<%BANNER%>

An Exploratory Video-on-Demand Analysis: Identifying Early Adopters and Attitudes toward Potential Advertising


PAGE 1

AN EXPLORATORY VIDEO-ON-DEMAND ANALYSIS: IDENTIFYING EARLY ADOPTERS AND ATTITUDES TOWARD POTENTIAL ADVERTISING By STEPHEN WILLIAM MARSHALL A THESIS PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLOR IDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF ARTS IN MASS COMMUNICATION UNIVERSITY OF FLORIDA 2004

PAGE 2

Copyright 2004 by Stephen William Marshall

PAGE 3

This document is dedicated to the graduate st udents of the University of Florida as well as to my parents and family for all their love and support. Most importantly, this research is dedicated to couch potatoes across the globe and to my lovely doggie, Ozwaldo.

PAGE 4

ACKNOWLEDGMENTS First, I would like to thank my wonderful committee for not only guiding me through this research but also for challenging me. The faith you all have shown in me I often did not have in myself. Additionally, I thank my committee for exposing me to a career path unknown to me. I will always remember and thank each of you for this adventure. Second, I would like to thank my family and friends for supporting me through this process. I love and appreciate all the support through my good and bad times. These have been changing times and I love each of you dearly. Last, I wish the world peace. God is love. iv

PAGE 5

TABLE OF CONTENTS page ACKNOWLEDGMENTS .................................................................................................iv LIST OF TABLES ............................................................................................................vii ABSTRACT .......................................................................................................................ix CHAPTER 1 INTRODUCTION........................................................................................................1 2 LITERATURE REVIEW.............................................................................................5 The Video-On-Demand Background............................................................................5 Theoretical Grounding................................................................................................17 Diffusion of Innovations.............................................................................................18 3 METHODOLOGY.....................................................................................................40 Data Collection...........................................................................................................40 Survey Design.............................................................................................................41 Data Analysis..............................................................................................................44 4 RESULTS...................................................................................................................45 Descriptive Statistics..................................................................................................45 Research Question 1...................................................................................................53 Research Question 2...................................................................................................55 Research Question 3...................................................................................................60 Research Question 4...................................................................................................66 Research Question 5...................................................................................................73 5 SUMMARY AND CONCLUSIONS.........................................................................75 APPENDIX VIDEO-ON-DEMAND TELEVISION SURVEY....................................84 REFERENCES..................................................................................................................89 BIOGRAPHICAL SKETCH.............................................................................................93 v

PAGE 6

6

PAGE 7

LIST OF TABLES Table page 4-1 Age Distribution.......................................................................................................45 4-2 Income Distribution..................................................................................................46 4-3 Marital Status...........................................................................................................47 4-4 Occupation Distribution...........................................................................................47 4-5 Education..................................................................................................................48 4-6 Awareness, Use and Subscribers of VOD................................................................48 4-7 VOD Functions........................................................................................................49 4-8 Advertising Scenarios and Models...........................................................................50 4-9 Media Use................................................................................................................51 4-10 Social and Innovativeness Characteristics...............................................................52 4-11 Technology Ownership............................................................................................53 4-12 Preferred VOD Options Between Users and Non-Users of VOD............................54 4-13 Mean Comparison of VOD Options by VOD Users................................................55 4-14 Ad Scenario #1.........................................................................................................56 4-15 Ad Scenario #2.........................................................................................................57 4-16 Ad Scenario #3.........................................................................................................57 4-17 Ad Scenario #4.........................................................................................................58 4-18 Ad Scenario #5.........................................................................................................58 4-19 Ad Scenario #6.........................................................................................................58 4-20 Willingness to Provide Personal Information..........................................................59 vii

PAGE 8

4-22 Mean Media Use Scores Between Those Aware and Not Aware of VOD..............61 4-23 Social/Innovative Mean Scores Between Those Aware and Not Aware of VOD...62 4-24 Technology Ownership Mean Scores Between those Aware and Not Aware of VOD.........................................................................................................................64 4-25 Demographic Scores Between those Aware and Not Aware of VOD.....................64 4-26 Media Usage Comparisons between Subscribers and Non-subscribers of VOD.....67 4-27 Social/Innovative Characteristics Comparisons between Subscribers and Non-subscribers of VOD..................................................................................................68 4-28 Technology Ownership Comparisons between Subscribers and Non-subscribers of VOD.....................................................................................................................69 4-29 Demographics Comparisons between Subscribers and Non-subscribers of VOD..70 4-30 Viewing TV Content Anytime Comparisons between Subscribers and Non-subscribers of VOD..................................................................................................72 4-31 VOD Functionality Comparisons between Subscribers and Non-subscribers of VOD.........................................................................................................................73 4-32 Multiple Regression Data.........................................................................................74 viii

PAGE 9

Abstract of Thesis Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Master of Arts in Mass Communication AN EXPLORATORY VIDEO-ON-DEMAND ANALYSIS: IDENTIFYING EARLY ADOPTERS AND ATTITUDES TOWARD POTENTIAL ADVERTISING By Stephen William Marshall August 2004 Chair: Chang-Hoan Cho Major Department: Mass Communications This study attempts to identify consumers who are currently aware and likely to adopt video-on-demand technology. This research also studies attitudes toward possible advertising models within the video-on-demand business. Diffusion of innovations was the theoretical grounding for this research. A survey was used to collect the data and statistical analysis was used to conclude results. Results found respondents did not like proposed advertising models and were highly unlikely to exchange personal information for a discount on services. Additionally, results from the survey found those interested in video-on-demand are younger and better educated. ix

PAGE 10

CHAPTER 1 INTRODUCTION Video-on-demand (VOD) technologies have come a long way in the last decade. From the early testing done in the 1990s to the penetration in major cities across the United States today; VOD is here to stay. After years of planning and failed promises, the cable industry is finally able to deliver reliable and economical VOD services (Schiesel, 2002). With the recent news of News Corps purchase of DirecTV and Comcasts offer for Disney, it appears that interactive TV is finally turning into a reality. Three major cable operators heavily involved in interactive television ventures include Charter Communications, Cablevision Systems Corp. and Insight Communications Co. These companies are relying on these services to help generate more subscriptions to VOD, interactive television and high-definition services (Applebaum, 2004). Video-on-demand will become a major player in interactive television, video entertainment and education thanks to the technological advances of the digital cable industry. Cable companies have spent over $50 billion upgrading networks from analog to digital, and cable companies need a return on their hefty investment (Brister, 2001). Video-on-demand technologies will give cable companies a strategic advantage and point-of-difference when compared to competing services. VOD will be a savor for the cable industry and will give digital cable a competitive advantage when compared to satellite or other delivery platforms. 1

PAGE 11

2 In todays high-tech world, we take for granted the luxury of being able to watch what we want to watch, when we want to watch it (Schiesel, 2002). Video-on-demand allows us to do just that. The introduction of the VCR and later the DVD gave us the ability with some degree but each are not without consequences and drawbacks. The VCR was often difficult for the user to master. The VCR was designed to have many recording features but uses became limited due to lengthy manuals and design complexities. In addition, the VCR involved extra connections to the television and the purchase of videotapes for recording was also required. Even with these difficulties, the VCR received widespread adoption because viewers enjoyed pre-recording programs for later playback and renting poplar movies. In fact, the VCR created an entire new industry of home entertainment and viewing flexibility, essentially bringing the movie theater environment into the home. The DVD has followed the same entertainment footsteps of the VCR and has widely been adopted due to the technologys ease of use and increased video and sound quality. The DVD adds the digital quality aspect to video entertainment, providing sharp pictures, improved sound, improved ease of use and extra search and scene features not found in VCR technology. The main drawback for the DVD is the machine is typically an output machine enabling viewers to only watch rented or purchased prerecorded material (although newer models currently allow for recording but these models are still quite expensive). Video-on-demand offers these same qualities with even greater improvements. VOD is easy to use, has the digital quality of DVDs, creates an asynchronous environment (much like the VCR) and requires no additional connections or equipment.

PAGE 12

3 The technology is accessed through the same cable set-top box used to receive cable programming. The user accesses VOD with their cable remote and the VOD menu resembles the cable and DVD on-screen guide already familiar to many subscribers. Based on past performance and relative similarities, it appears the next wave of home video technology will come in the form of VOD. Cable companies, through the groundwork of digital cable are planning to deliver VOD technology to subscribers and technology experts feel adoption will occur quickly. The Yankee Group, a technology research firm, estimates by the end of 2002 about seven million homes around the nation will have access to VOD. This is an increase of 100 percent from the three million VOD households in 2001 (Schiesel, 2002). The Yankee Group also estimates VOD revenues for U.S. operators will be just under $2 billion by the end of 2005 (Iler, 2001). VOD has also received positive results from subscribers in current test markets. In Cleveland Ohio, Adelphia Cable currently has 50,000 customers able to receive VOD services. Of those customers, 35,000 have used the service. Other cable operators such as Comcast, AT&T, and Time Warner Cable have begun to launch or test VOD services hoping to use VOD as a competitive advantage to satellite and broadcast, and other competing technologies. This exploratory research has two goals. First, to use the constructs found in the diffusion of innovations literature to identify the early adopters of VOD. Diffusion of innovations theory has been widely used and accepted as a means to identify the early adopter variables within a social system and predict characteristics of early adopters. The

PAGE 13

4 second goal of this research is to measure the attitudes of consumers in regards to possible advertising scenarios within the VOD structure. Specifically, this exploratory research will explore these research questions: What types of consumers are currently aware of VOD technology? What factors indicate early adopters of VOD technology? What factors indicate intent to adopt VOD technology? What functions of VOD technology are found to be most preferred? What is the most positive prospective advertising model for VOD? This research has great significance for academics as well as practitioners. For academics, this research will add to the body of knowledge regarding diffusion of innovations theory as applied to digital technology. For the practitioner, identifying the potential VOD adopter is critical for positive product roll out and strategic planning. By identifying the traits and characteristics of potential early adopters, companies can target the early adopters of VOD more aggressively. AOL Time Warner Chief Executive Officer Gerald Levin sums up the significance of this research by stating, I believe all television will be distributed in either real time or on demand. It is the ultimate in consumer choice, convenience and control (Jordan, 2002.)

PAGE 14

CHAPTER 2 LITERATURE REVIEW The Video-On-Demand Background The digital cable environment is extremely competitive and VOD technology gives digital cable a distinct competitive advantage. Digital cable is dealing with competing pressures from digital satellite, analog cable and broadcast television. Currently only digital cable can deliver the VOD technology, giving digital cable a strategic point-of-difference competitive advantage. Although digital cable offers more channels than analog cable, the amount of channels has not been alluring enough to encourage many consumers to upgrade their service to digital (Brister, 2002). In addition, satellite is stealing many high-paying cable subscribers from digital cable since satellite can offer the same amount of programming. Digital cable marketers need to integrate the VOD application to add value and reduce consumer turnover (Schiesel, 2002). VOD technology can reduce churn or numbers associated with people dropping the digital service by creating a point-of-difference. Digital churn continues to be a problem for digital cable and the VOD competitive advantage will help companies keep or upgrade their current customers because the service is not available to satellite or analog systems (Stump, 2002). VOD technology offered by digital cable creates an incentive for analog users to make the switch and upgrade their current service (Cummings, 2002). A seven month VOD trial by Time Warner Cable in the Cincinnati market found 67% of the digital subscribers were more likely to retain their service because of free Scripps 5

PAGE 15

6 Network (DIY, Food Network, HGTV) on-demand content (Figler, 2002). Free-on-demand or FOD is being used as a digital churn reducer (Baumgartner, 2002). Also, FOD is being used to induce trial of VOD programming and technology, enabling the viewer to experience the added benefits of the service without a fee (Stump, 2002). The first major test of video-on-demand was an experiment conducted by Time Warner Cable in late 1994. Time Warner used Orlando Florida as a test market. The equipment was not advanced and was much more expensive than the equipment used today. In 1994, Time Warners cost per stream (per household) was around $13,000 with each set-top box costing over $5,000 (Schiesel, 2002). It was not economically feasible for cable companies to continue VOD service at this high cost rate. These costs are much lower today due to video compression, digital infrastructure, and overall improvements involving equipment. In order for VOD to work, cable operators spent millions of dollars updating their systems to provide two-way communication from the cable headend to the cable subscriber. The current cost per stream for VOD service is about $475 and the cost is estimated to fall below $300 by early 2004. The two-way communication from digital technology overcomes the barriers of analog one-way communication, which traditionally limited the cable subscriber to receiving to the same programming the operator was sending the other subscribers (Schiesel, 2002). The VOD functionality offers viewers an asynchronous environment for viewing video entertainment and information. Users of VOD technology can access programs at anytime and are able to stop, fast forward, and rewind programs much like the features found on a traditional VCR, DVD, or digital video recorder (DVR). VOD delivers

PAGE 16

7 exceptional entertainment quality comparable to the audio and video quality found on DVD and offered through digital cable. Selecting a VOD program is much easier than renting a DVD or setting the typical VCR to record. To select a VOD program, the viewer selects a program title from the electronic program guide with her or his cable remote control. By selecting the program, a request for the program is placed and sent to the cable system head-end. The request is then processed by the VOD server and recorded by the cable system. After the request is processed, the video is streamed from the storage library at the cable headend, to the network, and finally to the household. Although this sounds involved, all actions occur in real time and the viewer receives the video content moments after making the selection (Brister, 2001). Since VOD allows the television viewer new possibilities and functionalities when navigating through programming, the first research question will address the following: RQ1: What functions of VOD technology are found to be most preferred? Cable companies are currently offering VOD services in three varieties: free-on-demand (FOD), movies-on-demand (MOD), and subscriptions-on-demand (SVOD). Free-on-demand offerings can be viewed immediately without additional cost. Movies-on-demand involve recently released movies and resemble some of the same features of earlier pay-per-view and video store rentals. Subscription-on-demand allows viewers choose when they will watch programs from premium channels (HBO or Showtime) or SVOD offers additional content from the previously mentioned free-on-demand services (Cummings, 2002). Lastly, there has been some industry discussion to the idea of education-on-demand selections but these selections could fall in one of the previously

PAGE 17

8 mentioned categories, based on the cost or package offered to the consumer (Figler, 2002). Cable systems are test marketing the VOD product in a variety of ways. This research found Comcast Corp., Time Warner Cable, and Cablevision doing the most publicized tests, although others are involved. It is important to review these tests to strategically identify variables for developing potential advertising models. Comcast has done the most extensive testing published to date. Comcast is headquartered in Philadelphia and is a developer, manager, and operator of broadband cable networks providing basic cable, digital cable, and high-speed Internet service. Comcast is the third largest cable system in the U.S. and serves more than 8.5 million subscribers in six geographic regions (PR Newswire, 2002). Comcast is quickly rolling out VOD technology in the Philadelphia market where the company is in the process of building servers and infrastructure to offer 1,500 hours of on-demand programming. They hoped to have the VOD system ready to go by the end of 4 th quarter 2002 (Baumgartner, 2002). Comcasts VOD pricing strategy is meant to drive digital penetration in basic-only households through low-cost programming tiers. One tier entitled, on-demand classic will cost $9.95 a month and include content from several basic cable networks including a kids package of channels (Nickelodeon, Disney, and Cartoon Network), 15 basic networks, electronic program guide, digital music channels and discounts for movie-on-demand purchases. The second package entitled, On-demand Plus would sell for $14.95 a month and include the same as classic along with SVOD packages from HBO, Showtime, Starz Group LLC, as well as 40 basic channels. Currently, Comcast has some

PAGE 18

9 form of VOD up and running in 19 markets, covering 3 million homes and 600,000 digital subscribers (Stump, 2002). Time Warner Cable (TWC) has also announced a VOD roll out. By the end of 2002, TWC stated VOD would be available throughout New York City, TWCs largest cable market. TWC has 1.2 million subscribers in New York City and plans to have 1,300 hours of programming available at any one time. To put this amount of programming in perspective, 1,300 hours of programming is the equivalent of almost two months of watching television (Schiesel, 2002). Cablevision Systems Corp is adding value to its digital tier with free-on-demand content from the third party content provider Mag Rack along with VOD affiliations with Fox and local PBS affiliates (Baungartner, 2002). Viewing and usage data for VOD was difficult to locate since most reports are proprietary or require a large amount of money to view. At the time of this research, some data was available regarding the viewing habits of Starz VOD subscribers. According to Starz vice president Greg DePrez, early research shows there is as much usage of VOD during the day as in primetime. VOD usage tends to peak around 3pm, presumably when the children return from school and peak again at 8pm. VOD activity for Starz tends to be heavier on the weekends compared to weekdays. These weekend and weekday viewing findings differ from the habits found in the typical television viewer (Stump, 2002a). Content providers have a variety of reasons for getting involved with VOD service. VOD is the ideal platform to increase brand awareness for cable networks and programming distribution (Figler, 2002). Many television-programming networks have

PAGE 19

10 prepackaged content ready for delivery via VOD while other companies will strive to make content specifically tailored for VOD service. Strong cable brands will help drive VOD usage and cable operators must form strategic relationships with these brands (Stump, 2002). While VOD is a programming service not subject to the time restraints of the typical programming model, cable operators and content providers must remain sensitive to basic cable viewing patterns (Stump, 2002). The typical VOD user will not use VOD all the time. From current industry research, content providers fall into three categories: premium content providers, basic cable content providers and niche network content providers. Each offers different content for the VOD service and each will fulfill a different viewer need. The premium content provider is made up of traditional premium cable brands such as HBO, Showtime Inc., Starz, and the Movie Channel as well as adult entertainment providers such as Erotic Networks and Playboy. Both HBO and Showtime currently offer VOD content in various test markets. HBO offers a subscription-on-demand service and has tested fees of $3.95, $6.95 and $9.95 for the service. For HBO, the service charge is in addition to the typical fee charged to the viewer for having access to HBO. These prices are currently being tested on Time Warner Cable systems and the prices vary based on the market demand (Stump, 2002a). Showtime expects to have more than 35 launches of VOD by the end of 2002. The launch agreements are with several cable companies including Cox Communications, Charter Communications, Time Warner Cable, Comcast, Cablevision, Adelphia and

PAGE 20

11 Insight Communications. Showtime on Demand will carry about 120 hours of programming per month consisting of 45 percent series programs, 20 percent family programs, and 15 percent adult programs (Stump, 2002a). Adult programming has always been a staple of the pay-per-view offerings found on a cable system. On average, adult programming nets cable systems at least 60 cents per subscriber compared to the 15 cents per subscriber generated by new release movies. Erotic Networks is a leader in adult on-demand content and is now carried on five of the top ten cable systems. Erotic Networks plans to expand the networks reach to about 4 million homes and the network plans to offer about 60 hours of programming per month (Figler, 2002). In competition with Erotic Networks is the Playboy Network. Playboy currently offers about 70 hours of on-demand content per month. The Playboy Network currently reaches about 60 percent of the VOD enabled homes nationwide (Figler, 2002). It would be much harder to find a cable network not planning on offering some form of VOD content. Cable networks already providing VOD content include but are not limited to A&E, AMC, BBC, CNN, Court TV, Discovery, DIY, Erotic Networks, ESPN, Food Network, Fox Sports, FX, Golf Channel, Hallmark, HGTV, History, IFC, MTV Networks, National Geographic, NBC, Outdoor Life Network, Oxygen, Speed, Sundance, TBN, TBS, TMC, WE, and the Weather Channel (Figler, 2002). Some basic cable content providers are giving content away to help induce VOD usage while other companies are looking for a monthly fee. For example, the Scripps Network, including cable brands HGTV, Food Network, DIY and Fine Living has agreed with Time Warner Cable to provide 10 hours per month of free VOD content. The deal

PAGE 21

12 was meant to introduce customers to VOD (Figler, 2002). Free access to this VOD programming will induce trial, help build loyalty for the Scripps brands and should benefit both parties. ESPN plans to license 150 hours of library programming and replays of some college football and college basketball games to VOD distributor TVN Entertainment Corp. However, not all ESPN content will be free and the network is still working on how it will provide content to distributors (Romaso, 2002). The Discovery Network has publicly discussed how it will distribute its large amount of content and Discovery could become the VOD model for other popular cable brands. Discovery is offering two tiers of programming. The first tier is a free service called Discovery Choice 10. Choice 10 will offer a menu of ten different programming categories and will be advertiser supported. The Discovery Choice 10 package will have roughly 25 hours of programming and will be refreshed on a monthly basis (Baumgartner, 2002). Discoverys other service entitled, Discovery On Demand, could have up to 500 titles pulled from Discoverys 70,000 hour library. Discovery On Demand would be a subscription service although pricing has yet to be publicly discussed (Figler, 2002). VOD offers smaller cable networks and content providers, lacking the programming to fuel an entire network, the opportunity to build their brand due to the localized nature of the VOD distribution networks. Because VOD is a requested programming format, niche providers can tailor content to fulfill subscriber needs. Three examples of niche networks currently providing content to cable systems are Tech TV, Chaos Media Networks, and Mag Rack.

PAGE 22

13 Tech TV is an upstart cable network with repurposed programming especially tailored for the VOD audience. Tech TV plans to create instructional videos that consumers can access at anytime. A Tech TV program may teach the viewer how to set up a home computer network or what type of digital camera to buy (Figler, 2002). Another niche content provider is Chaos Media. Chaos Media focuses on non-movie genres such as education, fitness, household help and health. Chaos Media will offer custom exercise and diet plans as well as custom tracking for the plans. The Academy Channel, through Chaos Media, offers training and certification courses (Figler, 2002). The niche-programming provider receiving the most press is Mag Rack. Mag rack claims to be offering content in categories underserved in the cable marketplace. Mag Racks programming is based on relationships with over 30 different magazines. Programs are specific to special topics such as, Classic Cars, Birdsight, and Cooking with the Pros. Since the start of the company, Mag Rack has cost over $50 million to develop and the programmer uses original and acquired programming. Mag Rack charges a license fee per each VOD subscriber, much like traditional cable networks. Currently, Cablevision has allocated about 150 hours of storage capacity to Mag Rack content alone (Baumgartner, 2002). This research has discussed cable television providers, cable programming providers, and niche content providers but has yet to discuss the third party intermediary connecting all three. If deals are not made between the content provider and the cable system, a third party vendor will undoubtedly provide content and VOD system support.

PAGE 23

14 Third party vendors such as In Demand or TVN will bridge technology and content for the VOD subscriber. In Demand is the worlds largest provider of VOD and pay-per-view programming. In Demand works with approximately 1,900 affiliated systems and provides programming for the NBA, NHL, MLB, and NASCAR in addition to first-run movies and other professional sporting events ( www.indemand .com/about/who ). Shareholders for In Demand include AT&T, Time Warner Entertainment, Comcast and Cox. In Demand currently has 2.5 million subscribers and expects to double the number by the end of 2002. In Demand also has agreements with content programmers such as Court TV, ESPN, Fox, Hallmark, Sesame Workshop and Turner. In addition to sports and cable programming deals, In Demand has VOD distribution deals for movies. Movie producers holding agreements with In Demand include Dream-Works, MGM, Sony Columbia TriStar, Twentieth Century Fox and Universal. Currently, In Demand has access to 75 percent of Hollywoods recent releases in addition to the vast Hollywood library titles provide by the movie companies (Figler, 2002). TVN is another third party VOD vendor. TVN currently serves over 600 cable systems reaching over 50 million U.S. households and offers a comprehensive suite of pay-per-view and VOD programming along with equipment to encode, archive, transport and manage video. TVN was selected by Comcast Cable to support Comcasts VOD initiative and will support over 750 hours of VOD content monthly (PR Newswire, 2002). TVN has major licensing deals with content providers such as Artisan, MGM, RCN, and Sesame Workshop (Figler, 2002).

PAGE 24

15 Although widespread adoption of VOD could lead to many industry changes and issues, problems involving bandwidth, content allocation and advertising present the most current industry challenges. VOD will put much more demand and stress on the cable infrastructure than found in the current environment. Storage, managing, and ensuring systems run smoothly will become more difficult when demand increases the data moving from various servers and networks (Baumgartner, 2002). The current cable infrastructure is only wired to handle 10 percent of the households requesting VOD at any given time and this could be a potential stumbling block for operators if people adopt the VOD structure quicker than expected (Schiesel, 2002). Further, this usage overload could cause lock-ups in user homes, making the system seem unreliable and problematic (Cummings, 2002). Although these problems can be dealt with by increasing the server capacity and improving the general infrastructure, both improvements involve increased spending in time and resources (Schiesel, 2002). If these problems are not solved quickly, digital churn could occur, reducing the competitive advantage offered by VOD. The changing of content could be another challenging task for cable systems (Baumgartner, 2002). VOD content would need to be changed on a regular basis and if the system has multiple content vendors, the task could become confusing and complicated. Storage for all the content could also be a challenge. With VOD requiring so much content, the impact of delivery could be felt by the satellite carriers delivering the content to the cable system headend (Brown, 2002). Satellites are still the most economic way to deliver programming content to cable networks and current satellite bandwidth is limited.

PAGE 25

16 Last, advertisers are concerned about advertising in the VOD environment. The VCR and digital video recorder enabled homes to erase or fast forward commercials. Content providers need to keep advertisers in mind since they provide the primary funding for programming. Software can be installed on VOD technology platforms to ensure consumers cannot skip through commercials. This non-skip software is embedded in the video from the VOD cable server (Stump, 2002). Finally, this advantage could be a potential positive for advertisers looking to avoid the personal video recorders now being highly publicized by the satellite television providers and those advertisers looking to target specific niche markets. Even with these drawbacks, advertisers are still testing advertising possibilities. For example, Channing Dawson from the Scripts Network is testing the possibility of adding a billboard ad at the beginning of each show to tease a product along with placing typical spots throughout the program. Dawson also notes since VOD is a performance based medium, gathering the personal information of the viewer and the viewers habits will be essential (Stump, 2003). Nissan has been very proactive attempting to use VOD and interactive television advertising to reach and interact with a new audience. The ads are designed to be marrying content and new media in a new format. Additionally, Nissan claims to be the first car manufacturer to use this platform (Precision Marketing, 2004). The Direct Marketing Association recently conducted a study measuring attitudes among marketers regarding interactive television functions. The study found 40% of the respondents considering this new platform an important option for the marketing mix (Duncan, 2004).

PAGE 26

17 Advertisers spent over $54.5 billion on U.S. television advertising in 2003, more than any other medium. With the technology changing this one time stable environment, new advertising models and formats need to be developed. Consumers will not be interested in spending the estimated $250 a year for commercial-free television (Angwin et al, 2004). Funding content and providing a vehicle to reach potential consumers will be essential, leading to the next research question: RQ2: What is the most positive prospective advertising model for VOD? This industry background on the current status of VOD serves to illustrate the challenges and needs of this emerging industry. Cable systems, cable-programming providers, content providers and third party vendors are all spending millions of dollars on a technology still not widely adopted or even known to exist by the general public. The goal of this research is to recognize the characteristics of early adopters of VOD and identify their attitudes toward certain types of advertising. Diffusion of innovations is the perfect theory to help identify the key variables of those willing to adopt VOD technology and advertising due to the theorys long history and theoretical grounding. Theoretical Grounding Diffusion is the process by which an innovation is communicated through certain channels over time among the members of a social system (Rogers, 1995 p.5). Diffusion of innovations involves new ideas (Rogers, 1995). Much like information flow theory, diffusion of innovations is source dominated with the elite in a social system establishing the key point of view when deciding to diffuse an innovation. Everett Rogers has combined many of the elements found in original information flow research with personal influence and information found in the fields of anthropology, sociology and

PAGE 27

18 rural agricultural work (Baron and Davis, 2003). Because this theory is useful when investigating the flow of information within social units, and identifying and predicting the adoption characteristics of particular social groups, diffusion of innovations has been chosen for this VOD adoption research. Diffusion of Innovations Key variables and constructs for diffusion reside in early German and British schools of thought but considerable interest, research, and diffusion literature development did not begin until the mid-1960s. A paradigm in diffusion literature was developed in the mid-1960s due to the increased need for consumer behavior literature (Rogers, 1976). Rogers (1976) states the origins of diffusion research trace from the German-Austrian and British schools of diffusionism in anthropology and a French sociologist named Gabriel Tarde. Tarde (1903) pioneered the S-shaped diffusion curve found consist in diffusion research. Lastly, Rogers (1976) states the revolutionary paradigm for diffusion research came in the 1940s when Bryce Ryan and Neal Gross (1943) produced the most widely known diffusion research, investigating the diffusion of hybrid seed corn in two Iowa communities. Many of the beginning concepts involving diffusion of innovations can be found in the book written by Gabriel Tarde entitled; The Laws Of Imitation published in 1903. Jean-Gabriel Tarde was born in Sarlat, France in the year 1843. Tarde was a sociologist and directed his attention to social behavior and more specifically, to how people pass feelings and thoughts from group to group or from person to person (Williams). Tarde developed the theory of imitation, a 19 th century social learning theory and is known as one of the modern day learning theorists (Williams). In his book, The Laws Of Imitation, many of the same theoretical concepts matching diffusion are discussed such

PAGE 28

19 as universal repetition, social resemblance, society and the inadequacy of the economic and juristic conception, and logical influences (Tarde, 1904). Finally, Tarde proposed the S-shaped diffusion curve, or the statistically plotted rate of adoption within a social system and discussed the role of opinion leaders in the spreading of new ideas (Rogers, 1976). A study conducted by Ryan and Gross set forth a new approach in communication research and scholars have used their study involving the adoption of hybrid corn as a model for researching communication adoption (Rogers, 1976). Ryan and Gross (1943) selected the topic of hybrid corn because the diffusion of hybrid corn among Iowa farmers made it attractive. The study was attractive to the authors due the traditional conservative nature of the typical farmer and the recall of adoption was well within the memory span of the current farmers at the time. Data from the Ryan and Gross (1943) study was collected by personal interview within two Iowa farming communities. From the study, the authors found the typical farmer first became aware of the hybrid seed corn from salespeople (49%), but their neighbors ended up being the most persuasive source leading to adoption. Consistent with established diffusion research, the first farmers to adopt were of a higher socioeconomic status. The entire study lasted from 1934-1941 with a total sample of 259 interviewees. Of the 259 cases, 2 never accepted the change in corn and only three accepted in 1941 with all others in the sample previously adopting the new corn. The rate of adoption followed the stereotypical S-shaped diffusion curve, much like Tardes claim and consistent with the diffusion research of today. This study is known as the classic example of diffusion of innovation research (Rogers, 1976).

PAGE 29

20 The hybrid corn study set forth a new approach to communication research and an increasing number of scholars in all fields quickly took note of the study. Diffusion research began to emerge as a single body of constructs based on communication and human behavior. It was not until the 1960s that mass communication scholars began to investigate communication under these specific constructs, first applying them to news stories. While mass communication scholars were investigating the diffusion of news, marketers and consumer behavior researchers began to use the theory for understanding the consumer adoption process of new products. Marketing and consumer research became a strong force in the 1960s and diffusion of innovations seemed like a perfect fit for understanding how a company could launch a new product more efficiently. The theoretical and methodological background of diffusion research appeared to be the perfect basis for targeting and predicting consumer adoption for marketers (Rogers, 1976). Among those looking to use diffusion of innovations in the marketing environment was Frank M. Bass. Bass (1969) published a research paper entitled, A New Product Growth for Model Consumer Durables, and his research made a significant contribution to diffusion of innovations theory by constructing a mathematical model for new product growth in consumer durables. The model developed by Bass is a growth model for estimating or predicting the timing of adoption. Bass model is set within the theoretical framework of Rogers literature and has become a driving force in diffusion research in the field of consumer behavior and marketing. Bass states the probability of purchase at any time has a linear relationship to the number of previous buyers of a product. Product adoption will grow exponentially and then ultimately see exponential decay over time.

