of others about purchasing a particular product or good, for example giving donations to charities. But
customers can also be discouraged in relation with the expectations of others. Related to this conclusion of Ajzen, subjective norms can be divided into thoughts that are related positively, stimulate the purchase intention, or related negatively, which decreases the purchase
intention. Subjective norms related to negative thoughts, like drugs and smoking will decrease the purchase intention of customers.
Koram et al., (2011) examined the relation between social support and norms in the use of drugs. Drugs users are the most influenced by beliefs about peers in their social environment. The strong influence of perceived peer behavior is mostly focused on popular opinion leaders in the social environment. The beliefs of the drugs users about these persons will affect their behavioral intentions.
Also the effect of customers thoughts related to positive expectations will influence the purchase process. Recently, a study to young people’s donations has shown that subjective plays an important role in donating to charities (Knowles, Hyde, & White, 2012). Just like subjective is mentioned, attitude and past behavior are mentioned (in first and second place respectively).
Young people are more influenced by others expectations than old people are, which means that these effects will differ between ages. In relation to the music industry is
investigated which effect ‘beliefs about peers’ has on illegal music downloading. If customers have negative expectations of friends about illegal music downloading, it will decrease the amount of downloads (Levin, Dato-on, & Manolis, 2007). All these studies concluded that subjective norms are a powerful driver of purchase intention.
Based on previous studies, the effect of subjective norms will have a positive effect on purchase intention. For freemium music services it would be helpful to examine how strong the effect of subjective norm are on customers purchase intention. The following hypotheses is formulated:
Hypotheses 2: Subjective norms have a positive relation to purchase intention of the premium version
2.5 Moral awareness
Reynolds (2006) has previously definitions combined and believes that moral awareness represents “a person’s determination that a situation contains moral content and legitimately
can be considered from a moral point of view”. Reynolds explains moral awareness from the decision making process, whereby customers first identify the situation and then will take some action. An example is from the clothing industry where many customers buy clothes from big clothing producers like H&M and Primark, while these companies used child labor and poor working conditions for their employees. In some cases customers knew about these activities and boycott them. But there were also customers who knew about these activities and still bought clothes.
In the process of moral decision-making some argue that customers care about their moral behavior, while others mention that the moral aspects of their decision only partly will influence customers (Reynolds, 2006). If customers are moral aware of the information or situation, but if customers do not perceive them as important, it would hard to believe that these beliefs have influence on the decision (Jordan, 2009). Thus, consumer’s values have an effect on the impact of moral awareness on decision-making. Moral awareness related to the music industry is an important and recent topic (Chiou, Huang, & Lee, 2005). Since the introduction of the Internet, music piracy is a well-known topic. Music piracy is that customers download music for free without permission on the Internet. The introduction of the freemium business models in the music industry ensures that artists and other stakeholders get paid. Music industry appeals customers to be aware of the importance of paying for their music (Coyle, Gould, Gupta, & Gupta, 2009). In relation to the effect of moral awareness in the decision process in the music industry is investigated which effect moral awareness has on the three factors of the theory of planned behavior; purchase intention (Woo Jin Choi & Winterich, 2013), brand attitude (Hartmann & Apaolaza-Ibáñez, 2012) and subjective norms (Chan & Bishop, 2013).
2.5.1 Moral awareness on purchase intention
In most studies the effect of moral awareness on purchase intention is positive. Customers with high moral awareness are more likely to purchase a certain product. Tan (2002)
concluded that moral awareness influence consumer’s decision to subscribe for music streaming services instead of illegal downloading. In the music streaming service industry, in relation with other industries, is the power of moral judgment in the decision process stronger according to Tan. Wang, Yeh and Liao (2013) support the positive relationship between moral awareness and purchase intention as well. Their study is focused on online content services and the factors that affect customers to purchase.
As earlier announced, the concept of moral awareness in relation with purchase intention in the music industry is more complex because customers can easily illegal download music for free. Chen, Shang and Lin (2008) found that morality has no significant influence on customers who have the intention to download illegal music. Although this study is focused on illegal downloading instead of purchasing music streaming services, the
customers in this study are less influenced by moral norms.
Customers are more appealed on their moral duty in the music industry because of the free alternative of illegal downloading. Nevertheless, in many studies the moral
awareness has proven to have a positive effect on purchase intention. Therefore, the following hypothesis is investigated in this study.
