IGI global handbook of research on global diffusion of broadband data transmission jan 2008 ISBN 1599048515 pdf

Handbook of Research on Global Diffusion of Broadband Data Transmission Yogesh K Dwivedi Swansea University, UK Anastasia Papazafeiropoulou Brunel University, UK Jyoti Choudrie University of Hertfordshire, UK Volume I INFORMATION SCIENCE REFERENCE Hershey • New York Acquisitions Editor: Development Editor: Senior Managing Editor: Managing Editor: Copy Editor: Typesetter: Cover Design: Printed at: Kristin Klinger Kristin Roth Jennifer Neidig Sara Reed Jeannie Porter, Maria Boyer Sean Woznicki Lisa Tosheff Yurchak Printing Inc Published in the United States of America by Information Science Reference (an imprint of IGI Global) 701 E Chocolate Avenue, Suite 200 Hershey PA 17033 Tel: 717-533-8845 Fax: 717-533-8661 E-mail: cust@igi-global.com Web site: http://www.igi-global.com and in the United Kingdom by Information Science Reference (an imprint of IGI Global) Henrietta Street Covent Garden London WC2E 8LU Tel: 44 20 7240 0856 Fax: 44 20 7379 0609 Web site: http://www.eurospanonline.com Copyright © 2008 by IGI Global All rights reserved No part of this publication may be reproduced, stored or distributed in any form or by any means, electronic or mechanical, including photocopying, without written permission from the publisher 3URGXFWRUFRPSDQ\QDPHVXVHGLQWKLVVHWDUHIRULGHQWL¿FDWLRQSXUSRVHVRQO\,QFOXVLRQRIWKHQDPHVRIWKHSURGXFWVRUFRPSDQLHVGRHV not indicate a claim of ownership by IGI Global of the trademark or registered trademark Library of Congress Cataloging-in-Publication Data Handbook of research on global diffusion of broadband data transmission / Yogesh K Dwivedi, Anastasia Papazafeiropoulou, and Jyoti Choudrie, editors p cm Summary: “This book explores broadband adoption and the digital divide through a global perspective, it provides research on constructs such as relative advantage, utilitarian outcomes, hedonic outcomes, and service quality From over 100 noted experts in nearly 30 countries, WKLVZRUNDOORZVSROLF\PDNHUV,QWHUQHWVHUYLFHSURYLGHUVDQGRWKHUVWRJDLQPXOWLFXOWXUDOLQVLJKWLQWRZKDWIDFWRUVLQÀXHQFHFRQVXPHUV¶ decisions to adopt broadband” Provided by publisher ISBN 978-1-59904-851-2 (hardcover) ISBN 978-1-59904-852-9 (e-book) Internet users Attitudes Digital divide Internet service providers Consumer behavior Broadband communication systems Globalization Social aspects I Dwivedi, Yogesh Kumar II Papazafeiropoulou, Anastasia III Choudrie, Jyoti TK5103.4.H36 2008 ¶GF 2007052995 British Cataloguing in Publication Data A Cataloguing in Publication record for this book is available from the British Library All work contributed to this book set is original material The views expressed in this book are those of the authors, but not necessarily of the publisher If a library purchased a print copy of this publication, please go to http://www.igi-global.com/reference/assets/IGR-eAccess-agreement.pdf for information on activating the library's complimentary electronic access to this publication To my Mother and Father for their love and blessings…And to my lovingly and much awaited ‘Shagun’ To Athina To my dearest, wonderful, encouraging and supportive Mum, Dad and Bobby Editorial Advisory Board Nikhilesh Dholakia University of Rhode Island, USA Guy Fitzgerald Brunel University, UK Sergio Godoy Universidad Catolica de Chile, Chile Heejin Lee University of Melbourne, Australia Catherine Middleton Ryerson University, Canada Challa Radhakumari Sri Sathya Sai University, India Aradhana Srivastava PRIA (Participatory Research in Asia), India Viswanath Venkatesh University of Arkansas, USA Michael D Williams Swansea University, UK Vishanth Weerakkody Brunel University, UK Table of Contents Foreword xxxi Preface xxxiv Acknowledgment xlvi Volume I Chapter I Broadband Adoption and Diffusion (BAD): A Framework / Yogesh K Dwivedi and Anastasia Papazafeiropoulou Section I National Policies Division I Africa Chapter II South Africa: The Long Walk to Broadband Freedom / Justin Henley Beneke 13 Division II Asia Chapter III Bridging the Digital Divide Through Broadband Deployment / Challa Radhakumari 30 Chapter IV Broadband Policy, Market Competition, and User Adoption in Taiwan / Yu-li Liu 47 Chapter V ICT Competency of Bangladesh to Face Broadband Diffusion / Anwarul Islam and K.C Panda 60 Chapter VI Socio-Cultural Interpretations to the Diffusion and Use of Broadband Services in a Korean Digital Society / Dal Yong Jin 78 Chapter VII Structural Changes and Regulatory Challenges in the Japanese Telecommunications Industry / Hidenori Fuke 90 Division III Australia and New Zealand Chapter VIII $QDO\VLVRIWKH,QÀXHQFHRI$XVWUDOLD¶V*RYHUQPHQW3ROLF\RQ%URDGEDQG,QWHUQHW$FFHVV Qiuyan Fan 109 Chapter IX Broadband for the Mass Market / Roger Saunders 126 Chapter X Competition, Regulation, and Broadband Diffusion: The Case of New Zealand / Bronwyn Howell 139 Division IV Europe Chapter XI Digital Divide and Broadband Access: The Case of an Italian Region / Enrico Ferro, J Ramon Gil-Garcia and Natalie Helbig 160 Chapter XII Improving Broadband Access in Rural Areas / Ingjerd Skogseid 177 Chapter XIII Metropolitan Broadband Networks: Design and Implementation Aspects, and Business Models / Antonios Alexiou, Christos Bouras, John Papagiannopoulos and Dimitris Primpas 196 Chapter XIV Small World: The Irish Broadband Experience / Diana Wilson, Kevin O’Reilly, and Dave Murray 211 Chapter XV Social, Political, and Ethical Responsibility in Broadband Adoption and Diffusion: $*HUPDQ&DVH6WXG\Axel Schulz, Bernd Carsten Stahl, and Simon Rogerson 227 Division V North America Chapter XVI Competition in Broadband Provision and the Digital Divide / Wei-Min Hu and James E Prieger 241 Chapter XVII *RYHUQPHQWDODQG&XOWXUDO)DFWRUVLQ%URDGEDQG$GRSWLRQ Elizabeth Fife, Laura Hosman,and Francis Pereira 260 Chapter XVIII Regulation and the Deployment of Broadband / James E Prieger and Sunhwa Lee 278 Section II Consumer-User Behaviors Division I Australia Chapter XIX Factors Affecting Broadband Adoption for Mainstream Consumers / Peter Adams 306 Division II Europe Chapter XX Developing a Dynamic View of Broadband Adoption / Herbert Daly, Adrina Ortiz, Yogesh K Dwivedi, Ray J Paul, J Santos, and J.M Sarriegi 322 Chapter XXI Employing the Content Validity Approach for Improving the Content of the Broadband Adoption Survey Instrument / Yogesh K Dwivedi, Banita Lal, and Khalil Khoumbati 337 Chapter XXII Inside the Microcosm: A Case Study of a Wireless Internet Hotspot / Pierre Vialle, Olivier Epinette, and Olivier Segard 349 Chapter XXIII 7KH8VHVDQG*UDWL¿FDWLRQVRI%URDGEDQG,QWHUQHW Karianne Vermaas and Lidwien van de Wijngaert 366 Division III Middle East Chapter XXIV Factors Affecting Attitudes towards Broadband Adoption in the Kingdom of Saudi Arabia / Vishanth Weerakkody 380 Division IV North America Chapter XXV Characteristics of Farm and Rural Internet Use in the United States / Peter L Stenberg and Mitchell Morehart 395 Division V South America Chapter XXVI Broadband User Behavior Characterization / Humberto T Marques Neto, Leonardo C.D Rocha, Pedro H.C Guerra, Jussara M Almeida, Wagner Meira Jr., and Virgilio A.F Almeida 408 Volume II Chapter XXVII Precisions about the Broadband Divide in Chile / Sergio Godoy and M Soledad Herrera 427 Section III Small and Medium-Sized Enterprises (SMEs) Division I Asia Chapter XXVIII A Survey on the Adoption and Usage of Broadband Internet / Roya Gholami, John Lim, and Sang-Yong Tom Lee 448 Division II Europe Chapter XXIX Broadband Access and Broadband-Based Applications: An Empirical Study of the Determinants of Adoption Among Italian SMEs / Massimo G Colombo, Luca Grilli, and Cinzia Verga 466 Chapter XXX Broadband Diffusion to SMEs in the UK / Oluwasola Oni and Anastasia Papazafeiropoulou 481 Chapter XXXI Environmental Drivers of E-Business Strategies Among SMEs / Alessandro Arbore and Andrea Ordanini 493 Chapter XXXII ([SORULQJ60(V¶$GRSWLRQRI%URDGEDQGLQWKH1RUWKZHVWRI(QJODQGBoumediene Ramdani and Peter Kawalek 504 Chapter XXXIII External Pressures for Adoption of ICT Services among SMEs / Andrea Ordanini and Alessandro Arbore 524 Section IV Impact on Emerging Applications Division I Entertainment Industry Chapter XXXIV IPTV Business Model Analysis / Kate Carney Landow, Michelle Fandre, Raghu Nambiath, Ninad Shringarpure, Harvey Gates, Artur Lugmayr, and Scott Barker 538 Chapter XXXV The Impact of the Internet on the Law and Economics of the United States Motion Picture Industry / Stanford L Levin, John B Meisel, and Timothy S Sullivan 563 Division II Health Industry Chapter XXXVI Broadband for Health