PAGE 30

21 Bass found the consumer durables tested in his model were in agreement with his proposed model and states his model works well for understanding the process of new product adoption and long-range forecasting. Bass states there are two main influences persuading potential consumers, internal factors such as word of mouth or interpersonal communication and external factors such as the mass media (Mahajan and Muller, 1990). Bass model is still used today and continues to be a mathematical framework for adoption. Much of the information in this research involving the classic diffusion model and was taken from the Everett M. Rogers (1995) book entitled, Diffusion of Innovations, Fourth Edition. Although some research within this section has been credited to other sources, Rogers is the most definitive source on diffusion of innovations theory and his book generalizes thousands of previous research articles. The Rogers (1995) text provides the theoretical definitions found in this research involving the classification of adopters, characteristics of adopters and consequences of innovations. Rogers theoretical and methodological background will be used for identifying early adopters of VOD technology. The classic diffusion model (Rogers, 1995) consists of the innovation or new idea, the communication channels the new idea is delivered through, the time it takes for the innovation to be adopted or rejected, and the social system the idea is communicated within. The first construct in the classic diffusion model is the idea or innovation. Rogers (1995) states an innovation can be a new idea, practice or object for adoption and not all innovations are equivalent. Because innovations are all different, they will posses

PAGE 31

22 different rates of adoption. Some innovations present a relative advantage to an existing idea and the innovation becomes perceived as better. Social prestige, convenience, and increased satisfaction can all create a relative advantage. Compatibility is another element of the innovation examining to the degree the innovation is consistent with existing values, past experiences and the needs of potential adopters. Other notable factors of an innovation include complexity or how difficult the new idea is to understand, trialability or the ability to test the idea, and observability or the ability for the results to be seen by others. The communication channel is the second construct in Rogers (1995) diffusion model. The way a new idea passes between individuals or within social groups is through the communication channel. Participants within social groups share and create information with one another to reach a mutual understanding. This communication should not be thought of as linear act but as a two-way process of information convergence. Diffusion research has generalized that most people do not evaluate an innovation based on scientific studies but rely more on subjective evaluation conveyed from other individuals, like themselves, who have previously adopted the innovation. The diffusion literature also states more successful diffusion communication occurs when individuals or groups have similar attributes, beliefs, education, or social status. This communication can move individuals closer or further away from the innovation through the message delivered by the change agent. Rogers (1995) states time is the third element impacting the diffusion process. The time dimension involves the information-decision process, the innovativeness of the individual and the rate of innovation adoption.

PAGE 32

23 Rogers states (1995) the rate to which an individual or group passes from first knowledge, to attitude formation, to the adoption or rejection decision, and finally to confirm the idea encompasses the innovation-decision process. The process had five main steps; knowledge, persuasion, decision, implementation and confirmation. When an individual first learns of an innovation they gain knowledge. Persuasion involves the favorable of unfavorable opinion toward the idea. Based on persuasion, the individual will make a decision to adopt or reject the new idea. Implementation activities are related to adopting or rejecting the new idea and confirmation occurs when the individual seeks reinforcement. It is important to note this time step can not only lead to adoption or rejection but also the discontinuance of an innovation. Discontinuance occurs when the individual becomes dissatisfied with an innovation or when the innovation is replaced. Rogers (1995) states the degree of innovativeness of an individual or group is the second part of the timing aspect. The innovativeness of an individual reflects the earliness or lateness the individual adopts an innovation compared to other members of the social system. Research indicates individuals within adopter categories have more in common with each other then with members of other adopter categories. The third segment of the time aspect illustrated by Rogers (1995) involves the total rate of adoption. The relative speed to which a new idea is adopted is the rate of adoption. This rate of adoption typically forms an S-shaped curve when statistically plotted. The slope steepness of the curve is set by how quickly or slowly an innovation is ultimately adopted within a social system. The fourth aspect to the classic diffusion model involves the social system surrounding the innovation. Rogers (1995) defines the social system as interrelated parts

PAGE 33

24 involved in joint problem solving to finding a common goal. Social systems consist of groups, individuals, and organizations. The structure, norms and change agents of a social system impact diffusion. Within each social system exists hierarchical positions of influence, with higher hierarchical positions leading the lower through various means. The communication structure within the socials system sets the pattern for communication flow and the structure of a social system can facilitate or impede the diffusion process. Rogers (1995) states the most innovative member of a social system is very often looked upon as a deviant from the system and may play a limited role in the diffusion process due to credibility. On the other hand, according to Rogers (1995) opinion leaders and change agents have the ability to increase the rate of the entire adoption process. Opinion leaders are able to influence individual attitudes and behavior informally to increase diffusion. Opinion leaders are exposed to many forms of communication within a social structure, have higher social status, and are typically more innovative. Opinion leaders are often directly influenced and learn of innovations from individuals known as change agents. Change agents are individuals who attempt to influence innovative decisions to their desired conclusion. They are often professional or expert members of a social system and are able to influence opinion leaders in leading diffusion campaigns. These four basic aspects exemplify how the adoption process moves through a social system and what variables influence adoption. Mass adoption of VOD technologies has yet to take place. By understanding the communication of an innovation, the time and elements of adoption and the social aspects of a system, marketers can get a perspective understanding as to how the information regarding consumer VOD technology will flow

PAGE 34

25 through our social system. Although marketers are interested in the potential flow, knowing the characteristics of first adopters will ultimately set the model in motion. Ryan and Gross (1943) classified segments of Iowa farmers in relation to the amount of time it took them to adopt hybrid corn seed. These adoption types fit into Rogers (1995) ideal adoption types. Although some individuals are exceptions to the adopter types, these generalizations offer parameters for variables when identifying potential VOD adopters. Rogers (1995) identifies innovators as the first adoption group. On average, innovators represent 2.5 percent of a social system. Venturesomeness is an important trait for the innovators. Innovators typically deal with a high degree of uncertainty and idea complexity. They are prone to experience and accept occasional setbacks. Innovators are often the gatekeepers within a system because they are first to adopt and bring an innovation into the system (Rogers, 1995). Acceptance of an innovation by innovators and by the early adopters can set the stage for the ultimate success of a product (Danko and MacLachlan, 1983). Rogers (1995) states early adopters represent 13.5 percent of the average social system and they are more integrated in the social system than the innovators. The early adopters hold the greatest degree of opinion leadership and this category is typically sought by change agents due to their impact in speeding up the total adoption process. Early adopters can also serve as role models for a social system since they possess many similar traits of the other members in a social system. Innovation moves quickly through the early adopter group because adoption by this group reduces uncertainty for the entire social group conveying a subjective positive evaluation to near-peers through

PAGE 35

26 interpersonal networks (Rogers, 1995). Quick acceptance by early adopters usually equals rapid market growth and high profits for innovative products (Danko and MacLachlan, 1983). Rogers (1995) finds the average early majority group represents 34 percent of the individuals within a social system. This group adopts new ideas just before the average member of the system. This group is vital to the adoption process since they provide interconnectedness in the systems interpersonal communication networks. They are followers and adopt innovations but seldom lead. The late majority represents 34 percent of a social system and Rogers (1995) states this group adopts ideas just after the average member in a social system decides to adopt. For the late majority members, innovations are approached with a skeptical and cautious nature and late majority members wait for the majority of members of a social system to adopt prior to adopting. The weight from system social norms and pressure from peers is necessary to motivate adoption. This group typically has scarce resources and needs the uncertainty of an innovation removed so their limited resources are not wasted (Rogers, 1995). Pressures from previous adopters aid in the timing of the adoption process for this group (Bass, 1969). Laggards, representing 16 percent of a social system are the final group of adopters identified by Rogers (1995). Laggards are the last group to adopt an innovation. They are typically isolated within the social system and possess no opinion influence or leadership. Laggards always look for information based on the past and they must be certain a new idea will not fail before they adopt. They are typically of lower economic class and education.

PAGE 36

27 Realizing the strength of each adoption group is important when attempting to spread a new innovation through a social system. Although the idea of an innovation for a social system is presumed to be positive, innovations can have consequences. Rogers (1995) states innovations can cause certain changes within a social system and these changes could impact a system in a variety of ways. Most presumably, these changes occur due to the adoption or rejection of an innovation. Some consequences of innovations are desirable while others can be undesirable. These consequences can occur if the innovation is functional for the intended purpose or possibly dysfunctional or used for unintended purposes. Innovations can also have direct or indirect consequences. Direct or indirect consequences can occur immediately in response to the adoption of the innovation or can be an unforeseen result of the innovation adoption. Lastly, innovations can have anticipated and unanticipated results. These consequences vary based on the intent of the innovation and sometimes the unanticipated results can be detrimental. Rogers (1995) states there are many criticisms of diffusion research. He found diffusion research often has a pro-innovation bias. This bias is the idea the innovation being studied should be diffused and adopted by the social system. This bias can be illustrated in a tremendous amount of diffusion research because the change agent or group interested in the innovation is the client for the study and has a valid interest in the innovations diffusion. These agencies often spend large amounts of money on studies and do not wish to see their money wasted on a poor study.

PAGE 37

28 Another criticism involves recall. Recall is heavily relied upon in diffusion research and is a methodological enemy of diffusion research. Recall is often clouded by time from when the adoption of an innovation actually occurred and when the recall for a study occurs, thus reducing methodological accuracy. Time is another enemy of adoption recall since many studies are not longitudinal but are just snapshots of the adoption process. Often the authors of research want to find the current status of an innovation adoption process and overlook the longitudinal aspect. Lastly, reliance on correlational analysis of survey data can often lead to a false sense of causality. If the authors of a study refrain from properly operationalizing their variables, the analysis becomes meaningless. Further, intervening variables are always involved in survey research and these variables must be accounted for in the final adoption analysis. Proving exact causality is difficult and is a negative for all types of quantitative survey research. Rogers (1995) lists distinct categories involving the variables related to innovativeness. Early adopters of innovations are typically found to have similar traits in socioeconomic status, personality values, and communication behavior that distinguish them from other members of a social system. Acknowledging and operationalizing these traits is key for recognizing early adopters of VOD technologies. These traits will be operationized by the author for the survey instrument to measure early adopters of VOD technologies. Taken directly from Rogers 1995 publication, the three categories and traits are: Socioeconomic Characteristics (Rogers 1995, p. 269): Earlier Adopters are not different from later adopters in age

PAGE 38

29 Earlier adopters have more years of formal education Earlier adopters are more likely to be literate Earlier adopters have higher social status Earlier adopters have a greater degree of upward social mobility Earlier adopters have higher social status Personality Variables (Rogers 1995, p. 272-273): Earlier adopters have greater empathy Earlier adopters may be less dogmatic Earlier adopters have a greater ability to deal with abstractions Earlier adopters have a greater rationality Earlier adopters have a more favorable attitude toward science Earlier adopters have a more favorable attitude toward change Earlier adopters are better able to cope with uncertainty and risk Earlier adopters are less fatalistic Earlier adopters have higher aspirations Communication Behavior (Rogers 1995, p. 273-274): Earlier adopters have more social participation Earlier adopters are more highly interconnected through interpersonal networks Earlier adopters are more cosmopolite Earlier adopters have more change agent contact Earlier adopters have greater exposure to mass media communication Earlier adopters have greater exposure to interpersonal communication Earlier adopters seek information on innovations more actively Earlier adopters have greater information on new innovations Earlier adopters have a higher degree of opinion leadership The review of relevant diffusion literature contained in this research proposal is not exhaustive. Diffusion studies have produced a body of literature impossible to thoroughly cover in one research study. The adoption of new digital technology leads the pertinent criteria for the diffusion research selections included in this literature review. William Danko and James MacLachlan (1983) produced empirical research examining the specifics and indicators of individuals adopting personal computers. The implications of their study found differences in media choice, advertising message

PAGE 39

30 content, and distribution for the early adopting group when compared to the late adopters. As for the studys findings, the authors state most early adopters of computers were highly educated males and suggested the advertising messages be tailored to a male audience. Danko and MacLachlan (1983) made a few advertising suggestions for targeting early adopters of computers. The authors suggested messages delivered in cerebral vehicles such as Psychology Today and Scientific American or by direct mail would be efficient since time spent with television was minimal for this group. Lastly, the authors suggest most early adopters will buy from mail-order type services since they are more adventuresome and less reliant on support. Although this article is dated and had a main goal of determining an appropriate advertising positioning strategy, it does illustrate some of the qualities of the early adopter category and the diffusion of personal computers. Digital satellite programming provider Directv used Frank Bass model to help predict subscriber rates for the satellite provider prior to launch. Bass et als (2001) research project entitled, DIRECTV: Forecasting Diffusion of a New Technology Prior to Product Launch applied Bass 1969 model to predict diffusion for the new satellite provider. Directv is a subsidiary of the Hughes Corporation under General Motors. At the time the study began, the company was developing technology to provide programming to subscribers via satellite and compete with cable television. The study had three distinct research questions: deciding what pricing and programming to offer, who would be the first to buy, and how many would be the first to buy? After a review of research, Directv launched a concept test with the main purpose of obtaining data for forecasting Directv market diffusion over the next 10 years. On the basis of the data collected from the

PAGE 40

31 concept test, Frank Bass used his model (1969) to forecast the number of U.S. subscribers for Directv and when they would subscribe. Because the study was longitudinal, the authors compared the actual diffusion of Directv to the original Bass prediction. The findings illustrated Bass model was successful in forecasting the diffusion of Directv for the first five-year period from 1994 to 1999. Studies involving Internet adoption are pertinent to this study due to the newness and innovativeness found in Internet technology. A study produced by Atkins and Jeffres (1998) profiled Internet adopters in terms of social locators, media use habits, and their orientation toward adopting new technologies. Their findings supported some of the early adopter notions of demographics and technology uses found in diffusion theory. The study examined demographic as well as technology needs. The authors found communication needs were a stronger predictor than demographic information alone for those adopting the Internet. Atkins and Jeffres (1998) also feel further diffusion research should refine measures of cosmopoliteness and localiteness. Still, their findings stated Internet adopters are typically young and educated. They also felt the Internet was still in the early stages of adoption. Lastly, technology orientation or the need for innovativeness was one of the points of difference when comparing adopters to non-adopters. Since digital cable is the delivery method for VOD, research involving digital cable adoption adds valuable insight to the current study. Myung-Hyun Kang (2002) produced research exploring the factors associated with the early adoption of digital cable. The purpose of the study was to understand and predict digital cable adoption by identifying a profile of early digital subscribers along with implications for industry structure and marketing. Kangs study looked at early adopters and adoptive innovativeness as

PAGE 41

32 dependent variables and demographics, media use, technology ownership, innovative attitudes, and satisfaction as independent variables. The study found most hypotheses from diffusion theory were supported with a few significant exceptions. Kang found income was not related to digital cable subscribership. The author believes the reason for this could be found in the increased cost of upgrading to digital service. The study did find high consumption of television did relate to digital cable adoption and those who currently subscribe to premium channels would be better targets for digital cable. Lastly, digital cable companies who appear innovative and form strong positive relationships with consumers will be adopted more quickly. The next technological step from VCRs is the current growth of digital video recorder market. Douglas Ferguson and Elizabeth Perse (2001) conducted a study entitled, Enhanced Television Viewing with Digital Video Recorders (DVRs): Audience satisfaction in an Asynchronous Television Environment. This article is quite pertinent to the current study since VOD services also provide viewing in the asynchronous environment. From their exploratory findings, DVR owners are clearly early adopters and enjoy the uses and benefits from the technology. Timeshifting was linked to viewing satisfaction, with having the ability to record programs airing at inconvenient times as a most valued feature. The menu driven program selecting and recording feature was also found to be a valued feature. Lastly, DVR owners found the machines much easier to operate than VCRs and the functions of the DVR allowed users easier operation regarding storage and playback. Dupagne (1999) explored characteristics of potential high-definition television adopters based on the diffusion of innovation theory. The purpose of this study was to

PAGE 42

33 assess consumer predispositions toward high-definition television and profile potential adopters based on demographic characteristics, mass media use, ownership of home entertainment products and importance of high-definition television attributes. Awareness, interest and purchase intention were the dependent variable measured. Based on this research, most profile characteristics were consistent with previous diffusion studies. Innovators and early adopters of high-definition television were more likely to be younger and have higher incomes. The early adopters were also more interested in the attributes found in this new technology. Demographics and importance of television attributes were found to be stronger predictors of high-definition television compared mass media use tendencies or ownership of other technology. Lin (1998) examined the adoption rate and adopter types of personal computer consumers. Her study looked at media use patterns, use of other existing communication technology ownership in addition to demographics and other social characteristics such as the need for innovativeness. Lin has made a significant contribution to diffusion research by adding the need for innovativeness as an important construct to measure and include in diffusion research. According to Lin, the need for innovativeness characteristic is an important contributor to the overall adoption profile of a consumer and should be considered. In her personal computer study, Lin (1998) found the need for innovativeness characteristic was the strongest predictor among adopters, followed by likely-adopters and non-adopters of computers. Lin found likely adopters were those who were oriented toward the need for innovativeness but lack the financial resources to participate in the

PAGE 43

34 early adopter phase. Media use, education and gender were not found to be a predictive but age and income were found to be representative to the past diffusion studies. In another diffusion technology study, Lin and Jeffres (1998) explored audience intentions to experiment with or adopt multimedia video technologies. These technologies would allow a consumer to use voice, data, video channels and communication through a single interactive coaxial cable system. They measured interest in experimenting with new technologies and interest in adopting new multi-channel cable services. For predictive variables, they measured demographics, media use, and satisfaction with media content. Need for innovativeness was also measured. The study found demographic variables were largely irrelevant toward interest in experimenting with the new system. Lower satisfaction with current television content and lighter viewing levels were found to be significant predictors of adoption. Lin and Jeffres state existing media use patterns, along with media content satisfaction, could be strong variables to help determine functional substitutions between an existing medium and an emerging medium. They also state personality traits such as innovativeness could be instrumental to examine differing levels of adoption and interest. Atkin, Jeffres and Neuendorf (1998) examined Internet adoption as telecommunication behavior. The intent of their study was to profile Internet adopters in terms of social locators, media use habits, communication needs and their orientation toward adopting new technologies. The profile was built off of measuring variables such as demographics, communication needs, communication activities, technology relations and media consumption behavior. Their study found some support for the early adopter profiles derived from diffusion theory, such as early adopters being young and educated.

PAGE 44

35 This study also found innovativeness to be an important predictor for the adoption of innovations. Results from this study failed to confirm the expectation that attitudinal variables, those addressing communication needs, are more explanatory than demographics. Neuendorf and Atkin (1998) assessed the use of telephone-based audio innovations and profiled audience users and utilities for audio text services (1-900 services) and fax services. They examined use of these services by measuring demographics, quality of life, social communication activity, and media use. Consistent with diffusion literature, typical heavy fax and audio info users were found to be younger. Inconsistent with diffusion literature, demographics and other traditional social indicators are not uniformly important in the prediction of innovation use. The authors state the modest role played by demographics reinforces past findings suggesting that their explanatory influence has weakened over time. Eastlick (1996) examined consumer intention to adopt interactive teleshopping. The purpose of this study was to identify personal innovativeness characteristics, shopping attitudes, and product use patterns that contribute to intention to engage in interactive teleshopping. The study also looked to examine attitudes toward the innovation characteristics and examine relationships among the variables. Attitude toward interactive teleshopping was measured as well as the intent to adopt teleshopping. The author looked to explain these dependent variables by measuring demographics, shopping orientations, perceived uncertainty, and personal innovativeness. The results from the study show that overall opinions of interactive teleshoppings innovation attributes as well as several priority acquisition patterns contributed most to predicting

PAGE 45

36 subjects attitudes toward interactive teleshopping. Demographics were not predictive but relative advantage, ease of use, trialability and observability were consistent with classic diffusion research. It is important to note some of the studies included in this literature review are not diffusion articles. Even though they are not diffusion research, they are still important to address due to the added relevance or insight they contribute to the current study. Video fragmentation is another area of interest for this proposal. A study done by Carolyn Lin (1994) is very important to include even though Lins study did not use diffusion of innovation as theoretical grounding. Lins study entitled; Audience fragmentation in a Competitive Video Marketplace examined the competitive nature of the fragmented video marketplace. This fragmented marketplace presents challenges and opportunities for cable companies and advertisers and the findings and variables are important for measuring the adoption of VOD services. In Lins (1994) study, strong differences were found in behavioral, environmental and motivational indicators when comparing premium-cable homes to non-cable homes. The study found cable homes are typically heavier and more satisfied viewers. Premium households were found to have the most viewing satisfaction, followed by basic cable and concluding with non-cable households. The study also found high channel switching in premium and basic cable homes and low switching in non-cable homes. Cable homes were found to have more sophisticated viewers than non-cable homes and cable homes were more likely than non-cable to be engaged in both viewing-decision reevaluation or multiple channel viewing during commercial breaks. Lin states this channel switching suggests cable homes are much more active when making a viewing selection.

PAGE 46

37 In addition, Lin found viewing exposure and planning were important measured concepts in this research. Weekday and weekend viewing was found to be similar for non-cable and basic cable homes. However, homes subscribing to premium cable viewed 30 to 45 more minutes per day. Program preplanning was higher for those households subscribing to cable than those non-cable households. As previously mentioned, premium households were more likely to view multiple programs followed by basic then non-cable households. Last, demographic data collected from this study states premium cable homes reside in the youngest age group and wealthiest group. Roughly half of the basic cable subscribers were middle age and 61 percent of non-cable households were in the lowest economic group. Non-cable homes were found to have a lower level of education while over half (62%) of the basic cable subscribers had completed college. All factors defined by Lin are particularly important to note for the current proposed VOD research. Another study particularly relevant to this proposal but not based on the theory of diffusion of innovations was a study done by Randy Jacobs (1995) entitled, Exploring the Determinants of Cable Television Subscriber Satisfaction. In this study, Jacobs focused on the factors associated with subscriber satisfaction. Jacobs suggests satisfaction is directly linked with disconnection or subscriber upgrading/downgrading behaviors. Subscriber evaluations of cable system performance are important to the development of overall subscriber satisfaction. Programming variety and quality were also found to be important for overall satisfaction. With the onset of competition, Jacobs eludes programming evaluations could move to the forefront of satisfaction formulation. Since technological advancements have enhanced service reliability and picture quality, these

PAGE 47

38 advancements are less of a satisfaction indicator because they are now expected outcomes and no longer points-of-difference between competing entities. Jacob also found monthly cost, system size, and the number of channels offered was not related to satisfaction. Previous VCR adoption studies add insight to the current proposal for investigating the uses and asynchronous benefits of VOD. Klopfenstein and Spears (1991) examined the relationship between VCR owners and attitudes about watching network television, recording television, and overall satisfaction with VCR ownership. These variables were measured in order to determine whether VCR approval would increase or decrease based on time of ownership. The notion of discontinuance found in diffusion theory was not found to have occurred in VCR ownership at the time of this study. The study found the longer the VCR was in the home, the more likely the respondents were to record television programming. In addition, owners of VCRs for the longest period of time were more likely to delete commercials, record while sleeping, and record programs while not at home. Research involving demographic data found income was the only factor related to first VCR adopters. First adopters were also technolphiles supporting the diffusion theory assumption that early adopters are unique At the time of the research, only one scholarly VOD article could be found. Ronald Rizzuto and Michael O. Wirth (2002) developed a pioneering study entitled, The Economics of Video On Demand: A Simulation Analysis investigating if VOD (i.e., movies on demand) would be economically viable. They found economic aspects of VOD could be viable without major impacts on peak utilization rates. They identified three key VOD factors for economic viability: movie buy rates, Hollywood and cable operator revenue splits, and peak utilization rates. By doing diffusion research in the area

PAGE 48

39 of VOD adoption, projecting perspective utility rates could be useful for concluding economic viability based on the model proposed by Rizzuto and Wirth. Identifying and predicting the social adoption characteristics of this new VOD technology is important for cable companies, content providers and advertisers. By identifying these characteristics, advertisers can better tailor messages to reach those whom are interested in VOD technology; with the ultimate goal of enticing them to try VOD and hopefully increase the speed of various social groups through the adoption curve. Based on the theoretical grounding found in diffusion of innovations and the various variables found in other technology research included in this literature review, this exploratory study will examine the final three research questions: RQ3: What types of consumers are currently aware of VOD technology? RQ4: What factors indicate early adopters of VOD technology? RQ5: What factors indicate intent to adopt VOD technology?

PAGE 49

CHAPTER 3 METHODOLOGY Data Collection Due to time and financial restraints, a nonprobability sample was used. Respondents for this convenience sample were recruited during the Easter weekend of April 16-18, 2003 and were asked to take a self-administered survey. Written survey methodology was the chosen method of data collection because written surveys are easy to administer to a large population and surveys are particularly useful when needing to describe the characteristics of a large population (Babbie, 2001). The survey was administered on Clearwater Beach, Florida. Respondents were approached while relaxing on the beach and bottled water was given to respondents in appreciation for their cooperation. The survey took approximately five minutes to complete. A total of 190 respondents were asked to complete the questionnaire yielding a total of 170 surveys. The primary researcher for this study recruited the respondents by introducing himself and explaining how their cooperation would aid in his thesis research. Respondents were also assured their information would not be used for commercial or marketing purposes. Any questions regarding the survey were addressed upon the respondents completion of the survey. Survey research is the collection of information used to better understand or predict some aspect of respondents attitudes or behaviors (Davis, 1997). Surveys were used in this research due to their ability to be distributed to numerous respondents at the same 40

PAGE 50

41 time. Since this was a convenience sample, the ability to distribute multiple instruments simultaneously was essential. Overall, this made the survey the most obvious data collection instrument of choice. Survey Design The survey questionnaire focused on nine specific areas, either derived from previous diffusion of innovation research or important for rating the proposed exploratory advertising models or questions. The survey included questions regarding the following areas of interest: social and innovative attributes, attitude toward asynchronous viewing of differing television programming, VOD functionality, media usage, VOD awareness, knowledge, use and likeliness to subscribe VOD services, likelihood to provide personal information or be exposed to targeted advertising, rating possible advertising scenarios, technology ownership, and demographic information. The first section of the questionnaire was set up to measure the social and innovativeness characteristics of the respondents. The questions were modeled after previous technology diffusion of innovations studies (Lin 1998; Jeffres & Lin 1998; Atkin, Jeffres & Neuendorf 1998; Neuendoft & Atkin 1998; Kang 2002; Eastlick 1996). Respondents were asked to rate a number of statements on a seven point scale with one being strongly disagree to seven being strongly agree regarding social activities and need for innovativeness (see appendix for questionnaire). The second section of the questionnaire was designed to measure one of the attributes of VOD technology. Conceivably, a variety of programs would be or are currently available on a cable video server for the consumer to view at anytime. This set of questions asked the consumer to rate how desirable programming would be from four different sources: broadcast (e.g. ABC, CBS, NBC, FOX), cable (e.g. CNN, MTV,

PAGE 51

42 ESPN, HGTV), premium (HBO, Showtime, Starz), or Pay-per-view new release movies. Respondents were asked to rate each choice on a one to seven scale with one being not important and seven being very important. The third section of the questionnaire was designed to measure another attribute of VOD technology. As previously discussed in Chapter 2, VOD allows the consumer to stop, fast-forward, rewind, and pause television programming content. Respondents were asked to rate this attribute on a one to seven scale with one being not important and seven being very important. This section was designed to measure the usage of media. Respondents were asked to rate on a one to seven scale with one being not a user at all to seven being a heavy user of the following media: newspaper, radio, television, magazine, video rental, movie going and Internet. Previous diffusion of innovation research regarding technology has used similar measures (Kang, 2002; Dupagne, 1999; Lin 1998; Jeffres & Lin 1998; Atkin, Jeffres & Neuendorf 1998; Neuendoft & Atkin 1998; Eastlick 1996). This section of the questionnaire was designed to measure the dependent variables of VOD at various levels of involvement. VOD awareness was measured by asking respondents if they were aware of VOD (yes or no.) If the respondent was aware, they were asked to rate their knowledge of VOD technology on a seven-point scale with one being not at all knowledgeable and seven being highly knowledgeable. VOD use was asked by requesting if the respondent had ever previously used VOD (yes or no). Last, respondents were asked if they currently subscribe to VOD (yes or no) and if not, their intention to subscribe rated on a seven-point scale with one being not likely and seven being most likely.

PAGE 52

43 These exploratory questions were used to measure respondents likelihood to give personal information to advertisers and their willingness to be exposed to targeted advertising. Each was measured by asking the respondent to rate their attitude based on a seven-point scale, with one being not at all likely and seven being most likely. These exploratory questions were designed to measure respondent attitudes toward possible VOD advertising based on pay-per-view scenarios. Pay-per-view scenarios were used since it was assumed, regardless of the respondents VOD knowledge, that they have had some prior experience with pay-per-view content. Each question asked the respondent to rate on a seven-point scale with one being not at all likely and seven being most likely, how likely they would participate in the scenario. The first question asked if the respondent would pay $2.99 for the movie in exchange for being exposed to ten minutes of advertising commercials. The second question asked if the respondent would consider paying $1.99 for twenty minutes of commercials spread throughout the movie. The third question asked how likely the respondent would pay $2.99 for a pay-per-view movie with text messaging moving across the bottom of the screen during the entire movie. The forth question asked the respondent to rate paying $1.99 for a pay-per-view movie with text messaging moving across the bottom during the entire length of the movie. Respondents were asked whether they subscribe or own a number of technology products and services. These products or services included: Digital video recorder, personal computer, high-speed Internet, video game system, cell phone, home theater system, high definition television, large screen television (35 or larger), DVD player, personal digital assistant, cable TV with a set-top box, cable TV without a set-top box,

PAGE 53

44 direct broadcast satellite, and premium cable television. Measuring technology ownership is consistent with previous diffusion of innovations research (Dupagne, 1999; Lin, 1998; Kang, 2002). Demographic questions included gender, education, marital status, age, household size, occupation and income. Demographic variables are essential to diffusion of innovation research and have been previously measured in other technology diffusion studies (Rogers, 1995; Kang, 2002; Dupagne, 1999; Lin 1998; Jeffres & Lin 1998; Jeffres & Atkin, 1996; Atkin, Jeffres & Neuendorf 1998; Neuendoft & Atkin 1998; Eastlick 1996). Data Analysis Different statistical methodological processes were applied to each research question. For analyzing what types of consumers are aware of VOD and for indicating factors important for early adopters of VOD, or RQ1 and RQ2, descriptive statistics were used in conjunction with t-tests and chi-square tests, analyzing differences of means and statistical significance. For analyzing RQ3 or the relationship of factors contributing to intention to adopt VOD technology in the next six months, multiple regression analysis was implemented. To analyze what functions of VOD technology were found to be most preferred as well as the analysis of the most positive prospective advertising model for VOD or RQ 4 and 5, t-tests were used to analyze differences in means. SPSS for Windows, Release 11.5.0 was used to run all descriptive and inferential statistics.