Hypotheses 3: Moral awareness has a positive influence to purchase intention of the premium version
2.5.2 Moral awareness on brand attitude
Hartmann and Apaolaza-Ibáñez (2012) conclude that advertising campaigns directed at the environment concern of the organization, not only have moral benefits but also brand attitude benefits. If customers agree with the environment concerns and the approach of the
organization to solve these problems, customers improve their attitude toward the brand. This result is consistent with other studies related to the effect of morale intentions on the attitude toward the brand (Guo, 2013; Woo Jin Choi & Winterich, 2013). In the music industry
the relation between moral awareness and brand attitude is perhaps more difficult. Customers have the possibility to download music for free or subscribe themselves for streaming music services. Cesareo and Pastore (2014) mention that the moral judgment of customers in the music industry is more complicated than in other industries because they can easily switch to illegal music downloading. Cesareo and Pastore conclude that moral judgment and attitude toward the service are important factors in the purchase process for music subscription services, but it is unclear if moral judgment of customers increases the brand attitude a consumer has about the music streaming service. To see whether this effect is present, the following hypotheses is formulated:
Hypotheses 4: Moral awareness has a positive influence on brand attitude of freemium music services
2.5.3 Moral awareness on subjective norms
In the following paragraph is the effect of moral awareness on subjective norms examined for streaming music services. Although both concepts are individually investigated in relation to the purchase process, the effect of moral awareness on subjective norms is not yet
investigated.
Lacobucci, Daly, Lindell and Griffin (2013) investigated the effect of professional values, self-esteem and ethical confidence in the decision making process in hospitals for nurses. This study conclude that if nurses are convinced of their moral judgment in a particular situation it will increase their self-esteem and take action. Hwang and Lee (2012) did research into the purchase intention of online customers for e-commerce companies. They concluded that if customers have high self-esteem in their buying decision they are less influenced by subjective norms. From these two studies can be seen that a high moral
awareness creates more confidence. If customers have confidence it will lower the effect of subjective norms.
Huijts, Molin, and Steg (2012) did research into the psychological factors that influence sustainable technology acceptance. In this research is mentioned that strong personal norms decrease the effect of social influence in a particular situation. Although moral awareness is not the same as personal norms, it implies that strong norms decrease the effect of social influence.
Although previous studies have not explicitly done research on the effects of moral awareness on subjective norms, based on the studies of Lacobucci et al., (2013), Hwang and Lee (2012) and Huijts, Molin and Steg (2012) is a negative effect expected. Therefore the following hypothesis is formulated:
Hypotheses 5a: Moral awareness has a negative influence on subjective norms of freemium music services
The effect of moral awareness on subjective norms is suggested to be negative. In the following hypotheses is the moderator purchase intention used to examine if the relationship is affected. This moderating effect of purchase intention is not yet investigated into the relationship between moral awareness and subjective norms. However, the following studies will help to determine which moderating effect purchase intention could have. According to Huy Tuu and Ottar Olsen (2012) customers build certainty and knowledge in their intention to purchase. If customers eventually have the intention to purchase the product, their intention is based on trust and knowledge and is hard to change. Chiu, Hsu, Lai, and Chang (2012) support this statement that customers with high purchase intentions are hard to change, this study suggests that these people have developed an intention and want to achieve it too. These two studies support the statement of most researches that purchase intention is a good predictor for purchase behavior (Fishbein & Ajzen, 1975; Wang et al., 2013). Thus, if customers have high purchase intention customers are less focused on other factors. In case of the effect of moral awareness and subjective norms, purchase intention will negatively affect this relationship. Low purchase intention will increase the effect of moral awareness on
subjective norms, and high purchase intention will decrease the effect of moral awareness on subjective norms. The following hypotheses is formulated:
Hypotheses 5b: Purchase intention has a negative moderating effect on the relationship between moral awareness and subjective norms
2.6 Product positioning strategy
In the freemium music services it is difficult to create competitive advantage (Kumar, 2014). An important challenge is to understand how companies can use product strategies to influence the relationship between the drivers of purchase intention. In this paragraph the effect of product strategies on the drivers of purchase intention will be discussed. First the product strategy is defined and then the concept of competitive advantage will be explained. After that, the focus will be on two different product positioning strategies; cost leadership and differentiation. At last, the moderating effects of the different product strategies on the relationship between the drivers of purchase intention and purchase intention are discussed.
Product strategy is the pattern of managerial actions that explains how a firm achieves and maintains competitive advantage through positioning a product in the market (Zott & Amit, 2008). An important aspect in this definition of product strategies is how companies position themselves in the market in order to create competitive advantage.
Competitive advantage is the process of value creation, which results in advantages compared to other competitors. There are different views of how firms can generate
competitive advantages. The position view is one of the most common ones in the current literature and therefore used in this study (Kotler & Keller, 2012). In the position view companies create a competitive advantage if they create value by their product positioning. One of the important authors in this field is Porter. Porter (1980) described two different product-positioning strategies for companies in the market. Companies can either use a cost leadership or differentiation strategy. According to Porter if companies want to create
competitive advantage it is important to choose one strategy. Cost leadership focuses on low cost of operation, driven by efficiency and economies of scope, scale and size. Differentiation strategy is the differentiation of a product, in order to create more value for the customers. A sub strategy linked to these main strategies is the niche strategy, whereby firms focus on a niche market instead of the overall market. This strategy is mainly used in combination with the generic strategies and therefore not further mentioned in this study.