in Developing Countries / Aradhana Srivastava 581 Chapter XXXVII Improving Health Services via Advanced ICT Networks / Peter Farr, Isabelle Ellis, and John Royle 593 Chapter XXXVIII Remote Patient Monitoring in Residential Care Homes: Using Wireless and Broadband Networks / Tanja Bratan, Malcolm Clarke, Joanna Fursse, and Russell Jones 604  while the ITU had recommended at least 1.5 Mbps $UHSRUWE\&KLOH¶V0LQLVWU\RI7HOHFRPPXQLFDWLRQVSXEOLVKHGLQODWHGH¿QHGEURDGEDQGDV any permanent connection with a transfer speed of 128 Kbps at least Ricardo Baeza, an expert who publicly disagreed with the report, argued there was no broadband in Chile, since only 12.5% of connections exceeded 1.0 Mbps (La Tercera, 2005) Meanwhile a report by the Latin American branch of Cisco Systems published in January 2007 stated that 45.5% of Internet connections were of broadband, reaching a total of 1,034,000 lines (Cisco Systems, 2007; Mouse.cl, 2007) Even though bit-rates offered by providers have been increasing in Chile in recent years (as well as “broadband” connections), confusion persists 7KHUHSRUWE\&LVFRLQ&KLOHNHSWGH¿QLQJ broadband as any “always-on” connection even if they were lower than 128 kbps for either downloading or uploading (Cisco Systems, 2007) Therefore, for the sake of simplicity, we will understand as broadband those connections advertised as such by established, legally accountable local providers, provided they are “always on” in contrast to dial-up, pay-per-minute modem links Respondents surveyed by WIP-Chile in 2003, 2004, and 2006 would have been unable to distinguish beyond WKLVGH¿QLWLRQ:HWKHUHIRUHVKDUHWKHGH¿QLWLRQRI broadband used by the Ministry of Telecommunications, which distinguishes between “dedicated” (“always-on”, a.k.a broadband) and “commuted” (dial-up, modem) connections.6 By December 2005, 80.7% of connections were dedicated; this ZDVWKHODWHVWRI¿FLDO¿JXUHLQ&KLOHDYDLODEOHDW the time of writing.7 429 Precisions about the Broadband Divide in Chile RESULTS AND DISCUSSION Measuring the Divides Using the Odds/Ratio Method 7KLVVHFWLRQDWWHPSWVWRGH¿QHWKHGLJLWDOGLYLGH in Chile by using ratios of indicators drawn from :,3&KLOH¶VVXUYH\VRIDQG Operationalization of the digital divide can be done in different ways, yet that would yield noncomparable results (Herrera, 2006) One possibility consists of calculating differences in terms of percentage points concerning a certain indicator, such as access to the Internet between the most favored group and the least favored one Another involves calculating a ratio instead 7KH¿UVWDOWHUQDWLYHLQGLFDWHVWKHUHDOGLIIHUHQFH LQDFFHVVEXWGRHVQRWUHÀHFWWKHRYHUDOOOHYHORI development of this variable within the country For example, take Country A in which the least favored group has 2% of Internet access and a most favored group with 22% This 20% gap would be exactly the same as that of Country B in which the least favored and the most favored groups show a 70% and 90% of access respectively Yet the meaning RIWKH¿JXUHLVFRPSOHWHO\GLIIHUHQW In contrast, calculating a ratio implies comparing one group against the other in relative terms This has the advantage of taking into account the stage of technological development or access to information technologies (IT) of the country anaO\]HG)ROORZLQJWKHSUHYLRXVH[DPSOH&RXQWU\$¶V UDWLRLV  DQG&RXQWU\%¶VLV   So the conclusion is that the gap is considerably lower in B, despite the percentage point difference is the same for both Ratio-based gaps are often wider in countries where the general levels of access are lower Based on the concept of a ratio, one can calculate the difference in percentage points between a country and a referential one used as a yardstick Alternatively, a group of countries can be contrasted For instance, Chile reduced its technological gap with respect to the United States from 77% in 2001 to 70% in 2005 The Santiago Chamber of Commerce (CCS, 2006), through its ICATI Index,8 GHWHUPLQHGWKDW&KLOH¶VWHFKQRORJLFDOJDSYLVjYLV 430 the developed nations fell from 61% in 2000 to 55% in 2005 (CCS, 2006) A similar calculation can be done with respect to the OECD countries, which include the most advanced economies as well as middle-income ones comparable to Chile (Mexico, Turkey, Hungary, Poland, Slovakia, and the Czech Republic) This gap reduction has been more