‗The influence of channel expansion predictors on
the perceived media richness of Twitter.‘
By Laura Kranenburg
1
‗The influence of channel expansion predictors on
the perceived media richness of Twitter.‘
By
Laura Kranenburg
Rijksuniversiteit Groningen, Faculty of Economics and Business
Master thesis Business Administration, specialization Marketing
August, 2012
Stadionweg 41-3, 1077 RW Amsterdam +31640406404 lmrkranenburg@gmail.com Student number 1622242First supervisor: Prof. Dr. J.C. Hoekstra Second Supervisor: Dr. J.A. Voerman
2 MANAGEMENT SUMMARY
The suitability of a communication channel to convey the intended message is defined by perceived media richness. When a channel conveys a message exactly as intended the channel is determined as rich. Perceived media richness helps people with selecting a certain channel for communication (Daft & Lengel, 1984). These communication channels are evaluated differently per person. Where one person finds a channel to be perfect for its designated communication, another person might disagree. Their perceptions of the media richness of a channel can therefore differ. However, these differences can be reduced as the channel expansion theory (Carlson & Zmud, 1999) states that the perceived media richness of a channel can be influenced. This channel expansion theory (Carlson & Zmud, 1999) states there are five variables that influence the perceived media richness of a channel: 1) experience with the channel, 2) experience with the topic of discussion, 3) experience with the organizational context, 4) experience with one‘s communication partner, and 5) perceived social influence. These variables are called the ‗channel expansion predictors‘.
The channel expansion theory is tested for the channels e-mail, telephone and face-to-face media (Carlson & Zmud, 1999; Timmerman & Madhavapeddi, 2008; D‘Urso & Rains, 2008). This research determines if the channel expansion theory also holds for the channel Twitter. Twitter is used more and more in the external communication of companies, however not much research has been performed on Twitter yet. The first research question therefore is: ―What is the
influence of the 1) experience with the channel, 2) experience with the topic, 3) experience with the organizational context, 4) experience with one’s communication partner, and 5) the perceived social influence on the perceived media richness of Twitter?”
As experience can change with increased usage (Rogers, 1983, Adam, Nelson & Todd, 1992), some of the investigated relationships might be affected by the intensity of use. The moderating role of intensity of use is therefore investigated: “Does the intensity of use enhance the
relationship between 1) experience with the channel, and 2) experience with the communication partner and the perceived media richness of Twitter?”
3
recommending the channel Twitter to others, and 2) recommending the Twitter page to others?”
The data collection for this research took place by means of a questionnaire which was distributed in three ways. The questionnaire was sent via the Twitterpage of @aexnl, as well as within the company to all the Twitterusers, and via an Economic faculty by sending the questionnaire to professors and a finance class. These three groups have the highest chance of being on Twitter and/ or have affiliation with the content placed upon the Twitter page. The sampling frame consists of 2840 people of whom a total 79 responded. 68 respondents filled in the questionnaire completely and correct. The sample size (N=68) led to a response rate of 2,4%. A single and multiple regression analysis is used to find that three of the five channel expansion predictors have a positive influence on the perceived media richness of Twitter. The most influential factors on the perceived media richness of Twitter in order of importance are experience with one‘s communication partner, experience with the channel, and experience with the topic of discussion. For experience with the organizational context, and perceived social influence no significant relationships are found. The study investigates if intensity of use plays a moderating role on the relationships between experience with the channel and experience with one‘s communication partner, and the perceived media richness of Twitter. No support for this moderating role was found. A relationship between the perceived media richness of Twitter and the recommending of the channel Twitter to others, and recommending the Twitter page to others is assumed. Support is found for a positive relationship between the perceived media richness of Twitter and recommending the channel Twitter and recommending the Twitter page to others.
4 PREFACE
The reason for conducting this research is to finish my Master of Science in Marketing at the University of Groningen. The finishing of this Master Thesis will officially put an end to my student life and is the beginning of a new phase in life. As a user of Social Media I find it an interesting field of research. This combined with my internship led to the subject of investigation for this thesis. I would like to thank my supervisor prof. dr. Janny C. Hoekstra for her constant guidance during the entire process. I would also like to thank my second supervisor dr. J.A. Voerman for her feedback and insights to finalize writing my thesis.
Laura Kranenburg.
5 TABLE OF CONTENTS
1. INTRODUCTION ... 7
1.1 Background ... 7
1.2 Research questions ... 8
1.3 Managerial and academic relevance ... 9
1.4 Structure ... 10
2. THEORETICAL FRAMEWORK ... 11
2.1 Channel expansion theory ... 11
2.1.1 Channel expansion predictors ... 11
2.1.2 Previous research ... 12
2.1.3 Channel expansion theory applied on Twitter ... 13
2.2 Intensity of use ... 14 2.3 Word-of-mouth... 15 2.4 Conceptual model ... 17 3. RESEARCH DESIGN ... 18 3.1 Data collection... 18 3.2 Sample ... 18 3.3 Measurement of constructs... 19 3.3.1. Reflective scales ... 21 3.3.2. Formative scale ... 22 3.4 Method of analysis ... 23 4. RESULTS ... 25
4.1 Role of Channel Expansion Predictors ... 25
4.2 Moderating role of Intensity of Use ... 26
6
5. CONCLUSION ... 28
5.1 Conclusion Channel Expansion Predictors ... 28
5.2 Conclusion moderating role of Intensity of Use ... 29
5.3 Conclusion recommending to others ... 29
7 1. INTRODUCTION
Text messaging, a message with a maximum of 160 characters sent via one mobile phone to another, is used in our daily lives to communicate with each other. When text messaging was introduced, for some people it appeared to be a great solution for sending short messages and receiving a rapid response. However, for other people this might not have been the case.
When a communication channel conveys a message exactly as intended, the channel is determined as rich (Daft & Lengel, 1984). Channels are perceived differently in their richness; face-to-face communication is perceived as more rich, where written letters are perceived as more lean (Daft & Lengel, 1984). As is shown in the example above, different people can also have different perceptions concerning the richness of a channel.