2.6.1 Brand attitude
Differentiation strategies have more influence on high brand attitude than on cost leadership and they are more focused on the needs of customers. Brand attitude will be positively influenced if the product satisfies the needs of the consumer. The effect of strategies on brand attitude is widely investigated, but the effect in freemium business models not yet.
Dodd and Lindley (2003) investigated the effect of the differentiation strategy on store concepts and brand attitude. If companies use the differentiation approach in the store concepts it positively influences the brand attitude of customers. González-Benito and Martos-Partal (2012) investigated the relationship between brands positioning and attitude. If brands differentiate their products, customers will perceive higher quality of the products. The perception of quality influences how customers rank the brand image, and therefore their attitude. In both studies the differentiation strategy is linked to a positive brand attitude.
Yi and Yoo (2011) investigated the effect of non-monetary and monetary product strategies on brand attitude. A non-monetary strategy stands for companies focusing on status, and monetary stands for companies that focus on price. If companies focus on price, it will affect the brand attitude on the long term negatively. Although price focused strategies damage the brand attitude in the long term, the effect differs between customers. If
customers are deal proneness, price promotions are less damaging for the brand attitude. The effect of differentiation strategies on brand attitude to purchase intention in relation to freemium business models is not yet investigated. Based on earlier studies is
assumed that differentiation strategies have a positive moderating effect, and cost leadership strategies a negative moderating affect. The following hypothesis are formulated:
Hypotheses 6: Differentiation strategies have a more positive moderating effect on the relationship between brand attitude and purchase intention than low cost strategies 2.6.2 Subjective norms
Subjective norm is the social pressure to perform a certain behavior. A company’s product positioning strategy (differentiation or low cost) influences the subjective norms in the purchase decision. However, several results are found about the effect of the product positioning in relation to subjective norms.
If companies use differentiation strategies, it will increase the value of the product (Achtenhagen, Melin, & Naldi, 2013). If customers believe that their friends value the product then subjective norms influence the decision process (Fishbein & Ajzen, 1975). Although this conclusion mentions the positive effect of differentiation strategies on subjective norms, several studies conclude that the effect of differentiation strategies is negative. Ha and Janda (2012) concluded that customers are not much influenced by subjective norms if they
purchase a differentiated product like energy-efficient electronic products. According to their statement companies should focus more on improving the attitude of customers about green products and less on customers beliefs about others in relation to green products. Lee, Trimi, and Kim (2013) studied the effects of innovation and imitation products on the adaption of new technology. Innovated products are new to the market, while imitated products imitate the innovation product and are offered for a lower price or have extra features for the same price. If customers buy innovated products, the influence of subjective norms is low. If customers buy imitated products, the influence of subjective norms is high. Customers who buy innovated products are more influenced by product design and characteristics while customers of imitated products favor the beliefs of others. Cost leadership strategies have a
negative effect on the beliefs customers have of others. If customers purchase a low priced product they think others interpret it negatively.
The influence of product strategy on subjective norms is divided. The effect of differentiation strategies is positive but low, while the effect of cost leadership is negative. In this study is investigated which effect both strategies have on subjective norms, the following hypothesis are formulated:
Hypotheses 7: Differentiation strategies have a more positive moderating effect on the relationship between subjective norms and purchase intention than low cost strategies 2.6.3 Moral awareness
If companies have a moral contribution, it is important for these companies to communicate it. When companies decide to use differentiation strategies, they will focus on marketing and product differentiation, in which the social contribution could play an important role. If
companies focus on cost leadership, they are more focused on reducing prices than on communicating their social contribution.
If customers can choose between two similar products, whereas one is promoted as a sustainable product, customers will purchase that product (Manaktola & Jauhari, 2007). But if customers need to sacrifice lower performance levels or higher prices they will not be as eager to choose the sustainable product. Thus, if products have the same price and quality the marketing of social contribution plays a role.
Even with the moderator variable of price, differentiating strategies focus on marketing which increases the moral awareness and this affects the purchase intention. Therefore, the following hypotheses is formulated:
Hypotheses 8: Differentiation strategies have a more positive moderating effect on the relationship between moral awareness and purchase intention than low cost strategies
Figure 1. Conceptual model
3 RESEARCH METHOD
In this chapter, the methodology of the research has been described for the questionnaire and the case study. For the questionnaire a description is given of the procedure, data collection, measurement of the variables and data analyses. For the case studies, the case selection and secondary data are discussed.