considerable in terms of Internet and mobile-phone access LQVWHDGRI¿[HGWHOHSKRQHOLQHVDQG3&RZQHUVKLS 318' ,Q&KLOH¶VSHUFDSLWDFRQnectivity to mobile phones and the Internet was 77% and 48% of the OECD average respectively According to the Ministry of Telecommunications, ZKLOH¿[HGWHOHSKRQHVVWDJQDWHGDWDERXWRI the population by the late 1990s, mobiles grew IURP DOPRVW QRQH LQ  WR  LQ WKH ¿UVW quarter of 2006 Based in this comparison in relative terms, Di Maggio, Hargittai, Russell, and Robinson (2001) calculated the odds ratio (or “advantage ratio”) as [pj/(1-pj)]/ [pk /(1-pk)], where pj is the probability of the most favored group of having access to IT and pk is the same probability for the less favored group An odds ratio of 6.3 means that the probability of access to IT in contrast to the probability of not having access is 6.3 higher for the most favored group compared to the least favored one (Herrera, 2006) Table 1, Table 2, and Table summarize the odds ratio between the most and least favored groups of Chileans surveyed by WIP Chile in 2003 and 2006 in terms of education (Table 1), income (Table 2), and age (Table 3) Preliminary evidence from WIP countries suggest these are the most relevant variables concerning use and access to the Internet.9 In each one of these tables, data from 2003 and 2006 are divided in rows indicating each type of gap we want to analyze, that is, users vs nonusers of the Internet, home users vs users elsewhere, and broadband home users vs all residential users Additionally, each table shows the contrast between the least and most favored group in terms of percentage point differences as well as ratios Let us start with Table 1, which compares the gap according to education level The least favored Precisions about the Broadband Divide in Chile Table Digital gap ratios in terms of education % least favored % most favored Percentage group group point Favored/not difference favored ratio 2006 2003 Secondary Odds ratio1 education or Higher less education Internet users / sample 27.4 72.4 45.0 2.6 7.0 Home users / users 31.4 68.4 37.1 2.2 4.7 Broadband home users / home users 56.9 59.0 * * * Internet users / sample 20.7 71.2 50.5 3.4 9.5 Home users / users 35.9 73.4 37.6 2.0 4.9 Broadband home users / home users 91.7 86.6 * * * Source: WIP-Chile surveys 2003 and 2006 'LIIHUHQFHVDUHQRWVWDWLVWLFDOO\VLJQL¿FDQWDWYDOXHVS Odds ratio: [pj /(1-pj )]/ [pk /(1-pk )], where pj is the probability of the most favored group of having access to IT and pk is the same probability for the less favored group group reached secondary education or less, whereas their counterparts went to higher education (technical, college, or university) Data did not show a difference between broadband home users and the remaining home users, but it is interesting to note that the odds ratio deteriorated from 2003 to 2006 when analyzing Internet users over overall population Table examines the gap according to income As usual, there is an important difference between the rich and the poor Yet the fast spread of broadband in Chile, which has been even faster among the lower-income groups, account for lower ratios in terms of home broadband connections between 2003 and 2006 Finally, Table looks at the age gap Despite Internet users in general tend to be younger than nonusers, there are interesting differences: whereas the older appear in disadvantage when both “Internet users above sample” and “broadband home users above home users” categories are analyzed, they become the most favored group in the home user category Aged people are more likely than the young to be home users who go online either at school and/or at public places in a higher proportion (WIP-Chile, 2006) Of course, wealthier children and adolescents are heavy broadband users at home as well, but they are a minority among all Web users 7KH8QLWHG1DWLRQV¶'HYHORSPHQW3URJUDPLQ &KLOH 318'   FRQ¿UPV WKDW WKH LQFRPH based gap is narrower among the young, since most of them enjoy Web access at schools regardless of their socioeconomic level Nevertheless, the wealthier youngsters who attend to private, feepaying schools get online mostly at their homes, while those attending to public-sector schooling so at school where usage is more restricted These three types of gap measurement—percentage point difference, ratio, and