PAGE 54

CHAPTER 4 RESULTS Descriptive Statistics The convenience sample yielded 170 adults made of 28% males (n=48) and 72% females (n=121.) According to the 2000 U.S. Census, 49% of the national population is male and 51% of the national population is female. Therefore, caution should be taken to generalize this studys results onto the population because of the skewed gender. As for age, 23% were 18-24 years old, 28% were 25-34 years old, 24% were 35-44 years old, 18% were 45-54 years old, 5% were 55-64 years old and only one person was over the age of 65. The mean age was 36 years old (std. deviation of 12.1). Compared to the general population derived from 2000 U.S. Census data, the current sample is skewed much younger. Since the sample is skewed much younger, caution should again be taken when projecting data onto the general population. Age frequencies and percents are found in table 4-1. Table 4-1. Age Distribution Sample US Census Age Frequency Percent Percent 18-24 38 23% 10% 25-34 46 28% 13% 35-44 40 24% 16% 45-54 30 18% 14% 55-64 9 5% 9% 65+ 1 less than 1% 13% 45

PAGE 55

46 For household income, 7% of the total sample made less than $19,000 per year, 18% made between $20,000 and $39,000 per year, 19% made between $40,000 and $59,000 per year, 18% made between $60,000 and $79,000 per year, 14% made between $80,000 and $99,000 per year and 20% of the households made over $100,000 per year. According to the 2000 U.S. Census, the general population consists of more people making less than $19,000 per year. Therefore, our sample has a higher income bias and this should be noted when projecting data to the general population. Income frequencies and percents are listed in table 4-2. Table 4-2. Income Distribution Sample US Census Income Frequency Percent Percent <$19,000 11 7% 23% $20k-$39k 31 18% 15% $40k-$59k 31 19% 18% $60k-$79K 30 18% 13% $80k-$99k 23 14% 8% >$100k 33 20% 13% Of total respondents in sample, 41% were single or considered other meaning they did not consider themselves one of the choices, 49% were married, 9% were divorced, less than 1% was widowed. This finding differs from the U.S. Census numbers because comparatively the sample is skewed with too many singles. The mean sample household size was 3, close the general population household size of 2.59, provided by the U.S. census. Frequencies and percents for marital status are presented in Table 4-3.

PAGE 56

47 Table 4-3. Marital Status Sample US Census Status Frequency Percent Percent Single/Other 68 41% 27% Married 82 49% 57% Divorced 15 9% 10% Widow 1 less than 1% 7% As for occupation, 3% were laborers, 1% were machine technicians, 2% were craftsmen, 7% were clerical, 12% were in sales, 3% were administrators, 37% were considered professionals, 6% were managers, 29% were other or occupation not listed. Since the U.S. Census categories did not exactly match the sample categories, conclusions were drawn to match respective categories. Compared to general population data from the U.S. Census, the sample lacked service and construction positions, matched close regarding sales and office occupations as well as management and professional, but was deficient meeting the other category. Frequencies and percents of total sample and U.S. Census percents are listed in Table 4-4. Table 4-4. Occupation Distribution Sample US Census Occupation Frequencies Percents Category Percents Laborer 5 3% Machine/Service Tech 2 1% Craftsman 3 2% Service and Construction Positions 24% Clerical 11 7% Sales 19 12% Administrator 5 3% Sales and Office Occupations 27% Professional 61 37% Manager 10 6% Management & Professional 34% Other 49 29% Other 15%

PAGE 57

48 For respondent education, 1% had less than a high school education, 26% had finished high school, 57% finished college, and 16% finished graduate school. Compared to the U.S. Census data from 2000, our sample is not representative as far as respondents having an education less than high school level. The current sample also has respondents with college degrees and graduate degrees compared to the general population. Frequencies and percents are included in Table 4-5. Table 4-5. Education Sample U.S. Census Education Frequency Percent Percent Less Than High School 2 less than 1% 20% High School 43 26% 29% College 94 57% 43% Graduate School 27 16% 9% The study revealed that 58% of the respondents were aware of VOD technology. Of those respondents who were aware of VOD (n=94), only 43% had actually used VOD and only 26% subscribe to VOD services. Table 4-6 displays awareness, use and subscribers of VOD. Table 4-6. Awareness, Use and Subscribers of VOD Yes No Aware of VOD 99 (58%) 68 (40%) Used VOD 40 (43%) 54 (57%) Subscribe to VOD 24 (26%) 69 (74%) Respondents who are aware of VOD but do not subscribe were asked to rate their intent to subscribe to VOD services over the next six months. Respondents rated their intent to subscribe a M=2.06 (n=66, Std. Dev 1.402) from a seven-point scale with one being not likely at all and seven being most likely. This result seems to indicate the three

PAGE 58

49 quarters of those who are aware of VOD are not very likely to subscribe to VOD type services within the next six months. As for the functions of VOD, rated on a seven-point scale by total respondents with one being not important and seven being very important revealed having the ability to pause, fast-forward, and rewind content had a mean score of M=4.9 revealing an above average importance to the respondents. Perhaps the ability to use television much like a VCR appealed to the sample. The ability to watch programming at anytime (asynchronous), rated on the same scale, revealed interesting results. The ability to watch broadcast network content was rated the highest with the average for broadcast network content at M=4.97. Cable was rated second highest with the average for cable network content at M=4.68. Premium content had the third highest score with the average a M=3.91. Pay-per-view new release movies was rated the lowest with a M=3.56 score. This finding is interesting since the bulk of VOD services currently offered are in the realm of pay-per-view and premium content. Broadcast programming and cable programming were rated most important and yet they are not as widely available as the others. Table 4-7 shows the mean and standard deviation scores for VOD functions. Table 4-7. VOD Functions VOD Functions Mean Std. Deviation Ability to Stop, Fast-Forward, Rewind, Pause TV Content 4.91 1.94 Watch Broadcast Network Programming Anytime 4.97 1.83 Watch Cable Network Programming Anytime 4.68 1.78 Watch Premium Network Programming Anytime 3.91 2.09 Watch Pay-Per-View Network Programming Anytime 3.56 2.19 Rated on a 1-7 scale with 1 being not important to 7 very important

PAGE 59

50 The total sample found the advertising models less than desirable based on the average ratings from the seven-point scale with one being not at all likely and seven being most likely. The average rating from respondents to pay $2.99 per movie with a total of ten minutes of standard commercials was M=2.15. The average rating for twenty minutes of commercials at $1.99 was M=1.93. For banner or ticker advertising throughout the entire feature, respondents rated the $2.99 charge M=1.85 and the $1.99 charge M=2.2. Clearly, the respondents are not very likely to do any of the proposed adverting scenarios but the banner/ticker advertising scenario with respondents paying $1.99 for the program was found to be most popular of the choices. In a general sense, the respondents found the idea of releasing personal information to advertisers for a discount or being exposed to targeted advertising for a discount on pay-per-view movies less desirable as well. The average score for releasing personal information was M=2.6, based on a seven-point scale with one being not likely at all and seven being most likely. The average score for exposure to targeted advertising was slightly higher at 2.84. Both mean scores are similar and again it appears respondents do not seem likely to want targeted advertising and discounts delivered by releasing personal information. All mean scores and standard deviations are included in table 4-8. Table 4-8. Advertising Scenarios and Models VOD Advertising Scenarios and Models Mean Std. Deviation Willingness to Provide Personal Information 2.61 1.79 Willingness to be Exposed to Targeted Advertising 2.84 1.81 Willingness to Pay $2.99 for 10 minutes of Advertising (commercials) 2.15 1.57 Willingness to Pay $1.99 for 20 minutes of Advertising (commercials) 1.93 1.5 Willingness to Pay $2.99 for Banner or Ticker Type Advertising 1.85 1.51 Willingness to Pay $1.99 for Banner or Ticker Type Advertising 2.2 1.86 Rated on a 1-7 scale with 1 being not at all likely to 7 being most likely

PAGE 60

51 Respondents were asked to rate their usage of newspapers, radio, television, magazines, video rental, movie going and the Internet. Ratings were based on a one to seven scale with one being not a user at all to seven being a heavy user. Television had the highest rating at M=5.61 with radio second at M=5.57, Internet third at M=5.42, newspaper forth at M=4.57, magazines fifth at M=4.21, video rental sixth at 3.91 and the movie going usage mean usage score was M=3.57. It appears the respondents in our sample are electronically savvy since television, radio and Internet ranked in the top three. This finding leads to a possible conclusion that our sample uses more electronic media when compared to other types of media. The mean scores and standard deviations can be found in Table 4-9. Table 4-9. Media Use Media Mean Std. Deviation Television Use 5.61 1.49 Radio Use 5.57 1.47 Internet Use 5.42 1.87 Newspaper Use 4.57 1.97 Magazine Use 4.21 1.6 Video Rental Use 3.91 1.73 Movie Going Use 3.57 1.7 Rated on a 1-7 scale with 1 do not use at all to 7 use very heavily For the social and innovativeness characteristics, respondents were asked to rate a number of social activities on a one to seven scale with one being strongly disagree and seven being strongly agree. I like to participate in social activities had the highest mean score at M=5.81 while I am in good shape financially had the lowest at M=4.71. All respondents scores were above average leading to the conclusion this sample is biased toward having above average social and innovative characteristics. Table 4-10

PAGE 61

52 displays all the mean and standard deviation scores for the social and innovative activities. Table 4-10. Social and Innovativeness Characteristics Social and Innovative Characteristics Mean Std. Deviation I like to learn about new ideas 5.82 1.41 I like to participate in social activities 5.81 1.54 I like to explore new technologies 5.42 1.53 I like to try new products 5.41 1.51 I enjoy interacting with my neighbors 5.12 1.67 I keep up with new technologies 5.04 1.46 I am willing to take risks in order to try new things 4.88 1.48 I am in good shape financially 4.71 1.43 Rated on a 1-7 scale with 1 being strongly disagree to 7 strongly agree Respondents were asked if they owned a number of new technology items. Of those items, a majority of the respondents own a personal computer (89%), a cellular phone (87%), and/or a DVD player (79%). Personal digital assistants were the least owned technology (16%) by the total sample. Overall this sample is very technological savvy. The sample has an extraordinarily high number of computer owners since according to the National Telecommunication and Information Association the computer ownership average for the general population is 56.7% (NTIA, 2002). Further, 34.4% of the general population subscribe to high speed Internet services compared to the 58% of our sample (URL wire, 2003). This leads to another important point of caution regarding projecting these findings into the general population since the current sample seems to be very technologically savvy. Total ownership frequency and percents are listed in table 4-11.

PAGE 62

53 Table 4-11. Technology Ownership Technology Ownership Frequency Percent Personal Computer 152 89% Cell Phone 148 87% DVD 134 79% High-speed Internet 99 58% Video Game System 89 52% Cable Television without Set-Top-Box 72 42% Cable Television with Set-Top-Box 71 42% Large Television (35" or larger) 68 40% Digital Video Recorder 63 37% Home Theatre 58 34% High-Definition Television 38 22% Direct Broadcast Satellite 34 20% Personal Digital Assistant 27 16% Research Question 1 Research question 1 looked to examine what functions of VOD were found to be most preferred. All measured respondents were aware of VOD. Independent sample t-tests were performed yielding only one significant result and one result of marginal significance. The mean score of those who have used VOD (M=4.72) was statistically significantly higher than the mean score of those who had not used VOD (M=3.87) when analyzing the ability to view premium network programming anytime, (t=-1.913, d.f.= 92, p=.03). For pay-per-view, a marginal statistical significance relationship was discovered (t=-1.51, d.f.=92, p=.067). Those who had used VOD (M=4.5) rated the viewing pay-per-view anytime higher than those non-users (M=3.78). Therefore, users of VOD prefer the ability to view premium network and pay-per-view programming over other VOD options. All results of preferred VOD options between users and non-users can be found in table 4-12.

PAGE 63

54 Table 4-12. Preferred VOD Options Between Users and Non-Users of VOD Used VOD VOD Functionality (Rewind, FF, PauseTV Content) Yes No t df p Mean 5.3 5.0 -.766 92 .223 Standard Deviation 1.77 1.94 Watch Broadcast Network Programming Anytime Mean 4.93 5.09 .426 92 .336 Standard Deviation 1.789 1.96 Watch Cable Network Programming Anytime Mean 4.72 4.61 -.308 92 .380 Standard Deviation 1.71 1.82 Watch Premium Network Programming Anytime Mean 4.72 3.87 -1.913 92 .03* Standard Deviation 2.15 2.14 Watch Pay-Per-View Programming Anytime Mean 4.5 3.78 -1.51 92 .067 Standard Deviation 2.25 2.3 indicates significant mean difference at p<.05 One-tailed t-test In addition to the comparing option between users and non-users, comparisons were also made examining the mean scores of each VOD option among the current users. Paired sample t-tests were used to statistically compare the means and each option was compared against another. Generally, VODs functionality (M=5.3) was rated the highest with the ability to watch broadcast networks (M=4.93) having the second highest mean score. The mean scores for having the ability to watch cable programming (M-4.72) and premium cable programming (M-4.72) were the same. The only difference of means found statistically significant (t=-2.327, d.f.=39, p=.013) was the comparison of the highest mean, VOD functionality (M=5.3) and the ability for VOD users to watch pay-per-view programming anytime (M=4.5). The results are shown in Table 4-13

PAGE 64

55 Table 4-13. Mean Comparison of VOD Options by VOD Users Mean Std. Devation t df p VOD Functionality (Rewind, FF, PauseTV Content) 5.3 1.79 -1.228 39 0.114 Watch Broadcast Network Programming Anytime 4.93 1.79 Watch Cable Network Programming Anytime 4.72 1.79 0.831 39 0.205 Watch Broadcast Network Programming Anytime 4.93 1.79 Watch Premium Network Programming Anytime 4.72 1.71 0 39 0.5 Watch Cable Network Programming Anytime 4.72 1.79 Watch Pay-Per-View Programming Anytime 4.5 4.72 0.577 39 0.284 Watch Premium Network Programming Anytime 4.72 2.15 VOD Functionality (Rewind, FF, PauseTV Content) 5.3 1.786 -2.327 39 0.013* Watch Pay-Per-View Programming Anytime 4.5 2.253 indicates significant mean difference at p<.05 One-tailed t-test Research Question 2 Research question 2 analyzed the exploratory advertising models based on pay-per-view movies as well as attitudes regarding giving up personal information and receiving targeted advertising. The goal of this first part of this research question was to identify which would be the most preferred advertising scenario. Independent t-tests were run to compare mean rating scores between the commercial scenario and then between the ticker/banner scenarios. Last, all mean rating scores for the advertising models were compared against each other. All mean rating scores were derived from the total sample population and a total of six advertising scenarios were analyzed.

PAGE 65

56 The first comparison, Ad Scenario #1, examined the difference in means, based on a one to seven scale with one being not at all likely and seven being most likely, between paying $2.99 for 10 minutes of advertising (M=2.15) and paying $1.99 for 20 minutes of advertisings (M=1.93).In the first comparison it appears respondents were not very likely to approve of either advertising method for a discount on the movie. Still, paying $2.99 for 10 minutes of television advertising was rated higher and the relationship was found to be statistically significant compared than paying $1.99 for 20 minutes of advertising, (t=2.57, d.f.=167, p=.006). The results are shown in table 4-14. Table 4-14. Ad Scenario #1 $2.99 for 10 Minutes of Advertising $1.99 for 20 Minutes of Advertising t df p Mean 2.15 1.93 2.57 167 .006* Standard Deviation 1.57 1.5 indicates significant mean difference at p<.05 One-tailed t-test The second scenario, Ad Scenario #2, compared the mean rating score toward paying the $2.99 ticker or banner advertising option with the mean rating score of paying $1.99 for the ticker or banner advertising option. The mean score for the $1.99 ticker option (M=2.2) was higher and statistically significantly different than the mean rating for the $2.99 option (M=1.85), (t=-4.08,d.f.=167, p=.001). However, the score was still low and considered an unlikely advertising model by the respondents based on the one to seven scale. The t-test results are listed in table 4-15. Ad Scenario #3 again uses t-tests to compare the mean ratings of $2.99 for 10 minutes of television commercials and the mean rating of paying $2.99 for ticker advertising. The mean for paying $2.99 for the commercials (M=2.15) was slightly higher and the relationship was found to be statistically significant than the mean score

PAGE 66

57 Table 4-15. Ad Scenario #2 $2.99 For Ticker Advertising $1.99 for Ticker Advertising t df p Mean 1.85 2.2 -4.08 167 .001* Standard Deviation 1.5 1.86 indicates significant mean difference at p<.05 One-tailed t-test for paying $2.99 for ticker advertising (M=1.57), (t=2.22, d.f.=167, p=.014). Respondents prefer commercial advertising compared to banner advertising at the $2.99 price level. Still, based on these low mean scores it appears respondents were not favorable toward these advertising scenarios. The results are displayed in table 4-16. Table 4-16. Ad Scenario #3 $2.99 for 10 Minutes of Advertising $2.99 For Ticker Advertising t df p** Mean 2.15 1.85 2.22 167 .014* Standard Deviation 1.57 1.51 indicates significant mean difference at p<.05 One-tailed t-test Ad Scenario #4 looked at the mean differences between paying $1.99 for 20 minutes of commercial advertising during a pay-per-view movie and paying $1.99 for ticker advertising during a pay-per-view movie. The mean score for paying $1.99 for ticker advertising (M=2.2) was higher than paying $1.99 for 20 minutes of commercial (M=1.93), (t=-1.892, d.f.= 167, p=.03). The relationship between the two means was also found to be statistically significant. At the $1.99 pricing point, respondents found the ticker advertising marginally favorable. The results are listed in table 4-17. Ad Scenario #5 compared the mean scores of paying $2.99 for 10 minutes of commercial advertising with paying $1.99 for ticker advertising. The t-test did not reveal the mean scores were significantly different. Although the mean score for $1.99 ticker

PAGE 67

58 Table 4-17. Ad Scenario #4 $1.99 for 20 Minutes of Advertising $1.99 For Ticker Advertising t df p** Mean 1.93 2.2 -1.892 167 .03* Standard Deviation 1.5 1.86 indicates significant mean difference at p<.05 One-tail t-test advertising was the rated the highest by the respondents, the difference in means is not substantial. The findings for Ad Scenario #5 are listed in table 4-18. Table 4-18. Ad Scenario #5 $2.99 for 10 Minutes of Advertising $1.99 For Ticker Advertising t df p Mean 2.15 2.2 -.338 167 .368 Standard Deviation 1.57 1.86 indicates significant mean difference at p<.05 One-tailed t-test Ad Scenario #6 compared the mean rating scores for respondents paying $1.99 for 20 minutes (M=1.93) of commercial advertising during a pay-per-view program and paying $2.99 for ticker advertising (M=1.85) during the pay-per-view program. According to the mean scores, these were the least popular advertising scenarios. The findings were not statistically significant and are displayed in table 4-19. Table 4-19. Ad Scenario #6 $1.99 for 20 Minutes of Advertising $2.99 For Ticker Advertising t df p** Mean 1.93 1.85 .607 167 .273 Standard Deviation 1.5 1.5 indicates significant mean difference at p<.05 **Sig. 1-tailed

PAGE 68

59 Important to advertisers is knowing if respondents are willing to provide personal information so advertisers can send specific, targeted advertising. Using a one to seven scale with one being not at all likely and seven being most likely mean scores were derived and t-tests were used to compare mean scores of those respondents who are aware and not aware of VOD. A comparison was also done of respondents who are aware. This comparison examined those who have used VOD and those who have not used VOD. Overall findings were not statistically significant. The mean averages were also quite low, concluding respondents were not likely to release personal information for a discount on pay-per-view movies. The results are shown in figure 4-20. Table 4-20. Willingness to Provide Personal Information Aware of VOD Yes No t df p Mean 2.6 2.54 -.222 161 .412 Standard Deviation 1.75 1.86 indicates significant mean difference at p<.05 Used of VOD Yes No t df p Mean 2.6 2.85 .663 91 .255 Standard Deviation 1.63 1.91 indicates significant mean difference at p<.05 Attitude ratings were also collected to determine if respondents would like to receive advertising specific to their interests. t-tests were done to determine a difference in mean between those who are currently aware of VOD and also between those who are aware and have used VOD. These mean scores were also quite low concluding respondents were not likely to be willing to be exposed to targeted advertising. Interestingly enough, those who are not aware of VOD had the highest approval rate toward targeted advertising (M=3.11) compared to all others. No statistically significant

PAGE 69

60 difference in means could be concluded in either case. The mean scores and t-test results are displayed in table 4-21. Table 4-21. Willingness to be Exposed to Targeted Advertising Aware of VOD Yes No t df p Mean 2.92 2.63 -.998 161 .16 Standard Deviation 1.88 1.69 indicates significant mean difference at p<.05 Used of VOD Yes No t df p Mean 2.6 3.11 1.31 91 .096 Standard Deviation 1.89 1.85 Research Question 3 For the third research question, this study attempts to discern the characteristics between those who are aware of VOD and those who are not aware of VOD technology. To answer this question, depending on level of measurement, independent sample t-tests and chi-square tests were conducted. t-tests were conducted for interval or continuous variables such as the social/innovative characteristics and media use while chi-square tests were conducted with categorical variables such as demographic data and technology ownership. Our first variable was media use. For media use, independent t-tests were conducted. There were two significant findings between those in the sample who are aware of VOD and those who are not. The mean for television use was higher for those who are aware of VOD (M=5.76) compared to those who are not aware (M=5.35), and the result was statistically significant (t= -1.735, d.f.=165, p=.043). The second statistically significant mean difference was from Internet usage. Those who are aware of

PAGE 70

61 VOD rated Internet usage higher (M=5.75) compared to those who are not aware of VOD (M=5.03), (t= -2.505, d.f. 165, p=.006). The findings suggest that people who are more aware of VOD tend to use television and the Internet more than those who are not aware of VOD. This result implies those who are aware of VOD were possibly exposed to VOD through the use of television or the Internet. Total media use mean scores are shown in table 4-22. Table 4-22. Mean Media Use Scores Between Those Aware and Not Aware of VOD Aware Not Aware t df p Newspaper Use Mean score 4.67 4.37 -.958 165 .170 Std. Deviation 1.86 2.14 Radio Use Mean score 5.61 5.46 -.647 165 .260 Std. Deviation 1.41 1.56 Television Use Mean score 5.76 5.35 -1.735 165 .043* Std. Deviation 1.37 1.61 Magazine Use Mean score 4.36 4 -1.438 164 .076 Std. Deviation 1.71 1.41 Video Rental Mean score 3.99 3.78 -.770 165 .221 Std. Deviation 1.75 1.71 Movie Going Mean score 3.71 3.38 -1.203 165 .116 Std. Deviation 1.75 1.67 Internet Mean score 5.75 5.03 -2.505 165 .006* Std. Deviation 1.58 2.12 indicates statistically significant mean difference at p<.05 One-tailed t-tests Social and innovative characteristics scores were analyzed to discern any significant difference between those in the sample who are currently aware of VOD and those who are not aware of VOD. Independent group t-tests were run with no statistically

PAGE 71

62 significant differences in mean scores found. These results appear to tell us that there are no important social or innovative characteristics separating those who are aware of VOD and those who are not aware of VOD. The results are displayed in table 4-23. Table 4-23. Social/Innovative Mean Scores Between Those Aware and Not Aware of VOD Yes No t df p I like to try new products Mean score 5.47 5.37 -.460 165 .323 Std. Deviation 1.47 1.49 I like learning about new technologies Mean score 5.42 5.43 .009 165 .497 Std. Deviation 1.59 1.46 I like to socialize Mean score 5.83 5.85 .104 165 .4585 Std. Deviation 1.53 1.47 I like to socialize with my neighbors Mean score 5.12 5.18 .221 164 .413 Std. Devation 1.67 1.63 I am in good financial standings Mean score 4.69 4.69 .019 165 .4925 Std. Devation 1.43 1.44 I like to learn about new ideas Mean score 5.91 5.78 -.614 164 .27 Std. Devation 1.33 1.43 I like to learn about new technologies Mean score 5.09 4.96 -.581 165 .281 Std. Devation 1.51 1.43 I like to take risks Mean score 4.92 4.84 -.351 165 .36 Std. Devation 1.52 1.38 indicates statistically significant mean difference at p<.05 One-tailed t-tests For technology ownership, chi-square tests were run on the categorical variables examining the relationship between ownership of various new technology items to VOD awareness. Seven statistically significant relationships were found. Those who currently subscribe to high speed Internet services were significantly more aware of VOD than

PAGE 72

63 those who do not, (x = 3.898, d.f.=1, p=.035). High definition television ownership was the second significant finding. There were significant differences found in both those who own and do not own high definition televisions and awareness of VOD, (x = 4.039, d.f. =1, p=.033). Those who do not own big screen televisions were significantly less aware of VOD compared to all others, (x= 3.322, d.f.=1, p=.048). Those respondents who own DVD players were found to be significantly more aware of VOD compared to those who do not own DVD players, (x= 5.189, d.f.=1, p=.019). Those respondents who subscribe to cable with a set-top box were significantly more aware of VOD compared to those who do not, (x= 7.662, d.f.=1, p=.004). Interestingly enough, those subscribing to cable without a set-top box were also significantly more aware of VOD, (x= 3.389, d.f=1, p=.046). Last, those who subscribe to premium cable were significantly more aware of VOD technology compared to those respondents who do not subscribe, (x=5.342, d.f.=1, p=.015). In short, the results show those who are aware of VOD tend to own high-speed Internet, high definition television, large screen TVs, DVD players, subscribe to cable television with or without a set-top-box and subscribe to premium cable services. This finding suggests those respondents who own television and Internet related products tend to be more aware of VOD services. As for the demographic characteristics, education was the only demographic characteristic significantly related to VOD awareness. Those respondents who attended college were significantly more aware of VOD technology than those who did not attend college, (x=10.148, d.f.=3, p=.009). All demographic comparisons and statistical information can be found in table 4-25.

PAGE 73

64 Total results for technology ownership can be found in table 4-24. Table 4-24. Technology Ownership Mean Scores Between those Aware and Not Aware of VOD Aware of VOD Not Aware of VOD Ownership Yes No Yes No x df p DVR 32 67 29 37 2.293 1 .089 PC 91 8 58 8 1.245 1 .196 High-Speed Internet 64 35 33 34 3.898 1 .035* Video Game System 51 48 36 31 .079 1 .452 Cell Phone 86 13 59 8 .051 1 .509 Home Theater System 37 62 20 47 1.003 1 .202 High Definition Television 28 71 10 57 4.039 1 .033* Large Screen TV (35" or larger) 45 54 21 46 3.322 1 .048* DVD Player 84 15 47 20 5.189 1 .019* Personal Digital Assistant 17 82 9 58 .423 1 .336 Cable TV with set-top-box 51 48 20 47 7.662 1 .004* Cable TV without set-top-box 36 63 34 33 3.389 1 .046* Direct Broadcast Satellite 19 80 13 54 .001 1 .563 Premium Cable 52 47 23 44 5.342 1 .015* indicates significant mean difference at p<.05 One-tailed t-test Table 4-25. Demographic Scores Between those Aware and Not Aware of VOD Gender Aware of VOD Not Aware of VOD x df p Male 30 18 .230 1 .382 Female 69 49 indicates significant mean difference at p<.05 Education Aware of VOD Not Aware of VOD x df p Less Than High School 1 1 10.148 3 .009* High School 17 26 College 62 32 Graduate School 19 8 indicates significant mean difference at p<.05

PAGE 74

65 Table 4-25. Continued Aware of VOD Not Aware of VOD x df p Single 43 24 3.818 4 .216 Married 48 34 Divorced 8 7 Widow 0 1 Other 0 1 indicates significant mean difference at p<.05 Age Aware of VOD Not Aware of VOD 18-24 23 15 25-34 28 18 35-44 26 14 45-54 18 12 55-64 3 6 65+ 0 1 One-tailed t-test (t= -.774, df=163, p=.22) Household Size Aware of VOD Not Aware of VOD 1 11 10 2 38 9 3 8 15 4 27 17 5 12 11 6 3 3 One-tailed t-test (t= -1.312, df=162, p=.095) Occupation Aware of VOD Not Aware of VOD x df p Laborer 4 1 10.352 8 .241 Machine/Service Tech 1 1 Craftsman 2 1 Clerical 6 5 Sales 10 9 Administrator 3 2 Professional 44 17 Manager 7 3 Other 22 27

PAGE 75

66 Table 4-25. Continued Aware of VOD Not Aware of VOD x df p <$19,000 6 5 .613 5 .987 $20k-$39k 20 11 $40k-$59k 19 12 $60k-$79K 18 12 $80k-$99k 10 13 >$100k 14 19 indicates significant mean difference at p<.05 One-tailed t-test Research Question 4 Research question 4 examined what factors indicate early adopters of VOD technology. This question attempts to discern the characteristics between those who are aware of VOD and subscribe from those who are not subscribers of VOD technology. To answer this question, descriptive statistics were analyzed using independent sample t-tests and chi-square tests were used. t-tests were conducted for interval or continuous variables such as the social/innovative characteristics, importance of viewing content anytime, VOD functionality (ability to fast forward, rewind, pause) and media use while chi-square tests were conducted with categorical variables such as demographic data and technology ownership. For the media usage score comparisons between subscribers and non-subscribers, independent t-tests were run. There were no significant differences found between subscribers and non-subscribers of VOD and media use. For this study, there is no relationship between media usage and the likelihood of VOD adoption. Compared with previous studiesThe findings are displayed in table 4-26.