3.1 Procedure
For this research an online survey has been used to investigate the influence of brand attitude, level of involvement and level of awareness on purchase intention. The survey strategy will have a deductive approach. “A deductive approach is concerned with developing hypotheses based on existing theory, and then designing a research strategy to test the hypothesis” (Wilson, Johns, Miller, & Pentecost, 2010).
Data collection will be gathered with quantitative research methods. Quantitative methods are chosen because of explanatory basis of the study, limited numbers of studies,
and the possibility to analyze relationships and differences and make conclusions related to numerical data (Boeije, ’t Hart, & Hox, 2009). An online questionnaire is applied to collect the data because it helps to describe, predict and explain social phenomena like the purchase decisions process. Benefits of online questionnaires are the possibility of reaching a large number of respondents, low in costs and the high reliability. The reason high reliability is a benefit is because respondents freely choose to participate, which is a positive factor with regard to the quality of the responses (Lefever, Dal, & Matthíasdóttir, 2007). Despite the benefits, online questionnaires exclude non-Internet users and can cause problems when questions are not clear formulated or understood. Still, we will apply the online questionnaire because of the benefits and we will test it carefully so that any errors are corrected.
3.2 Data collection
The sample population was gathered through a convenience sample. “A statistical method of drawing representative data by selecting people because of the ease of their volunteering or selecting units because of their availability or easy access” (Murray, Rugeley, Mitchell, & Mondak, 2013). A convenience sample reduces the external validity of the research, because of the risk the sample does not represent the population. An Advantage is the possibility to collect data in a short time period with a low budget.
The population of the sample for the survey exists of persons with a music freemium account, like Spotify or others. Respondents could have a free or a premium account. All respondents are approached via Facebook and email, and therefore not exactly clear which people have been approached. It is likely that through Facebook mainly younger people are approached. Because the respondents that are approached originate from the researchers’ network, it is possible that there’s a relatively higher education level, which can affect the study.
The data collection for the survey has started at the 19th of April and lasted until the 4th of may 2015. In total 164 questionnaires were submitted, and 130 were completed. Out
of the complete surveys 6 persons have been deleted because they were using YouTube or iTunes as music service, which is not a freemium music service. In total 124 respondents are useful for the analysis.
The following information is about the demographics of the respondents. The demographics show that 48% of the respondents were male and 52% were female. The range of age of the respondents was between the 19 and 69 year (M=29,7, SD=11,9). 75% of the respondents were between the age of 19 and 29, 6% belongs in the group of 30-39, 6% in 40-49, 11% in 50-59 and 2% in 60+. Because the respondents were approached through the researchers network of Facebook and shared by friends and parents, the number in the age categories 19-29 and 50-59 is relatively higher. Estimation of response rate is probably high because of the convenience sample (Murray et al., 2013). All surveys were anonymously returned via the use of Qualtrics.
3.3 Operation variables
Selected measurement items must represent the concept. To ensure content validity, measurement items in this study were mainly adapted from prior studies.
3.3.1 Brand attitude
Brand attitude is one of the independent variables and defined as customers’ overall evaluation of a brand (Bellman, Potter, Treleaven-Hassard, Robinson, & Varan, 2011). The scale for brand attitude construct consists of a four item assessing overall brand evaluation on a five point likert scale (Yoo & MacInnis, 2005). The four items are not likable-likable, negative-positive, bad-good, and unfavorable-favorable. For all of the items the following applies: the higher their scores, the higher the brand attitude. The brand attitude variable is operationally defined as the mean score of the items, a high score indicates a positive feeling about a certain brand.
3.3.2 Subjective norms
Subjective norm is a second independent variable. Subjective norms are the influence of others in someone’s social environment on his behavioral intentions, the beliefs of people and weighted by the importance of their opinions, will influence behavioral intention (Fishbein & Ajzen, 2011). The construct is measured with three items, related to the scale of (Ajzen & Sheikh, 2013). The three items are related to the importance of thoughts, value of opinions and how other behaves. These statements were rated on a five point likert scale from very unlikely (1) to very likely (5). If customers have high scores, it means subjective norms have more impact on their purchase intention. The subjective norms variable is operationally defined as the mean score of the items, a high score indicates more value to the opinion of someone else.
3.3.3 Moral awareness
Moral awareness is the third independent variable. Moral awareness in the music industry is the consciousness that customers have about their actions. The construct moral awareness consists of four items on a five point likert scale (Bateman et al., 2012). The four items are; definitely not-definitely, unlikely-likely, improbable-probable, impossible-possible. For all of the items the following applies: the higher their scores, the higher the moral awareness. The moral awareness variable is operationally defined as the mean score of the items, a high score indicates a positive value about the ethical aspects of a decision.