1.1 Background
The perceived media richness of a communication channel determines the suitability of a channel to convey the intended message. It helps people to select a certain channel for their communication purposes (Daft & Lengel, 1984). One person‘s view of a channel such as e-mail can be very positive, it is the ideal channel for him/her, suitable for all kinds of communication. However, another person believes that this channel is incredibly slow, not user-friendly, impersonal and is not suitable for the job that he/she wants to use it for. Two people can therefore have two different perceptions of the same channel. When a company selects a channel for its external communication, it could be that this channel is not perceived as the most suitable type by its customers. However, according to the channel expansion theory the perceived media richness of a channel can be influenced (Carlson & Zmud, 1999). With a positively changed perceived media richness the channel can eventually be determined as suitable to convey the intended message.
8 knowing what experiences are the most important in influencing the perceived media richness of a certain channel, one can be more selective on what to focus on to positively change the perceived media richness.
Timmerman & Madhavapeddi (2008), and D‘Urso & Rains (2008) show support for the channel expansion theory. Carlson & Zmud (1999) tested their theory longitudinal and cross-sectional on the channel e-mail. Timmerman & Madhavapeddi (2008) replicated this research and complemented with the channels telephone and face-to-face media. The theory was supported on all three types of media finding a positive relation between some of the knowledge-building experiences, perceived social influence and perceived media richness.
Over a period of time more and more communication channels, such as Social Media, have emerged. Social Media is a way of communicating through the Internet. It ―combines features of one-way media and two-way media. Like one-way media, information is broadcasted from one source to a (potentially unknown) audience. But like two-way media, individuals can react and respond to this communication through the same channels‖ (Hogan & Quan-Haase, 2010). In a short period of time Social Media has gained a lot of users and multiple types of Social Media, such as Facebook, LinkedIn, and Twitter have arisen. The different types of Social Media have their own characteristics and are used for different purposes: ―Social networking, social bookmarking, video-sharing, picture-sharing, professional networking, user forums, weblogs, and micro-blogging‖ (Fischer & Reuber, 2011). Generalizations for all types of Social Media can therefore be difficult to make. According to comScore Media Matrix (2006), every second Internet user in the United States has visited at least one of the top 15 social networking sites (Trusov, Bucklin & Pauwels, 2009). Twitter has 400 million tweets per day (Twitter, 2012), where Facebook has 1.6 billion views a day (Facebook, 2012). However, despite this large reach and usage the way in which the use of Social Media can be influenced is not investigated much. 1.2 Research questions
9 another channel. The specific type of Social Media that is used for this research is Twitter.
Twitter displays personal messages with a maximum of 140 characters. When using Twitter a form of micro-blogging is used. Micro-blogs have a set of shared characteristics: short text messages, instant message delivery, and subscriptions to receive updates (Jansen et al., 2009). These characteristics are not found in the characteristics of the channels already tested before, which could indicate different results for this channel in comparison with the previously tested ones. Therefore, one of the research questions of this investigation is ―What is the influence of
the 1) experience with the channel, 2) experience with the topic, 3) experience with the organizational context, 4) experience with one’s communication partner, and 5) the perceived social influence on the perceived media richness of Twitter?”
Experiences can improve with increased usage (Rogers, 1983). It might therefore affect some of the investigated relationships. This research therefore also investigates “Does the intensity of use
enhance the relationship between 1) experience with the channel, and 2) experience with the communication partner and the perceived media richness of Twitter?”
The Internet provides consumers with the possibility to find and give advice by engaging in electronic word-of-mouth (eWOM) (Hennig-Thurau, Gwinner, Walsh, Germler, 2004). Micro-blogging is found to be a new form of eWOM (Jansen, Zang, Sobel & Chowdury, 2009) as people use micro-blogs to talk online about their daily activities and to seek and share word-of-mouth (Java, Song, Finin & Tseng, 2007; Jansen et al., 2009). This research investigates if the perceived media richness of Twitter will positively influence the sharing of word-of-mouth. Therefore, the final research question of this investigation is “Does the perceived media richness
of Twitter positively influence 1) recommending the channel Twitter to others, and 2) recommending the Twitter page to others?”
1.3 Managerial and academic relevance
10 experiences, social influence and the perceived media richness of Twitter isof relevance to companies as they can try to focus on the most influential experiences and enhance them. This positively affects the perceived media richness of the channel, which leads to a better evaluation of the selected communication channel Twitter. In the micro-blogging field not much has been published. Java et al. (2007) studied the topological and geographical properties of Twitter‘s social network, while McFedries (2007) and Milstein, Chowdhury, Hochmuth, Lorica & Magoulas (2008) presented an overview of micro-blogging and Twitter. Micro-blogging in an educational setting is investigated (Ebner & Schiefner, 2008; Grosseck & Holotescu, 2008), however no research has been performed that applied the channel expansion theory to the
channel of Twitter. This study therefore contributes to the existing literature as it adds a new type of channel where the channel expansion theory is tested upon, and it provides new insights into micro-blogging.
1.4 Structure
11 2. THEORETICAL FRAMEWORK
This chapter starts with a brief overview of the theories that are used in this research. From this, a series of hypotheses are derived that are aimed at determining the relationships between the knowledge-building experiences of the channel expansion theory, the perceived social influence and the perceived media richness of the channel Twitter. The moderating role of intensity of use on the relationship between experience with the channel, and experience with the communication partner and the perceived media richness of Twitter is investigated. As well as determining the influence of the perceived media richness of Twitter on recommending the channel Twitter and the Twitter page to others. The conceptual model is presented at the end of the chapter, summarizing all hypotheses.
2.1 Channel expansion theory
Media richness defines a channel‘s capacity to carry rich information. It is based on four characteristics: 1) speed of feedback, 2) ability to present individually tailored messages, 3) ability to communicate multiple cues, and 4) the capability of the channel to use natural language to convey subtleties (Daft & Lengel, 1984). All channels differ in richness, a channel defined as more rich is for example able to deliver feedback in a more rapid pace and can offer multiple cues. In the media richness theory a ranking of the richness of channels is made, i.a. including the channels face-to-face communication and written letters. The perceived media richness of a channel helps people to determine which channel to use for certain communication purposes. (Daft & Lengel, 1984).
The channel expansion theory defines that the perceived media richness is influenced by knowledge-building experiences and perceived social influence. When the experiences and social influence increase, there is a positive influence on the perceived media richness, leading to a richer perception of the channel. The channel expansion theory determines what influences the perceived media richness, while the media richness theory determines what the characteristics of the perceived media richness are.