odds ratio—are rather consistent among them Income and educaWLRQDFFRXQWIRUWKHELJJHVWJDSFRQ¿UPLQJRWKHU studies showing the high dependence of IT access on economic factors (see WIP-Chile, 2006) The probability of being an Internet user as well as a broadband home user is higher in the upper income group In 2003, the main gap in terms of income was related to broadband Yet this changed—in 2006, 72% of the lower income group had this technology at home This is probably related to lower prices of connections and aggressive marketing strategies of providers (mostly cable TV operators) 431 Precisions about the Broadband Divide in Chile Table Digital gap ratios according to income using the WIP Chile surveys of 2003 and 2006 % least favored % most favored Percentage group group point Favored/not High-upper difference favored ratio Odds ratio1 2006 2003 middle income Lower income (ABC1-C2 (C3-D groups) groups) Internet users / sample 22.1 69.7 47.6 3.1 8.1 Home users / users 22.3 77.2 54.9 3.5 11.8 Broadband home users / home users 4.9 72.2 67.3 14.6 50.1 Internet users / sample 31.5 73.3 41.9 2.3 6.0 Home users / users 23.8 86.8 63.0 3.6 21.1 Broadband home users / home users 71.8 91.5 19.7 1.3 4.2 'LIIHUHQFHVDUHQRWVWDWLVWLFDOO\VLJQL¿FDQWDWYDOXHVS Odds ratio: [pj /(1-pj )]/ [pk /(1-pk )], where pj is the probability of the most favored group of having access to IT and pk is the same probability for the less favored group % least favored % most favored group group point Favored/not Odds Older: Younger: difference favored ratio ratio1 46 - 60 years 12 - 29 years Internet users / sample 12.1 61.5 49.4 5.1 11.6 Broadband home users / home users 64.3 57.9 23.1 75.1 Broadband home users / home users 84.1 86.2 2003 Internet users / sample Home users (vs nonhome users) 2006 2006 2003 Table Digital gap ratios according to age using the WIP Chile surveys of 2003 and 2006 Home users (vs nonhome users) Younger: Older: 12 - 29 years 46 - 60 years 50.7 70.4 51.5 79.6 Percentage * * * 52.0 3.3 10.0 * * * 19.8 1.4 2.3 28.0 1.5 3.7 'LIIHUHQFHVDUHQRWVWDWLVWLFDOO\VLJQL¿FDQWDWYDOXHVS Odds ratio: [pj /(1-pj )]/ [pk /(1-pk )], where pj is the probability of the most favored group of having access to IT and pk is the same probability for the less favored group 432 Precisions about the Broadband Divide in Chile Factors Affecting the Digital Divides Depending on which digital divide we are referring to, we can hypothesize that “pro-inclusion” factors are different For example, in the United States, the biggest age divide occurs in terms of the general level of access to the Internet, whereas the biggest income gap exists in terms of the usage of high-speed, broadband Internet (Di Maggio et al., 2001) In Chile, as said before, the main education and age gap occurs between Internet users and nonusers A higher income generally ensures Internet access at home Based on the most important factors pointed out by literature which explain the differences in access, a regression analysis was done to differentiate among the three divides mentioned above and thus assess their importance The variables analyzed include education, income, working status, age, gender, household composition, online social networks, knowledge of the English language, psychological aspects (such as technological motivation), and usage of other communication media (Di Maggio & Hargittai, 2001; Godoy & Herrera, 2004; Herrera, 2005; Keniston, 2004; Lenhart et al., 2003; Norris, 2001; OECD, 2001) According to literature, two main dimensions explain the differentiated access to the Internet: individual resources and social opportunities (De Haan, 2004) We now explain each of them in more detail $PRQJWKHLQGLYLGXDOUHVRXUFHVD¿UVWW\SH relates to material ones such as money and time Income and IT costs affect the ability to purchase these products—the higher the income and the lower the cost, the higher the chance of acquiring them (De Haan, 2004) In general, in countries where Internet penetration is low, the online population is more elitist and thus the socioeconomic gap is wider Even in technologically advanced countries such as Japan and the United States, this break is still highly relevant, yet its relative LPSRUWDQFHLVGHFOLQLQJEHFDXVHRIWKHLQÀXHQFH of other factors (Chen & Wellman, 