3.3.4 Purchase intention
Purchase intention is the dependent variable and is the consumer’s intention to buy a particular brand or product. The construct consists of three items on a five point likert scale, based on the research of Wang, Yeh, and Liao (2013). The purchase intention was
measured by asking ‘I plan to pay for online freemium music services in the future’, ‘I intend to purchase online freemium music services in the future’, and ‘I predict I would buy online freemium music services in the future’ with anchors ranging from ‘very unlikely’ to ‘very likely’. For all of the items the following applies: the higher the score, the higher the purchase
intention. The purchase intention variable is operationally defined as the mean score of the items, a high score indicates a positive intention to buy a product.
3.4 Data analysis
Partial Least Square (PLS) is used in this research because of the ability to analyze complex models and for when the sample size is small (Chin, 1998). Although the model is not too complex, the sample size is relatively small. PLS enables researchers to answer interrelated questions in single, systematic and comprehensive analysis by modeling the relationships among multiple independent and dependent variables simultaneously (Gefen, Straub, & Boudreau, 2000). This is in contrast to other analytical techniques such as ANOVA in SPSS, which can only test one relationship at a time. SMART PLS software is used for the analysis. Factor loadings, Cronbachs alpha, reliability and validity were obtained by running the
sample through the software.
3.5 Case selection
The case study chosen in this research compares the different product strategies in the music streaming industry. The purpose of this case study is “to investigate the phenomena in depth to provide rich description and understanding” (Yin, 2013). As the literature review indicates, the product strategies in different industries have different effects on purchase intentions. To understand which affect this will have in relation to the music industry, case studies were used.
The companies in this case study are chosen based on the survey. The music streaming organizations are Spotify, Deezer and SoundCloud. These organizations were chosen because they are the most prevalent in the questionnaire. The reason that the same organizations are selected from the questionnaire is to establish a connection between the quantitative and qualitative study in the overall conclusion.
In order to examine the effect of product strategies on the different drivers of purchase intention, the construct product strategies need to be operationalized. Product
strategy is a latent variable because it cannot be directly observed. Therefore structural equation modeling has been used (Baumgartner & Homburg, 1996). Structural Equation Modeling (SEM) is an approach to research scientific constructs with many facets that cannot directly be observed. The different product strategies in this research are the low cost and differentiation strategy. For both strategies SEM revealed some concepts that conceptualize the strategy (Kotha & Vadlamani, 1995). For differentiation it is new product development, quality control procedures, quality of products, brand identification, influence distribution channel, customer service capability, highly trained personnel, high priced products and building reputation. For low cost strategies it is related to the concepts of operating
efficiencies, concern for cost reduction, price advertising, high inventory levels and refining existing products.
In the data collection for the case studies different sources of secondary data have been used. Secondary data is information that has been collected for a purpose other than your current research project but has some relevance and utility for your research (Kotler & Keller, 2012). Advantages of secondary data are; time saving and comparison between different data. However, secondary data can be outdated or not accurate for the specific research. Because of the extensive databases into the fields of product strategies, short time period of the study and an extensive quantitative study is chosen to focus on secondary data.
Different databases are used to examine the effects of product strategies like AMADEUS, Business source premier, and Warc. In these databases a search has been conducted for other case studies, industry and company reports, academic articles and reports from international private research companies. The conceptualizations of the
competitive strategies are used as search terms in the different databases and to define the product strategies of the different companies into low cost or differentiation.
4 RESULTS
This chapter is about the analyses and results of this research. First, the results of the measurement model are presented and discussed. Second, the results of the structural model are presented and discussed. At last, the focus will be on the moderating effects of this study.
4.1 Model evaluation
The statistical output of SMART PLS is analyzed according to the book of Hair, Hult, Ringle, & Sarstedt (2014). According to their approach a research model can be analyzed in two stages, the assessment of the measurement model and the assessment of the structural model. In the first step each variable is tested for reliability and validity. A reliable
measurement ensures that the findings will be the same if the experiment is done again under the same conditions (Bryman, 2012). If the measurement is not reliable it will be hard to conclude if the results are valuable because another study somewhere else will probably find other results. Validity is to test whether the construct measure what it intends to
measure. The measurement model is evaluated by examining the composite reliability, convergent validity and discriminant validity. In the second step the structural model is analyzed with the beta-coefficient for the relationships and explained variance for the construct. At last, the t-statistics are used to evaluate if the results are significant.
4.2 Measurement model
The measurement model is used to test the constructs of the model on the reliability and validity. According to Hair et al., (2014) in order to test the reliability and validity of the model, they recommend the convergent validity and the reliability for each variable.
4.2.1 Reliability and Validity
Reliability refers to the ability to reproduce the results over and over again under the same conditions. By analyzing the correlation between the variable and the construct we can find the reliability for each variable. The loadings explain in which extent the variable is able to measure a part of the construct. According to Hair et al., (2014) each construct should at least explain a substantial part of each variable variance. If a variable is higher than 0,7 then it would be considered as reliable. In table 1.1 we mentioned all the reliability and validity numbers for the construct model.