PAGE 76

67 Table 4-26. Media Usage Comparisons between Subscribers and Non-subscribers of VOD Media Type Subscriber Non-Subscriber t df p Newspaper Use Mean score 4.67 4.8 .299 91 .383 Std. Deviation 1.79 1.86 Radio Use Mean score 5.25 5.74 1.46 91 .074 Std. Deviation 1.45 1.4 Television Use Mean score 5.92 5.78 -.414 91 .34 Std. Deviation 1.06 1.45 Magazine Use Mean score 4.25 4.5 .615 90 .27 Std. Deviation 1.94 1.63 Video Rental Mean score 4.08 3.97 -.272 91 .393 Std. Deviation 1.67 1.77 Movie Going Mean score 3.63 3.72 .233 91 .408 Std. Deviation 1.86 1.78 Internet Mean score 6 5.77 -.629 91 .266 Std. Deviation 1.53 1.56 indicates significant mean difference at p<.05 One-tailed t-test Much like the media usage score comparisons between subscribers and non-subscribers, the social/innovative characteristic comparisons found no significant mean score differences between subscribers and non-subscribers of VOD. Further, there appears to be no relationship between subscribers, non-subscribers and social/innovative characteristics. The findings are displayed in table 4-27. For technology ownership, chi-square tests were run on the categorical variables examining the relationship between ownership of various new technology items to VOD subscribers. Three statistically significant differences were discovered. Of those

PAGE 77

68 Table 4-27. Social/Innovative Characteristics Comparisons between Subscribers and Non-subscribers of VOD Subscriber Non-Subscriber t df p I like to try new products Mean score 5.75 5.45 -.86 91 .2 Std. Deviation .989 1.605 I like learning about new technologies Mean score 5.75 5.42 -.89 91 .19 Std. Deviation 1.36 1.63 I like to socialize Mean score 6.25 5.78 -1.35 91 .09 Std. Deviation 1.19 1.54 I like to socialize with my neighbors Mean score 4.79 5.23 1.13 91 .13 Std. Devation 1.81 1.59 I am in good financial standings Mean score 4.38 4.64 .74 91 .23 Std. Devation 1.34 1.02 I like to learn about new ideas Mean score 5.92 5.94 .08 91 .47 Std. Devation 1.02 1.46 I like to learn about new technologies Mean score 5.12 5.03 -.26 91 .4 Std. Devation 1.54 1.58 I like to take risks Mean score 4.71 4.91 .56 91 .3 Std Devation 1.27 1.64 indicates significant mean difference at p<.05 One-tailed t-test significant differences, more people who are non-subscribers also have cell phones, (x=7.61, d.f.=1, p=.011). This finding is probably due to the fact that most all respondents who are aware of VOD also own cell phones (83%) and more people overall are non-subscribers. Those who subscribe to cable without a set-top box are less likely to subscribe to VOD services, (x=10.416, d.f=1, p=.001). This finding is due to the fact in order to have VOD service, you must have a set-top box thus non-adopters would not have a set-top-box. Last, those who subscribe to direct broadcast satellite are statistically

PAGE 78

69 less likely to subscribe to VOD services, (x=9.354, d.f.= 1, p=.004). A possible interpretation of this finding could involve the simple fact that VOD technology is not available to direct broadcast satellite subscribers. Total mean scores and chi-square values can be seen in table 4-28. Table 4-28. Technology Ownership Comparisons between Subscribers and Non-subscribers of VOD Subscriber Non-Subscriber Ownership Yes No Yes No x2 df p DVR 12 12 21 48 2.98 1 .071 PC 21 3 67 2 3.23 1 .106 High-Speed Internet 16 8 44 25 .065 1 .502 Video Game System 14 10 35 34 .414 1 .343 Cell Phone 17 7 64 5 7.61 1 .011* Home Theater System 13 11 23 46 3.26 1 .060 High Definition Television 10 14 19 50 1.66 1 .151 Large Screen TV (35" or larger) 12 12 28 41 .645 1 .286 DVD Player 20 4 62 7 .756 1 .302 Personal Digital Assistant 2 22 14 55 1.787 1 .153 Cable TV with set-top-box 16 8 34 35 2.17 1 .108 Cable TV without set-top-box 2 22 31 38 10.416 1 .001* Direct Broadcast Satellite 9 15 7 62 9.354 1 .004* Premium Cable 17 7 34 35 3.341 1 .055 indicates significant mean difference at p<.05 One-tailed t-test Comparing the demographic variables of subscribers and non-subscribers, only two statistically significant characteristics were found: education and marital status. For education, a significant difference was found where respondents with only a high school education were less likely to be subscribers of VOD, (x=7.085, d.f.=1, p=.014). Marital status also yielded a statistical difference with single respondents less likely to subscribe to VOD services, (x= 7.421, d.f.=1, p=.012). The gender of the respondents yielded a marginally statistically significant difference with 40% of the men subscribing and 22% of the women subscribing (x=3.6, d.f.=1, p=.054). This difference could be because

PAGE 79

70 there are two and half times as many women in the sample. All demographic variables and counts are listed in table 4-29. Table 4-29. Demographics Comparisons between Subscribers and Non-subscribers of VOD Subscribe to VOD Yes No x df p Male 10 15 3.6 1 .054 Female 14 49 indicates significant mean difference at p<.05 One-tailed t-test Subscribe to VOD Yes No x2 df p Less Than High School 0 0 7.085 1 .014* High School 3 16 College 20 37 Graduate School 1 16 indicates significant mean difference at p<.05 One-tailed t-test Subscribe to VOD Yes No x2 df p Single 5 36 7.451 1 .012* Married 15 28 Divorced 4 5 Widow 0 0 Other 0 0 indicates significant mean difference at p<.05 One-tailed t-test Subscribe to VOD Age Yes No Mean S.D. t df p 18-24 0 21 36.25 9.166 -0.913 90 0.182 25-34 12 17 35-44 6 17 45-54 6 10 55-64 0 3 65+ 0 0 indicates significant mean difference at p<.05 One-tailed t-test

PAGE 80

71 Table 4-29. Continued Subscribe to VOD Household Size Yes No mean S.D. t df p 1 3 8 3.17 1.523 0.693 90 0.245 2 8 25 3 2 8 4 5 18 5 5 8 6 1 1 Subscribe to VOD Occupation Yes No x df p Laborer 0 3 8.536 8 .191 Machine/Service Tech 0 1 Craftsman 1 0 Clerical 1 5 Sales 2 8 Adminsitrator 0 3 Professional 11 28 Manager 1 8 Other 8 13 indicates significant mean difference at p<.05 One-tailed t-test Subscribe to VOD Yes No x df p <$19,000 1 0 1.684 5 .446 $20k-$39k 6 13 $40k-$59k 4 14 $60k-$79K 6 11 $80k-$99k 3 8 >$100k 4 13 indicates significant mean difference at p<.05 VOD allows viewers to watch television content on-demand and it is important to examine the importance of this capability between those who currently subscribe to VOD and those who do not currently subscribe. For all television programs, VOD subscribers

PAGE 81

72 tend to believe VODs viewing anytime feature is important compared with the non-subscribers. However, of the four programming choices, only the pay-per-view movie mean average for subscribers (M=5.21) was higher with statistical significance than the non-subscribers (M=3.71), (t=-2.86, d.f.=91, p=.003). Therefore, the result generally suggests that viewing anytime is important, especially for pay-per-view programs. Results can be found in table 4-30 for all television content program producers. Table 4-30. Viewing TV Content Anytime Comparisons between Subscribers and Non-subscribers of VOD Subscribe to VOD On-demand Programmning Yes No t df p Broadcast Network Programs Mean 5.17 4.96 -.468 91 .32 Std. Deviation 1.58 1.99 Cable Network Programs Mean 5.12 4.52 -1.45 91 .076 Std. Deviation 1.70 1.78 Premium Network Programs Mean 4.79 4.09 -1.38 91 .086 Std. Deviation 2.19 2.15 Pay-Per-View Programs Mean 5.21 3.71 -2.86 91 .003* Std. Deviation 2.14 2.23 indicates statistically significant mean difference at p<.05 One-tailed t-test The functionality of VOD, or the ability to stop, rewind, fast forward and pause television content was also measured between subscribers and non-subscribers of VOD. The mean score of importance of VOD functionality for subscribers (M=5.71) was than non-subscribers (M=4.86), (t=1.922, d.f.=91, p=.029). Therefore, this result confirms that the function of stop, rewind, fast forward, and pause is very important for VOD usage among VOD subscribers. The results can be found in Table 4-31.

PAGE 82

73 Table 4-31. VOD Functionality Comparisons between Subscribers and Non-subscribers of VOD Subscribe to VOD VOD Functionality Yes No t df p Mean 5.71 4.86 1.922 91 .029* Std. Deviation 1.40 2.01 indicates significant mean difference at p<.05 One-tailed significance Research Question 5 Research question 5 attempted to indicate factors contributing to intent to adopt VOD technology in the next six months. Respondents included in this analysis had to be aware of VOD but not current subscribers. Multiple regression analysis was used to indicate the relative power of variables to explain intent to adopt VOD technology. The independent variables measured included the sum of all the social/innovative characteristics scores, media use scores, technology ownership scores, importance of viewing television content anytime, importance of VOD functionality, gender, income, education and household size. The dependent variable was the rating of intent to adopt VOD technology over the next six months. Twenty-seven independent variables were measured yielding five statistically significant variables (R=.55, p<.001). The multiple regression model explained 30% of the variation of the mean toward intent to adopt VOD over the next six months. The most important variable to predict intent to adopt VOD was Internet Usage (Beta=.348, p=.002) followed by respondents having the ability to watch cable networks anytime (Beta=.279, p=.031). Interestingly enough, a negative relationship was found with those respondents who currently subscribe to premium cable (Beta=-.233, p=.044). The results indicate people are more likely to adopt VOD technology if they currently use the

PAGE 83

74 Internet, value watching cable programming at their own discretion but possible do not subscribe to premium cable services. The multiple regression results are listed in figure 4-32. Table 4-32. Multiple Regression Data Model Summary R R Square df F p .550 .302 5 5.2 .000 Variable B Std. Error Beta t p Internet Usage .292 ..091 .348 3.201 .002* Video Game System Ownership .528 .310 .19 1.704 .094 Premium Cable -.649 .316 -.054 -.417 .687 Watch Premium Channels Anytime -.036 .085 -.244 -2.284 .026* Watch Cable Networks Anytime .227 .103 .279 2.212 .031* indicates p<.05

PAGE 84

CHAPTER 5 SUMMARY AND CONCLUSIONS The current study adds to the body of knowledge regarding the diffusion of innovations theoretical framework applied to the adoption of new technologies. Additionally this study adds to the body of knowledge in regards to video-on-demand adoption and the findings display significant results for practitioners in the fields of advertising and telecommunication. The first research question looked to examine what aspects of VOD functionality were found to be most preferred among those respondents who are currently aware of VOD technology. This research question compared the average scores of those who have previously used VOD to those who have not previously used VOD. Aspects of VOD functionality include VCR type capabilities (ability to rewind, fast-forward, and pause video content) in addition to the ability to watch broadcast, cable, premium (HBO, Showtime), and pay-per-view content at the views discretion. This studys findings conclude that the VCR type functionality was rated the highest of all VOD attributes, although no statistical significant difference between users and non-users was found. However, preference among users was found to be statistically significantly higher involving the ability to watch premium network programming and marginally statistically significant among users in regards to pay-per-view programming. These findings suggest users prefer the VCR functions of VOD in addition to the ability to watch premium and pay-per-view video content. 75

PAGE 85

76 This study also compared the mean scores within of each aspect of VOD functionality among current users to examine if any statistical differences were present. The findings suggest VOD users prefer the VCR functionality above all other aspects of functionality. For the practitioner looking to communicate the attributes and benefits of using VOD, communicating the VCR functionality aspects of VOD could prove to be most positive. Perhaps a message tailored toward displaying or showing how the functionality works could prove to be a unique selling point for VOD. Additionally, another unique selling point could be the viewers ability to watch pay-per-view and premium content asynchronously. The findings from research question one add to the formulation for creating message strategy and this insight could aid in VOD positioning. Research question two examined respondent attitudes as to which exploratory prospective advertising model for VOD would be found to be most positive. Two items were measured. Respondent attitudes toward exploratory advertising models, based on the current pay-per-view format were examined in addition to respondent attitudes toward giving up personal information and receiving targeted advertising. The goal of the first part of research question two was to identify which exploratory advertising scenario would be most preferred. Comparisons were made between paying $2.99 for pay-per-view content containing10 minutes of commercial insertion advertising compared to paying $1.99 for 20 minutes of commercial insertion advertising. Comparisons were also made for paying $2.99 for ticker type advertising (during the entire pay-per-view event) compared to paying $1.99 for ticker advertising. Last, comparisons were made between each advertising scenario to discern any significant

PAGE 86

77 differences among scenarios. However, it should be noted all scores for each proposed advertising scenario were quite low, falling closer to not at all likely on the 1 to 7 scale with 1 being not at all likely and 7 being most likely. Total respondents rated the $2.99 for 10 minutes of commercial advertising the highest of the two commercial insertion scenarios. The finding was found to be statistically significantly different compared to the $1.99 for 20 minutes of commercial insertion scenario. As for the ticker type advertising scenarios, respondents rated paying $1.99 for ticker advertising higher than paying $2.99 for ticker advertising. This finding was statistically significantly different. Overall, respondents gave the $1.99 ticker advertising scenario a marginally higher rating but the rating was not statistically different from the $2.99 commercial advertising scenario. Overall these findings suggest respondent were not impressed by any of the proposed scenarios but they did find the $2.99 commercial insertion scenario and the $1.99 ticker scenario most preferred. Perhaps the practitioner should look to these advertising scenarios as examples of what not to do. Although the respondents found the $2.99 for 10 minutes of traditional commercial insertion and the $1.99 ticker scenario most favorable, both were still rated low and described by respondents as a least likely scenario. These four tested scenarios should be used as a boilerplate for further research and testing to find a more desirable or less-rejected advertising model. The second part of research question two involved the respondents willingness to give up personal information and receive targeted advertising in exchange for a discount on pay-per-view programming or other television services. Although the attitude scores were not found to be statistically significantly different, the scores do shed light on the

PAGE 87

78 respondents attitudes toward releasing personal information and receiving targeted advertising at a discount. In regards to releasing personal information for a discount and willingness to be exposed to targeted advertising, those respondents aware of VOD were not found to be any more or less willing to do either compared to those who are not aware of VOD. All scores were in the least likely range on the 1 to 7 scale and there were no significant differences found. For the practitioner, these results are not promising. Personal information is essential for creating and planning targeted advertising. Creating a message especially for a particular consumer will become more of a necessity as our current media landscape continues to fragment. Even with the fragmentation, technology allows us to reach consumers in new, non-traditional ways. Since most respondents stated they were least likely toward releasing personal information and receiving targeted advertising, practitioners should find other means or exchanges to collect this valuable data. The third research question attempts to discern the characteristics between those who are currently aware of VOD and those who are not aware of VOD technology. This study looked at respondent sample characteristics regarding social and innovativeness characteristics, media usage, technology ownership and demographic information to see how those who are aware of VOD differ from those who are currently not aware of VOD. In regards to social and innovativeness characteristics, this study could not find a single characteristic different between those who are aware and those who are not. As far as media use, those who are aware of VOD were found to use television and the Internet more compared to those who are not aware of VOD. This finding is consistent with

PAGE 88

79 Chan-Olmsted and Changs (2003) finding that Internet use was related to digital television knowledge. There were a number of significant differences found between those who are aware and those who are not aware in regards to technology ownership. Those who are aware of VOD were found to subscribe more to high speed Internet access, own more high definition televisions, own large screen televisions, own DVD players, subscribe to cable service with a set-top-box, and subscribe to premium cable services. Education level was the only demographic characteristic statistically different comparing those respondents aware of VOD with respondents unaware of VOD. Respondents who were aware of VOD were typically more educated, attaining undergraduate or graduate degrees. Research question four examined what factors indicate early adopters of VOD technology. The factors include social and innovative characteristics, importance of viewing content anytime, VOD functionality (VCR type control), media use, technology ownership and demographics. Scores for social and innovative characteristics were not found to be significantly different between current adopters and non-adopters. This was a surprise since many previous technology adoption studies have found a positive relationship between early adopters and varying degrees of social/innovative characteristics (Atkins & Jeffres 1998; Kang 2002; Lin 1998; Lin & Jeffres 1998; Neuendorf, Atkin, & Jeffres 1998; Eastlick, 1996). Regarding viewing videoing content asynchronously, early adopters rated viewing programs anytime higher than non-adopters, especially when considering pay-per-view programming. The functionality aspect of VOD (ability to fast-forward, rewind, pause) was also found to be an important factor to early adopters compared to non-adopters. For

PAGE 89

80 technology ownership, this study found non-adopters owned more cell phones, subscribe to cable service without a set-top-box, and subscribe to direct broadcast satellite. This technology aspect was consistent with Dupagne (1999) who found early adopters did not own more technology items but the finding is inconsistent with other studies where technology ownership was considerably higher for early adopters (Chan-Olmsted & Chang, 2003; Kang 2002). There were no significant differences found for media use between adopters and non-adopters which is consistent with some previous technology adoption studies where media use differences were weak overall (Atkin & Jeffres, 1996). As for demographics, only education and marital status were found to be significantly different between adopters and non-adopters. Sex was also found to have a marginal significant difference between adopters and non-adopters. These findings infer early adopters as more educated, married and male compared to non-adopters. These demographic findings are somewhat consistent with previous technology studies. Early adopters are typically more educated (Danko & MacLachlan, 1983; Atkins & Jeffres 1998; Kang 2002; Dupagne 1999; Neuendoft, Atkin & Jeffres, 1998) and typically male (Chan-Olmsted & Chang, 2003). The classic diffusion paradigm by Everett Rogers (1995) states early adopters seek information or innovations more actively, have a higher degree of opinion leadership and are more highly interconnected through interpersonal networks. This was clearly not the case in the current study due to the lack of difference in the social and innovative characteristics between those who are aware of VOD and those who are not. Rogers (1995) states early adopters have greater exposure to mass media. Here, the findings from this study were not consistent with Rogers since the overall scores for media usage were

PAGE 90

81 not above average for each media type; with newspaper use, radio use, magazine use, and movie going each rating lower for early adopters compared to non-adopters. Rogers (1995) and previous technology diffusion studies look to technology ownership as another key indicator and the current study was not consistent with those previous studies. Respondents aware of VOD were not more likely to own new technology items, signifying a more technological savvy group. As for the demographic findings, Rogers (1995) states early adopters have more formal years of education. This was true in the current study since those respondents aware of VOD were found to attain more years of education. Overall this study does add to the body of knowledge of diffusion research, staying consistent with some aspects of the paradigm, concluding the theory is consistent. For the practitioner, this study does add insight into ways to reach and target potential early adopters. This study found early adopters enjoy viewing programs on a flexible schedule and VOD functionality as especially important functions. Perhaps steps could be taken in a media campaign to highlight these important attributes. Additionally, the findings from this study infer VOD early adopters tend to be skewed more male, educated and married. These demographic findings could help to establish a potential target base for message design and targeting of the media campaign. Research question five attempts to pinpoint which factors are found to be most significant when trying to predict adoption of VOD technology within the next six months. Internet usage, the ability to watch cable programming anytime and subscribing to premium cable were found to be the most significant contributors. For the practitioner, these three variables could be excellent means of reaching and communicating the attributes and benefits of VOD to potential adopters. Potential

PAGE 91

82 adopters could be reached on the Internet or by advertising on cable. Additionally, premium cable services such as HBO or Showtime could communicate to subscribers to benefits of VOD for their programs. This study is not without limitations. First, the data was collected via convenience sample. This nonprobability recruiting measure was used due to lack of research funds. This method does not permit control over the representiveness of the sample or the ability to project these findings into the general population (Babbie, p. 179). Additionally, the sample was generally small and has a female skew. It is the researchers belief that the finding related to social and innovative characteristics suffered due to the fact all respondents were above average in these categories since the survey was administered on the beach. It is believed more outgoing and social people would venture to the beach compared to those who do not. This research also suffers from the same limitations found in other diffusion research as stated by Rogers (1995) regarding recall and the use of correlational analysis. Recall can be clouded by time which can affect the accuracy of respondents recall. Correlational analysis of survey data often leads to a sense of causality, if the operationalization of variables is not done correctly. Although the current study is based on the operationalization of variables from previous studies, some methodological differences are still present. Future research should look at VOD adoption utilizing random probability sampling techniques, looking at adoption longitudinally and perhaps redoing the study at a later date. First, using a larger, more robust probability sample would give the findings more validity and would allow these findings to be projected into the population. Second,

PAGE 92

83 implementing a longitudinal study would aid in confirming these findings. Last, with VOD penetration currently low and with rollouts planned in the future, perhaps more accurate data regarding awareness and adoption could be formulated at a later date. Overall, future research should help reduce the noted general limitations of diffusion research as well as add to the body of knowledge regarding diffusion of innovations research.

PAGE 93

APPENDIX VIDEO-ON-DEMAND TELEVISION SURVEY University of Florida College of Journalism and Communication Please answer all questions below except those you are asked to skip. Please give only ONE answer to each question, unless specified as choose all that apply. Mark your answer with an X in the bracket or write in the space provided. Your input will help us to understand the future of television entertainment. Thank you for your time. 1. Please tell us to what degree you agree with the following statements (On a 1-7 scale with 1 being strongly disagree and 7 strongly agree). Strongly Disagree Strongly Agree I like to try 1 2 3 4 5 6 7 new products I like to explore 1 2 3 4 5 6 7 new technologies I like to participate 1 2 3 4 5 6 7 in social activities I enjoy interacting 1 2 3 4 5 6 7 with my neighbors I am in good 1 2 3 4 5 6 7 shape financially I like to learn 1 2 3 4 5 6 7 about new ideas I keep up with 1 2 3 4 5 6 7 new technologies I am willing to take risks in order to 1 2 3 4 5 6 7 try new things 84

PAGE 94

85 2. How important is it for you to be able to view your favorite (or any) program at ANYTIME from the following sources? (Rate on a 1-7 scale with 1 being not important at all & 7 being very important) Not Important at All Very Important Broadcast Program Choices 1 2 3 4 5 6 7 (e.g. ABC, NBC, FOX, CBS) Cable Program Choices 1 2 3 4 5 6 7 (e.g. CNN, MTV, ESPN, HGTV) Premium Program Choices 1 2 3 4 5 6 7 (e.g. HBO, Showtime, Starz) Pay-Per-View or 1 2 3 4 5 6 7 New Release Movies Please Turn Page Over 3. How important is having the ability to pause, fast forward, rewind and stop television programs at anytime much like you would if you were viewing a videotape? (Rate on a 1-7 scale with 1 being not important at all & 7 being very important) Not Important Very Important 1 2 3 4 5 6 7 4. What is your usage of the following media? (Rate on a 1-7 scale with 1 being not a user at all & 7 being a heavy user) Do Not Use at All Use Very Heavily 1 2 3 4 5 6 7 Newspaper Radio Television Magazine Video Rental Movie going Internet

PAGE 95

86 5. Are you aware of video-on-demand television services? YES NO If NO , Please skip to question 6. If you answered YES to the above question, please answer the next four questions. How knowledgeable are you about video-on-demand? (Please answer using the 1-7 scale with1 being not at all knowledgeable and 7 being highly knowledgeable) Not at all Highly Knowledgeable Knowledgeable 1 2 3 4 5 6 7 Have you ever used video-on-demand before? YES NO Do you currently subscribe to video-on-demand type services? YES NO If you answered NO what is your intention to subscribe to video-ondemand type services in the next six months? (On a 1-7 scale with 1 being not at all likely and 7 being most likely to subscribe) Not at Most All likely Likely 1 2 3 4 5 6 7 Please Turn to Next Page 6. Based on pay-per-view movies available through your cable or satellite television provider, please indicate your attitude toward the following questions: Would you be willing to provide personal information about your household (such as income, occupation, and hobbies) to advertisers for a discount on pay-per-view movies or cable services? Not at all likely Most likely 1 2 3 4 5 6 7 Would you be willing to be exposed to advertising of products or services that might interest you for a discount on pay-per-view movies or cable services? Not at all likely Most likely 1 2 3 4 5 6 7

PAGE 96

87 7. Based on pay-per-view movies available through your cable or satellite television provider, please indicate your attitude toward the following scenarios and questions: If the price of a new release, pay-per-view movie were $3.99, would you be willing to: Pay $2.99 for the same movie in exchange for being exposed to 10 minutes of commercials spread throughout the movie (approximately 20 television commercials)? Not at all likely Most likely 1 2 3 4 5 6 7 Pay $1.99 for the same movie in exchange for being exposed to 20 minutes of advertising spread throughout the movie (approximately 40 minutes of commercials)? Not at all likely Most likely 1 2 3 4 5 6 7 Pay $2.99 for the same movie with text advertising messages moving across the bottom of the screen during the entire length of the movie? Not at all likely Most likely 1 2 3 4 5 6 7 Pay $1.99 for the same movie with text advertising messages moving across the bottom of the screen during the entire length of the movie? Not at all likely Most likely 1 2 3 4 5 6 7 Please Turn Page Over

PAGE 97

88 8. Please check the items or services that you currently own or subscribe to (Please check ALL items you own or subscribe to): Digital Video Recorder (e.g., Tivo, Replay TV, or other Personal Video Recorder) Personal Computer High-Speed Internet Video Game System Cell Phone Home Theater System High Definition Television Large Screen TV (35 or larger) DVD Player Personal Digital Assistant (e.g. Palm Pilot or HandSpring) Cable TV service with set-top box Cable TV service without set-top box Direct Broadcast Satellite (e.g. Direct TV or DishNetwork) Premium Cable TV Channels (e.g. HBO, Showtime, Starz) 9. Please tell us about yourself: Gender: Male Female Highest completed education: Less than high school High School College Graduate School Marital Status: Single Married Divorced Widow Other Age: I was born in 19___ Household Size (including you) ___ Occupation: Laborer Machine or Service Worker Craftsman Professional Clerical Sales Administrator Manager Other Household income: Less than $19,999 $20,000-$39,999 $40,000-$59,999 $ 60,000-$79,999 $80,000-$99,999 $100,000 or more Thank You Very Much for Your Time.

PAGE 98

REFERENCES Angwin, J., Grant, P., & Wingfield, N. (2004). Hot-Button Topic: In Embracing Digital Recorders, Cable Companies Take Big Risk; Viewers Flock to the Devices, But Advertisers May Flee; Debating Ad-Skip Feature; Time Warners Meteorite. The Wall Street Journal, April 26. A-1. Applebaum, S. (2004). Interactive TV Shows Signs of Life. Advertising Age, May 8. 75, 10-12. Atkin, D. J., & Jefferes, L. W. (1998). Understanding Internet Adoption as Telecommunication Behavior. Journal of Broadcasting & Electronic Media, 42, 4. Babbie, E. (2001). The Practice of Social Research: 9 th edition. Wadesworth, Thomason Learning, United States. Baron, S. J., & Davis D. K. (2003). Mass Communication Theory: Foundations, Ferment, and Future (3 rd edition). Wadsworth, Belmont CA. Baumgartner, J. (2002). Digital Dangle. CED, August 1, 14. Bass, F. M. (1969). A New Product Growth for Model Consumer Durables. Management Science, 16, 215-227. Bass, F, M., Gordon, K., Ferguson, T. L., & Githens, M. L. (2001). DIRECTV: Forecasting Diffusion of a New Technology Prior to Product Launch. Interfaces 31:3, 82-93. Brister, C. (2001). Video On Demand; Hollywood Puts Cable Operators On Hold Over Licensing Fees. Atlanta Journal and Constitution, September 26, 5F. Brown, P. J. (2002). Cable Looks to the Heavens. Broadcasting & Cable, November 18. Cummings, J. (2002). Video On Demand Lets Cable Subscribers Have It Their Way. Cox News Service, November 18. Danko, W. D., & MacLachlan, J. M. (1983). Research to Accelerate the Diffusion of a New Invention. Journal of Advertising Research, 23, 3. 89

PAGE 99

90 Davis, J.J. (1997). Advertising Research: Theory and Practice. Upper Saddle River, NJ: Prentice Hall. Dupagne, M. (1999). Exploring the Characteristics of Potential High-Definition Television Adopters. Journal of Media Economics, 12(1), 35-50. Eastlick, M. A. (1996). Consumer Intention to Adopt Interactive Teleshopping. Marketing Science Institute, report summary # 96-113. Ferguson, D.A., & Perse, E. M. (2001). Enhanced Television Viewing with Digital Video Recorders (DVRs): Audience Satisfaction in an Asynchronous Television Environment. Paper submitted in the Open Competition, to the Communication Technology Division of the Association for Education in Journalism and Mass Communication. Figler, A. (2002). Movies Lead, All Others Follow. Cable World, October 28, 40. Iler, D. (2001). VOD Shining Brightly in Cable Universe. Broadband Week, August 6, 18. Jacobs, R. (1995). Exploring the Determinants of Cable Television Subscriber Satisfaction. Journal of Broadcasting & Electronic Media, 39, 2. Jeffres, L.W., & Atkin, D.J. (1996). Predicting Use of Technologies for Communication and Consumer Needs. Journal of Broadcasting & Electronic Media, 40, 318-330. Jordan, J. (2001). HBO Video On Demand Tests Columbia Market. Associated Press State and Local Wire, June 29. Kang, M. (2002). Digital Cable: Exploring Factors Associated With Early Adoption. Journal of Media Economics, 15(3) 193-207. Klopfenstein, B.C., & Spears, S.C. (1991). VCR Attitudes and Behaviors by Length of Ownership. Journal of Broadcasting & Electronic Media, Volume 35, Issue 4. Lin, C. A. (1994). Audience Fragmentation in a Competitive Video Marketplace. Journal of Advertising Research, 34, 6. Lin, C. A. (1998). Exploring Personal Computer Adoption Dynamics. Journal of Broadcasting and Electronic Media, 42, 95-112. Lin, C. A., & Jeffres, L. W. (1998). Factors Influencing the Adoption of Multimedia Cable Technology. Journal of Mass Communication Quarterly, 75(2), 341-352. Mahajan, V., & Muller, E. (1990). New Product Diffusion Models in Marketing: A Review and Directions for Research. Journal of Marketing, 54, 1.

PAGE 100

91 Neuendorf, K.A., Atkin, D. J., & Jeffres, L. W. (1998). Understanding Adopters of Audio Information Innovations. Journal of Broadcasting and Electronic Media, Winter 1998, 80-93. National Telecommunications and Information Administration (NTIA) (2002). A NATION ONLINE: How Americans are Expanding their Use of the Internet, Washington, D.C. February. http://www.ntia.doc.gov/ntiahome/dn/index.html Precision Marketing (2004). Nissan Taps into TV Shows for Interactive Ad Campaign. Precision Marketing. London, March 12, 3. PR Newswire (2002). Comcast and TVN Expand VOD Relationship; TVNs ADONISS System to Support VOD Rollout in Philadelphia Market. PR Newswire Association. July 15. Rizzuto, R.J. & Wirth, M. O. (2002). The Economics of Video On Demand: A Simulation Analysis. The Journal of Media Economics, 15(3), 209-225. Rogers, E.M. (1976). New Product Adoption and Diffusion. Journal of Consumer Research. 2 March 1976, 290-301. Rogers, E.M. (1995). Diffusion of Innovations (4 th edition). New York: Free Press. Romaso, A. (2002). VOD On ESPN. Broadcasting & Cable, September 19. Ryan, B., & Gross, N. C. (1943). The Diffusion of Hybrid Seed Corn in Two Iowa Communities. Rural Sociology, 8: 15-24. Scanlon, M. (2002). Ramping Up Isnt Likely. Cable World, October 28, 40. Schiesel, S. (2002). Video on Demand Is Finally Taking Hold. New York Times, November 25, section C, page 4, column 1. Smith, J. (2001). Video On Demand Awaits Economic Upturn. Rocky Mountain News, December 31, p1B. Stump, M. (2002). Comcasts VOD Approach: Keep the Package Simple. Multichannel News, March 11, 1. Stump, M. (2002a). Showtime, Starz Tie SVOD Bundles. Multichannel News, July 15, 46. Stump, M. (2003). Ads on SVOD? Works for Magazines. Multichannel News, March 10, 2003.

PAGE 101

92 Tarde, G. (1903). The Laws of Imitation, translated from the second edition by Elsie Clews Parsons: New York, Holt & Company. URL Wire (2003). May Bandwidth Report US Broadband Penetration Breaks 35%. http://www.urlwire.com/news/060903.html. Williams, G., Gabriel Tarde and the Imitation of Deviance. accessed from Google November 23, 2002. http://jom-emit.cfpm.org/2000/vol4/marsden_p.html

PAGE 102

BIOGRAPHICAL SKETCH Stephen Marshall researches culture and technology issues and differences in regards to consumer advertising and telecommunication. Stephens background consists of four years with Nielsen Media Research and various other positions in the fields of telecommunication production. In addition to his other fancy jobs, Stephen has also cooked chicken and washed dishes to make a living. Stephen is continuing his education at the University of Florida with goals of receiving a doctorate and teaching others what little he knows. Have a nice day and bye-bye. 93


Permanent Link: http://ufdc.ufl.edu/UFE0003280/00001

Material Information

Title: An Exploratory Video-on-Demand Analysis: Identifying Early Adopters and Attitudes toward Potential Advertising
Physical Description: Mixed Material
Copyright Date: 2008

Record Information

Source Institution: University of Florida
Holding Location: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
System ID: UFE0003280:00001

Permanent Link: http://ufdc.ufl.edu/UFE0003280/00001

Material Information

Title: An Exploratory Video-on-Demand Analysis: Identifying Early Adopters and Attitudes toward Potential Advertising
Physical Description: Mixed Material
Copyright Date: 2008

Record Information

Source Institution: University of Florida
Holding Location: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
System ID: UFE0003280:00001


This item has the following downloads:


Full Text












AN EXPLORATORY VIDEO-ON-DEMAND ANALYSIS:
IDENTIFYING EARLY ADOPTERS AND ATTITUDES TOWARD POTENTIAL
ADVERTISING















By

STEPHEN WILLIAM MARSHALL


A THESIS PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
MASTER OF ARTS IN MASS COMMUNICATION

UNIVERSITY OF FLORIDA


2004

































Copyright 2004

by

Stephen William Marshall

































This document is dedicated to the graduate students of the University of Florida as well
as to my parents and family for all their love and support. Most importantly, this research
is dedicated to couch potatoes across the globe and to my lovely doggie, Ozwaldo.















ACKNOWLEDGMENTS

First, I would like to thank my wonderful committee for not only guiding me

through this research but also for challenging me. The faith you all have shown in me I

often did not have in myself. Additionally, I thank my committee for exposing me to a

career path unknown to me. I will always remember and thank each of you for this

adventure.