2004) +DYLQJIUHHWLPHLVDOVRLQÀXHQWLDO'H+DDQ and Huysmans (2002) found that Web users in the Netherlands had an average of three hours of extra spare time than nonusers after controlling background characteristics This relates both to marital status and paternity—those who enjoy more free time are mostly single with no children According to Raban (2004), single persons go online more frequently than married ones On the other hand, other studies discovered that marital status is affected by gender since single women are less online than those who are married because the latter have higher incomes than the former Subsequently, married females are more likely to access to the Internet at their homes (Robinson, Di Maggio, & Hargittai, 2003) Another type of individual resources consists on cognitive ones, also called human capital A ¿UVWRQHLVWKHVRFDOOHG³LQIRUPDF\´UHVRXUFHV that is, the ability to handle information required to make sense of the data available through digital technologies In general, education is a proxy for these variables, given their high correlation (De Haan, 2004; Robinson et al., 2003) Another resource is related to the knowledge of the languages in which most online content is produced, mainly English, yet in this analysis we did not consider this factor because we had not comparable questions along the years An additional factor is the disposition towards technology, since mental barriers can hamper both access and adoption (De Haan, 2004) 2WKHU W\SHV RI LQÀXHQWLDO LQGLYLGXDO IDFWRUV are age and gender In particular, the relationship between age and Internet access is particularly strong In 2002 in the United States, some 80% of individuals aged to 15 years were Internet users The percentage fell to 65% and 60% among those aged 25 and 55 respectively After that age, the rate of users diminishes abruptly (Di Maggio, Hargittai, Celeste, & Shafer, 2003) This age gap is high in most countries, and it has even increased since new users are mainly young (Chen & Wellman, 2004) Meanwhile the link between gender and Internet usage was stronger at the beginning of the expansion of IT Initially men had more access to the Web than women, mostly because of their higher share of the (paid) workforce But after the grow- 433 Precisions about the Broadband Divide in Chile ing access at home and school, the gender gap has decreased (Di Maggio et al., 2003) Nevertheless, in most countries apart from the United States, this gap still remains and it is even increasing in countries such as Germany and Italy (Chen & Wellman, 2004) Additionally, there are social contexts to consider The household, the school, and the workplace represent opportunities for IT access Households not only are endowed with economic resources enabling persons to be online (or not), but also are structured in ways that some of its members can help the rest to adopt the technology (De Haan, 2004) For instance, youngsters who have Web access at their schools or colleges can stimulate the purchase of a computer at their home, and teach their elders how to get online In the United States, families with children have more computers than those without them (Di Maggio et al., 2003) In Chile, a middle-income country often categorized within the so called Third World, Web access at schools is a very important factor explaining the access to the Internet among the young (Godoy & Herrera, 2004) Besides, schools also provide knowledge and skills required for taking advantage of IT Being employed also stimulates the access to IT (De Haan, 2004) The workplace is increasingly becoming more technologically endowed as economies move from manufacturing to services and then to information—even in Chile, according to the BIT-Chile project (Godoy, Herrera, Sepulveda, Lever, & Myrick, 2006) Businesses increasingly use IT to link the different steps of the value chain, linking consumers, suppliers, partners, and/or centers of production The production process is also increasingly more dependent on information, ZKLOHRUJDQL]DWLRQDOVWUXFWXUHVEHFRPHÀDWWHUDQG PRUHÀH[LEOH'L0DJJLRHWDO*RGR\HW al., 2006) All these changes make workers more dependent on digital technologies for their work In the United States, according to the 2001 Current Population Survey, 65% of employees are Internet users in contrast with 37% of those who