Table 1.1 Factor loadings
Construct Factor Loadings
Brand attitude 1 0.932 2 0.918 3 0.951 4 0.903 Subjective norms 1 0.573 2 0.813 3 0.776 Moral awareness 1 0.932 2 0.918 3 0.951 4 0.903 Purchase intention 1 0.880 2 0.915 3 0.892
Table 1.1 shows the results for each construct. The factor loading for item 1 of subjective norms is below 0,7. Therefore, the relationship between the indicator and construct is not consistent and this item is unreliable for the analysis of this study. Although this will increase the reliability measurements, it will decrease the number of indicators from three to two. If we analyze the question we could conclude that they measure how much value someone attach to the opinion of others “SN1: Most people, whom are important to me, do not matter whether
the composite reliability and convergent validity will be taken into account. All other factor loadings are above 0.7, which means that these relations are reliable.
4.2.2 Composite reliability and Convergent validity
For the internal reliability the composite reliability is tested. Although in most cases
Cronbachs Alpha is used for the internal reliability, this measurement is not most suitable for PLS models (Hair et al., 2014). The composite reliability is more suitable for this kind of analyses. The recommended number for composite reliability is above 0.7. The Average Variance Extracted (AVE) is a criterion for the convergent reliability. Convergent reliability is that all the indicators represent the same construct. An AVE value of at least 0,5 is sufficient. In table 1.2 the composite reliability is shown, which indicates the internal consistency of the constructs. Brand attitude, Subjective norms, Moral awareness and Purchase intention score all above 0.7 on the composite reliability, which indicate that the internal consistency of the constructs is reliable.
According to the factor loadings we need to delete item 1 of subjective norms. If we delete item 1 then we will increase the composite reliability from 0.768 to 0.867, AVE from 0.531 to 0.766, and Cronbachs alpha from 0.541 to 0.695. Although the reliability and validity measures were already sufficient, it increased significantly. In the following analyses we deleted the item from the conceptual model. Brand attitude, Subjective norms, Moral
awareness and Purchase intention score all above 0.7 on composite reliability, which means that the internal consistency of the constructs is consistent and thus reliable. All constructs have also a higher AVE score than 0.5, this means that all indicators represent the construct sufficient enough. The Cronbachs Alpha is above 0.7 for all constructs, although Hair et al., (2014) attach less value to this measurement in Smart PLS, it still support the results of the AVE and composite reliability, which is good.
Table 1.2 Reliability and Validity measures
Construct AVE Composite reliability Cronbachs alpha
Brand attitude 0.858 0.960 0.947
Subjective norms 0.766 0.867 0.695
Moral awareness 0.870 0.964 0.950
Purchase intention 0.802 0.924 0.883
4.2.3 Conclusion reliability and validity
The reliability and validity measures for the model were positive. In the beginning the factor loadings only loaded below minimum for item 1 of subjective norms, all other items were above the minimum of the factor loadings. In the measurement of the composite reliability and convergent validity we conclude that if we delete item 1 of subjective norms out of the model, it will increase the reliability and validity measures. Based on the output of the factor loadings and the reliability and validity measures the item was deleted. All other constructs were reliable and valid. This enables the research study to proceed with the analyses with the structural model.
4.3 Structural model
The structural model is related to several criteria. The coefficient of determination, denoted as R2 is the first and most important criteria in the structural model. According to Hair et al., (2014) R2 values of 0,67 are substantial, 0,33 are moderate and 0,19 are weak for
endogenous variable. If an endogenous variable is only explained by one or two variables, a moderate R2 is enough. The variance explained in the relations serve as an indication of how good the model of performance is.
A second criterion is the path coefficients of the structural model. These path
coefficients can be interpreted as standardized regression coefficients (𝛽). The re-sampling technique bootstrapping can determine whether the coefficients are significant or not. In bootstrapping, subsamples are created with observations randomly drawn from the original set of data, which provides t-statistics to conclude if the results are significant of the path
model. The critical t-statistic number that is related to the degrees of freedom (123) and two tailed (5%) is 1.976. This means that if the t-statistics is lower than 1,976, the hypothesis should be rejected.
In table 2.1 we show the path coefficient and the t-values. In table 2.2 we show the explained variance of the model.