2.1.1 Channel expansion predictors
12 one‘s communication partner, and 4) experience with the organizational context (Carlson & Zmud, 1999). The process of learning the features and the limitations of a channel is included in the experience with using a channel. Experience with the topic of discussion defines the complexity in the communication. Having more experience with the topic can make a conversation easier. With more experience with a topic, the knowledge of that topic increases, which leads to people being more able to understand more difficult, or detailed conversations concerning that topic. When having multiple interactions with a communication partner, the experience with one‘s communication partner will increase. When people learn traits, peculiarities, and communicative practices from another, it improves the understanding of the way in which the partner communicates. Experience with organizational context refers to understanding of for example symbols. If there is understanding of the use of language that is understood and communicated by organizational members, it will improve the understanding of the external communication of a company. These experiences influence the perceived media richness because with an increased experience, there is a better understanding of the capabilities of a channel which improves a perception of a channel. A person can have a perception of a channel, while never having used it. If the experience with the channel increases, so does the understanding of the channel. The person would be more able to define all the capabilities and will be better able to define the richness of a channel.
The perceived media richness is also influenced by perceived social influence, i.e. influence coming from friends, colleagues, and others that talk about their experiences, evaluations, and opinions concerning all kind of things (Schmitz & Fulk, 1991). From this moment on these five variables: 1) experience with using a channel, 2) experience with the topic of discussion, 3) experience with one‘s communication partner, 4) experience with the organizational context, and 5) perceived social influence are all referred to as the ‗channel expansion predictors‘.
2.1.2 Previous research
13 perceived media richness. In the cross-sectional study the perceived social influence did not find support, however support was found in the longitudinal study.
Support for the channel expansion theory was found with D‘Urso & Rains (2008), and Timmerman & Madhavapeddi (2008). They conducted the research for the channel e-mail as Carlson & Zmud (1999) did, and added the channels of telephone and face-to-face. The research of Timmerman & Madhavapeddi (2008) found the experiences with a channel, communication partner, and topic to have a positive relationship with the perceived media richness of e-mail, telephone, and face-to-face media. D‘Urso & Rains (2008) found that ―as users‘ levels of knowledge-building experiences with a medium and communications partner increased, so did perceptions of e-mail richness‖.
2.1.3 Channel expansion theory applied on Twitter
14 and therefore it is assumed that the same relationships will exist for the channel Twitter as do for the channels telephone, e-mail, face-to-face. Therefore,
Hypothesis 1a: Experience with the channel is positively related to the perceived media richness of Twitter.
Hypothesis 1b: Experience with the topic of discussion is positively related to the perceived media richness of Twitter.
Hypothesis 1c: Experience with one’s communication partner is positively related to the perceived media richness of Twitter.
Hypothesis 1d: Experience with the organizational context is positively related to the perceived media richness of Twitter.
In the research of Carlson & Zmud (1999) a relationship between perceived social influence and perceived media richness of the channel e-mail was not found in the cross-sectional study, however support was found in the longitudinal study. With the coming of Internet, which became more booming after the research of Carlson & Zmud (1999), the options for people to find and gather information has expanded (Hennig-Thurau et al., 2004). This makes it much easier to find opinions from others and to be influenced in that way. Therefore,
Hypothesis 1e: Perceived social influence is positively related to the perceived media richness of Twitter.
2.2 Intensity of use
15 The in-depth knowledge and use of the different tools supplied by a channel is also defined as intensity of use (Boys & Marsden, 2003; Bormann, Uphold & Maynard, 2009). A study
measuring the impact of internet use on business-to-business marketing measured the intensity of Internet with the use of Internet tools, for example the use of e-mail, and the World-Wide-Web (WWW) (Avlonitis & Karayanni, 2000).Applying this to Twitter, the tools are ‗retweet‘ and ‗mention‘. In this research the two dimensions of intensity of use are combined. Therefore not only the frequency, as Carlson & Zmud (1999) use it, but also the use of tools is measured with the intensity of use.
Rogers (1983) found that as the intensity of use changes, the user gains experience with the technology in question. Adams, Nelson & Todd (1992) state that more frequent users may have a broader experience with a computer, indicating that usage has an influence on experience.
Extensive knowledge and use of a channel is therefore assumed to affect the relationship between the experience with a channel and the perceived media richness.
Hypothesis 2a: The intensity of use enhances the relationship between experience with the channel and the perceived media richness of Twitter.
When making more extensive use of a channel, this could lead to an increased amount of sent messages. This could lead to more communication, experience, with a communication partner, indicating that intensity of use can influence the relationship between experience with one‘s communication partner and the perceived media richness. Therefore,
Hypothesis 2b: The intensity of use enhances the relationship between experience with one’s communication partner and the perceived media richness of Twitter.
2.3 Word-of-mouth
16 consumers about the ownership, usage, or characteristics of particular goods and services and/or their sellers‘ (Westbrook 1987). Recent studies from Gruen, Osmonbekov & Czaplewski (2006), and Wangenheim & Bayón (2007) agree with these definitions. WOM plays an important role in shaping the attitudes and behaviors of consumers (Brown & Reingen, 1987), people are more receptive for the opinions of others, instead of reading about things and assuming it is good. With Internet it became more easy to not only spread the word, but also to find information coming from other people concerning a certain product or topic, eWOM. Online user reviews started to substitute the old-fashioned offline WOM communication (Chevalier & Mayzlin, 2006). EWOMcommunication is ―any positive or negative statement made by a potential, actual, or former customers about a product or company, which is made available to a multitude of people and institutions via the Internet. It is immediate, has a significant reach, is credible by being in print, and is accessible by others‖ (Hennig-Thurau et al. 2004). WOM can positively or negatively influence decisions (Buttle, 1998), however Park & Lee (2009) found negative eWOM to have a greater effect than positive eWOM. The Internet does not only provide
opportunities for consumers, it also provides opportunities for firms to take advantage of WOM (Trusov et al., 2009). Trusov et al. (2009) anticipate that ―WOM referrals lead to new sign-ups and that new sign-ups lead to more WOM referrals‖, creating a vicious circle. When applying this to Twitter new sign-ups are new followers of your Twitter page. In this research the WOM referrals that one has received are measured with the perceived social influence. However, there are two sides of word-of-mouth referrals. The WOM that one would give to other people, such as recommending Twitter or the Twitter page to others, is not yet measured. The likelihood that people will spread WOM depends on their satisfaction level (De Matos and Rossi, 2008) When people are satisfied they will be more willing to tell friends and family about their positive experience (Maxham & Netemeyer, 2002b). Experience is thus positively linked to WOM. Therefore,
Hypothesis 3a: Perceived media richness of Twitter will positively influence recommending the channel Twitter to others.