not work (Di Maggio et al., 2003) Concerning social resources, an important one refers to online social networking According to 434 De Haan, people will be more likely to buy and use new IT if more persons in their social networks share them (De Haan, 2004, p 79) In Chile we also found that having relatives or friends online is a strong predictor of Internet adoption (Herrera, 2005) Usage of other (complementary) communication technologies is also relevant, particularly mobile phones (Di Maggio et al., 2001), in contrast to television In fact, Rice and Katz (2003) note that mobile phone subscribers are more likely to use the Internet than non subscribers Furthermore, the longer a person is a mobile subscriber, the longer that person is a Web user as well Yet despite levels of usage may be similar, the populations concerned are not completely equivalent—especially in developing countries like Chile, where cellular handsets are far more widespread than PCs (70% and 40%, respectively) Subsequently, factors predicting the usage of both technologies are not equal Determining the Factors Affecting the Three Digital Divides Based in these considerations, the following independent variables from the WIP-Chile surveys of 2003 and 2006 were analyzed in order to determine WKHLULQÀXHQFHHGXFDWLRQLQFRPHZRUNLQJVWDWXV age, gender, household composition (whether the interviewed lived with a son/daughter below 18 years of age), perceived importance of IT in general, perceived political importance of the Internet, perceived importance of the Internet as a source of information and entertainment, and usage of mobile phones As there were changes in the WIP questionnaire along the years, these were the only questions that remained constant and were therefore used in the analysis Subsequently we developed three main models of logistic regression, each one for each type of digital divide Table summarizes the percentages of total users, residential users, and broadband users separately according to each category of independent variables As in previous tables, 2003 and 2006 are compared Income relates with all variables in both years: the higher the income, the higher the probability of being a broadband home user Precisions about the Broadband Divide in Chile Table Bivaried relationships between independent and dependent variables 2003 2006 Broadband home Internet user Home Broadband home user Home user user Internet user user Below secondary 27.4 31.4 56.9 20.7 35.9 91.7 Secondary level 23.2 45.7 49.6 39.3 41.9 85.9 university) 72.4* 68.4* 59.0 71.2* 73.4* 86.6 ABC1C2 (high) 69.7* 77.2* 72.2* 73.3* 86.8* 91.5* C3 (middle) 31.8 46.4 33.9 48.7 53.4 81.7 D (low) 22.1 22.3 4.9 31.5 23.8 71.8 Works or studies 12.3 61.7 53.2 20.8 71.1 86.7 studies 40.6* 62.8 56.5 62.5* 57.4 86.3 Male 40.2 56.1 56.4 50.8 62.6 84.9 Female 35.5 55.0 56.8 45.7 54.9 88.1 12-29 61.4* 50.8 57.9 75.1* 51.5 86.2 Higher (technical, Education Income Neither works nor Working Status Gender 30-45 29.2 62.7 51.5 38.5 65.2 88.1 Age 46-60 12.1 70.4 64.3 23.1 79.6* 84.1 Lives with a son/ Yes 37.3 58.3 86.3 daughter aged less than 18 No Perceived importance Pessimistic 36.5 61.6 56.8 55.7* 59.0 86.7 45.5 58.6 88.6 of IT Optimistic 39.1 53.3 57.1 49.5 58.6 85.1 Disagree 49.2* 54.7 57.6 46.2 58.1 82.2 are at a disadvantage Agree 36.2 56.5 56.7 50.8 58.6 88.1 By using the Disagree 52.1* 59.7* 58.2 57.8* 59.3 86.8 21.1 39.8 45.9 36.9 58.7 85.8 46.9* 57.4 58.3 59.6* 59.7 88.6 3HRSOHZKRGRQ¶W access to the Internet Internet people will understand politics better Agree By using the Internet Disagree people will have more say in government Agree 34.4 52.4 49.6 39.4 56.5 81.2 Yes 46.7* 66.3* 59.7 55.6* 61.0* 87.5 Has a mobile phone No 27.6 34.5 44.6 25.7 45.4 75.9 Importance of Not important/a little 23.5 46.8 46.1 35.3 52.9 81.9 information Important 44.7* 57.6 58.4 54.5* 60.3 87.2 Importance of Not important/a little 34.6 57.0 50.9 44.8 57.5 86.7 Important 44.9* 53.7 64.5 57.7* 60.7 85.9 Internet as a source of Internet as a source of entertainment &KLVTXDUHGWHVWVWDWLVWLFDOO\VLJQL¿FDQWDWOHYHOS7KLVVWDWLVWLFDOWHVWZDVSHUIRUPHGDWLQGLYLGXDOOHYHOIURPFRQWLQJHQF\ tables of each “independent variable” with the respective “dependent variables” For example: education (below secondary, secondary level, higher) by Internet user (user, non user) above total sample; education by home user (residential user, non residential user) above users; education by broadband home user (broadband user, modem user) above home users 435 Precisions about the Broadband Divide in Chile 7DEOH5HJUHVVLRQFRHI¿FLHQWVSUHGLFWLQJWKHJDSEHWZHHQXVHUVQRQXVHUVDQG Independent YEAR variables Income Schooling 2003 High and uper middle ABC1C21 2006 B Exp(B) Sig B Exp(B) Sig 1.088 2.970 001 1.763 5.828 000 Middle C31 347 1.415 273 454 1.574 028 Complete or incomplete secondary2 1.236 3.443 072 013 1.013 965 Higher 4.011 55.185 000 1.108 3.028 001 12-29 years3 1.885 6.584 000 2.845 17.199 000 30-45 years3 436 1.547 241 908 2.479 000 Working status Works or studies4 706 2.025 025 1.143 3.137 000 Mobile phone Has5 686 1.986 011 634 1.886 004 Importance of Important for information, not important for 1.028 2.794 001 783 2.189 001 Internet entertainment6 1.164 3.201 001 704 2.022 002 Age Important for information and entertainment6 Cox & Snell R square 439 381 Nagelkerke R square 612 509 Category of comparison: lower income D Category of comparison: lower than secondary level Category of comparison: 46-60 years Category of comparison: does not work nor study  Category of comparison: does not have mobile phone Category of comparison: no or little importance of Internet as a source of information and entertainment Despite education being correlated to income, it GLGQRWVKRZDVWDWLVWLFDOO\VLJQL¿FDQWUHODWLRQVKLS with broadband usage at home Table also reveals that those who work and study are more likely to be online This is somehow related to age: the young are more likely to be Internet users as they have access to it at their schools On the other hand, neither gender nor the perceived importance of IT are related to Internet usage Table also reveals that the lack of children in the household increases the probability of being a Web user, even after controlling by age This apparently contradicts some conclusions from other studies commented above (Di Maggio et al., 2003) Therefore, having children seems to reduce the free time available as well as the extra-family sociability When we checked the relationship between marital 436 status and Internet usage controlled by age, we found that not having a partner was linked to more usage only among the young This is consistent ZLWK5DEDQ¶V(/LYLQJ3URMHFW ZKLFKIRXQG that marital status was the most important factor explaining the difference between “heavy users” and “light users.” The former group was mostly composed by unmarried persons who were not absorbed by the highly demanding child-caring Age has different effects according to which gap is analyzed: the lower the age, the higher the probability of being a user Yet more aged users get connected in a higher proportion at their homes, since they have less chance to so at work or at school Contrary to what may appear logical, users give less importance to the political implications of the Internet as well as to the very fact of be- Precisions about the Broadband Divide in Chile 7DEOH  5HJUHVVLRQ FRHI¿FLHQWV SUHGLFWLQJ WKH JDS EHWZHHQ UHVLGHQWLDO XVHUQRQ UHVLGHQWLDO user, 2003 and 2006 Independent YEAR variables 2003 B Income Schooling Sig B Exp(B) Sig High and uper middle ABC1C2 1.466 4.331 003 2.745 15.569 000 Middle C31 394 1.482 408 1.175 3.237 000 -.162 850 905 069 1.071 878 426 1.530 755 675 1.965 141 119 1.126 826 -.315 730 408 -.239 788 660 -.380 684 357 Complete or incomplete secondary Higher2 Age 2006 Exp(B) 12-29 years 30-45 years3 Working status Works or studies -.417 659 412 -.796 451 035 Mobile phone Has5 736 2.088 049 004 1.004 991 Importance of Important for information, not important for 325 1.385 467 173 1.188 580 Internet entertainment6 292 1.339 526 276 1.318 371 Important for information and entertainment6 Cox & Snell R square 142 272 Nagelkerke R square 194 366 Category of comparison: lower income D Category of comparison: lower than secondary level Category of comparison: 46-60 years Category of comparison: does not work nor study  Category of comparison: does not have mobile phone Category of comparison: no or little importance of Internet as a source of information and entertainment ing connected (they agree less with the statement “persons who have no access to the Internet are at DGLVDGYDQWDJH´