Table 2.1 Bootstrapping output Causal relationship Path coefficient (𝜷) Standard error T-Statistics Significance 1. BA à PI 0.009 0.139 0.065 0.948 2. PI à SN -0.048 0.089 0.547 0.585 3. MA à PI -0.117 0.107 1.096 0.275 4. MA à BA 0.160 0.093 1.728 0.086 5. MA à SN -0.266 0.078 3.422 0.001 6. Interaction effect -0.289 0.092 3.097 0.002
Table 2.2 Explained Variances
Construct R Square
Purchase intention 0.013
Brand attitude 0.026
Subjective norms 0.162
In figure 2, the conceptual framework is shown for the model with the path coefficients and the explained variances, in order to perceive a better understanding of the total view of the correlations. The explained variance of the first model is 0.067, which is a week effect according to Hair (2014).
Figure 2. Conceptual model (*=significant at p<0.05)
Hypotheses 1, 2 and 3 were not supported. Hypotheses 1 supposes that brand attitude has a positive effect on purchase intention. Although the effect was positive, the results is not significant (𝛽 =.009, t = 0.065, p >.05). Hypotheses 2 assume that subjective norms have a positive effect on purchase intention. During the analyses the effect was turned around. Instead subjective norms influence purchase intention, purchase intention influenced subjective norms. Also, the effect proved to be negative instead of positive. Despite this modification in the study, the effect was still found not significant (𝛽 =-.048, t = 0.065, p >.05). Hypotheses 3 suppose that the effect of moral awareness on purchase intention is positive. The effect of moral awareness on purchase intention was found positive but non-significant (𝛽 =-.117 , t = 1.096, p >.05).
Hypotheses 4 suppose that the effect of moral awareness on brand attitude is
positive, however the hypothesis was not supported. The effect of moral awareness on brand attitude was found positive but non-significant (𝛽 =.160, t = 1.728, p >.05).
Hypotheses 5 and 6 were supported. Hypotheses 5 implies that moral awareness has a negative effect on subjective norms (𝛽 =-.266, t = 3.422, p <.05). This indicates that
people, who value the moral aspect in their decision, are less influenced by the opinions of others. Moral awareness accounted for 16.2% of variance in subjective norms. Hypotheses 6 was also supported, the negative effect of purchase intention on the relationship between moral awareness and subjective norms (𝛽 =-.289, t = 3.097, p <.05). This indicates that customers with a low purchase intention are more influenced by moral awareness in relation with subjective norms, than people with high purchase intention.
4.4 Moderating effects
Occupation and age were both tested for a moderating effect on the different models. Moderating effect occurs when a variable affects the relationship between an independent and dependent variable.
4.4.1 Occupation
In the first moderating effect is tested which effect occupation has on the relationship between the dependent and the independent variables. For both groups (students and working people) PLS and Bootstrapping were used. Afterwards we could compare the path coefficients and significance of both groups in order to conclude if there are any differences.
Table 3.1 Working people moderating effect
Relationships Path Coefficient Standard Error T-Statistics Significance
BA à PI 0.228 0.296 0.772 0.442 PI à SN -0.205 0.124 1.648 0.102 MA à PI -0.166 0.156 1.065 0.289 MA à BA 0.064 0.145 0.443 0.658 MA à SN -0.272 0.127 2.145 0.034 Interaction -0.404 0.139 2.900 0.004
Table 3.2 Students moderating effect
Relationships Path Coefficient Standard Error T-Statistics Significance
BA à PI -0.026 0.172 0.152 0.880 PI à SN 0.076 0.206 0.371 0.712 MA à PI -0.152 0.133 1.137 0.258 MA à BA 0.220 0.119 1.851 0.067 MA à SN -0.212 0.137 1.540 0.126 Interaction -0.173 0.145 1.191 0.236
In table 4.1 and 4.2 there are two significant differences between students and working people. First the relationship between moral awareness on subjective norms is discussed. The path coefficients for working people (𝛽 = -.272, t=2.145, p>.05) are higher than for students (𝛽 = -.212, t=1.540, p<.05) and the effect for working people is significant. This implies that the effect of moral awareness on subjective norms is stronger for working people. If working people have a high moral awareness, they will be less influenced by subjective norms. Secondly, the path coefficient for the interaction effect is higher for working people (𝛽 = -.404, t=2.900, p<.05) than for students (𝛽 = -.173, t=1.191, p>.05) and this result is significant for working people. This implies that working people with low purchase intention are less influenced by the opinion of others if they have a high moral awareness and working people with low moral awareness are more influenced by the opinion of others.
4.4.2 Gender
In the second moderation, the effect of gender is tested. In this moderation effect is
examined which effect gender has on the different relationships. Table 4.1 and 4.2 show the result of this analysis.