17 2.4 Conceptual model
The conceptual model summarizes all the before mentioned hypotheses and is shown in figure 1. The entire model will be controlled by age.
Figure 1: Conceptual model
+H2b + H3b +H3a + + H1e +H2a + H1d +H1c + H1a + H1b Experience with the channel
Experience with the topic of discussion
Experience with one‘s communication partner Experience with the organizational context Perceived Social Influence
18 3. RESEARCH DESIGN
In this chapter, the research methods that are used in the data collection and the data analysis are presented. The data collection process is described in paragraph 3.1, following with an explanation of the population and sample in paragraph 3.2. The measurement of the constructs is described in paragraph 3.3, making a distinction between the reflective scales and the formative scale used in the research, and discussing the issue of multicollinearity. In the final section, paragraph 3.4, the method of analysis is defined.
3.1 Data collection
The data collection took place by means of a questionnaire and is distributed using three methods. The first method is to collect the data via the Twitter page of NYSE Euronext, @aexnl. This is an exchange that facilitates trade in cash and derivative products. There is little direct communication with the end-user and the information channels are difficult to get relevant information from. Twitter is used in this context, as the @aexnl-page provides relevant and value-adding content, as well as a possibility for interaction between the exchange and the end-users. The second method of distribution was sending the questionnaire internally in the company to all the Twitter users. The third method was distributing the questionnaire in a finance class and spreading it to professors of an Economic faculty. These three groups have the highest chance of being on Twitter and/ or have affiliation with the content placed upon the Twitter page.
3.2 Sample
The population of this research consists of all people that use Twitter. The sampling frame consists of the 2700 followers of the Twitter page, 20 Twitter users within the company, and 20 professors and around 100 students in the Economic Studies field. The questionnaire was sent with a link in a tweet multiple times a week during a period of five weeks on the Twitter page. Retweets (forwarding someone else‘s message via your own Twitter account) of the tweet which included the link to the questionnaire have occurred, leading to a higher reach. In total, 79 people responded, 68 respondents filled in the questionnaire completely and correct. The sample size (N=68) represented a response rate of 2,4%.
19 2006). The sample has a higher male percentage than average. The male/female ratio Twitter user is 47% male, 53% female (Sysomos research), this sample also has a higher male percentage than this average.
3.3 Measurement of constructs
All constructs for the questionnaire are measured with using existing formative and reflective scales by Daft & Lengel (1984), Carlson & Zmud (1999), and Schmitz & Fulk (1991). All channel expansion predictor variables and recommendations to others are measured with a reflective scale. Intensity of use is measured with a formative scale. Adaptations are made where needed, and in some cases new items are included. Table 1 provides an overview of the constructs and their reliabilities.
TABLE 1 Constructs and reliability
Construct Items Nr. of
items
Source Cronbach’s
Alpha
Experience with Twitter 5 Carlson & Zmud
(1999)
.943 I am very experienced using
I feel that Twitter is easy to use I feel competent using Twitter I understand how to use all of the features of the Twitter system
I feel comfortable using Twitter
I feel that I am a novice using the Twitter system**
Experience with exchange related information
1 Carlson & Zmud (1999)
I feel that I am experienced with exchange related information
I feel that I am well-versed in the concepts associated with exchange related concepts** I do not feel knowledgeable about exchange related information **
Experience with @aexnl 5 Carlson & Zmud
(1999)
.937 Overall, I feel I know @aexnl
well
20
with @aexnl I feel comfortable
communicating with @aexnl I feel involved with @aexnl
I do not trust @aexnl** I feel that I am not familiar with @aexnl**
I feel more comfortable communicating in a formal manner with @aexnl rather than in an informal
manner**
Experience with the exchange 5 Carlson & Zmud
(1999)
.975 I am well-versed in the internal
affairs of the exchange
I am familiar with the culture of the exchange
I use a lot of exchange related jargon in my job
I know just as much as people that work on the exchange When communicating with the exchange I am able to use organization-specific jargon and cultural references
Perceived social influence 8 Schmitz & Fulk
(1991),
Carlson & Zmud (1999)
.921
My colleagues frequently use Twitter to communicate My colleagues have expressed to me how useful Twitter is My friends frequently use Twitter to communicate My friends have expressed to me how useful Twitter is My family frequently uses Twitter to communicate My family has expressed to me how useful Twitter is
My supervisor has expressed to me how useful Twitter is My supervisor frequently uses Twitter to communicate
Perceived Media Richness 4 Daft & Lengel (1984) .918
Twitter allows @aexnl and me to give and receive timely feedback
21
to communicate a variety of different cues (such as emotional tone, attitude, or familiarity) in our messages Twitter allows @aexnl and me to use rich and varied language in our messages
Intensity of use 5 NEW Formative scale
I have received …… messages from the Twitter page of @aexnl during a week. How often a day do you take a look at the messages that are sent by the Twitter page of @aexnl?
How often do you look at the Twitter page of @aexnl per week?
How often do you retweet per week?
How often do you mention others per week?
I have sent …….. messages to the Twitter page of @aexnl during a week.**
Recommending to others 2 NEW Reflective scale
I would recommend Twitter to others
I would recommend @aexnl to others
** Excluded from further research
3.3.1. Reflective scales
22 (Nunnally, 1978; Churchill, 1979; Malhotra, 2006). With coefficient alphas all higher than .90 the scales are defined as reliable. The reliabilities for all constructs are shown in table 1. High correlations are present, therefore multicollinearity poses a threat and is tested upon (Cooper & Schindler, 2010). Field (2000) states that the tolerance should be higher than .10 and the Variance of Inflation (VIF) should be lower than 10. Even though high correlations are shown between for example recommending the channel Twitter and recommending the Twitter page, there is no case of multicollinearity for any of the constructs as the tolerances and VIFscores are acceptable for all constructs. The scores of the tolerance and VIF are shown in appendix 2.