Second, I would like to thank my family and friends for supporting me through this

process. I love and appreciate all the support through my good and bad times. These have

been changing times and I love each of you dearly.

Last, I wish the world peace. God is love.

















TABLE OF CONTENTS




A C K N O W L E D G M E N T S ................................................................................................. iv

LIST OF TABLES ...... ......... ...... ...................... ............vii

ABSTRACT.................. .................. ix

CHAPTER

1 INTRODUCTION ................... ............................ .......... .. .......... 1

2 LITERATURE REVIEW .................................................. ...............5

The Video-On-Demand Background..............................................5
Theoretical Grounding.............................................. 17
Diffusion of Innovations........................................................ ......... .. ...... 18

3 M E TH O D O L O G Y ..............................................................40

Data Collection ................................................40
Survey Design..........................................41
Data Analysis........................................44

4 RESULTS .................................................45

Descriptive Statistics .................................................... ........45
Research Question 1 ................................................53
Research Question 2 ................................................... .........55
Research Question 3 ................................................60
Research Question 4 ................................................... .........66
Research Question 5 ........................................ .........73

5 SUMMARY AND CONCLUSIONS ...................................... ..............................75

APPENDIX VIDEO-ON-DEMAND TELEVISION SURVEY............... ...............84

REFEREN CES ....................................................... 89

BIOGRAPHICAL SKETCH .................................................. ........93



v


































































6
















LIST OF TABLES

Table page

4-1 A ge D distribution ................... .......................................................... ........ ...............45

4-2 Incom e D istribution............................................46

4-3 M marital Status ................ ............................... 47

4-4 O occupation D distribution ................................................. ............... 47

4-5 Education................................ ........48

4-6 Awareness, U se and Subscribers of VOD ............................................................. 48

4-7 V O D Functions ....................................................... 49

4-8 Advertising Scenarios and M odels.................................. ................. 50

4-9 Media Use .......................................................51

4-10 Social and Innovativeness Characteristics ........................................52

4-11 Technology Ow nership ................................................ ............... 53

4-12 Preferred VOD Options Between Users and Non-Users of VOD............................54

4-13 Mean Comparison of VOD Options by VOD Users..............................................55

4-14 Ad Scenario #1 ..................... ................................56

4-15 Ad Scenario #2 ........................ ............ ........ ...............57

4-16 A d Scenario #3 ...................... ................................57

4-17 A d Scenario #4 ................................................................58

4-18 Ad Scenario #5 ........................ ............ ....... ...............58

4-19 A d Scenario #6 ................................................................58

4-20 Willingness to Provide Personal Information ................. ............................59









4-22 Mean Media Use Scores Between Those Aware and Not Aware of VOD...........61

4-23 Social/Innovative Mean Scores Between Those Aware and Not Aware of VOD ...62

4-24 Technology Ownership Mean Scores Between those Aware and Not Aware of
VOD .............................................................. 64

4-25 Demographic Scores Between those Aware and Not Aware of VOD...................64

4-26 Media Usage Comparisons between Subscribers and Non-subscribers of VOD.....67

4-27 Social/Innovative Characteristics Comparisons between Subscribers and Non-
subscribers of VOD ................. ...... ..................................... ...... 68

4-28 Technology Ownership Comparisons between Subscribers and Non-subscribers
of VOD ...................................... .................................. ........ 69

4-29 Demographics Comparisons between Subscribers and Non-subscribers of VOD ..70

4-30 Viewing TV Content Anytime Comparisons between Subscribers and Non-
subscribers of VOD ................. ...... ..................................... ...... 72

4-31 VOD Functionality Comparisons between Subscribers and Non-subscribers of
V O D ...............................................................................7 3

4-32 Multiple Regression Data..............................................74















Abstract of Thesis Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Master of Arts in Mass Communication


AN EXPLORATORY VIDEO-ON-DEMAND ANALYSIS:
IDENTIFYING EARLY ADOPTERS AND ATTITUDES TOWARD POTENTIAL
ADVERTISING

By

Stephen William Marshall

August 2004

Chair: Chang-Hoan Cho
Major Department: Mass Communications

This study attempts to identify consumers who are currently aware and likely to

adopt video-on-demand technology. This research also studies attitudes toward possible

advertising models within the video-on-demand business. Diffusion of innovations was

the theoretical grounding for this research. A survey was used to collect the data and

statistical analysis was used to conclude results. Results found respondents did not like

proposed advertising models and were highly unlikely to exchange personal information

for a discount on services. Additionally, results from the survey found those interested in

video-on-demand are younger and better educated.














CHAPTER 1
INTRODUCTION

Video-on-demand (VOD) technologies have come a long way in the last decade.

From the early testing done in the 1990s to the penetration in major cities across the

United States today; VOD is here to stay. After years of planning and failed promises, the

cable industry is finally able to deliver reliable and economical VOD services (Schiesel,

2002).

With the recent news of News Corp's purchase of DirecTV and Comcast's offer for

Disney, it appears that interactive TV is finally turning into a reality. Three major cable

operators heavily involved in interactive television ventures include Charter

Communications, Cablevision Systems Corp. and Insight Communications Co. These

companies are relying on these services to help generate more subscriptions to VOD,

interactive television and high-definition services (Applebaum, 2004).

Video-on-demand will become a major player in interactive television, video

entertainment and education thanks to the technological advances of the digital cable

industry. Cable companies have spent over $50 billion upgrading networks from analog

to digital, and cable companies need a return on their hefty investment (Brister, 2001).

Video-on-demand technologies will give cable companies a strategic advantage and

point-of-difference when compared to competing services. VOD will be a savor for the

cable industry and will give digital cable a competitive advantage when compared to

satellite or other delivery platforms.









In today's high-tech world, we take for granted the luxury of being able to watch

what we want to watch, when we want to watch it (Schiesel, 2002). Video-on-demand

allows us to do just that. The introduction of the VCR and later the DVD gave us the

ability with some degree but each are not without consequences and drawbacks.

The VCR was often difficult for the user to master. The VCR was designed to have

many recording features but uses became limited due to lengthy manuals and design

complexities. In addition, the VCR involved extra connections to the television and the

purchase of videotapes for recording was also required. Even with these difficulties, the

VCR received widespread adoption because viewers enjoyed pre-recording programs for

later playback and renting poplar movies. In fact, the VCR created an entire new industry

of home entertainment and viewing flexibility, essentially bringing the movie theater

environment into the home.

The DVD has followed the same entertainment footsteps of the VCR and has

widely been adopted due to the technology's ease of use and increased video and sound

quality. The DVD adds the digital quality aspect to video entertainment, providing sharp

pictures, improved sound, improved ease of use and extra search and scene features not

found in VCR technology. The main drawback for the DVD is the machine is typically an

"output" machine enabling viewers to only watch rented or purchased prerecorded

material (although newer models currently allow for recording but these models are still

quite expensive).

Video-on-demand offers these same qualities with even greater improvements.

VOD is easy to use, has the digital quality of DVDs, creates an asynchronous

environment (much like the VCR) and requires no additional connections or equipment.









The technology is accessed through the same cable set-top box used to receive cable

programming. The user accesses VOD with their cable remote and the VOD menu

resembles the cable and DVD on-screen guide already familiar to many subscribers.

Based on past performance and relative similarities, it appears the next wave of home

video technology will come in the form of VOD.

Cable companies, through the groundwork of digital cable are planning to deliver

VOD technology to subscribers and technology experts feel adoption will occur quickly.

The Yankee Group, a technology research firm, estimates by the end of 2002 about seven

million homes around the nation will have access to VOD. This is an increase of 100

percent from the three million VOD households in 2001 (Schiesel, 2002). The Yankee

Group also estimates VOD revenues for U.S. operators will be just under $2 billion by

the end of 2005 (Iler, 2001).

VOD has also received positive results from subscribers in current test markets. In

Cleveland Ohio, Adelphia Cable currently has 50,000 customers able to receive VOD

services. Of those customers, 35,000 have used the service. Other cable operators such as

Comcast, AT&T, and Time Warner Cable have begun to launch or test VOD services

hoping to use VOD as a competitive advantage to satellite and broadcast, and other

competing technologies.

This exploratory research has two goals. First, to use the constructs found in the

diffusion of innovations literature to identify the early adopters of VOD. Diffusion of

innovations theory has been widely used and accepted as a means to identify the early

adopter variables within a social system and predict characteristics of early adopters. The









second goal of this research is to measure the attitudes of consumers in regards to

possible advertising scenarios within the VOD structure.

Specifically, this exploratory research will explore these research questions:

* What types of consumers are currently aware of VOD technology?
* What factors indicate early adopters of VOD technology?
* What factors indicate intent to adopt VOD technology?
* What functions of VOD technology are found to be most preferred?
* What is the most positive prospective advertising model for VOD?


This research has great significance for academics as well as practitioners. For

academics, this research will add to the body of knowledge regarding diffusion of

innovations theory as applied to digital technology. For the practitioner, identifying the

potential VOD adopter is critical for positive product roll out and strategic planning. By

identifying the traits and characteristics of potential early adopters, companies can target

the early adopters of VOD more aggressively. AOL Time Warner Chief Executive

Officer Gerald Levin sums up the significance of this research by stating, "I believe all

television will be distributed in either real time or on demand. It is the ultimate in

consumer choice, convenience and control" (Jordan, 2002.)














CHAPTER 2
LITERATURE REVIEW

The Video-On-Demand Background

The digital cable environment is extremely competitive and VOD technology

gives digital cable a distinct competitive advantage. Digital cable is dealing with

competing pressures from digital satellite, analog cable and broadcast television.

Currently only digital cable can deliver the VOD technology, giving digital cable a

strategic point-of-difference competitive advantage.

Although digital cable offers more channels than analog cable, the amount of

channels has not been alluring enough to encourage many consumers to upgrade their

service to digital (Brister, 2002). In addition, satellite is stealing many high-paying cable

subscribers from digital cable since satellite can offer the same amount of programming.

Digital cable marketers need to integrate the VOD application to add value and reduce

consumer turnover (Schiesel, 2002).

VOD technology can reduce churn or numbers associated with people dropping the

digital service by creating a point-of-difference. Digital churn continues to be a problem

for digital cable and the VOD competitive advantage will help companies keep or

upgrade their current customers because the service is not available to satellite or analog

systems (Stump, 2002). VOD technology offered by digital cable creates an incentive for

analog users to make the switch and upgrade their current service (Cummings, 2002). A

seven month VOD trial by Time Warner Cable in the Cincinnati market found 67% of the

digital subscribers were more likely to retain their service because of free Scripps









Network (DIY, Food Network, HGTV) on-demand content (Figler, 2002). "Free-on-

demand" or FOD is being used as a digital chum reducer (Baumgartner, 2002). Also,

FOD is being used to induce trial of VOD programming and technology, enabling the

viewer to experience the added benefits of the service without a fee (Stump, 2002).

The first major test of video-on-demand was an experiment conducted by Time

Warner Cable in late 1994. Time Warner used Orlando Florida as a test market. The

equipment was not advanced and was much more expensive than the equipment used

today. In 1994, Time Warner's cost per stream (per household) was around $13,000 with

each set-top box costing over $5,000 (Schiesel, 2002). It was not economically feasible

for cable companies to continue VOD service at this high cost rate. These costs are much

lower today due to video compression, digital infrastructure, and overall improvements

involving equipment.

In order for VOD to work, cable operators spent millions of dollars updating their

systems to provide two-way communication from the cable headend to the cable

subscriber. The current cost per stream for VOD service is about $475 and the cost is

estimated to fall below $300 by early 2004. The two-way communication from digital

technology overcomes the barriers of analog one-way communication, which

traditionally limited the cable subscriber to receiving to the same programming the

operator was sending the other subscribers (Schiesel, 2002).

The VOD functionality offers viewers an asynchronous environment for viewing

video entertainment and information. Users of VOD technology can access programs at

anytime and are able to stop, fast forward, and rewind programs much like the features

found on a traditional VCR, DVD, or digital video recorder (DVR). VOD delivers









exceptional entertainment quality comparable to the audio and video quality found on

DVD and offered through digital cable.

Selecting a VOD program is much easier than renting a DVD or setting the typical

VCR to record. To select a VOD program, the viewer selects a program title from the

electronic program guide with her or his cable remote control. By selecting the program,

a request for the program is placed and sent to the cable system head-end. The request is

then processed by the VOD server and recorded by the cable system. After the request is

processed, the video is streamed from the storage library at the cable headend, to the

network, and finally to the household. Although this sounds involved, all actions occur in

real time and the viewer receives the video content moments after making the selection

(Brister, 2001).

Since VOD allows the television viewer new possibilities and functionalities when

navigating through programming, the first research question will address the following:

RQ1: What functions of VOD technology are found to be most preferred?

Cable companies are currently offering VOD services in three varieties: free-on-

demand (FOD), movies-on-demand (MOD), and subscriptions-on-demand (SVOD).

Free-on-demand offerings can be viewed immediately without additional cost. Movies-

on-demand involve recently released movies and resemble some of the same features of

earlier pay-per-view and video store rentals. Subscription-on-demand allows viewers

choose when they will watch programs from premium channels (HBO or Showtime) or

SVOD offers additional content from the previously mentioned free-on-demand services

(Cummings, 2002). Lastly, there has been some industry discussion to the idea of

education-on-demand selections but these selections could fall in one of the previously









mentioned categories, based on the cost or package offered to the consumer (Figler,

2002).

Cable systems are test marketing the VOD product in a variety of ways. This

research found Comcast Corp., Time Warner Cable, and Cablevision doing the most

publicized tests, although others are involved. It is important to review these tests to

strategically identify variables for developing potential advertising models.

Comcast has done the most extensive testing published to date. Comcast is

headquartered in Philadelphia and is a developer, manager, and operator of broadband

cable networks providing basic cable, digital cable, and high-speed Internet service.

Comcast is the third largest cable system in the U.S. and serves more than 8.5 million

subscribers in six geographic regions (PR Newswire, 2002). Comcast is quickly rolling

out VOD technology in the Philadelphia market where the company is in the process of

building servers and infrastructure to offer 1,500 hours of on-demand programming.

They hoped to have the VOD system ready to go by the end of 4th quarter 2002

(Baumgartner, 2002).

Comcast's VOD pricing strategy is meant to drive digital penetration in basic-only

households through low-cost programming tiers. One tier entitled, "on-demand classic"

will cost $9.95 a month and include content from several basic cable networks including

a kids package of channels (Nickelodeon, Disney, and Cartoon Network), 15 basic

networks, electronic program guide, digital music channels and discounts for movie-on-

demand purchases. The second package entitled, "On-demand Plus" would sell for

$14.95 a month and include the same as "classic" along with SVOD packages from HBO,

Showtime, Starz Group LLC, as well as 40 basic channels. Currently, Comcast has some









form of VOD up and running in 19 markets, covering 3 million homes and 600,000

digital subscribers (Stump, 2002).

Time Warner Cable (TWC) has also announced a VOD roll out. By the end of

2002, TWC stated VOD would be available throughout New York City, TWC's largest

cable market. TWC has 1.2 million subscribers in New York City and plans to have 1,300

hours of programming available at any one time. To put this amount of programming in

perspective, 1,300 hours of programming is the equivalent of almost two months of

watching television (Schiesel, 2002).

Cablevision Systems Corp is adding value to its digital tier with free-on-demand

content from the third party content provider Mag Rack along with VOD affiliations with

Fox and local PBS affiliates (Baungartner, 2002).

Viewing and usage data for VOD was difficult to locate since most reports are

proprietary or require a large amount of money to view. At the time of this research,

some data was available regarding the viewing habits of Starz VOD subscribers.

According to Starz vice president Greg DePrez, early research shows there is as much

usage of VOD during the day as in primetime. VOD usage tends to peak around 3pm,

presumably when the children return from school and peak again at 8pm. VOD activity

for Starz tends to be heavier on the weekends compared to weekdays. These weekend and

weekday viewing findings differ from the habits found in the typical television viewer

(Stump, 2002a).

Content providers have a variety of reasons for getting involved with VOD service.

VOD is the ideal platform to increase brand awareness for cable networks and

programming distribution (Figler, 2002). Many television-programming networks have









prepackaged content ready for delivery via VOD while other companies will strive to

make content specifically tailored for VOD service. Strong cable brands will help drive

VOD usage and cable operators must form strategic relationships with these brands

(Stump, 2002). While VOD is a programming service not subject to the time restraints of

the typical programming model, cable operators and content providers must remain

sensitive to basic cable viewing patterns (Stump, 2002). The typical VOD user will not

use VOD all the time.

From current industry research, content providers fall into three categories:

premium content providers, basic cable content providers and niche network content

providers. Each offers different content for the VOD service and each will fulfill a

different viewer need.

The premium content provider is made up of traditional premium cable brands such

as HBO, Showtime Inc., Starz, and the Movie Channel as well as adult entertainment

providers such as Erotic Networks and Playboy.

Both HBO and Showtime currently offer VOD content in various test markets.

HBO offers a subscription-on-demand service and has tested fees of $3.95, $6.95 and

$9.95 for the service. For HBO, the service charge is in addition to the typical fee

charged to the viewer for having access to HBO. These prices are currently being tested

on Time Warner Cable systems and the prices vary based on the market demand (Stump,

2002a).

Showtime expects to have more than 35 launches of VOD by the end of 2002. The

launch agreements are with several cable companies including Cox Communications,

Charter Communications, Time Warner Cable, Comcast, Cablevision, Adelphia and









Insight Communications. "Showtime on Demand" will carry about 120 hours of

programming per month consisting of 45 percent series programs, 20 percent family

programs, and 15 percent adult programs (Stump, 2002a).

Adult programming has always been a staple of the pay-per-view offerings found

on a cable system. On average, adult programming nets cable systems at least 60 cents

per subscriber compared to the 15 cents per subscriber generated by new release movies.

Erotic Networks is a leader in adult on-demand content and is now carried on five of the

top ten cable systems. Erotic Networks plans to expand the network's reach to about 4

million homes and the network plans to offer about 60 hours of programming per month

(Figler, 2002).

In competition with Erotic Networks is the Playboy Network. Playboy currently

offers about 70 hours of on-demand content per month. The Playboy Network currently

reaches about 60 percent of the VOD enabled homes nationwide (Figler, 2002).

It would be much harder to find a cable network not planning on offering some

form of VOD content. Cable networks already providing VOD content include but are

not limited to A&E, AMC, BBC, CNN, Court TV, Discovery, DIY, Erotic Networks,

ESPN, Food Network, Fox Sports, FX, Golf Channel, Hallmark, HGTV, History, IFC,

MTV Networks, National Geographic, NBC, Outdoor Life Network, Oxygen, Speed,

Sundance, TBN, TBS, TMC, WE, and the Weather Channel (Figler, 2002).

Some basic cable content providers are giving content away to help induce VOD

usage while other companies are looking for a monthly fee. For example, the Scripps

Network, including cable brands HGTV, Food Network, DIY and Fine Living has agreed

with Time Warner Cable to provide 10 hours per month of free VOD content. The deal









was meant to introduce customers to VOD (Figler, 2002). Free access to this VOD

programming will induce trial, help build loyalty for the Scripps brands and should

benefit both parties.

ESPN plans to license 150 hours of library programming and replays of some

college football and college basketball games to VOD distributor TVN Entertainment

Corp. However, not all ESPN content will be free and the network is still working on

how it will provide content to distributors (Romaso, 2002).

The Discovery Network has publicly discussed how it will distribute its large

amount of content and Discovery could become the VOD model for other popular cable

brands. Discovery is offering two tiers of programming. The first tier is a free service

called "Discovery Choice 10." Choice 10 will offer a menu of ten different programming

categories and will be advertiser supported. The Discovery Choice 10 package will have

roughly 25 hours of programming and will be refreshed on a monthly basis

(Baumgartner, 2002). Discovery's other service entitled, "Discovery On Demand,"

could have up to 500 titles pulled from Discovery's 70,000 hour library. Discovery On

Demand would be a subscription service although pricing has yet to be publicly discussed

(Figler, 2002).

VOD offers smaller cable networks and content providers, lacking the

programming to fuel an entire network, the opportunity to build their brand due to the

localized nature of the VOD distribution networks. Because VOD is a requested

programming format, niche providers can tailor content to fulfill subscriber needs. Three

examples of niche networks currently providing content to cable systems are Tech TV,

Chaos Media Networks, and Mag Rack.









Tech TV is an upstart cable network with repurposed programming especially

tailored for the VOD audience. Tech TV plans to create instructional videos that

consumers can access at anytime. A Tech TV program may teach the viewer how to set

up a home computer network or what type of digital camera to buy (Figler, 2002).

Another niche content provider is Chaos Media. Chaos Media focuses on non-

movie genres such as education, fitness, household help and health. Chaos Media will

offer custom exercise and diet plans as well as custom tracking for the plans. The

Academy Channel, through Chaos Media, offers training and certification courses

(Figler, 2002).

The niche-programming provider receiving the most press is Mag Rack. Mag rack

claims to be offering content in categories underserved in the cable marketplace. Mag

Rack's programming is based on relationships with over 30 different magazines.

Programs are specific to special topics such as, "Classic Cars," "Birdsight," and

"Cooking with the Pros." Since the start of the company, Mag Rack has cost over $50

million to develop and the programmer uses original and acquired programming. Mag

Rack charges a license fee per each VOD subscriber, much like traditional cable

networks. Currently, Cablevision has allocated about 150 hours of storage capacity to

Mag Rack content alone (Baumgartner, 2002).

This research has discussed cable television providers, cable programming

providers, and niche content providers but has yet to discuss the third party intermediary

connecting all three. If deals are not made between the content provider and the cable

system, a third party vendor will undoubtedly provide content and VOD system support.









Third party vendors such as In Demand or TVN will bridge technology and content for

the VOD subscriber.

In Demand is the world's largest provider of VOD and pay-per-view programming.

In Demand works with approximately 1,900 affiliated systems and provides

programming for the NBA, NHL, MLB, and NASCAR in addition to first-run movies

and other professional sporting events (www.indemand.com/about/who). Shareholders

for In Demand include AT&T, Time Warner Entertainment, Comcast and Cox. In

Demand currently has 2.5 million subscribers and expects to double the number by the

end of 2002. In Demand also has agreements with content programmers such as Court

TV, ESPN, Fox, Hallmark, Sesame Workshop and Turner. In addition to sports and cable

programming deals, In Demand has VOD distribution deals for movies. Movie producers

holding agreements with In Demand include Dream-Works, MGM, Sony Columbia

TriStar, Twentieth Century Fox and Universal. Currently, In Demand has access to 75

percent of Hollywood's recent releases in addition to the vast Hollywood library titles

provide by the movie companies (Figler, 2002).

TVN is another third party VOD vendor. TVN currently serves over 600 cable

systems reaching over 50 million U.S. households and offers a comprehensive suite of

pay-per-view and VOD programming along with equipment to encode, archive, transport

and manage video. TVN was selected by Comcast Cable to support Comcast's VOD

initiative and will support over 750 hours of VOD content monthly (PR Newswire,

2002). TVN has major licensing deals with content providers such as Artisan, MGM,

RCN, and Sesame Workshop (Figler, 2002).









Although widespread adoption of VOD could lead to many industry changes and

issues, problems involving bandwidth, content allocation and advertising present the most

current industry challenges.

VOD will put much more demand and stress on the cable infrastructure than found

in the current environment. Storage, managing, and ensuring systems run smoothly will

become more difficult when demand increases the data moving from various servers and

networks (Baumgartner, 2002). The current cable infrastructure is only wired to handle

10 percent of the households requesting VOD at any given time and this could be a

potential stumbling block for operators if people adopt the VOD structure quicker than

expected (Schiesel, 2002). Further, this usage overload could cause lock-ups in user

homes, making the system seem unreliable and problematic (Cummings, 2002). Although

these problems can be dealt with by increasing the server capacity and improving the

general infrastructure, both improvements involve increased spending in time and

resources (Schiesel, 2002). If these problems are not solved quickly, digital churn could

occur, reducing the competitive advantage offered by VOD.

The changing of content could be another challenging task for cable systems

(Baumgartner, 2002). VOD content would need to be changed on a regular basis and if

the system has multiple content vendors, the task could become confusing and

complicated. Storage for all the content could also be a challenge. With VOD requiring

so much content, the impact of delivery could be felt by the satellite carriers delivering

the content to the cable system headend (Brown, 2002). Satellites are still the most

economic way to deliver programming content to cable networks and current satellite

bandwidth is limited.









Last, advertisers are concerned about advertising in the VOD environment. The

VCR and digital video recorder enabled homes to erase or fast forward commercials.

Content providers need to keep advertisers in mind since they provide the primary

funding for programming. Software can be installed on VOD technology platforms to

ensure consumers cannot skip through commercials. This non-skip software is embedded

in the video from the VOD cable server (Stump, 2002). Finally, this advantage could be a

potential positive for advertisers looking to avoid the personal video recorders now being

highly publicized by the satellite television providers and those advertisers looking to

target specific niche markets.

Even with these drawbacks, advertisers are still testing advertising possibilities. For

example, Channing Dawson from the Scripts Network is testing the possibility of adding

a billboard ad at the beginning of each show to tease a product along with placing typical

spots throughout the program. Dawson also notes since VOD is a performance based

medium, gathering the personal information of the viewer and the viewer's habits will be

essential (Stump, 2003).

Nissan has been very proactive attempting to use VOD and interactive television

advertising to reach and interact with a new audience. The ads are designed to be

marrying content and new media in a new format. Additionally, Nissan claims to be the

first car manufacturer to use this platform (Precision Marketing, 2004).

The Direct Marketing Association recently conducted a study measuring attitudes

among marketers regarding interactive television functions. The study found 40% of the

respondents considering this new platform an important option for the marketing mix

(Duncan, 2004).









Advertisers spent over $54.5 billion on U.S. television advertising in 2003, more

than any other medium. With the technology changing this one time stable environment,

new advertising models and formats need to be developed. Consumers will not be

interested in spending the estimated $250 a year for commercial-free television (Angwin

et al, 2004). Funding content and providing a vehicle to reach potential consumers will be

essential, leading to the next research question:

RQ2: What is the most positive prospective advertising model for VOD?

This industry background on the current status of VOD serves to illustrate the

challenges and needs of this emerging industry. Cable systems, cable-programming

providers, content providers and third party vendors are all spending millions of dollars

on a technology still not widely adopted or even known to exist by the general public.

The goal of this research is to recognize the characteristics of early adopters of

VOD and identify their attitudes toward certain types of advertising. Diffusion of

innovations is the perfect theory to help identify the key variables of those willing to

adopt VOD technology and advertising due to the theory's long history and theoretical

grounding.

Theoretical Grounding

"Diffusion is the process by which an innovation is communicated through certain

channels over time among the members of a social system" (Rogers, 1995 p.5). Diffusion

of innovations involves new ideas (Rogers, 1995). Much like information flow theory,

diffusion of innovations is source dominated with the elite in a social system establishing

the key point of view when deciding to diffuse an innovation. Everett Rogers has

combined many of the elements found in original information flow research with

personal influence and information found in the fields of anthropology, sociology and









rural agricultural work (Baron and Davis, 2003). Because this theory is useful when

investigating the flow of information within social units, and identifying and predicting

the adoption characteristics of particular social groups, diffusion of innovations has been

chosen for this VOD adoption research.

Diffusion of Innovations

Key variables and constructs for diffusion reside in early German and British

schools of thought but considerable interest, research, and diffusion literature

development did not begin until the mid-1960s. A paradigm in diffusion literature was

developed in the mid-1960s due to the increased need for consumer behavior literature

(Rogers, 1976). Rogers (1976) states the origins of diffusion research trace from the

German-Austrian and British schools of diffusionism in anthropology and a French

sociologist named Gabriel Tarde. Tarde (1903) pioneered the S-shaped diffusion curve

found consist in diffusion research. Lastly, Rogers (1976) states the "revolutionary

paradigm" for diffusion research came in the 1940s when Bryce Ryan and Neal Gross

(1943) produced the most widely known diffusion research, investigating the diffusion of

hybrid seed com in two Iowa communities.

Many of the beginning concepts involving diffusion of innovations can be found in

the book written by Gabriel Tarde entitled; "The Laws Of Imitation" published in 1903.

Jean-Gabriel Tarde was born in Sarlat, France in the year 1843. Tarde was a sociologist

and directed his attention to social behavior and more specifically, to how people pass

feelings and thoughts from group to group or from person to person (Williams). Tarde

developed the theory of imitation, a 19th century social learning theory and is known as

one of the modem day learning theorists (Williams). In his book, "The Laws Of

Imitation," many of the same theoretical concepts matching diffusion are discussed such









as universal repetition, social resemblance, society and the inadequacy of the economic

and juristic conception, and logical influences (Tarde, 1904). Finally, Tarde proposed the

S-shaped diffusion curve, or the statistically plotted rate of adoption within a social

system and discussed the role of opinion leaders in the spreading of new ideas (Rogers,

1976).

A study conducted by Ryan and Gross set forth a new approach in communication

research and scholars have used their study involving the adoption of hybrid corn as a

model for researching communication adoption (Rogers, 1976). Ryan and Gross (1943)

selected the topic of hybrid corn because the diffusion of hybrid corn among Iowa

farmers made it attractive. The study was attractive to the authors due the traditional

conservative nature of the typical farmer and the recall of adoption was well within the

memory span of the current farmers at the time.

Data from the Ryan and Gross (1943) study was collected by personal interview

within two Iowa farming communities. From the study, the authors found the typical

farmer first became aware of the hybrid seed corn from salespeople (49%), but their

neighbors ended up being the most persuasive source leading to adoption. Consistent

with established diffusion research, the first farmers to adopt were of a higher

socioeconomic status. The entire study lasted from 1934-1941 with a total sample of 259

interviewees. Of the 259 cases, 2 never accepted the change in corn and only three

accepted in 1941 with all others in the sample previously adopting the new corn. The rate

of adoption followed the stereotypical S-shaped diffusion curve, much like Tarde's claim

and consistent with the diffusion research of today. This study is known as the classic

example of diffusion of innovation research (Rogers, 1976).









The hybrid corn study set forth a new approach to communication research and an

increasing number of scholars in all fields quickly took note of the study. Diffusion

research began to emerge as a single body of constructs based on communication and

human behavior. It was not until the 1960s that mass communication scholars began to

investigate communication under these specific constructs, first applying them to news

stories. While mass communication scholars were investigating the diffusion of news,

marketers and consumer behavior researchers began to use the theory for understanding

the consumer adoption process of new products. Marketing and consumer research

became a strong force in the 1960s and diffusion of innovations seemed like a perfect fit

for understanding how a company could launch a new product more efficiently. The

theoretical and methodological background of diffusion research appeared to be the

perfect basis for targeting and predicting consumer adoption for marketers (Rogers,

1976).

Among those looking to use diffusion of innovations in the marketing environment

was Frank M. Bass. Bass (1969) published a research paper entitled, "A New Product

Growth for Model Consumer Durables," and his research made a significant contribution

to diffusion of innovations theory by constructing a mathematical model for new product

growth in consumer durables. The model developed by Bass is a growth model for

estimating or predicting the timing of adoption. Bass' model is set within the theoretical

framework of Rogers' literature and has become a driving force in diffusion research in

the field of consumer behavior and marketing. Bass states the probability of purchase at

any time has a linear relationship to the number of previous buyers of a product. Product

adoption will grow exponentially and then ultimately see exponential decay over time.









Bass found the consumer durables tested in his model were in agreement with his

proposed model and states his model works well for understanding the process of new

product adoption and long-range forecasting. Bass states there are two main influences

persuading potential consumers, internal factors such as word of mouth or interpersonal

communication and external factors such as the mass media (Mahajan and Muller, 1990).