Table 4.1 Men moderating effect
Relationship Path Coefficient Standard Error T-Statistics Significance
BA à PI -0.084 0.180 0.467 0.641 PI à SN -0.024 0.176 0.138 0.891 MA à PI 0.012 0.159 0.076 0.939 MA à BA 0.060 0.141 0.421 0.674 MA à SN -0.337 0.108 3.135 0.002 Interaction -0.278 0.147 1.894 0.060
Table 4.2 Women moderating effect
Model 2. Path Coefficient Standard Error T-Statistics Significance
BA à PI 0.067 0.192 0.350 0.727 PI à SN -0.075 0.177 0.426 0.671 MA à PI -0.237 0.112 2.126 0.035 MA à BA 0.259 0.109 2.367 0.019 MA à SN -0.236 0.137 1.721 0.088 Interaction -0.217 0.167 1.298 0.197
In table 4 three significant differences between genders are determined. First, the path coefficient for moral awareness on purchase intention is higher for women (𝛽 = -.237, t=2.126, p <.05) than for men (𝛽 = .012, t=0.076, p >.05). This relationship is negative for women and slightly positive for men, and it is only significant for women. The effect of moral awareness on the purchase intention is stronger for women. Second, the path coefficient of moral awareness on brand attitude is higher for women (𝛽 = .259, t=2.367, p <.05) than for men (𝛽 = .060, t=0.421, p >.05) and the result is significant for women. This implies that the effect of moral awareness on brand attitude is stronger for women. Women are more
affected by the moral aspect in the purchase intention than men. Third, the path coefficient of the relationship of moral awareness on subjective norms is higher for men (𝛽 = -.337,
t=3.135, p <.05) than for women (𝛽 = -.236, t=2.126, p >.05). The result for men is also significant, which means that the relationship for men is stronger. If men focus more on moral awareness in a decision, their focus on subjective norms decreases.
4.5 Summary
In table 5 is shown which relationships are tested including the results of each relationship. In this table is seen that out of the six hypotheses two of them are accepted. The negative effect of moral awareness on subjective norms is supported, and the effect of moral
awareness on subjective norms with moderator purchase intention is supported. This means that people, who value the moral aspect in their decision, are less influenced by the opinions of others. Customers with a low purchase intention are more influenced by moral awareness in relation to subjective norms, than people with high purchase intention. The variance explained for the constructs in the second model was low, purchase intention =0.013, brand attitude =0.026 and subjective norms =0.162.
Occupation and gender were tested as moderating variables for both models. Differences between students and working people were found for the relationships of moral awareness on subjective norms and the interaction effect of moral awareness on subjective norms with moderator purchase intention. The difference in moral awareness on subjective norms implies that working people are less influenced by subjective norms if they have a high moral awareness, and this effect is not supported for students. The difference for the interaction effect is that working people with low purchase intention are influenced by moral awareness on subjective norms, in contrast to students.
Differences between men and women are determined by moral awareness on purchase intention, brand attitude and subjective norms. Women are negatively affected by moral awareness in their purchase intention and negatively affected by moral awareness in their opinion about a certain brand, in contrast to men. Men are negative influenced by moral awareness on subjective norms, in contrast to women.
Table 5 Summary of the research results
No. Hypothesis Analyses Result
H1. Brand attitude has a positive influence on purchase intention of freemium music services
T-statistic = 0.065, p=0.948
Rejected
H2. Subjective norms has a positive influence on purchase intention of freemium music services
(hypotheses reversed)
T-statistic = 0.547, p=0.585
Rejected
H3. Moral awareness has a positive influence on the purchase intention of freemium music services
T-statistic = 1.096, p=0.275
Rejected
H4. Moral awareness has a positive influence on brand attitude of freemium music services
T-statistic = 1.728, p=0.086
Rejected
H5. Moral awareness has a negative influence on subjective norms of freemium music services
T-statistic = 3.422, p=0.001
Accepted
H6. Moral awareness has a negative influence on subjective norms, with moderator purchase intention T-statistic = 3.097, p=0.002 Accepted Occupat ion
a. The relationship between brand attitude and purchase intention moderated by occupation b. The relationship between purchase intention and subjective norms moderated by occupation c. The relationship between moral awareness and purchase intention moderated by occupation d. The relationship between moral awareness and brand attitude moderated by occupation
e. The relationship between moral awareness and subjective norms moderated by occupation f. Interaction effect between moral awareness on subjective norms, moderated by purchase intention and occupation
a. No difference in 𝛽 b. No difference in 𝛽 c. No difference in 𝛽 d. No difference in 𝛽 e. Difference in 𝛽 f. Difference in 𝛽 - - - - - -
Gender a. The relationship between brand attitude and purchase intention moderated by gender b. The relationship between purchase intention and subjective norms moderated by gender c. The relationship between moral awareness and purchase intention moderated by gender
d. The relationship between moral awareness and brand attitude moderated by gender
e. The relationship between moral awareness and subjective norms moderated by gender
a. No difference in 𝛽 b. Difference in 𝛽 c. Difference in 𝛽 d. No difference in 𝛽 e. No difference in 𝛽 - - - - -