3.3.2. Formative scale
A formative scale is used for the construct intensity of use, including five items. These items reflect both the frequency of use, based on Carlson and Zmud (1999), as the more extensive use of a channel (Avlonitis & Karayanni, 2000). High correlations between formative indicators might occur, but are generally not expected. Factor analysis and Cronbach‘s Alpha are not appropriate to evaluate a formative measure (Rossiter, 2002; Christophersen & Konradt, 2010). A correlation test is therefore used to test the suitability of the scales for further analyses. A correlation of .30 or lower is acceptable to continue the research with. If the correlation exceeds the level of .30, the variables are said to measure the same construct (Cohen, 1988; Nurosis, 1993)The correlation table for the items of intensity of use is shown in table 2. The table shows that retweets and mentions, and read tweets and looking on the page have correlations that are higher than .30. However, further analyses are conducted as the items were considered to be highly correlated when creating the scale, and any elimination of relevant formative indicators results in a reduction of scale validity (Christopherson & Konradt, 2010).
TABLE 2
Correlations Intensity of Use
Variable Mean Standard
23 3.4 Method of analysis
In this paragraph the statistical methods for analyzing the data are described. For the data analysis the statistical software program SPSS is used.
Simple and multiple regression analysis is used for testing the hypotheses. The following three models are estimated, where model 1 is linked to hypotheses 1a, 1b, 1c, 1d, and 1e, model 2 is linked to hypotheses 2a and 2b, and model 3 is linked to hypotheses 3a and 3b.
(1): PMR = 0 + 1Exp_Tw + 2Exp_@aexnl + 3Exp_info + 4Exp_exch + 5PSI +
6AGE+ ε
(2a): PMR = 0 + 1Exp_Tw + 2Exp_@aexnl + 3Exp_info + 4Exp_exch + 5PSI + 6IOU + 7IOU*Exp_Tw + 8AGE+ ε
(2b): PMR = 0 + 1Exp_Tw + 2Exp_@aexnl + 3Exp_info + 4Exp_exch + 5PSI + 6IOU + 7IOU*Exp_@aexnl + AGE+ ε
(2c): PMR = 0 + 1Exp_Tw + 2Exp_@aexnl + 3Exp_info + 4Exp_exch + 5PSI + 6IOU + 7IOU*Exp_Tw+ 8IOU*Exp_@aexnl + AGE+ ε
(3a): Recomm_Tw = = 0 + 1PMR+ 2AGE+ ε
(3b): Recomm_@aexnl = = 0 + 1PMR + AGE + ε In which:
PMR = Perceived Media Richness of Twitter Exp_Twi = Experience with Twitter
Exp_@aexnl = Experience with @aexnl
Exp_info = Experience with exchange related information Exp_exch = Experience with the exchange
PSI = Perceived Social Influence IOU = Intensity of Use
IOU*Exp_Tw = Intensity of Use * Experience with Twitter
24 Recomm_@aexnl = Recommending @aexnl to others
25 4. RESULTS
In this chapter the results of the regression analyses are presented. The hypotheses are tested and the results are used to determine if the hypotheses are supported or not. A summary of all hypotheses can be found at the end of the chapter.
4.1 Role of Channel Expansion Predictors
The regression results for all the hypotheses are shown in table 3. H1a hypothesizes a positive relationship between experience with the channel and the perceived media richness of Twitter. This hypothesis is supported ( = .323; p < .01). Support is also found for H1b ( =.314; p < .05) and H1c ( =.412; p <.01), confirming that both experience with topic of discussion and experience with the communication partner are positively related to the perceived media richness of Twitter. Experience with the communication partner has the strongest effect on the perceived media richness of Twitter ( =.412). Experience with the channel has the second largest positive influence ( = .323), and experience with the topic of discussion has a =.314.
No support was found for H1d ( =-.197; p >.10) and H1e ( = .025; p >.10 ) indicating that there is no significant evidence that experience with the organizational context and perceived social influence are positively related to the perceived media richness of Twitter. The model has a determination coefficient of R² = .487 and has been controlled for by age, however no significance (p>.10) was shown indicating that age has no proven influence on the relationship between the channel expansion predictors and the perceived media richness of Twitter.
TABLE 3
Regression results (standardized coefficients)
26 Perceived Media Richness 3a, 3b .608** .686** Intensity of Use 2 .561 .155 .411 Intensity of Use*Experience with Twitter 2a -.599 -.777 Intensity of Use*Experience with @aexnl 2b .140 .499 Age All .055 .014 .014 .014 .161 .145 R² .526 .520 .511 .524 .470 .370 (Adjusted R²) (.487) (.455) (.446) (.449) (.462) (.360) F-Value 13.738** 8.048 7.775 7.020 58.604** 38.714**
Standardized betas are used
a) Dependent variable: Perceived Media Richness of Twitter b) Dependent variable: Recommending the channel Twitter to others c) Dependent variable: Recommending the Twitter page to others * p < .05
** p < .01
4.2 Moderating role of Intensity of Use
The regression model of the interaction term intensity of use showed insignificant results. Hypothesis 2a and 2b were therefore not supported. While considering the small sample size, the models were computed into one, Model 2c, to find out if significance could improve. H2a (p >.10) and H2b (p >.10) have still found no support, showing no proof that intensity of use enhances the relationship between experience with the channel, experience with the communication partner and the perceived media richness of Twitter. All three models were controlled for by age, however no significant results were shown (p >.10).
4.3 Role of recommending to others
The perceived media richness of Twitter is assumed to influence recommending the channel Twitter, and recommending the communication‘s partner, the Twitter page, to others. This was tested in model 3a and 3b. Both H3a ( = .608; p <.01) and H3b ( = .686; p <.01) found support. Recommending the channel to others ( = .608) and recommending the communication‘s partner to others ( = .686) are both positively influenced by the perceived media richness.
27
TABLE 4 Summary of hypotheses
Hypothesis Supported/ not supported
1a Experience with the channel is positively related to the perceived media richness of Twitter.
Supported 1b Experience with the topic of discussion is positively related to the
perceived media richness of Twitter.
Supported 1c Experience with one’s communication partner is positively related to the
perceived media richness of Twitter.
Supported 1d Experience with the organizational context is positively related to the
perceived media richness of Twitter.
Not supported 1e Perceived social influence is positively related to the perceived media
richness of Twitter.