Bass' model is still used today and continues to be a mathematical framework for

adoption.

Much of the information in this research involving the classic diffusion model and

was taken from the Everett M. Rogers (1995) book entitled, "Diffusion of Innovations,

Fourth Edition." Although some research within this section has been credited to other

sources, Rogers is the most definitive source on diffusion of innovations theory and his

book generalizes thousands of previous research articles. The Rogers (1995) text provides

the theoretical definitions found in this research involving the classification of adopters,

characteristics of adopters and consequences of innovations. Rogers' theoretical and

methodological background will be used for identifying early adopters of VOD

technology.

The classic diffusion model (Rogers, 1995) consists of the innovation or new idea,

the communication channels the new idea is delivered through, the time it takes for the

innovation to be adopted or rejected, and the social system the idea is communicated

within.

The first construct in the classic diffusion model is the idea or innovation. Rogers

(1995) states an innovation can be a new idea, practice or object for adoption and not all

innovations are equivalent. Because innovations are all different, they will posses









different rates of adoption. Some innovations present a relative advantage to an existing

idea and the innovation becomes perceived as better. Social prestige, convenience, and

increased satisfaction can all create a relative advantage. Compatibility is another element

of the innovation examining to the degree the innovation is consistent with existing

values, past experiences and the needs of potential adopters. Other notable factors of an

innovation include complexity or how difficult the new idea is to understand, trialability

or the ability to test the idea, and observability or the ability for the results to be seen by

others.

The communication channel is the second construct in Rogers' (1995) diffusion

model. The way a new idea passes between individuals or within social groups is

through the communication channel. Participants within social groups share and create

information with one another to reach a mutual understanding. This communication

should not be thought of as linear act but as a two-way process of information

convergence. Diffusion research has generalized that most people do not evaluate an

innovation based on scientific studies but rely more on subjective evaluation conveyed

from other individuals, like themselves, who have previously adopted the innovation. The

diffusion literature also states more successful diffusion communication occurs when

individuals or groups have similar attributes, beliefs, education, or social status. This

communication can move individuals closer or further away from the innovation through

the message delivered by the change agent.

Rogers (1995) states time is the third element impacting the diffusion process. The

time dimension involves the information-decision process, the innovativeness of the

individual and the rate of innovation adoption.









Rogers states (1995) the rate to which an individual or group passes from first

knowledge, to attitude formation, to the adoption or rejection decision, and finally to

confirm the idea encompasses the innovation-decision process. The process had five main

steps; knowledge, persuasion, decision, implementation and confirmation. When an

individual first learns of an innovation they gain knowledge. Persuasion involves the

favorable of unfavorable opinion toward the idea. Based on persuasion, the individual

will make a decision to adopt or reject the new idea. Implementation activities are related

to adopting or rejecting the new idea and confirmation occurs when the individual seeks

reinforcement. It is important to note this time step can not only lead to adoption or

rejection but also the discontinuance of an innovation. Discontinuance occurs when the

individual becomes dissatisfied with an innovation or when the innovation is replaced.

Rogers (1995) states the degree of innovativeness of an individual or group is the

second part of the timing aspect. The innovativeness of an individual reflects the earliness

or lateness the individual adopts an innovation compared to other members of the social

system. Research indicates individuals within adopter categories have more in common

with each other then with members of other adopter categories.

The third segment of the time aspect illustrated by Rogers (1995) involves the total

rate of adoption. The relative speed to which a new idea is adopted is the rate of adoption.

This rate of adoption typically forms an S-shaped curve when statistically plotted. The

slope steepness of the curve is set by how quickly or slowly an innovation is ultimately

adopted within a social system.

The fourth aspect to the classic diffusion model involves the social system

surrounding the innovation. Rogers (1995) defines the social system as interrelated parts









involved in joint problem solving to finding a common goal. Social systems consist of

groups, individuals, and organizations. The structure, norms and change agents of a social

system impact diffusion. Within each social system exists hierarchical positions of

influence, with higher hierarchical positions leading the lower through various means.

The communication structure within the socials system sets the pattern for

communication flow and the structure of a social system can facilitate or impede the

diffusion process. Rogers (1995) states the most innovative member of a social system is

very often looked upon as a deviant from the system and may play a limited role in the

diffusion process due to credibility.

On the other hand, according to Rogers (1995) opinion leaders and change agents

have the ability to increase the rate of the entire adoption process. Opinion leaders are

able to influence individual attitudes and behavior informally to increase diffusion.

Opinion leaders are exposed to many forms of communication within a social structure,

have higher social status, and are typically more innovative. Opinion leaders are often

directly influenced and learn of innovations from individuals known as change agents.

Change agents are individuals who attempt to influence innovative decisions to

their desired conclusion. They are often professional or expert members of a social

system and are able to influence opinion leaders in leading diffusion campaigns.

These four basic aspects exemplify how the adoption process moves through a

social system and what variables influence adoption. Mass adoption of VOD technologies

has yet to take place. By understanding the communication of an innovation, the time and

elements of adoption and the social aspects of a system, marketers can get a perspective

understanding as to how the information regarding consumer VOD technology will flow









through our social system. Although marketers are interested in the potential flow,

knowing the characteristics of first adopters will ultimately set the model in motion.

Ryan and Gross (1943) classified segments of Iowa farmers in relation to the

amount of time it took them to adopt hybrid corn seed. These adoption types fit into

Rogers' (1995) ideal adoption types. Although some individuals are exceptions to the

adopter types, these generalizations offer parameters for variables when identifying

potential VOD adopters.

Rogers (1995) identifies innovators as the first adoption group. On average,

innovators represent 2.5 percent of a social system. Venturesomeness is an important trait

for the innovators. Innovators typically deal with a high degree of uncertainty and idea

complexity. They are prone to experience and accept occasional setbacks. Innovators are

often the gatekeepers within a system because they are first to adopt and bring an

innovation into the system (Rogers, 1995). Acceptance of an innovation by innovators

and by the early adopters can set the stage for the ultimate success of a product (Danko

and MacLachlan, 1983).

Rogers (1995) states early adopters represent 13.5 percent of the average social

system and they are more integrated in the social system than the innovators. The early

adopters hold the greatest degree of opinion leadership and this category is typically

sought by change agents due to their impact in speeding up the total adoption process.

Early adopters can also serve as role models for a social system since they possess many

similar traits of the other members in a social system. Innovation moves quickly through

the early adopter group because adoption by this group reduces uncertainty for the entire

social group conveying a subjective positive evaluation to near-peers through









interpersonal networks (Rogers, 1995). Quick acceptance by early adopters usually

equals rapid market growth and high profits for innovative products (Danko and

MacLachlan, 1983).

Rogers (1995) finds the average early majority group represents 34 percent of the

individuals within a social system. This group adopts new ideas just before the average

member of the system. This group is vital to the adoption process since they provide

interconnectedness in the system's interpersonal communication networks. They are

followers and adopt innovations but seldom lead.

The late majority represents 34 percent of a social system and Rogers (1995) states

this group adopts ideas just after the average member in a social system decides to adopt.

For the late majority members, innovations are approached with a skeptical and cautious

nature and late majority members wait for the majority of members of a social system to

adopt prior to adopting. The weight from system social norms and pressure from peers is

necessary to motivate adoption. This group typically has scarce resources and needs the

uncertainty of an innovation removed so their limited resources are not wasted (Rogers,

1995). Pressures from previous adopters aid in the timing of the adoption process for this

group (Bass, 1969).

Laggards, representing 16 percent of a social system are the final group of adopters

identified by Rogers (1995). Laggards are the last group to adopt an innovation. They are

typically isolated within the social system and possess no opinion influence or leadership.

Laggards always look for information based on the past and they must be certain a new

idea will not fail before they adopt. They are typically of lower economic class and

education.









Realizing the strength of each adoption group is important when attempting to

spread a new innovation through a social system. Although the idea of an innovation for a

social system is presumed to be positive, innovations can have consequences.

Rogers (1995) states innovations can cause certain changes within a social system

and these changes could impact a system in a variety of ways. Most presumably, these

changes occur due to the adoption or rejection of an innovation. Some consequences of

innovations are desirable while others can be undesirable. These consequences can occur

if the innovation is functional for the intended purpose or possibly dysfunctional or used

for unintended purposes.

Innovations can also have direct or indirect consequences. Direct or indirect

consequences can occur immediately in response to the adoption of the innovation or can

be an unforeseen result of the innovation adoption.

Lastly, innovations can have anticipated and unanticipated results. These

consequences vary based on the intent of the innovation and sometimes the unanticipated

results can be detrimental.

Rogers (1995) states there are many criticisms of diffusion research. He found

diffusion research often has a pro-innovation bias. This bias is the idea the innovation

being studied should be diffused and adopted by the social system. This bias can be

illustrated in a tremendous amount of diffusion research because the change agent or

group interested in the innovation is the client for the study and has a valid interest in the

innovation's diffusion. These agencies often spend large amounts of money on studies

and do not wish to see their money wasted on a poor study.









Another criticism involves recall. Recall is heavily relied upon in diffusion research

and is a methodological enemy of diffusion research. Recall is often clouded by time

from when the adoption of an innovation actually occurred and when the recall for a

study occurs, thus reducing methodological accuracy.

Time is another enemy of adoption recall since many studies are not longitudinal

but are just "snapshots" of the adoption process. Often the authors of research want to

find the current status of an innovation adoption process and overlook the longitudinal

aspect.

Lastly, reliance on correlational analysis of survey data can often lead to a false

sense of causality. If the authors of a study refrain from properly operationalizing their

variables, the analysis becomes meaningless. Further, intervening variables are always

involved in survey research and these variables must be accounted for in the final

adoption analysis. Proving exact causality is difficult and is a negative for all types of

quantitative survey research.

Rogers (1995) lists distinct categories involving the variables related to

innovativeness. Early adopters of innovations are typically found to have similar traits in

socioeconomic status, personality values, and communication behavior that distinguish

them from other members of a social system. Acknowledging and operationalizing these

traits is key for recognizing early adopters of VOD technologies. These traits will be

operationized by the author for the survey instrument to measure early adopters of VOD

technologies. Taken directly from Rogers' 1995 publication, the three categories and

traits are:

* Socioeconomic Characteristics (Rogers 1995, p. 269):
* Earlier Adopters are not different from later adopters in age









* Earlier adopters have more years of formal education
* Earlier adopters are more likely to be literate
* Earlier adopters have higher social status
* Earlier adopters have a greater degree of upward social mobility
* Earlier adopters have higher social status


Personality Variables (Rogers 1995, p. 272-273):

* Earlier adopters have greater empathy
* Earlier adopters may be less dogmatic
* Earlier adopters have a greater ability to deal with abstractions
* Earlier adopters have a greater rationality
* Earlier adopters have a more favorable attitude toward science
* Earlier adopters have a more favorable attitude toward change
* Earlier adopters are better able to cope with uncertainty and risk
* Earlier adopters are less fatalistic
* Earlier adopters have higher aspirations


Communication Behavior (Rogers 1995, p. 273-274):

* Earlier adopters have more social participation
* Earlier adopters are more highly interconnected through interpersonal networks
* Earlier adopters are more cosmopolite
* Earlier adopters have more change agent contact
* Earlier adopters have greater exposure to mass media communication
* Earlier adopters have greater exposure to interpersonal communication
* Earlier adopters seek information on innovations more actively
* Earlier adopters have greater information on new innovations
* Earlier adopters have a higher degree of opinion leadership

The review of relevant diffusion literature contained in this research proposal is not

exhaustive. Diffusion studies have produced a body of literature impossible to thoroughly

cover in one research study. The adoption of new digital technology leads the pertinent

criteria for the diffusion research selections included in this literature review.

William Danko and James MacLachlan (1983) produced empirical research

examining the specifics and indicators of individuals adopting personal computers. The

implications of their study found differences in media choice, advertising message









content, and distribution for the early adopting group when compared to the late adopters.

As for the study's findings, the authors' state most early adopters of computers were

highly educated males and suggested the advertising messages be tailored to a male

audience. Danko and MacLachlan (1983) made a few advertising suggestions for

targeting early adopters of computers. The authors suggested messages delivered in

"cerebral" vehicles such as Psychology Today and Scientific American or by direct mail

would be efficient since time spent with television was minimal for this group. Lastly, the

authors suggest most early adopters will buy from mail-order type services since they are

more adventuresome and less reliant on support. Although this article is dated and had a

main goal of determining an appropriate advertising positioning strategy, it does illustrate

some of the qualities of the early adopter category and the diffusion of personal

computers.

Digital satellite programming provider Directv used Frank Bass' model to help

predict subscriber rates for the satellite provider prior to launch. Bass et al's (2001)

research project entitled, "DIRECTV: Forecasting Diffusion of a New Technology Prior

to Product Launch" applied Bass' 1969 model to predict diffusion for the new satellite

provider. Directv is a subsidiary of the Hughes Corporation under General Motors. At the

time the study began, the company was developing technology to provide programming

to subscribers via satellite and compete with cable television. The study had three distinct

research questions: deciding what pricing and programming to offer, who would be the

first to buy, and how many would be the first to buy? After a review of research, Directv

launched a concept test with the main purpose of obtaining data for forecasting Directv

market diffusion over the next 10 years. On the basis of the data collected from the









concept test, Frank Bass used his model (1969) to forecast the number of U.S. subscribers

for Directv and when they would subscribe. Because the study was longitudinal, the

authors compared the actual diffusion of Directv to the original Bass prediction. The

findings illustrated Bass' model was successful in forecasting the diffusion of Directv for

the first five-year period from 1994 to 1999.

Studies involving Internet adoption are pertinent to this study due to the newness

and innovativeness found in Internet technology. A study produced by Atkins and Jeffres

(1998) profiled Internet adopters in terms of social locators, media use habits, and their

orientation toward adopting new technologies. Their findings supported some of the early

adopter notions of demographics and technology uses found in diffusion theory. The

study examined demographic as well as technology needs. The authors found

communication needs were a stronger predictor than demographic information alone for

those adopting the Internet. Atkins and Jeffres (1998) also feel further diffusion research

should refine measures of cosmopoliteness and localiteness. Still, their findings stated

Internet adopters are typically young and educated. They also felt the Internet was still in

the early stages of adoption. Lastly, technology orientation or the need for innovativeness

was one of the points of difference when comparing adopters to non-adopters.

Since digital cable is the delivery method for VOD, research involving digital cable

adoption adds valuable insight to the current study. Myung-Hyun Kang (2002) produced

research exploring the factors associated with the early adoption of digital cable. The

purpose of the study was to understand and predict digital cable adoption by identifying a

profile of early digital subscribers along with implications for industry structure and

marketing. Kang's study looked at early adopters and adoptive innovativeness as









dependent variables and demographics, media use, technology ownership, innovative

attitudes, and satisfaction as independent variables. The study found most hypotheses

from diffusion theory were supported with a few significant exceptions. Kang found

income was not related to digital cable subscribership. The author believes the reason for

this could be found in the increased cost of upgrading to digital service. The study did

find high consumption of television did relate to digital cable adoption and those who

currently subscribe to premium channels would be better targets for digital cable. Lastly,

digital cable companies who appear innovative and form strong positive relationships

with consumers will be adopted more quickly.

The next technological step from VCRs is the current growth of digital video

recorder market. Douglas Ferguson and Elizabeth Perse (2001) conducted a study

entitled, "Enhanced Television Viewing with Digital Video Recorders (DVRs): Audience

satisfaction in an Asynchronous Television Environment." This article is quite pertinent

to the current study since VOD services also provide viewing in the asynchronous

environment. From their exploratory findings, DVR owners are clearly early adopters and

enjoy the uses and benefits from the technology. Timeshifting was linked to viewing

satisfaction, with having the ability to record programs airing at inconvenient times as a

most valued feature. The menu driven program selecting and recording feature was also

found to be a valued feature. Lastly, DVR owners found the machines much easier to

operate than VCRs and the functions of the DVR allowed users easier operation

regarding storage and playback.

Dupagne (1999) explored characteristics of potential high-definition television

adopters based on the diffusion of innovation theory. The purpose of this study was to









assess consumer predispositions toward high-definition television and profile potential

adopters based on demographic characteristics, mass media use, ownership of home

entertainment products and importance of high-definition television attributes.

Awareness, interest and purchase intention were the dependent variable measured. Based

on this research, most profile characteristics were consistent with previous diffusion

studies. Innovators and early adopters of high-definition television were more likely to be

younger and have higher incomes. The early adopters were also more interested in the

attributes found in this new technology. Demographics and importance of television

attributes were found to be stronger predictors of high-definition television compared

mass media use tendencies or ownership of other technology.

Lin (1998) examined the adoption rate and adopter types of personal computer

consumers. Her study looked at media use patterns, use of other existing communication

technology ownership in addition to demographics and other social characteristics such

as the need for innovativeness.

Lin has made a significant contribution to diffusion research by adding the need for

innovativeness as an important construct to measure and include in diffusion research.

According to Lin, the need for innovativeness characteristic is an important contributor to

the overall adoption profile of a consumer and should be considered.

In her personal computer study, Lin (1998) found the need for innovativeness

characteristic was the strongest predictor among adopters, followed by likely-adopters

and non-adopters of computers. Lin found likely adopters were those who were oriented

toward the need for innovativeness but lack the financial resources to participate in the









early adopter phase. Media use, education and gender were not found to be a predictive

but age and income were found to be representative to the past diffusion studies.

In another diffusion technology study, Lin and Jeffres (1998) explored audience

intentions to experiment with or adopt multimedia video technologies. These

technologies would allow a consumer to use voice, data, video channels and

communication through a single interactive coaxial cable system. They measured interest

in experimenting with new technologies and interest in adopting new multi-channel cable

services. For predictive variables, they measured demographics, media use, and

satisfaction with media content. Need for innovativeness was also measured. The study

found demographic variables were largely irrelevant toward interest in experimenting

with the new system. Lower satisfaction with current television content and lighter

viewing levels were found to be significant predictors of adoption. Lin and Jeffres state

existing media use patterns, along with media content satisfaction, could be strong

variables to help determine functional substitutions between an existing medium and an

emerging medium. They also state personality traits such as "innovativeness" could be

instrumental to examine differing levels of adoption and interest.

Atkin, Jeffres and Neuendorf (1998) examined Internet adoption as

telecommunication behavior. The intent of their study was to profile Internet adopters in

terms of social locators, media use habits, communication needs and their orientation

toward adopting new technologies. The profile was built off of measuring variables such

as demographics, communication needs, communication activities, technology relations

and media consumption behavior. Their study found some support for the early adopter

profiles derived from diffusion theory, such as early adopters being young and educated.









This study also found innovativeness to be an important predictor for the adoption of

innovations. Results from this study failed to confirm the expectation that attitudinal

variables, those addressing communication needs, are more explanatory than

demographics.

Neuendorf and Atkin (1998) assessed the use of telephone-based audio innovations

and profiled audience users and utilities for audio text services (1-900 services) and fax

services. They examined use of these services by measuring demographics, quality of

life, social communication activity, and media use. Consistent with diffusion literature,

typical heavy fax and audio info users were found to be younger. Inconsistent with

diffusion literature, demographics and other traditional social indicators are not uniformly

important in the prediction of innovation use. The authors state the modest role played

by demographics reinforces past findings suggesting that their explanatory influence has

weakened over time.

Eastlick (1996) examined consumer intention to adopt interactive teleshopping.

The purpose of this study was to identify personal innovativeness characteristics,

shopping attitudes, and product use patterns that contribute to intention to engage in

interactive teleshopping. The study also looked to examine attitudes toward the

innovation characteristics and examine relationships among the variables. Attitude

toward interactive teleshopping was measured as well as the intent to adopt teleshopping.

The author looked to explain these dependent variables by measuring demographics,

shopping orientations, perceived uncertainty, and personal innovativeness. The results

from the study show that overall opinions of interactive teleshopping's innovation

attributes as well as several priority acquisition patterns contributed most to predicting









subjects' attitudes toward interactive teleshopping. Demographics were not predictive but

relative advantage, ease of use, trialability and observability were consistent with classic

diffusion research.

It is important to note some of the studies included in this literature review are not

diffusion articles. Even though they are not diffusion research, they are still important to

address due to the added relevance or insight they contribute to the current study.

Video fragmentation is another area of interest for this proposal. A study done by

Carolyn Lin (1994) is very important to include even though Lin's study did not use

diffusion of innovation as theoretical grounding. Lin's study entitled; "Audience

fragmentation in a Competitive Video Marketplace" examined the competitive nature of

the fragmented video marketplace. This fragmented marketplace presents challenges and

opportunities for cable companies and advertisers and the findings and variables are

important for measuring the adoption of VOD services. In Lin's (1994) study, strong

differences were found in behavioral, environmental and motivational indicators when

comparing premium-cable homes to non-cable homes. The study found cable homes are

typically heavier and more satisfied viewers. Premium households were found to have the

most viewing satisfaction, followed by basic cable and concluding with non-cable

households. The study also found high channel switching in premium and basic cable

homes and low switching in non-cable homes. Cable homes were found to have more

sophisticated viewers than non-cable homes and cable homes were more likely than non-

cable to be engaged in both viewing-decision reevaluation or multiple channel viewing

during commercial breaks. Lin states this channel switching suggests cable homes are

much more active when making a viewing selection.









In addition, Lin found viewing exposure and planning were important measured

concepts in this research. Weekday and weekend viewing was found to be similar for

non-cable and basic cable homes. However, homes subscribing to premium cable viewed

30 to 45 more minutes per day. Program preplanning was higher for those households

subscribing to cable than those non-cable households. As previously mentioned, premium

households were more likely to view multiple programs followed by basic then non-cable

households.

Last, demographic data collected from this study states premium cable homes

reside in the youngest age group and wealthiest group. Roughly half of the basic cable

subscribers were middle age and 61 percent of non-cable households were in the lowest

economic group. Non-cable homes were found to have a lower level of education while

over half (62%) of the basic cable subscribers had completed college. All factors defined

by Lin are particularly important to note for the current proposed VOD research.

Another study particularly relevant to this proposal but not based on the theory of

diffusion of innovations was a study done by Randy Jacobs (1995) entitled, "Exploring

the Determinants of Cable Television Subscriber Satisfaction." In this study, Jacobs

focused on the factors associated with subscriber satisfaction. Jacobs suggests satisfaction

is directly linked with disconnection or subscriber upgrading/downgrading behaviors.

Subscriber evaluations of cable system performance are important to the development of

overall subscriber satisfaction. Programming variety and quality were also found to be

important for overall satisfaction. With the onset of competition, Jacobs eludes

programming evaluations could move to the forefront of satisfaction formulation. Since

technological advancements have enhanced service reliability and picture quality, these









advancements are less of a satisfaction indicator because they are now expected outcomes

and no longer points-of-difference between competing entities. Jacob also found monthly

cost, system size, and the number of channels offered was not related to satisfaction.

Previous VCR adoption studies add insight to the current proposal for investigating

the uses and asynchronous benefits of VOD. Klopfenstein and Spears (1991) examined

the relationship between VCR owners and attitudes about watching network television,

recording television, and overall satisfaction with VCR ownership. These variables were

measured in order to determine whether VCR approval would increase or decrease based

on time of ownership. The notion of discontinuance found in diffusion theory was not

found to have occurred in VCR ownership at the time of this study. The study found the

longer the VCR was in the home, the more likely the respondents were to record

television programming. In addition, owners of VCRs for the longest period of time were

more likely to delete commercials, record while sleeping, and record programs while not

at home. Research involving demographic data found income was the only factor related

to first VCR adopters. First adopters were also "technolphiles" supporting the diffusion

theory assumption that early adopters are unique

At the time of the research, only one scholarly VOD article could be found.

Ronald Rizzuto and Michael 0. Wirth (2002) developed a pioneering study entitled, "The

Economics of Video On Demand: A Simulation Analysis" investigating if VOD (i.e.,

movies on demand) would be economically viable. They found economic aspects of

VOD could be viable without major impacts on peak utilization rates. They identified

three key VOD factors for economic viability: movie buy rates, Hollywood and cable

operator revenue splits, and peak utilization rates. By doing diffusion research in the area









of VOD adoption, projecting perspective utility rates could be useful for concluding

economic viability based on the model proposed by Rizzuto and Wirth.

Identifying and predicting the social adoption characteristics of this new VOD

technology is important for cable companies, content providers and advertisers. By

identifying these characteristics, advertisers can better tailor messages to reach those

whom are interested in VOD technology; with the ultimate goal of enticing them to try

VOD and hopefully increase the speed of various social groups through the adoption

curve.

Based on the theoretical grounding found in diffusion of innovations and the

various variables found in other technology research included in this literature review,

this exploratory study will examine the final three research questions:



RQ3: What types of consumers are currently aware of VOD technology?

RQ4: What factors indicate early adopters of VOD technology?

RQ5: What factors indicate intent to adopt VOD technology?














CHAPTER 3
METHODOLOGY

Data Collection


Due to time and financial restraints, a nonprobability sample was used.

Respondents for this convenience sample were recruited during the Easter weekend of

April 16-18, 2003 and were asked to take a self-administered survey. Written survey

methodology was the chosen method of data collection because written surveys are easy

to administer to a large population and surveys are particularly useful when needing to

describe the characteristics of a large population (Babbie, 2001).

The survey was administered on Clearwater Beach, Florida. Respondents were

approached while relaxing on the beach and bottled water was given to respondents in

appreciation for their cooperation. The survey took approximately five minutes to

complete. A total of 190 respondents were asked to complete the questionnaire yielding a

total of 170 surveys. The primary researcher for this study recruited the respondents by

introducing himself and explaining how their cooperation would aid in his thesis

research. Respondents were also assured their information would not be used for

commercial or marketing purposes. Any questions regarding the survey were addressed

upon the respondents' completion of the survey.

Survey research is the collection of information used to better understand or predict

some aspect of respondent's attitudes or behaviors (Davis, 1997). Surveys were used in

this research due to their ability to be distributed to numerous respondents at the same









time. Since this was a convenience sample, the ability to distribute multiple instruments

simultaneously was essential. Overall, this made the survey the most obvious data

collection instrument of choice.

Survey Design

The survey questionnaire focused on nine specific areas, either derived from

previous diffusion of innovation research or important for rating the proposed exploratory

advertising models or questions. The survey included questions regarding the following

areas of interest: social and innovative attributes, attitude toward asynchronous viewing

of differing television programming, VOD functionality, media usage, VOD awareness,

knowledge, use and likeliness to subscribe VOD services, likelihood to provide personal

information or be exposed to targeted advertising, rating possible advertising scenarios,

technology ownership, and demographic information.

The first section of the questionnaire was set up to measure the social and

innovativeness characteristics of the respondents. The questions were modeled after

previous technology diffusion of innovations studies (Lin 1998; Jeffres & Lin 1998;

Atkin, Jeffres & Neuendorf 1998; Neuendoft & Atkin 1998; Kang 2002; Eastlick 1996).

Respondents were asked to rate a number of statements on a seven point scale with one

being strongly disagree to seven being strongly agree regarding social activities and need

for innovativeness (see appendix for questionnaire).

The second section of the questionnaire was designed to measure one of the

attributes of VOD technology. Conceivably, a variety of programs would be or are

currently available on a cable video server for the consumer to view at anytime. This set

of questions asked the consumer to rate how desirable programming would be from four

different sources: broadcast (e.g. ABC, CBS, NBC, FOX), cable (e.g. CNN, MTV,









ESPN, HGTV), premium (HBO, Showtime, Starz), or Pay-per-view new release movies.

Respondents were asked to rate each choice on a one to seven scale with one being not

important and seven being very important.

The third section of the questionnaire was designed to measure another attribute of

VOD technology. As previously discussed in Chapter 2, VOD allows the consumer to

stop, fast-forward, rewind, and pause television programming content. Respondents were

asked to rate this attribute on a one to seven scale with one being not important and seven

being very important.

This section was designed to measure the usage of media. Respondents were asked

to rate on a one to seven scale with one being not a user at all to seven being a heavy user

of the following media: newspaper, radio, television, magazine, video rental, movie going

and Internet. Previous diffusion of innovation research regarding technology has used

similar measures (Kang, 2002; Dupagne, 1999; Lin 1998; Jeffres & Lin 1998; Atkin,

Jeffres & Neuendorf 1998; Neuendoft & Atkin 1998; Eastlick 1996).

This section of the questionnaire was designed to measure the dependent variables

of VOD at various levels of involvement. VOD awareness was measured by asking

respondents if they were aware of VOD (yes or no.) If the respondent was aware, they

were asked to rate their knowledge of VOD technology on a seven-point scale with one

being not at all knowledgeable and seven being highly knowledgeable. VOD use was

asked by requesting if the respondent had ever previously used VOD (yes or no). Last,

respondents were asked if they currently subscribe to VOD (yes or no) and if not, their

intention to subscribe rated on a seven-point scale with one being not likely and seven

being most likely.









These exploratory questions were used to measure respondents likelihood to give

personal information to advertisers and their willingness to be exposed to targeted

advertising. Each was measured by asking the respondent to rate their attitude based on a

seven-point scale, with one being not at all likely and seven being most likely.

These exploratory questions were designed to measure respondent attitudes toward

possible VOD advertising based on pay-per-view scenarios. Pay-per-view scenarios were

used since it was assumed, regardless of the respondent's VOD knowledge, that they

have had some prior experience with pay-per-view content. Each question asked the

respondent to rate on a seven-point scale with one being not at all likely and seven being

most likely, how likely they would participate in the scenario. The first question asked if

the respondent would pay $2.99 for the movie in exchange for being exposed to ten

minutes of advertising commercials. The second question asked if the respondent would

consider paying $1.99 for twenty minutes of commercials spread throughout the movie.

The third question asked how likely the respondent would pay $2.99 for a pay-per-view

movie with text messaging moving across the bottom of the screen during the entire

movie. The forth question asked the respondent to rate paying $1.99 for a pay-per-view

movie with text messaging moving across the bottom during the entire length of the

movie.

Respondents were asked whether they subscribe or own a number of technology

products and services. These products or services included: Digital video recorder,

personal computer, high-speed Internet, video game system, cell phone, home theater

system, high definition television, large screen television (35" or larger), DVD player,

personal digital assistant, cable TV with a set-top box, cable TV without a set-top box,









direct broadcast satellite, and premium cable television. Measuring technology ownership

is consistent with previous diffusion of innovations research (Dupagne, 1999; Lin, 1998;

Kang, 2002).

Demographic questions included gender, education, marital status, age, household

size, occupation and income. Demographic variables are essential to diffusion of

innovation research and have been previously measured in other technology diffusion

studies (Rogers, 1995; Kang, 2002; Dupagne, 1999; Lin 1998; Jeffres & Lin 1998; Jeffres

& Atkin, 1996; Atkin, Jeffres & Neuendorf 1998; Neuendoft & Atkin 1998; Eastlick

1996).

Data Analysis

Different statistical methodological processes were applied to each research

question. For analyzing what types of consumers are aware of VOD and for indicating

factors important for early adopters of VOD, or RQ 1 and RQ2, descriptive statistics were

used in conjunction with t-tests and chi-square tests, analyzing differences of means and

statistical significance. For analyzing RQ3 or the relationship of factors contributing to

intention to adopt VOD technology in the next six months, multiple regression analysis

was implemented. To analyze what functions of VOD technology were found to be most

preferred as well as the analysis of the most positive prospective advertising model for

VOD or RQ 4 and 5, t-tests were used to analyze differences in means. SPSS for

Windows, Release 11.5.0 was used to run all descriptive and inferential statistics.















CHAPTER 4
RESULTS

Descriptive Statistics


The convenience sample yielded 170 adults made of 28% males (n=48) and 72%

females (n=121.) According to the 2000 U.S. Census, 49% of the national population is

male and 51% of the national population is female. Therefore, caution should be taken to

generalize this study's results onto the population because of the skewed gender.