Not supported 2a The intensity of use enhances the relationship between experience with the
channel and the perceived media richness of Twitter.
Not supported 2b The intensity of use enhances the relationship between experience with
one’s communication partner and the perceived media richness of Twitter.
Not supported 3a Perceived media richness of Twitter will positively influence
recommending the channel Twitter to others.
Supported 3b Perceived media richness of Twitter will positively influence
recommending the Twitter page to others.
28 5. CONCLUSION
This paper conceptualizes and measures if the channel expansion theory holds for a new type of channel, Twitter. It defines the relationship between 1) experience with the channel, 2) experience with the topic of discussion, 3) experience with one‘s communication partner, 4) experience with the organizational context, and 5) the perceived social influence and the perceived media richness of Twitter. It also determines if the intensity of use has a moderating role on some of the relationships between the channel expansion predictors and the perceived media richness of Twitter. And it describes if the perceived media richness of Twitter influences recommending the channel Twitter and recommending the Twitter page to others. For three of the five channel expansion predictors support was found. Experience with one‘s communication partner had the most positive influence in the perceived media richness of Twitter, following with experience with the channel and experience with the topic of discussion. For the variables experience with the organizational context and perceived social influence no support was found. The moderating role of intensity of use between the experience with the channel and experience with one‘s communication partner and the perceived media richness of Twitter found no support. The perceived media richness of Twitter was found to have a positive influence on recommending the channel Twitter and recommending the Twitter page to others.
5.1 Conclusion Channel Expansion Predictors
29 richness of e-mail. The research of Timmermans & Madhavapeddi (2008) found a positive relationship with the channel, communication partner, and topic across the channels e-mail, telephone, and face-to-face media. These are the same results of this study.
5.2 Conclusion moderating role of Intensity of Use
The intensity of use was measured differently in the prior research of Carlson & Zmud (1999), the total messages processed found no support of a relation between channel use and the perceived richness of the channel e-mail. The intensity of use measures not only the frequency of messages but also the in-depth knowledge and use of the different tools supplied by the channel. No support was found for the moderating role of this variable. Measuring the moderating effect of the interaction terms separately received no significance, grouping the two models into one achieved the same results. The small sample could be a problem of the insignificance.
5.3 Conclusion recommending to others
Perceived media richness of Twitter was found to have a positive relationship with recommending the channel Twitter and recommending the Twitter page to others. From this can be derived that the media richness positively influences the recommendations to others. This was to be expected, considering the fact that a positive attitude usually leads to a positive word-of-mouth. In this case having a positive perceived media richness leads to a positive evaluation of the channel Twitter and the Twitter page, which is communicated to other people.
5.4 Limitations
Limitations in this study lie in the low response rate, only 2.4% of the sample frame responded accordingly. 68 respondents could be used for the research. This particularly led to problems measuring the moderating role of intensity of use. A larger sample would have probably led to more significant results, considering the fact that many variables were included.
30 Another limitation is the industry, within the exchange business it is very difficult to directly reach the end-consumer. It could be that in another industry where companies have more direct contact with its end-consumers this study has different results for the channel Twitter. The different results with Carlson & Zmud (1999) could also lie with the short-period of time. Where Carlson & Zmud (1999) did a cross-sectional and a longitudinal study, this study took place within 5 weeks. Prolonging this study could lead to different results. Also, a limitation is that it is a very specific target group that is investigated. The respondents have to use Twitter, but also know the @aexnl Twitter page and also have knowledge of the content placed upon the Twitter page.
The questionnaire includes self-report data, this could have led to a bias in the results. 5.5 Future research
A future recommendation for this study lies at doing the same research on Twitter pages in other industries, to see if they show the same results. It would also be interesting to make a comparison with the previous research of Daft & Lengel (1984) to see if the ranking of the channels (e.g. face-to-face and written letters) has still remained the same. Including new types of channels would be a good addition to the research. As more channels have come to exist over the years it is of relevance, also for companies, to have the perception of the channels defined. Differences in ranking may have occurred with the existence of new channels.
Social Media consists of different types of communication channels, from which Twitter is one. These channels all have their specialties, even though they also have a lot in common such as the rapid response. It could be of interest to test this theory on other Social Media channels such as Facebook to see if a certain common ground can be found.
Finding a way to enhance the knowledge-building experiences would be interesting to investigate. Determine what needs to be done to improve these experiences for different type of channels. It was found that as the experiences increase, so do the perceptions of media richness, however no study has offered a solution as to how to increase these experiences.
5.6 Implications
32 REFERENCES
ABA Bank Marketing, Time to tweet?, November 2009. P. 24-30.
Adams, D. A., Nelson, R.R., Todd, P.A., (1992), ‗Perceived usefulness, ease of use, and usage information technology: A Replication‘, MIS Quarterly, June
Arndt, J. (1967), ‗Word-of-mouth advertising and informal communication. In D. Cox (Ed.),
Risk taking and information handling in consumer behavior, Boston: Harvard University.
Avlonitis, G.J., Karayanni, D.A., (2000), ‗The Impact of Internet use on Business-to-Business Marketing. Examples from American and European Companies‘, Industrial Marketing
Management, 29, 441-459.
Bormann, J.E., Uphold, C.R., Maynard, C., (2009) ‗Predictors of complementary/alternative medicine users and intensity of use among men with HIV infection from two geographic areas in the United States‘, Journal of the Association of Nurses AIDS Care, 6, 468-480.
Boys, A., Marsden, J., (2003) ‗Perceived functions predicts intensity of use and problems in young polysubstance users‘, Addiction, 8, 951-963.
Brown, J.J., Reingen P.H., (1987) ‗Social Ties and Word-of-Mouth Referral Behavior‘, Journal
of Consumer Research, vol. 14, December.
Bun Lee, E., (2012), ‗Young, black, and connected: Facebook usage among African American college students‘, Journal of Black Studies, 43 (3) 336-354.
Buttle, F.A., (1998), ‗Word of mouth: understanding and managing referral marketing‘, Journal
of Strategic Marketing, 6, 241-254.
Carlson, J.R., Zmud, R.W., (1994) ‗Channel expansion theory: A dynamic view of media and information richness perceptions‘, Academy of Management Best Papers, Proceedings (D.P. Moore, Ed.): 280-284.
33 Chevalier, J.A., Mayzlin, D., (2006), ‗The effect of word of mouth on sales: online book
reviews‘, Journal of Marketing Research, 43 (3), 345-354.
Christophersen, T., Konradt, U., (2010), ‗The development of a formative and a reflective scale for the assessment of on-line store usability‘, Journal of Systemic, Cybernetics and Informatics, 6 (5), 36-41.
Churchill, G.A., (1979, February), ‗A paradigm for developing better measures of marketing construct‘, Journal of Marketing Research, 64-73.
Cialdini, R.B., (2001), ‗Influence: Science and practice’, 4th
Ed, Boston: Allyn&Bacon. Cohen, 1988, ‗Statistical power analysis for the behavioral sciences’ (2nd ed.). New Jersey: Lawrence Erlbaum.
Cooper, D.R., Schindler, P.S., (2010), ‗Business Research Methods’, McGraw-Hill Education, Europe, 11th Revised Edition.
Daft, R., Lengel, R. (1984) ‗Information richness: A new approach to managerial behavior and organization design.‘ In B. M. Staw & L. L. Cummings (Eds.), Research in organizational
behavior, vol. 6: 191-233. Greenwich, CT: JAI Press.
Daft, R., Lengel, R. (1986) ‗Organizational information requirements, media richness and structural design.‘ Management Science, 32, 554-571.
Daft, R., Lengel, R., Trevino, L. (1987) ‗Message equivocality, media selection, and manager performance: Implications for information systems.‘ MIS Quarterly, 17: 355-366.
De Matos, C. E., Rossi, C. A. V., (2008), ‗Word-of-mouth communications in marketing: a meta analytic review of the antecedents and moderators‘, Journal of the Academy of Marketing Science, 36: 578-596.
D‘Urso, S.C., & S. A. Rains (2008) ‗Examining the scope of channel expansion. A test of channel expansion theory with new and traditional communication media‘, Management
34 Ebner, M., Schiefner, M., (2008), ‗In Microblogging – more than fun?‘ In I. Arnedillo Sánchez & P. Isaías (Eds.). Proceedings of the IADIS Mobile Learning Conference (pp. 15—159). Lisbon, Portugal, IADIA.
Ellison, N.B., Steinfeld, C., Lampe, C., (2007), ‗The benefits of Facebook ‖friends‖: Social capital and college students‘ use of online social network sites‘, Journal of Computer-Mediated
Communication, 12(4), 1143-1168.
Fischer, E., Reuber, R.A., (2011), ‗Social Interaction via new social media: (How) can
interactions on Twitter affect effectual thinking and behaviour?‘, Journal Business Venturing, 26, 1-18.
Field, A., (2000), ‗Discovering Statistics Using SPSS for Windows’, Sage Publications, London. Grosseck, G., Holotesch, C., (2008, April), ‗Can we use Twitter for educational activities?‘ Paper presented at the Fourth International Science Conference eLearning and Software for Education, Bucharest, Romania.
Gruen, T.W., Osmonbekov, T., Czaplewski, A. J., (2006), ‗eWOM: the impact of customer-to-customer online know-how exchange on customer-to-customer value and loyalty‘, Journal of Business Review, 59 (4), 449-456.
Hennig-Thurau, T., Gwinner, K., Walsh, G., Germler, D. D. (2004) ‗Electronic word-of-mouth via consumer-opinion platforms: what motivates consumers to articulate themselves on the internet?‘, Journal of Interactive Marketing, Vol. 18 (1), 38-52.
Hogan, Bernie and Anabel Quan-Haase (2010), ―Persistence of change in Social Media‖,
Bulletin of Science Technology & Society, 30: 309.
Jansen, B. J., Zhang, M., Sobel, K., Chowdury, A., (2009), ‗Twitter power: Tweets as Electronic Word of Mouth‘, Journal of the American Society for information science and technology, 60 (11): 2169-2188.
Java, A., Song, X., Finin, T., Tseng, B., (2007), ‗Why we use Twitter: Understanding
35
First SNA-KDD 2007 Workshop on Webwining and Social Network Analysis (WebKDDSNA
-KDD ‘07) pp. 56-65. New York: ACM Press.
Katz, E., & Lazarsfeld, P. F., (1955), Personal influence, Glencoe: Free.
Kelman, H., (1958), ‗Compliance, identification, and internalization: Three processes of attitude change‘, Journal of Conflict Resolution, 1, 51-60.
Malhotra, N.K., (2006), ‗Marketing Research: An applied orientation.‘ Pearson Prentice Hall. Maxham, J.G., III, Netemeyer, R.G., (2002b), ‗Modeling customer perceptions of complaint handling over time: the effects of perceived justice on satisfaction and intent, Journal of Retailing, 78 (4), 239-252.
McFedries, P., (2007), ‗Technically speaking: All a-Twitter‘, IEEE Spectrumm. 44 (10), 84. Milstein, S., Chowdury, A., Hochmuth, G., Lorica, B., Magoulas, R., (2008), ‗Twitter and the micro-messaging revolution: Communication, connections, and immediacy – 140 characters at a time‘, Sabastopol, CA: O‘Reilly Media.
Nunnally, J., (1978), Psychometric theory, 2nd ed. New York: McGraw Hill. Nurosis, M., (1993), Statistical data analysis: SPSS Inc.
NYSE Euronext (2006) ‗Ontwikkelingen op de markt van particulieren beleggers in Nederland 1997-2006‘.
Park, C., Lee, T.M., (2009), ‗Information direction, website reputation and eWOM effect: A moderating role of product type‘, Journal of Business Research, 62 (1), 61-67.
Robertson, T., Gatignon, H., (1986), ‗Competitive effects on technology diffusion‘, Journal of
Marketing, 50(3), 1-12.
Rogers, E.M., (1983), ‗Diffusion of innovations’, New York: The Free Press.
Rossiter, J.R., (2002), ‗The C-OAR-SE Procedure for scale development in marketing‘,
36 Schmitz, J., Fulk J. (1991) ‗Organizational colleagues, media richness, and electronic mail—A test of the social influence model of technology use‘, Communication Research, 18: 487-523. Sysomos Research - Inside Twitter. An in-depth look at the 5% of most active users, August 2009.