As for age, 23% were 18-24 years old, 28% were 25-34 years old, 24% were 35-44

years old, 18% were 45-54 years old, 5% were 55-64 years old and only one person was

over the age of 65. The mean age was 36 years old (std. deviation of 12.1). Compared to

the general population derived from 2000 U.S. Census data, the current sample is skewed

much younger. Since the sample is skewed much younger, caution should again be taken

when projecting data onto the general population. Age frequencies and percent are found

in table 4-1.

Table 4-1. Age Distribution
Sample US Census
Age Frequency Percent Percent
18-24 38 23% 10%
25-34 46 28% 13%
35-44 40 24% 16%
45-54 30 18% 14%
55-64 9 5% 9%
65+ 1 less than 1% 13%









For household income, 7% of the total sample made less than $19,000 per year,

18% made between $20,000 and $39,000 per year, 19% made between $40,000 and

$59,000 per year, 18% made between $60,000 and $79,000 per year, 14% made between

$80,000 and $99,000 per year and 20% of the households made over $100,000 per year.

According to the 2000 U.S. Census, the general population consists of more people

making less than $19,000 per year. Therefore, our sample has a higher income bias and

this should be noted when projecting data to the general population. Income frequencies

and percent are listed in table 4-2.

Table 4-2. Income Distribution

Sample US Census
Income Frequency Percent Percent
<$19,000 11 7% 23%
$20k-$39k 31 18% 15%
$40k-$59k 31 19% 18%
$60k-$79K 30 18% 13%
$80k-$99k 23 14% 8%
>$100k 33 20% 13%


Of total respondents in sample, 41% were single or considered "other" meaning

they did not consider themselves one of the choices, 49% were married, 9% were

divorced, less than 1% was widowed. This finding differs from the U.S. Census numbers

because comparatively the sample is skewed with too many singles. The mean sample

household size was 3, close the general population household size of 2.59, provided by

the U.S. census. Frequencies and percent for marital status are presented in Table 4-3.









Table 4-3. Marital Status

US
Sample Census
Status Frequency Percent Percent
Single/Other 68 41% 27%
Married 82 49% 57%
Divorced 15 9% 10%
Widow 1 less than 1% 7%


As for occupation, 3% were laborers, 1% were machine technicians, 2% were

craftsmen, 7% were clerical, 12% were in sales, 3% were administrators, 37% were

considered professionals, 6% were managers, 29% were other or occupation not listed.

Since the U.S. Census categories did not exactly match the sample categories,

conclusions were drawn to match respective categories. Compared to general population

data from the U.S. Census, the sample lacked service and construction positions, matched

close regarding sales and office occupations as well as management and professional, but

was deficient meeting the other category. Frequencies and percent of total sample and

U.S. Census percent are listed in Table 4-4.

Table 4-4. Occupation Distribution
Sample US Census
Occupation Frequencies Percents Category Percents
Laborer 5 3%
Machine/Service Service and
Tech 2 1% Construction
Craftsman 3 2% Positions 24%
Clerical 11 7% Sales and
Sales 19 12% Office
Administrator 5 3% Occupations 27%
Professional 61 37% Management
&
Manager 10 6% Professional 34%
Other 49 29% Other 15%









For respondent education, 1% had less than a high school education, 26% had

finished high school, 57% finished college, and 16% finished graduate school. Compared

to the U.S. Census data from 2000, our sample is not representative as far as respondents

having an education less than high school level. The current sample also has respondents

with college degrees and graduate degrees compared to the general population.

Frequencies and percent are included in Table 4-5.

Table 4-5. Education
U.S.
Sample Census
Education Frequency Percent Percent
Less Than High less than
School 2 1% 20%
High School 43 26% 29%
College 94 57% 43%
Graduate School 27 16% 9%


The study revealed that 58% of the respondents were aware of VOD technology. Of

those respondents who were aware of VOD (n=94), only 43% had actually used VOD

and only 26% subscribe to VOD services. Table 4-6 displays awareness, use and

subscribers of VOD.

Table 4-6. Awareness, Use and Subscribers of VOD
Yes No
Aware of VOD 99 (58%) 68 (40%)
Used VOD 40 (43%) 54 (57%)
Subscribe to VOD 24 (26%) 69 (74%)


Respondents who are aware of VOD but do not subscribe were asked to rate their

intent to subscribe to VOD services over the next six months. Respondents rated their

intent to subscribe a M=2.06 (n=66, Std. Dev 1.402) from a seven-point scale with one

being not likely at all and seven being most likely. This result seems to indicate the three-









quarters of those who are aware of VOD are not very likely to subscribe to VOD type

services within the next six months.

As for the functions of VOD, rated on a seven-point scale by total respondents with

one being not important and seven being very important revealed having the ability to

pause, fast-forward, and rewind content had a mean score of M=4.9 revealing an above

average importance to the respondents. Perhaps the ability to use television much like a

VCR appealed to the sample.

The ability to watch programming at anytime (asynchronous), rated on the same

scale, revealed interesting results. The ability to watch broadcast network content was

rated the highest with the average for broadcast network content at M=4.97. Cable was

rated second highest with the average for cable network content at M=4.68. Premium

content had the third highest score with the average a M=3.91. Pay-per-view new release

movies was rated the lowest with a M=3.56 score. This finding is interesting since the

bulk of VOD services currently offered are in the realm of pay-per-view and premium

content. Broadcast programming and cable programming were rated most important and

yet they are not as widely available as the others.

Table 4-7 shows the mean and standard deviation scores for VOD functions.

Table 4-7. VOD Functions
Std.
VOD Functions Mean Deviation
Ability to Stop, Fast-Forward, Rewind, Pause TV Content 4.91 1.94
Watch Broadcast Network Programming Anytime 4.97 1.83
Watch Cable Network Programming Anytime 4.68 1.78
Watch Premium Network Programming Anytime 3.91 2.09
Watch Pay-Per-View Network Programming Anytime 3.56 2.19
Rated on a 1-7 scale with 1 being not important to 7 very important









The total sample found the advertising models less than desirable based on the

average ratings from the seven-point scale with one being not at all likely and seven

being most likely. The average rating from respondents to pay $2.99 per movie with a

total of ten minutes of standard commercials was M=2.15. The average rating for twenty

minutes of commercials at $1.99 was M=1.93. For banner or ticker advertising

throughout the entire feature, respondents rated the $2.99 charge M=1.85 and the $1.99

charge M=2.2. Clearly, the respondents are not very likely to do any of the proposed

adverting scenarios but the banner/ticker advertising scenario with respondents paying

$1.99 for the program was found to be most popular of the choices.

In a general sense, the respondents found the idea of releasing personal

information to advertisers for a discount or being exposed to targeted advertising for a

discount on pay-per-view movies less desirable as well. The average score for releasing

personal information was M=2.6, based on a seven-point scale with one being not likely

at all and seven being most likely. The average score for exposure to targeted advertising

was slightly higher at 2.84. Both mean scores are similar and again it appears respondents

do not seem likely to want targeted advertising and discounts delivered by releasing

personal information. All mean scores and standard deviations are included in table 4-8.

Table 4-8. Advertising Scenarios and Models
Std.
VOD Advertising Scenarios and Models Mean Deviation
Willingness to Provide Personal Information 2.61 1.79
Willingness to be Exposed to Targeted Advertising 2.84 1.81
Willingness to Pay $2.99 for 10 minutes of Advertising (commercials) 2.15 1.57
Willingness to Pay $1.99 for 20 minutes of Advertising (commercials) 1.93 1.5
Willingness to Pay $2.99 for Banner or Ticker Type Advertising 1.85 1.51
Willingness to Pay $1.99 for Banner or Ticker Type Advertising 2.2 1.86
Rated on a 1-7 scale with 1 being not at all likely to 7 being most likely









Respondents were asked to rate their usage of newspapers, radio, television,

magazines, video rental, movie going and the Internet. Ratings were based on a one to

seven scale with one being not a user at all to seven being a heavy user. Television had

the highest rating at M=5.61 with radio second at M=5.57, Internet third at M=5.42,

newspaper forth at M=4.57, magazines fifth at M=4.21, video rental sixth at 3.91 and the

movie going usage mean usage score was M=3.57. It appears the respondents in our

sample are electronically savvy since television, radio and Internet ranked in the top

three. This finding leads to a possible conclusion that our sample uses more electronic

media when compared to other types of media. The mean scores and standard deviations

can be found in Table 4-9.

Table 4-9. Media Use

Media Mean Std. Deviation
Television Use 5.61 1.49
Radio Use 5.57 1.47
Internet Use 5.42 1.87
Newspaper Use 4.57 1.97
Magazine Use 4.21 1.6
Video Rental Use 3.91 1.73
Movie Going Use 3.57 1.7
Rated on a 1-7 scale with 1 do not use at all to 7 use very heavily

For the social and innovativeness characteristics, respondents were asked to rate a

number of social activities on a one to seven scale with one being strongly disagree and

seven being strongly agree. "I like to participate in social activities" had the highest

mean score at M=5.81 while "I am in good shape financially" had the lowest at M=4.71.

All respondent's scores were above average leading to the conclusion this sample is

biased toward having above average social and innovative characteristics. Table 4-10









displays all the mean and standard deviation scores for the social and innovative

activities.

Table 4-10. Social and Innovativeness Characteristics
Social and Innovative Characteristics Mean Std. Deviation
I like to learn about new ideas 5.82 1.41
I like to participate in social activities 5.81 1.54
I like to explore new technologies 5.42 1.53
I like to try new products 5.41 1.51
I enjoy interacting with my neighbors 5.12 1.67
I keep up with new technologies 5.04 1.46
I am willing to take risks in order to try new
things 4.88 1.48
I am in good shape financially 4.71 1.43
Rated on a 1-7 scale with 1 being strongly disagree to 7 strongly agree

Respondents were asked if they owned a number of new technology items. Of

those items, a majority of the respondents own a personal computer (89%), a cellular

phone (87%), and/or a DVD player (79%). Personal digital assistants were the least

owned technology (16%) by the total sample. Overall this sample is very technological

savvy. The sample has an extraordinarily high number of computer owners since

according to the National Telecommunication and Information Association the computer

ownership average for the general population is 56.7% (NTIA, 2002). Further, 34.4% of

the general population subscribe to high speed Internet services compared to the 58% of

our sample (URL wire, 2003). This leads to another important point of caution regarding

projecting these findings into the general population since the current sample seems to be

very technologically savvy. Total ownership frequency and percent are listed in table 4-

11.









Table 4-11. Technology Ownership
Technology Ownership Frequency Percent
Personal Computer 152 89%
Cell Phone 148 87%
DVD 134 79%
High-speed Internet 99 58%
Video Game System 89 52%
Cable Television without Set-
Top-Box 72 42%
Cable Television with Set-Top-
Box 71 42%
Large Television (35" or larger) 68 40%
Digital Video Recorder 63 37%
Home Theatre 58 34%
High-Definition Television 38 22%
Direct Broadcast Satellite 34 20%
Personal Digital Assistant 27 16%


Research Question 1

Research question 1 looked to examine what functions of VOD were found to be

most preferred. All measured respondents were aware of VOD. Independent sample t-

tests were performed yielding only one significant result and one result of marginal

significance. The mean score of those who have used VOD (M=4.72) was statistically

significantly higher than the mean score of those who had not used VOD (M=3.87) when

analyzing the ability to view premium network programming anytime, (t=-1.913, d.f.=

92, p=.03). For pay-per-view, a marginal statistical significance relationship was

discovered (t=-1.51, d.f=92, p=.067). Those who had used VOD (M=4.5) rated the

viewing pay-per-view anytime higher than those non-users (M=3.78). Therefore, users of

VOD prefer the ability to view premium network and pay-per-view programming over

other VOD options. All results of preferred VOD options between users and non-users

can be found in table 4-12.









Table 4-12. Preferred VOD Options Between Users and Non-Users of VOD
Used VOD
VOD Functionality (Rewind, FF, PauseTV Content) Yes No t df p
Mean 5.3 5.0 -.766 92 .223
Standard Deviation 1.77 1.94
Watch Broadcast Network Programming Anytime
Mean 4.93 5.09 .426 92 .336
Standard Deviation 1.789 1.96
Watch Cable Network Programming Anytime
Mean 4.72 4.61 -.308 92 .380
Standard Deviation 1.71 1.82
Watch Premium Network Programming Anytime
Mean 4.72 3.87 -1.913 92 .03*
Standard Deviation 2.15 2.14
Watch Pay-Per-View Programming Anytime
Mean 4.5 3.78 -1.51 92 .067
Standard Deviation 2.25 2.3
* indicates significant mean difference at p<.05
One-tailed t-test


In addition to the comparing option between users and non-users, comparisons

were also made examining the mean scores of each VOD option among the current users.

Paired sample t-tests were used to statistically compare the means and each option was

compared against another. Generally, VOD's functionality (M=5.3) was rated the highest

with the ability to watch broadcast networks (M=4.93) having the second highest mean

score. The mean scores for having the ability to watch cable programming (M-4.72) and

premium cable programming (M-4.72) were the same. The only difference of means

found statistically significant (t=-2.327, d.f.=39, p=.013) was the comparison of the

highest mean, VOD functionality (M=5.3) and the ability for VOD users to watch pay-

per-view programming anytime (M=4.5). The results are shown in Table 4-13









Table 4-13. Mean Comparison of VOD Options by VOD Users
Std.
Mean Devation t df p
VOD Functionality (Rewind, FF,
PauseTV Content) 5.3 1.79 1.228 39 0.114
Watch Broadcast Network
Programming Anytime 4.93 1.79
Watch Cable Network Programming
Anytime 4.72 1.79 0.831 39 0.205
Watch Broadcast Network
Programming Anytime 4.93 1.79
Watch Premium Network
Programming Anytime 4.72 1.71 0 39 0.5
Watch Cable Network Programming
Anytime 4.72 1.79
Watch Pay-Per-View Programming
Anytime 4.5 4.72 0.577 39 0.284
Watch Premium Network
Programming Anytime 4.72 2.15
VOD Functionality (Rewind, FF,
PauseTV Content) 5.3 1.786 2.327 39 0.013*
Watch Pay-Per-View Programming
Anytime 4.5 2.253
indicates significant mean difference at p<.05 One-tailed t-test


Research Question 2

Research question 2 analyzed the exploratory advertising models based on pay-

per-view movies as well as attitudes regarding giving up personal information and

receiving targeted advertising.

The goal of this first part of this research question was to identify which would be

the most preferred advertising scenario. Independent t-tests were run to compare mean

rating scores between the commercial scenario and then between the ticker/banner

scenarios. Last, all mean rating scores for the advertising models were compared against

each other. All mean rating scores were derived from the total sample population and a

total of six advertising scenarios were analyzed.









The first comparison, Ad Scenario #1, examined the difference in means, based

on a one to seven scale with one being not at all likely and seven being most likely,

between paying $2.99 for 10 minutes of advertising (M=2.15) and paying $1.99 for 20

minutes of advertising (M=1.93).In the first comparison it appears respondents were not

very likely to approve of either advertising method for a discount on the movie. Still,

paying $2.99 for 10 minutes of television advertising was rated higher and the

relationship was found to be statistically significant compared than paying $1.99 for 20

minutes of advertising, (t=2.57, d.f.=167, p=.006). The results are shown in table 4-14.

Table 4-14. Ad Scenario #1
$2.99 for 10 $1.99 for 20
Minutes of Minutes of
Advertising Advertising t df p
Mean 2.15 1.93 2.57 167 .006*
Standard Deviation 1.57 1.5
indicates significant mean difference at p<.05
One-tailed t-test

The second scenario, Ad Scenario #2, compared the mean rating score toward

paying the $2.99 ticker or banner advertising option with the mean rating score of paying

$1.99 for the ticker or banner advertising option. The mean score for the $1.99 ticker

option (M=2.2) was higher and statistically significantly different than the mean rating

for the $2.99 option (M=1.85), (t=-4.08,d.f.=167, p=.001). However, the score was still

low and considered an unlikely advertising model by the respondents based on the one to

seven scale. The t-test results are listed in table 4-15.

Ad Scenario #3 again uses t-tests to compare the mean ratings of $2.99 for 10

minutes of television commercials and the mean rating of paying $2.99 for ticker

advertising. The mean for paying $2.99 for the commercials (M=2.15) was slightly

higher and the relationship was found to be statistically significant than the mean score









Table 4-15. Ad Scenario #2
$1.99 for
$2.99 For Ticker Ticker
Advertising Advertising t df p
Mean 1.85 2.2 -4.08167.001*
Standard Deviation 1.5 1.86
indicates significant mean difference at p<.05
One-tailed t-test

for paying $2.99 for ticker advertising (M=1.57), (t=2.22, d.f.=167, p=.014). Respondents

prefer commercial advertising compared to banner advertising at the $2.99 price level.

Still, based on these low mean scores it appears respondents were not favorable toward

these advertising scenarios. The results are displayed in table 4-16.

Table 4-16. Ad Scenario #3
$2.99 for 10 $2.99 For
Minutes of Ticker
Advertising Advertising t df p**
Mean 2.15 1.85 2.22 167 .014*
Standard Deviation 1.57 1.51
* indicates significant mean difference at p<.05
One-tailed t-test

Ad Scenario #4 looked at the mean differences between paying $1.99 for 20

minutes of commercial advertising during a pay-per-view movie and paying $1.99 for

ticker advertising during a pay-per-view movie. The mean score for paying $1.99 for

ticker advertising (M=2.2) was higher than paying $1.99 for 20 minutes of commercial

(M=1.93), (t=-1.892, d.f.= 167, p=.03). The relationship between the two means was also

found to be statistically significant. At the $1.99 pricing point, respondents found the

ticker advertising marginally favorable. The results are listed in table 4-17.

Ad Scenario #5 compared the mean scores of paying $2.99 for 10 minutes of

commercial advertising with paying $1.99 for ticker advertising. The t-test did not reveal

the mean scores were significantly different. Although the mean score for $1.99 ticker









Table 4-17. Ad Scenario #4
$1.99 for 20 $1.99 For
Minutes of Ticker
Advertising Advertising t df p**
Mean 1.93 2.2 -1.892 167 .03*
Standard Deviation 1.5 1.86
* indicates significant mean difference at p<.05
One-tail t-test

advertising was the rated the highest by the respondents, the difference in means is not

substantial. The findings for Ad Scenario #5 are listed in table 4-18.

Table 4-18. Ad Scenario #5
$2.99 for 10 $1.99 For
Minutes of Ticker
Advertising Advertising t df p
Mean 2.15 2.2 -.338 167 .368
Standard Deviation 1.57 1.86
* indicates significant mean difference at p<.05
One-tailed t-test

Ad Scenario #6 compared the mean rating scores for respondents paying $1.99 for

20 minutes (M=1.93) of commercial advertising during a pay-per-view program and

paying $2.99 for ticker advertising (M=1.85) during the pay-per-view program.

According to the mean scores, these were the least popular advertising scenarios. The

findings were not statistically significant and are displayed in table 4-19.


Table 4-19. Ad Scenario #6

$1.99 for 20 $2.99 For
Minutes of Ticker
Advertising Advertising t df p**
Mean 1.93 1.85 .607 167 .273
Standard Deviation 1.5 1.5
* indicates significant mean difference at p<.05
**Sig. 1-tailed









Important to advertisers is knowing if respondents are willing to provide personal

information so advertisers can send specific, targeted advertising. Using a one to seven

scale with one being not at all likely and seven being most likely mean scores were

derived and t-tests were used to compare mean scores of those respondents who are

aware and not aware of VOD. A comparison was also done of respondents who are

aware. This comparison examined those who have used VOD and those who have not

used VOD. Overall findings were not statistically significant. The mean averages were

also quite low, concluding respondents were not likely to release personal information for

a discount on pay-per-view movies. The results are shown in figure 4-20.

Table 4-20. Willingness to Provide Personal Information
Aware of VOD
Yes No t df p
Mean 2.6 2.54 -.222 161 .412
Standard Deviation 1.75 1.86
* indicates significant mean difference at p<.05

Used of VOD
Yes No t df p
Mean 2.6 2.85 .663 91 .255
Standard Deviation 1.63 1.91
* indicates significant mean difference at p<.05

Attitude ratings were also collected to determine if respondents would like to

receive advertising specific to their interests. t-tests were done to determine a difference

in mean between those who are currently aware of VOD and also between those who are

aware and have used VOD. These mean scores were also quite low concluding

respondents were not likely to be willing to be exposed to targeted advertising.

Interestingly enough, those who are not aware of VOD had the highest approval rate

toward targeted advertising (M=3.11) compared to all others. No statistically significant









difference in means could be concluded in either case. The mean scores and t-test results

are displayed in table 4-21.

Table 4-21. Willingness to be Exposed to Targeted Advertising

Aware of VOD
Yes No t df p
Mean 2.92 2.63 -.998 161 .16
Standard Deviation 1.88 1.69
indicates significant mean difference at p<.05

Used of VOD
Yes No t df p
Mean 2.6 3.11 1.31 91 .096
Standard Deviation 1.89 1.85


Research Question 3

For the third research question, this study attempts to discern the characteristics

between those who are aware of VOD and those who are not aware of VOD technology.

To answer this question, depending on level of measurement, independent sample t-tests

and chi-square tests were conducted. t-tests were conducted for interval or continuous

variables such as the social/innovative characteristics and media use while chi-square

tests were conducted with categorical variables such as demographic data and technology

ownership.

Our first variable was media use. For media use, independent t-tests were

conducted. There were two significant findings between those in the sample who are

aware of VOD and those who are not. The mean for television use was higher for those

who are aware of VOD (M=5.76) compared to those who are not aware (M=5.35), and

the result was statistically significant (t= -1.735, d.f.=165, p=.043). The second

statistically significant mean difference was from Internet usage. Those who are aware of









VOD rated Internet usage higher (M=5.75) compared to those who are not aware of VOD

(M=5.03), (t= -2.505, d.f. 165, p=.006). The findings suggest that people who are more

aware of VOD tend to use television and the Internet more than those who are not aware

of VOD. This result implies those who are aware of VOD were possibly exposed to VOD

through the use of television or the Internet. Total media use mean scores are shown in

table 4-22.

Table 4-22. Mean Media Use Scores Between Those Aware and Not Aware of VOD

Aware Not Aware t df p
Newspaper Use
Mean score 4.67 4.37 -.958 165 .170
Std. Deviation 1.86 2.14
Radio Use
Mean score 5.61 5.46 -.647 165 .260
Std. Deviation 1.41 1.56
Television Use
Mean score 5.76 5.35 -1.735 165 .043*
Std. Deviation 1.37 1.61
Magazine Use
Mean score 4.36 4 -1.438 164 .076
Std. Deviation 1.71 1.41
Video Rental
Mean score 3.99 3.78 -.770 165 .221
Std. Deviation 1.75 1.71
Movie Going
Mean score 3.71 3.38 -1.203 165 .116
Std. Deviation 1.75 1.67
Internet
Mean score 5.75 5.03 -2.505 165 .006*
Std. Deviation 1.58 2.12
indicates statistically significant mean difference at p<.05
One-tailed t-tests


Social and innovative characteristics scores were analyzed to discern any

significant difference between those in the sample who are currently aware of VOD and

those who are not aware of VOD. Independent group t-tests were run with no statistically









significant differences in mean scores found. These results appear to tell us that there are

no important social or innovative characteristics separating those who are aware of VOD

and those who are not aware of VOD. The results are displayed in table 4-23.

Table 4-23. Social/Innovative Mean Scores Between Those Aware and Not Aware of
VOD
Yes No t df p
I like to try new products
Mean score 5.47 5.37 -.460 165 .323
Std. Deviation 1.47 1.49
I like learning about new technologies
Mean score 5.42 5.43 .009 165 .497
Std. Deviation 1.59 1.46
I like to socialize
Mean score 5.83 5.85 .104 165 .4585
Std. Deviation 1.53 1.47
I like to socialize with my neighbors
Mean score 5.12 5.18 .221 164 .413
Std. Devation 1.67 1.63
I am in good financial standings
Mean score 4.69 4.69 .019 165 .4925
Std. Devation 1.43 1.44
I like to learn about new ideas
Mean score 5.91 5.78 -.614 164 .27
Std. Devation 1.33 1.43
I like to learn about new technologies
Mean score 5.09 4.96 -.581 165 .281
Std. Devation 1.51 1.43
I like to take risks
Mean score 4.92 4.84 -.351 165 .36
Std. Devation 1.52 1.38
indicates statistically significant mean difference at p<.05
One-tailed t-tests


For technology ownership, chi-square tests were run on the categorical variables

examining the relationship between ownership of various new technology items to VOD

awareness. Seven statistically significant relationships were found. Those who currently

subscribe to high speed Internet services were significantly more aware of VOD than









those who do not, (x2 = 3.898, d.f.=1, p=.035). High definition television ownership was

the second significant finding. There were significant differences found in both those

who own and do not own high definition televisions and awareness of VOD, (x2 = 4.039,

d.f =1, p=.033). Those who do not own big screen televisions were significantly less

aware of VOD compared to all others, (x2= 3.322, d.f.=l1, p=.048). Those respondents

who own DVD players were found to be significantly more aware of VOD compared to

those who do not own DVD players, (x2= 5.189, d.f.=l1, p=.019). Those respondents who

subscribe to cable with a set-top box were significantly more aware of VOD compared to

those who do not, (x2= 7.662, d.f.=l1, p=.004). Interestingly enough, those subscribing to

cable without a set-top box were also significantly more aware of VOD, (x2= 3.389,

d.f=l, p=.046). Last, those who subscribe to premium cable were significantly more

aware of VOD technology compared to those respondents who do not subscribe,

(x2 5.342, d.f.=l1, p=.015). In short, the results show those who are aware of VOD tend to

own high-speed Internet, high definition television, large screen TVs, DVD players,

subscribe to cable television with or without a set-top-box and subscribe to premium

cable services. This finding suggests those respondents who own television and Internet

related products tend to be more aware of VOD services.

As for the demographic characteristics, education was the only demographic

characteristic significantly related to VOD awareness. Those respondents who attended

college were significantly more aware of VOD technology than those who did not attend

college, (x2=10.148, d.f. =3, p=.009). All demographic comparisons and statistical

information can be found in table 4-25.









Total results for technology ownership can be found in table 4-24.

Table 4-24. Technology Ownership Mean Scores Between those Aware and Not Aware
of VOD

Aware of Not Aware
VOD ofVOD
Ownership Yes No Yes No x2 df p
DVR 32 67 29 37 2.293 1 .089
PC 91 8 58 8 1.245 1 .196
High-Speed Internet 64 35 33 34 3.898 1 .035*
Video Game System 51 48 36 31 .079 1 .452
Cell Phone 86 13 59 8 .051 1 .509
Home Theater System 37 62 20 47 1.003 1 .202
High Definition Television 28 71 10 57 4.039 1 .033*
Large Screen TV (35" or larger) 45 54 21 46 3.322 1 .048*
DVD Player 84 15 47 20 5.189 1 .019*
Personal Digital Assistant 17 82 9 58 .423 1 .336
Cable TV with set-top-box 51 48 20 47 7.662 1 .004*
Cable TV without set-top-box 36 63 34 33 3.389 1 .046*
Direct Broadcast Satellite 19 80 13 54 .001 1 .563
Premium Cable 52 47 23 44 5.342 1 .015*
* indicates significant mean difference at
p<.05
One-tailed t-test

Table 4-25. Demographic Scores Between those Aware and Not Aware of VOD
Aware of Not Aware of
Gender VOD VOD x2 df p
Male 30 18 .230 1 .382
Female 69 49
indicates significant mean difference at p<.05


Aware of Not Aware
Education VOD of VOD x2 df p
Less Than High School 1 1 10.148 3 .009*
High School 17 26
College 62 32
Graduate School 19 8
* indicates significant mean difference at p<.05









Table 4-25.
Aware of
VOD


Continued
Not Aware of
VOD


x2 df p


Single 43 24 3.818 4 .216
Married 48 34
Divorced 8 7
Widow 0 1
Other 0 1
* indicates significant mean difference at p<.05

Age Aware of VOD Not Aware of VOD
18-24 23 15
25-34 28 18
35-44 26 14
45-54 18 12
55-64 3 6
65+ 0 1
One-tailed t-test
(t= -.774, df=163, p=.22)

Household Size Aware of VOD Not Aware of VOD
1 11 10
2 38 9
3 8 15
4 27 17
5 12 11
6 3 3
One-tailed t-test
(t= -1.312,df=162,p=.095)


Not Aware of
Occupation Aware of VOD VOD x2 df p
Laborer 4 1 10.352 8 .241
Machine/Service Tech 1 1
Craftsman 2 1
Clerical 6 5
Sales 10 9
Administrator 3 2
Professional 44 17
Manager 7 3
Other 22 27









Table 4-25. C
Aware of Not Aware of
VOD VOD
<$19,000 6 5
$20k-$39k 20 11
$40k-$59k 19 12
$60k-$79K 18 12
$80k-$99k 10 13
>$100k 14 19
* indicates significant mean difference at p<.05
One-tailed t-test


continued

x2 df p
.613 5 .987


Research Question 4

Research question 4 examined what factors indicate early adopters of VOD

technology. This question attempts to discern the characteristics between those who are

aware of VOD and subscribe from those who are not subscribers of VOD technology. To

answer this question, descriptive statistics were analyzed using independent sample t-

tests and chi-square tests were used. t-tests were conducted for interval or continuous

variables such as the social/innovative characteristics, importance of viewing content

anytime, VOD functionality (ability to fast forward, rewind, pause) and media use while

chi-square tests were conducted with categorical variables such as demographic data and

technology ownership.

For the media usage score comparisons between subscribers and non-subscribers,

independent t-tests were run. There were no significant differences found between

subscribers and non-subscribers of VOD and media use. For this study, there is no

relationship between media usage and the likelihood of VOD adoption. Compared with

previous studies......The findings are displayed in table 4-26.









Table 4-26. Media Usage Comparisons between Subscribers and Non-subscribers of
VOD
Media Type Subscriber Non-Subscriber t df p
Newspaper Use
Mean score 4.67 4.8 .299 91 .383
Std. Deviation 1.79 1.86
Radio Use
Mean score 5.25 5.74 1.46 91 .074
Std. Deviation 1.45 1.4
Television Use
Mean score 5.92 5.78 -.414 91 .34
Std. Deviation 1.06 1.45
Magazine Use
Mean score 4.25 4.5 .615 90 .27
Std. Deviation 1.94 1.63
Video Rental
Mean score 4.08 3.97 -.272 91 .393
Std. Deviation 1.67 1.77
Movie Going
Mean score 3.63 3.72 .233 91 .408
Std. Deviation 1.86 1.78
Internet
Mean score 6 5.77 -.629 91 .266
Std. Deviation 1.53 1.56
indicates significant mean difference at p<.05
One-tailed t-test


Much like the media usage score comparisons between subscribers and non-

subscribers, the social/innovative characteristic comparisons found no significant mean

score differences between subscribers and non-subscribers of VOD. Further, there

appears to be no relationship between subscribers, non-subscribers and social/innovative

characteristics. The findings are displayed in table 4-27.

For technology ownership, chi-square tests were run on the categorical variables

examining the relationship between ownership of various new technology items to VOD

subscribers. Three statistically significant differences were discovered. Of those









Table 4-27. Social/Innovative Characteristics Comparisons between Subscribers and
Non-subscribers of VOD
Non-
Subscriber Subscriber t df p
I like to try new products
Mean score 5.75 5.45 -.86 91 .2
Std. Deviation .989 1.605
I like learning about new technologies
Mean score 5.75 5.42 -.89 91 .19
Std. Deviation 1.36 1.63
I like to socialize
Mean score 6.25 5.78 -1.35 91 .09
Std. Deviation 1.19 1.54
I like to socialize with my neighbors
Mean score 4.79 5.23 1.13 91 .13
Std. Devation 1.81 1.59
I am in good financial standings
Mean score 4.38 4.64 .74 91 .23
Std. Devation 1.34 1.02
I like to learn about new ideas
Mean score 5.92 5.94 .08 91 .47
Std. Devation 1.02 1.46
I like to learn about new technologies
Mean score 5.12 5.03 -.26 91 .4
Std. Devation 1.54 1.58
I like to take risks
Mean score 4.71 4.91 .56 91 .3
Std Devation 1.27 1.64
* indicates significant mean difference at p<.05
One-tailed t-test

significant differences, more people who are non-subscribers also have cell phones,

(x2=7.61, d.f.=l1, p=.011). This finding is probably due to the fact that most all

respondents who are aware of VOD also own cell phones (83%) and more people overall

are non-subscribers. Those who subscribe to cable without a set-top box are less likely to

subscribe to VOD services, (x2=10.416, d.f=l, p=.001). This finding is due to the fact in

order to have VOD service, you must have a set-top box thus non-adopters would not

have a set-top-box. Last, those who subscribe to direct broadcast satellite are statistically









less likely to subscribe to VOD services, (x2=9.354, d.f.= 1, p=.004). A possible

interpretation of this finding could involve the simple fact that VOD technology is not

available to direct broadcast satellite subscribers. Total mean scores and chi-square

values can be seen in table 4-28.

Table 4-28. Technology Ownership Comparisons between Subscribers and Non-
subscribers of VOD
Non-
Subscriber Subscriber
Ownership Yes No Yes No x2 df p
DVR 12 12 21 48 2.98 1 .071
PC 21 3 67 2 3.23 1 .106
High-Speed Internet 16 8 44 25 .065 1 .502
Video Game System 14 10 35 34 .414 1 .343
Cell Phone 17 7 64 5 7.61 1 .011*
Home Theater System 13 11 23 46 3.26 1 .060
High Definition Television 10 14 19 50 1.66 1 .151
Large Screen TV (35" or larger) 12 12 28 41 .645 1 .286
DVD Player 20 4 62 7 .756 1 .302
Personal Digital Assistant 2 22 14 55 1.787 1 .153
Cable TV with set-top-box 16 8 34 35 2.17 1 .108
Cable TV without set-top-box 2 22 31 38 10.416 1 .001*
Direct Broadcast Satellite 9 15 7 62 9.354 1 .004*
Premium Cable 17 7 34 35 3.341 1 .055
* indicates significant mean difference at p<.05
One-tailed t-test

Comparing the demographic variables of subscribers and non-subscribers, only

two statistically significant characteristics were found: education and marital status. For

education, a significant difference was found where respondents with only a high school

education were less likely to be subscribers of VOD, (x2=7.085, d.f.=l1, p=.014). Marital

status also yielded a statistical difference with single respondents less likely to subscribe

to VOD services, (x2= 7.421, d.f.=l1, p=.012). The gender of the respondents yielded a

marginally statistically significant difference with 40% of the men subscribing and 22%

of the women subscribing (x2=3.6, d.f.=l1, p=.054). This difference could be because









there are two and half times as many women in the sample. All demographic variables

and counts are listed in table 4-29.

Table 4-29. Demographics Comparisons between Subscribers and Non-subscribers of
VOD
Subscribe to VOD
Yes No x2 df p
Male 10 15 3.6 1 .054
Female 14 49
indicates significant mean difference at p<.05
One-tailed t-test


Subscribe to VOD


Yes No x2 df p
Less Than High School 0 0 7.085 1 .014*
High School 3 16
College 20 37
Graduate School 1 16


* indicates significant mean difference at p<.05
One-tailed t-test


Subscribe to VOD


Yes No x2 df p
Single 5 36 7.451 1 .012*
Married 15 28
Divorced 4 5
Widow 0 0
Other 0 0
indicates significant mean difference at p<.05
One-tailed t-test


Subscribe to VOD
Age Yes No Mean S.D. t df p
18-24 0 21 36.25 9.166 -0.913 90 0.182
25-34 12 17
35-44 6 17
45-54 6 10
55-64 0 3
65+ 0 0
* indicates significant mean difference at p<.05
One-tailed t-test









Table 4-29. Continued
Subscribe to
VOD

Yes No mean S.D.


t df p


1 3 8 3.17 1.523 0.693 90 0.245
2 8 25
3 2 8
4 5 18
5 5 8
6 1 1


Subscribe to VOD
Occupation Yes No x2 df p
Laborer 0 3 8.536 8 .191
Machine/Service Tech 0 1
Craftsman 1 0
Clerical 1 5
Sales 2 8
Adminsitrator 0 3
Professional 11 28
Manager 1 8
Other 8 13
indicates significant mean difference at p<.05
One-tailed t-test



Subscribe to VOD
Yes No x2 df p
<$19,000 1 0 1.684 5 .446
$20k-$39k 6 13
$40k-$59k 4 14
$60k-$79K 6 11
$80k-$99k 3 8
>$100k 4 13
indicates significant mean difference at p<.05


VOD allows viewers to watch television content on-demand and it is important to

examine the importance of this capability between those who currently subscribe to VOD

and those who do not currently subscribe. For all television programs, VOD subscribers


Household
Size









tend to believe VOD's viewing anytime feature is important compared with the non-

subscribers. However, of the four programming choices, only the pay-per-view movie

mean average for subscribers (M=5.21) was higher with statistical significance than the

non-subscribers (M=3.71), (t=-2.86, d.f.=91, p=.003). Therefore, the result generally

suggests that viewing anytime is important, especially for pay-per-view programs.

Results can be found in table 4-30 for all television content program producers.

Table 4-30. Viewing TV Content Anytime Comparisons between Subscribers and Non-
subscribers of VOD
Subscribe to VOD
On-demand Programmning Yes No t df p
Broadcast Network Programs
Mean 5.17 4.96 -.468 91 .32
Std. Deviation 1.58 1.99
Cable Network Programs
Mean 5.12 4.52 -1.45 91 .076
Std. Deviation 1.70 1.78
Premium Network Programs
Mean 4.79 4.09 -1.38 91 .086
Std. Deviation 2.19 2.15
Pay-Per-View Programs
Mean 5.21 3.71 -2.86 91 .003*
Std. Deviation 2.14 2.23
* indicates statistically significant mean difference at p<.05
One-tailed t-test


The functionality of VOD, or the ability to stop, rewind, fast forward and pause

television content was also measured between subscribers and non-subscribers of VOD.

The mean score of importance of VOD functionality for subscribers (M=5.71) was than

non-subscribers (M=4.86), (t=1.922, d.f.=91, p=.029). Therefore, this result confirms

that the function of stop, rewind, fast forward, and pause is very important for VOD

usage among VOD subscribers. The results can be found in Table 4-31.









Table 4-31. VOD Functionality Comparisons between Subscribers and Non-subscribers
of VOD

Subscribe to VOD
VOD Functionality Yes No t df p
Mean 5.71 4.86 1.922 91 .029*
Std. Deviation 1.40 2.01
indicates significant mean difference at p<.05
One-tailed significance

Research Question 5

Research question 5 attempted to indicate factors contributing to intent to adopt

VOD technology in the next six months. Respondents included in this analysis had to be

aware of VOD but not current subscribers. Multiple regression analysis was used to

indicate the relative power of variables to explain intent to adopt VOD technology. The

independent variables measured included the sum of all the social/innovative

characteristics scores, media use scores, technology ownership scores, importance of

viewing television content anytime, importance of VOD functionality, gender, income,

education and household size. The dependent variable was the rating of intent to adopt

VOD technology over the next six months.

Twenty-seven independent variables were measured yielding five statistically

significant variables (R=.55, p<.001). The multiple regression model explained 30% of

the variation of the mean toward intent to adopt VOD over the next six months. The most

important variable to predict "intent to adopt VOD" was "Internet Usage" (Beta=.348,

p=.002) followed by respondents having the ability to "watch cable networks anytime"

(Beta=.279, p=.031). Interestingly enough, a negative relationship was found with those

respondents who currently subscribe to premium cable (Beta=-.233, p=.044). The results

indicate people are more likely to adopt VOD technology if they currently use the









Internet, value watching cable programming at their own discretion but possible do not

subscribe to premium cable services. The multiple regression results are listed in figure 4-

32.


Table 4-32. Multiple Regression Data


Model Summary
R R Square df F p
.550 .302 5 5.2 .000


Variable B Std. Error Beta t p
Internet Usage .292 ..091 .348 3.201 .002*
Video Game System Ownership .528 .310 .19 1.704 .094
Premium Cable -.649 .316 -.054 -.417 .687
Watch Premium Channels Anytime -.036 .085 -.244 -2.284 .026*
Watch Cable Networks Anytime .227 .103 .279 2.212 .031*
* indicates p<.05














CHAPTER 5
SUMMARY AND CONCLUSIONS

The current study adds to the body of knowledge regarding the diffusion of

innovations theoretical framework applied to the adoption of new technologies.

Additionally this study adds to the body of knowledge in regards to video-on-demand

adoption and the findings display significant results for practitioners in the fields of

advertising and telecommunication.

The first research question looked to examine what aspects of VOD functionality

were found to be most preferred among those respondents who are currently aware of

VOD technology. This research question compared the average scores of those who have

previously used VOD to those who have not previously used VOD. Aspects of VOD

functionality include VCR type capabilities (ability to rewind, fast-forward, and pause

video content) in addition to the ability to watch broadcast, cable, premium (HBO,

Showtime), and pay-per-view content at the view's discretion. This study's findings

conclude that the VCR type functionality was rated the highest of all VOD attributes,

although no statistical significant difference between users and non-users was found.

However, preference among users was found to be statistically significantly higher

involving the ability to watch premium network programming and marginally statistically

significant among users in regards to pay-per-view programming. These findings suggest

users prefer the VCR functions of VOD in addition to the ability to watch premium and

pay-per-view video content.









This study also compared the mean scores within of each aspect of VOD

functionality among current users to examine if any statistical differences were present.

The findings suggest VOD users prefer the VCR functionality above all other aspects of

functionality.

For the practitioner looking to communicate the attributes and benefits of using

VOD, communicating the VCR functionality aspects of VOD could prove to be most

positive. Perhaps a message tailored toward displaying or showing how the functionality

works could prove to be a unique selling point for VOD. Additionally, another unique

selling point could be the viewer's ability to watch pay-per-view and premium content

asynchronously. The findings from research question one add to the formulation for

creating message strategy and this insight could aid in VOD positioning.

Research question two examined respondent attitudes as to which exploratory

prospective advertising model for VOD would be found to be most positive. Two items

were measured. Respondent attitudes toward exploratory advertising models, based on

the current pay-per-view format were examined in addition to respondent attitudes

toward giving up personal information and receiving targeted advertising.

The goal of the first part of research question two was to identify which exploratory

advertising scenario would be most preferred. Comparisons were made between paying

$2.99 for pay-per-view content containingl0 minutes of commercial insertion advertising

compared to paying $1.99 for 20 minutes of commercial insertion advertising.

Comparisons were also made for paying $2.99 for ticker type advertising (during the

entire pay-per-view event) compared to paying $1.99 for ticker advertising. Last,

comparisons were made between each advertising scenario to discern any significant









differences among scenarios. However, it should be noted all scores for each proposed

advertising scenario were quite low, falling closer to "not at all likely" on the 1 to 7 scale

with 1 being not at all likely and 7 being most likely.

Total respondents rated the $2.99 for 10 minutes of commercial advertising the

highest of the two commercial insertion scenarios. The finding was found to be

statistically significantly different compared to the $1.99 for 20 minutes of commercial

insertion scenario. As for the ticker type advertising scenarios, respondents rated paying

$1.99 for ticker advertising higher than paying $2.99 for ticker advertising. This finding

was statistically significantly different. Overall, respondents gave the $1.99 ticker

advertising scenario a marginally higher rating but the rating was not statistically

different from the $2.99 commercial advertising scenario. Overall these findings suggest

respondent were not impressed by any of the proposed scenarios but they did find the

$2.99 commercial insertion scenario and the $1.99 ticker scenario most preferred.

Perhaps the practitioner should look to these advertising scenarios as examples of

what not to do. Although the respondents found the $2.99 for 10 minutes of traditional

commercial insertion and the $1.99 ticker scenario most favorable, both were still rated

low and described by respondents as a "least likely" scenario. These four tested scenarios

should be used as a boilerplate for further research and testing to find a more desirable or

less-rejected advertising model.

The second part of research question two involved the respondent's willingness to

give up personal information and receive targeted advertising in exchange for a discount

on pay-per-view programming or other television services. Although the attitude scores

were not found to be statistically significantly different, the scores do shed light on the









respondent's attitudes toward releasing personal information and receiving targeted

advertising at a discount.

In regards to releasing personal information for a discount and willingness to be

exposed to targeted advertising, those respondents aware of VOD were not found to be

any more or less willing to do either compared to those who are not aware of VOD. All

scores were in the "least likely" range on the 1 to 7 scale and there were no significant

differences found.

For the practitioner, these results are not promising. Personal information is

essential for creating and planning targeted advertising. Creating a message especially for

a particular consumer will become more of a necessity as our current media landscape

continues to fragment. Even with the fragmentation, technology allows us to reach

consumers in new, non-traditional ways. Since most respondents stated they were "least

likely" toward releasing personal information and receiving targeted advertising,

practitioners should find other means or exchanges to collect this valuable data.

The third research question attempts to discern the characteristics between those

who are currently aware of VOD and those who are not aware of VOD technology. This

study looked at respondent sample characteristics regarding social and innovativeness

characteristics, media usage, technology ownership and demographic information to see

how those who are aware of VOD differ from those who are currently not aware of VOD.

In regards to social and innovativeness characteristics, this study could not find a

single characteristic different between those who are aware and those who are not. As far

as media use, those who are aware of VOD were found to use television and the Internet

more compared to those who are not aware of VOD. This finding is consistent with









Chan-Olmsted and Chang's (2003) finding that Internet use was related to digital

television knowledge. There were a number of significant differences found between

those who are aware and those who are not aware in regards to technology ownership.

Those who are aware of VOD were found to subscribe more to high speed Internet

access, own more high definition televisions, own large screen televisions, own DVD

players, subscribe to cable service with a set-top-box, and subscribe to premium cable

services. Education level was the only demographic characteristic statistically different

comparing those respondents aware of VOD with respondents unaware of VOD.

Respondents who were aware of VOD were typically more educated, attaining

undergraduate or graduate degrees.

Research question four examined what factors indicate early adopters of VOD

technology. The factors include social and innovative characteristics, importance of

viewing content anytime, VOD functionality (VCR type control), media use, technology

ownership and demographics.

Scores for social and innovative characteristics were not found to be significantly

different between current adopters and non-adopters. This was a surprise since many

previous technology adoption studies have found a positive relationship between early

adopters and varying degrees of social/innovative characteristics (Atkins & Jeffres 1998;

Kang 2002; Lin 1998; Lin & Jeffres 1998; Neuendorf, Atkin, & Jeffres 1998; Eastlick,

1996). Regarding viewing videoing content asynchronously, early adopters rated viewing

programs anytime higher than non-adopters, especially when considering pay-per-view

programming. The functionality aspect of VOD (ability to fast-forward, rewind, pause)

was also found to be an important factor to early adopters compared to non-adopters. For









technology ownership, this study found non-adopters owned more cell phones, subscribe

to cable service without a set-top-box, and subscribe to direct broadcast satellite. This

technology aspect was consistent with Dupagne (1999) who found early adopters did not

own more technology items but the finding is inconsistent with other studies where

technology ownership was considerably higher for early adopters (Chan-Olmsted &

Chang, 2003; Kang 2002). There were no significant differences found for media use

between adopters and non-adopters which is consistent with some previous technology

adoption studies where media use differences were weak overall (Atkin & Jeffres, 1996).

As for demographics, only education and marital status were found to be significantly

different between adopters and non-adopters. Sex was also found to have a marginal

significant difference between adopters and non-adopters. These findings infer early

adopters as more educated, married and male compared to non-adopters. These

demographic findings are somewhat consistent with previous technology studies. Early

adopters are typically more educated (Danko & MacLachlan, 1983; Atkins & Jeffres

1998; Kang 2002; Dupagne 1999; Neuendoft, Atkin & Jeffres, 1998) and typically male

(Chan-Olmsted & Chang, 2003).

The classic diffusion paradigm by Everett Rogers (1995) states early adopters seek

information or innovations more actively, have a higher degree of opinion leadership and

are more highly interconnected through interpersonal networks. This was clearly not the

case in the current study due to the lack of difference in the social and innovative

characteristics between those who are aware of VOD and those who are not. Rogers

(1995) states early adopters have greater exposure to mass media. Here, the findings from

this study were not consistent with Rogers since the overall scores for media usage were









not above average for each media type; with newspaper use, radio use, magazine use, and

movie going each rating lower for early adopters compared to non-adopters. Rogers

(1995) and previous technology diffusion studies look to technology ownership as

another key indicator and the current study was not consistent with those previous

studies. Respondents aware of VOD were not more likely to own new technology items,

signifying a more technological savvy group. As for the demographic findings, Rogers

(1995) states early adopters have more formal years of education. This was true in the

current study since those respondents aware of VOD were found to attain more years of

education. Overall this study does add to the body of knowledge of diffusion research,

staying consistent with some aspects of the paradigm, concluding the theory is consistent.

For the practitioner, this study does add insight into ways to reach and target

potential early adopters. This study found early adopters enjoy viewing programs on a

flexible schedule and VOD functionality as especially important functions. Perhaps steps

could be taken in a media campaign to highlight these important attributes. Additionally,

the findings from this study infer VOD early adopters tend to be skewed more male,

educated and married. These demographic findings could help to establish a potential

target base for message design and targeting of the media campaign.

Research question five attempts to pinpoint which factors are found to be most

significant when trying to predict adoption of VOD technology within the next six

months. Internet usage, the ability to watch cable programming anytime and subscribing

to premium cable were found to be the most significant contributors.

For the practitioner, these three variables could be excellent means of reaching and

communicating the attributes and benefits of VOD to potential adopters. Potential









adopters could be reached on the Internet or by advertising on cable. Additionally,

premium cable services such as HBO or Showtime could communicate to subscribers to

benefits of VOD for their programs.

This study is not without limitations. First, the data was collected via convenience

sample. This nonprobability recruiting measure was used due to lack of research funds.

This method does not permit control over the representiveness of the sample or the ability

to project these findings into the general population (Babbie, p. 179). Additionally, the

sample was generally small and has a female skew.

It is the researcher's belief that the finding related to social and innovative

characteristics suffered due to the fact all respondents were above average in these

categories since the survey was administered on the beach. It is believed more outgoing

and social people would venture to the beach compared to those who do not.

This research also suffers from the same limitations found in other diffusion

research as stated by Rogers (1995) regarding recall and the use of correlational analysis.

Recall can be clouded by time which can affect the accuracy of respondent's recall.

Correlational analysis of survey data often leads to a sense of causality, if the

operationalization of variables is not done correctly. Although the current study is based

on the operationalization of variables from previous studies, some methodological

differences are still present.

Future research should look at VOD adoption utilizing random probability

sampling techniques, looking at adoption longitudinally and perhaps redoing the study at

a later date. First, using a larger, more robust probability sample would give the findings

more validity and would allow these findings to be projected into the population. Second,






83


implementing a longitudinal study would aid in confirming these findings. Last, with

VOD penetration currently low and with rollouts planned in the future, perhaps more

accurate data regarding awareness and adoption could be formulated at a later date.

Overall, future research should help reduce the noted general limitations of diffusion

research as well as add to the body of knowledge regarding diffusion of innovations

research.
















APPENDIX
VIDEO-ON-DEMAND TELEVISION SURVEY

University of Florida College of Journalism and Communication

Please answer all questions below except those you are asked to skip. Please give only ONE answer
to each question, unless specified as "choose all that apply". Mark your answer with an X in the
bracket or write in the space provided. Your input will help us to understand the future of television
entertainment. Thank you for your time.


1. Please tell us to what degree you agree with the following statements (On a 1-7 scale with 1
being strongly disagree and 7 strongly agree).

Strongly Disagree Strongly Agree
I like to try 1 2 3 4 5 6 7
new products I O 0 0 0 0 0

I like to explore 1 2 3 4 5 6 7
new technologies D D D D D D D

I like to participate 1 2 3 4 5 6 7
in social activities D D D D D D D

I enjoy interacting 1 2 3 4 5 6 7
with my neighbors D O D D D D D

I am in good 1 2 3 4 5 6 7
shape financially D O D D D D D

I like to learn 1 2 3 4 5 6 7
about new ideas D D D D D D D


I keep up with 1 2 3 4 5 6 7

new technologies D D D D D D D

I am willing to take
risks in order to 1 2 3 4 5 6 7
try new things D D D D D D 0












2. How important is it for you to be able to view your favorite (or any) program at ANYTIME from
the following sources? (Rate on a 1-7 scale with 1 being not important at all & 7 being very
important)

Not Important at All Very Important
Broadcast Program Choices 1 2 3 4 5 6 7
(e.g. ABC, NBC, FOX, CBS) I O 0 0 0 0 0

Cable Program Choices 1 2 3 4 5 6 7
(e.g. CNN, MTV, ESPN, HGTV) O D D D D D

Premium Program Choices 1 2 3 4 5 6 7
(e.g. HBO, Showtime, Starz) D D D D D D D

Pay-Per-View or 1 2 3 4 5 6 7
New Release Movies D D D D D D D


IPlease Turn Page Over
3. important is having the ability to pause, fast forward, rewind and stop television programs
at anytime much like you would if you were viewing a videotape? (Rate on a 1-7 scale with 1
being not important at all & 7 being very important)

Not Important Very Important
1 2 3 4 5 6 7




4. What is your usage of the following media? (Rate on a 1-7 scale with 1 being not a
user at all & 7 being a heavy user)


1 2 3 4 5 6 7

Newspaper D D D D D D D
Radio D
Television D D D D D D D
Magazine D D D D D D D
Video Rental D D D D D D D
Movie going D D D D D D D
Internet D D D D D D D


Do Not Use at All


Use Very Heavily










5. Are you aware of video-on-demand television services?


YES


NO D If "NO," Please skip to question 6.


If you answered YES to the above question, please answer the next four questions.

How knowledgeable are you about video-on-demand? (Please answer using the 1-7
scale with being not at all knowledgeable and 7 being highly knowledgeable)


Not at all
Knowledgeable


Highly
Knowledgeable
1 2 3 4 5 6 7


Have you ever used video-on-demand before?
YES D NO 0

Do you currently subscribe to video-on-demand type services?
YES D NO D



If you answered NO, what is your intention to subscribe to video-on-
demand type services in the next six months? (On a 1-7 scale with 1
being not at all likely and 7 being most likely to subscribe)


Not at
All likely


Most
Likely


E 1 2 3 4 5 6 7


Please Turn to Next Page


6. Based on pay-per-view movies available through your cable or satellite television provider,
please indicate your attitude toward the following questions:

Would you be willing to provide personal information about your household (such as
income, occupation, and hobbies) to advertisers for a discount on pay-per-view movies or
cable services?


Not at all likely Most likely
1 2 3 4 5 6 7


Would you be willing to be exposed to advertising of products or services that might interest you
for a discount on pay-per-view movies or cable services?

Not at all likely Most likely
1 2 3 4 5 6 7












7. Based on pay-per-view movies available through your cable or satellite
television provider, please indicate your attitude toward the following scenarios
and questions:

If the price of a new release, pay-per-view movie were $3.99, would you be
willing to:

Pay $2.99 for the same movie in exchange for being exposed to 10 minutes of
commercials spread throughout the movie (approximately 20 television commercials)?

Not at all likely Most likely
1 2 3 4 5 6 7


Pay $1.99 for the same movie in exchange for being exposed to 20 minutes of
advertising spread throughout the movie (approximately 40 minutes of commercials)?

Not at all likely Most likely
1 2 3 4 5 6 7


Pay $2.99 for the same movie with text advertising messages moving across the bottom
of the screen during the entire length of the movie?

Not at all likely Most likely
1 2 3 4 5 6 7


Pay $1.99 for the same movie with text advertising messages moving across the bottom
of the screen during the entire length of the movie?

Not at all likely Most likely
1 2 3 4 5 6 7







^I Please Turn Page Over










8. Please check the items or services that you currently own or subscribe to
(Please check ALL items you own or subscribe to):

0 Digital Video Recorder (e.g., Tivo, Replay TV, or other Personal Video Recorder)
I Personal Computer
D High-Speed Internet
D Video Game System
D Cell Phone
D Home Theater System
D High Definition Television
D Large Screen TV (35" or larger)
D DVD Player
D Personal Digital Assistant (e.g. Palm Pilot or HandSpring)
D Cable TV service with set-top box
D Cable TV service without set-top box
D Direct Broadcast Satellite (e.g. Direct TV or DishNetwork)
D Premium Cable TV Channels (e.g. HBO, Showtime, Starz)


9. Please tell us about yourself:


Gender:
D Male D Female

Highest completed education:
D Less than high school D High School
D College D Graduate School

Marital Status: D Single D Married D Divorced D Widow D Other

Age: I was born in 19

Household Size (including you)

Occupation:
D Laborer D Machine or Service Worker D Craftsman D Professional
D Clerical D Sales OAdministrator D Manager D Other

Household income:


Less than $19,999
$40,000-$59,999
$80,000-$99,999


$20,000-$39,999
$ 60,000-$79,999
$100,000 or more


Thank You Very Much for Your Time.
















REFERENCES


Angwin, J., Grant, P., & Wingfield, N. (2004). Hot-Button Topic: In Embracing Digital
Recorders, Cable Companies Take Big Risk; Viewers Flock to the Devices, But
Advertisers May Flee; Debating Ad-Skip Feature; Time Warner's 'Meteorite'.
The Wall Street Journal, April 26. A-i.

Applebaum, S. (2004). Interactive TV Shows Signs of Life. Advertising Age, May 8. 75,
10-12.

Atkin, D. J., & Jefferes, L. W. (1998). Understanding Internet Adoption as
Telecommunication Behavior. Journal ofBroadcasting & Electronic Media, 42,
4.

Babbie, E. (2001). The Practice of Social Research: 9th edition. Wadesworth,
Thomason Learning, United States.

Baron, S. J., & Davis D. K. (2003). Mass Communication Theory: Foundations,
Ferment, and Future (3rd edition). Wadsworth, Belmont CA.

Baumgartner, J. (2002). Digital Dangle. CED, August 1, 14.

Bass, F. M. (1969). A New Product Growth for Model Consumer Durables. Management
Science, 16, 215-227.

Bass, F, M., Gordon, K., Ferguson, T. L., & Githens, M. L. (2001). DIRECTV:
Forecasting Diffusion of a New Technology Prior to Product Launch. Interfaces
31:3, 82-93.

Brister, C. (2001). Video On Demand; Hollywood Puts Cable Operators On Hold Over
Licensing Fees. Atlanta Journal and Constitution, September 26, 5F.

Brown, P. J. (2002). Cable Looks to the Heavens. Broadcasting & Cable,
November 18.

Cummings, J. (2002). Video On Demand Lets Cable Subscribers Have It Their Way.
Cox News Service, November 18.

Danko, W. D., & MacLachlan, J. M. (1983). Research to Accelerate the Diffusion of a
New Invention. Journal ofAdvertising Research, 23, 3.









Davis, J.J. (1997). Advertising Research: Theory and Practice. Upper Saddle River, NJ:
Prentice Hall.

Dupagne, M. (1999). Exploring the Characteristics of Potential High-Definition
Television Adopters. Journal of Media Economics, 12(1), 35-50.

Eastlick, M. A. (1996). Consumer Intention to Adopt Interactive Teleshopping.
Marketing Science Institute, report summary # 96-113.

Ferguson, D.A., & Perse, E. M. (2001). Enhanced Television Viewing with Digital
Video Recorders (DVRs): Audience Satisfaction in an Asynchronous Television
Environment. Paper submitted in the Open Competition, to the Communication
Technology Division of the Association for Education in Journalism and Mass
Communication.

Figler, A. (2002). Movies Lead, All Others Follow. Cable World, October28, 40.

Iler, D. (2001). VOD Shining Brightly in Cable Universe. Broadband Week, August 6,
18.

Jacobs, R. (1995). Exploring the Determinants of Cable Television Subscriber
Satisfaction. Journal ofBroadcasting & Electronic Media, 39, 2.

Jeffres, L.W., & Atkin, D.J. (1996). Predicting Use of Technologies for Communication
and Consumer Needs. Journal ofBroadcasting & Electronic Media, 40, 318-330.

Jordan, J. (2001). HBO Video On Demand Tests Columbia Market. AssociatedPress
State and Local Wire, June 29.

Kang, M. (2002). Digital Cable: Exploring Factors Associated With Early Adoption.
Journal of Media Economics, 15(3) 193-207.

Klopfenstein, B.C., & Spears, S.C. (1991). VCR Attitudes and Behaviors by Length of
Ownership. Journal ofBroadcasting & Electronic Media, Volume 35, Issue 4.

Lin, C. A. (1994). Audience Fragmentation in a Competitive Video Marketplace.
Journal of Advertising Research, 34, 6.

Lin, C. A. (1998). Exploring Personal Computer Adoption Dynamics. Journal of
Broadcasting and Electronic Media, 42, 95-112.

Lin, C. A., & Jeffres, L. W. (1998). Factors Influencing the Adoption of Multimedia
Cable Technology. Journal of Mass Communication Quarterly, 75(2), 341-352.

Mahajan, V., & Muller, E. (1990). New Product Diffusion Models in Marketing: A
Review and Directions for Research. Journal ofMarketing, 54, 1.









Neuendorf, K.A., Atkin, D. J., & Jeffres, L. W. (1998). Understanding Adopters of Audio
Information Innovations. Journal of Broadcasting and Electronic Media, Winter
1998, 80-93.

National Telecommunications and Information Administration (NTIA) (2002). A
NATION ONLINE: How Americans are Expanding their Use of the Internet,
Washington, D.C. February. http://www.ntia.doc.gov/ntiahome/dn/index.html

Precision Marketing (2004). Nissan Taps into TV Shows for Interactive Ad Campaign.
Precision Marketing. London, March 12, 3.

PR Newswire (2002). Comcast and TVN Expand VOD Relationship; TVN's ADONISS
System to Support VOD Rollout in Philadelphia Market. PR Newswire
Association. July 15.

Rizzuto, R.J. & Wirth, M. 0. (2002). The Economics of Video On Demand: A
Simulation Analysis. The Journal of Media Economics, 15(3), 209-225.

Rogers, E.M. (1976). New Product Adoption and Diffusion. Journal of Consumer
Research. 2 March 1976, 290-301.

Rogers, E.M. (1995). Diffusion of Innovations (4th edition). New York: Free Press.

Romaso, A. (2002). VOD On ESPN. Broadcasting & Cable, September 19.

Ryan, B., & Gross, N. C. (1943). The Diffusion of Hybrid Seed Corn in Two Iowa
Communities. Rural Sociology, 8: 15-24.

Scanlon, M. (2002). Ramping Up Isn't Likely. Cable World, October 28, 40.

Schiesel, S. (2002). Video on Demand Is Finally Taking Hold. New York Times,
November 25, section C, page 4, column 1.

Smith, J. (2001). Video On Demand Awaits Economic Upturn. Rocky Mountain News,
December 31, piB.

Stump, M. (2002). Comcast's VOD Approach: Keep the Package Simple. Multichannel
News, March 11, 1.

Stump, M. (2002a). Showtime, Starz Tie SVOD Bundles. Multichannel News, July 15,
46.

Stump, M. (2003). Ads on SVOD? Works for Magazines. Multichannel News, March
10, 2003.