Timmerman, C.E., Madhavapeddi, S.N. (2008). ‗Perceptions of organizational media richness: Channel expansion effects for electronic and traditional media across richness dimensions‘, IEEE
Transactions on Professional Communication, 51, 18-32
Trusov, M., Bucklin, R. E., Pauwels, K. (2009) ‗Effect of Word-of-Mouth Versus Traditional Marketing: Findings from an Internet Social Networking site‘, Journal of Marketing, Vol. 73, 90-102.
Wangenheim, F. v., Bayón, T. (2007), ‗The chain from customer satisfaction via word-of-mouth referrals to new customer acquisitions‘, Journal of the Academy of Marketing Science, 35 (2), 233-249.
Westbrook, R.A., Fornell, C, (1979), ‗Patterns of information source usage among durable goods buyers‘, Journal of Marketing Research, August.
Westbrook, R.A., (1987), ‗Product/consumption based affective responses and post-purchase processes‘, Journal of Marketing Research, 24 (3), 258-270.
Websites:
www.facebook.com (2012) Facts concerning Facebook
37 APPENDICES
APPENDIX 1 QUESTIONNAIRE
Deze enquête onderzoekt de waargenomen media rijkdom van het kanaal Twitter. Het beantwoorden van deze vragen is geheel anoniem. Een 7-point Likert scale zal gebruikt worden voor de beantwoording van de vragen, reikend van 1) helemaal niet mee eens tot 7) helemaal mee eens. |
Het invullen van deze enquête zal ongeveer 7 minuten van uw tijd nemen. Probeer alstublieft de enquête zo volledig mogelijk in te vullen.
Als u geïnteresseerd bent in het winnen van het boek Beleggingstips 2012 vul dan uw e-mailadres in bij de eerste vraag. Uw contactgegevens zullen alleen voor het selecteren van de winnaars van het boek worden gebruikt, niet voor andere doeleinden.
Veel plezier met het invullen van de enquête en alvast heel erg bedankt voor uw hulp!
Algemene vragen
1. Wilt u kans maken om het boek Beleggingstips 2012 te winnen? Vul dan hier uw e-mailadres in. 2. Wat is uw leeftijd?...
3. Wat is uw geslacht? O Man
O Vrouw
Totaal aantal berichten
1. Ik stuur ….. berichten naar de Twitter pagina van @aexnl per week. 2. Ik ontvang …. berichten van de Twitter pagina van @aexnl per week.
3. Hoe vaak per dag leest u de berichten die door de Twitter pagina van @aexnl worden verstuurd?...
4. Hoe vaak kijkt u naar de Twitter pagina van @aexnl per week?... 5. Hoe vaak retweet u?
O 0-10 keer per week O 11-20 keer per week O 21-30 keer per week O Meer dan 30 keer per week
6. Hoe vaak mentioned u anderen per week? O 1-20 keer per week
38
Ervaring met Twitter
1. Ik heb erg veel ervaring met het gebruik van Twitter.
2. Ik vind dat Twitter makkelijk is om te gebruiken.
3. Ik voel me bekwaam om Twitter te gebruiken
4. Ik begrijp hoe ik alle features van het Twitter systeem moet gebruiken.
5. Ik voel me op mijn gemak om Twitter te gebruiken.
6. Ik vind dat ik een beginner ben bij het gebruik van het Twitter systeem.
Ervaring met @aexnl
1. Over het algemeen gezien, vind ik dat ik bekend ben met @aexnl
2. Ik vertrouw @aexnl niet.
39 4. Ik voel me op mijn gemak om informele communicatie ( zoals afkortingen) te gebruiken met
@aexnl.
5. Ik vind dat ik niet bekend ben met @aexnl.
6. Ik voel me op mijn gemak om met @aexnl te communiceren.
7. Ik voel me betrokken bij @aexnl.
8. Ik voel me meer op mijn gemak om formeel te communiceren met @aexnl in plaats van informeel.
Ervaring met beurs gerelateerde informatie
1. Ik vind dat ik ervaren ben met beurs gerelateerd informatie.
2. Ik vind dat ik goed op de hoogte ben van alle concepten die geassocieerd zijn met beurs gerelateerde informatie.
40
Ervaring met de beurs ( NYSE Euronext)
1. Ik ben goed op de hoogte met de interne zaken van de beurs.
2. Ik ben bekend met de cultuur van de beurs.
3. Ik gebruik een heleboel beurs gerelateerde jargon in mijn werk.
4. Ik weet evenveel van de beurs als de mensen die op de beurs werken.
5. Wanneer ik communiceer met @aexnl ben ik in staat om beurs specifieke jargon en culturele referenties te gebruiken.
Waargenomen sociale invloed
1. Mijn collega’s gebruiken Twitter met regelmaat om te communiceren. O Ja
O Nee
2. Mijn collega’s hebben geuit hoe nuttig Twitter is. O Ja
O Nee
3. Mijn vrienden gebruiken Twitter met regelmaat om te communiceren. O Ja
O Nee
4. Mijn vrienden hebben geuit hoe nuttig Twitter is. O Ja
O Nee
5. Mijn familie gebruikt Twitter met regelmaat om te communiceren. O Ja
41 6. Mijn familie heeft geuit hoe nuttig Twitter is.
O Ja O Nee
7. Mijn meerdere heeft geuit hoe nuttig Twitter is. O Ja
O Nee
8. Mijn meerdere gebruikt Twitter met regelmaar om te communiceren. O Ja
O Nee
Waargenomen rijkdom van Twitter
1. Twitter zorgt ervoor dat @aexnl en ik op de juiste tijd feedback kunnen geven en ontvangen.
2. Twitter zorgt ervoor dat @aexnl en ik onze berichten op maat kunnen maken naar onze persoonlijke eisen.
3. Twitter zorgt ervoor dat @aexnl en ik verschillende signalen ( zoals emotionele toon, houding of formaliteit) kunnen communiceren via onze berichten.
4. Twitter zorgt ervoor dat @aexnl en ik rijk en gevarieerd taalgebruik kunnen gebruiken in onze berichten.
Aanbevelingen aan anderen
1. Ik zou het kanaal Twitter aanbevelen aan anderen.
42 APPENDIX 2
Tolerance and VIFscores
Constructs Tolerance VIF
1. Experience with Twitter .526 1.902 2. Experience with @aexnl .496 2.014 3. Experience with exchange related information .432 2.316
