Segmenting Internet shoppers based on their Web-usage-related lifestyle:a cross-cultural validationMalaika Brengman a,b,*,Maggie Geuens b,c ,Bert Weijters c ,Scott M.Smith d ,William R.Swinyard daLimburgs Universitair Centrum,Universitaire Campus,Gebouw D,BEDR-BBK,B-3590Diepenbeek,Belgium bGhent University,Department of Economics and Business Administration,Hoveniersberg 24,9000Ghent,BelgiumcVlerick Leuven Gent Management School,Bellevue 6,Ghent 9050,Belgium dMarriott School of Management,Brigham Young University,Provo,UT 84602USAAbstractOnline surveys in the United States and Belgium were conducted to cross-culturally validate the Internet shopper lifestyle scale [S.M.Smith,W.R.Swinyard,The identification of shopping behaviors among Internet users.World Marketing Congress,Cardiff Business School,2001].Special attention was devoted to sample,construct,and measurement equivalence.In both countries,the same six basic dimensions were found to underlie the scale:Internet convenience,perceived self-inefficacy,Internet logistics,Internet distrust,Internet offer,and Internet window-shopping.Except from having the same basic meaning and structure in Belgium as in the United States,the Web-usage-related lifestyle scale also led to the same segments in both countries.Four online shopping segments (tentative shoppers,suspicious learners,shopping lovers,and business users)and four online nonshopping segments (fearful browsers,positive technology muddlers,negative technology muddlers,and adventurous browsers)are profiled with regard to their Web-usage-related lifestyle,themes of Internet Usage,Internet attitude,psychographic,and demographic characteristics.D 2002Elsevier Inc.All rights reserved.Keywords:Internet segmentation;Internet lifestyle;Cross-cultural research;E-shoppers;Scale validation1.IntroductionRecent findings show that despite the prevalent problems of the industry,online shopping is continuing to undergo significant growth worldwide.Responsible for the growth is the growing number of users in emerging markets that are shopping online for the first time and the increasing con-fidence in online properties.According to a 36-country–42,000interview study by Taylor Nelson Sofres Interactive (TNSi),some 31%of the total adult population uses the Internet,and Internet shoppers worldwide have increased by 50%over the past year (Pastore,2001).Consequently,retailers can use their Internet presence to reach consumers all around the world (Quelch and Klein,1996;Jarvenpaa et al.,1999).It is therefore vital that vendors in cyberspace understand online consumers (Fram and Grady,1995).Asresearch on Internet consumer behavior is still very much in its infancy (Hoffman and Novak,1997;Vellido,2000),the identification of consumer segments has been highlighted as one of the important and necessary avenues of research needed in the field of e-commerce (Chang,1998).To come to a better understanding of the psychology of online shopping,Smith and Swinyard (2001)developed an ‘‘Internet shopper lifestyle’’measurement instrument that was the basis of their segmentation of American online shoppers and nonshoppers.Although lifestyle research has been said to be particularly interesting in cross-cultural research (Kahle,1999),a major criticism is that the cross-cultural validity of international life style instruments remains to be demonstrated (Brunsøand Grunert,1995).Replicating the study of Smith and Swinyard (2001)in Belgium enables us to test for cross-cultural differences in the basic meaning and structure of the Internet shopper lifestyle scale.Furthermore,in view of the fact that Belgium can be considered as an emerging market where consumers still hold a rather negative attitude toward online shopping0148-2963/$–see front matter D 2002Elsevier Inc.All rights reserved.doi:10.1016/S0148-2963(02)00476-9*Corresponding author.Tel.:+32-11-26-8637.E-mail address:malaika.brengman@luc.ac.be (M.Brengman).Journal of Business Research 58(2005)79–88(Geuens et al.,2000),it is interesting to investigate to what extent similar online segments can be distinguished in Belgium and the United States.2.Web-usage-related lifestyle:the relevance of lifestyle segmentationFollowing the idea that market segmentation can give a competitive edge,several demographic-and socioeco-nomic-based Internet segmentation attempts recently emerged(Crisp et al.,1997;Donthu and Garcia,1999). However,demographic and socioeconomic descriptors have been neither the most effective in developing segments (Wedel and Kamakura,2000)nor a good predictor of the propensity to buy online(Vellido,2000).Internet-related webographic characteristics seem more closely related to actual online purchase behavior.Those who actually purchase online appear to have been using the Internet for a longer time(Dahle´n,1999;Novak et al., 2000)to be more frequent Web users(Hoffman et al., 1996)and to spend more time on the Internet(Rangaswamy and Gupta,1999).Moreover,using propensity to adopt Internet shopping as a segmentation base,three segments can be discerned:a group that shops online,a group that has tried to shop online but did not succeed,and a group that has never tried and feels sceptical towards online shopping (Dahle´n,1999;Rangaswamy and Gupta,1999).Brengman and Geuens(2002)distinguish five Internet shopping adopter groups by classifying online shoppers into two groups based on their time of adoption and nonadopters into three groups based on their online shopping intentions. Again,webographics do not tell the full story.Individuals differ in important ways above and beyond demographics and webographics.Starting from consumers’motivations to use the Internet, McDonald(1996)segmented the Internet audience as‘‘avid adventurers,’’‘‘fact collectors,’’‘‘entertainment seekers,’’and‘‘social shoppers.’’Vellido et al.(2002)investigated consumers’opinion on online purchasing and online vendors that seem to consist of the underlying dimensions‘‘control and convenience,’’‘‘trust and security,’’‘‘affordability,’’‘‘ease of use,’’and‘‘effort/responsiveness.’’Using these dimensions as a segmentation base discerns seven segments:‘‘unconvinced,’’‘‘security conscious,’’‘‘undecided,’’‘‘con-vinced,’’‘‘complexity avoiders,’’‘‘cost conscious,’’and ‘‘customer service wary.’’These studies are very valuable, but one can go still further to understand online consumers.Based on the idea that‘‘the more you know and under-stand about consumers,the more effectively you can com-municate and market to them’’(Plummer,1974),the study of people’s values and lifestyles has become a standard tool for both social scientists and marketers around the world. Conceptual and methodological developments aside(Kahle, 1999),lifestyle segmentation instruments have shown value in segmenting markets,especially when combined with more product-specific variables,such as media usage (Wedel and Kamakura,2000).3.Internet shopper lifestyle scale measurementSmith and Swinyard(2001)have developed an instru-ment that encompasses interests and opinions towards the Internet,as well as Web-specific behaviors that increase the likelihood of obtaining relevant online segments.The instrument contains38Internet shopping psycho-graphic statements(Internet shopper lifestyle scale),14 measures of Internet behavior,and13themes of Internet usage.This questionnaire was mailed to a probability sample of4000American online households,of which 1738(43.5%)replied,and e-mailed to20,000e-mail addresses,of which2477(12.4%)replied.A comparative analysis across sample modes showed demographic differ-ences,but no difference in psychographic profiles.Principal components analysis(PCA)with Varimax rotation revealed six factors underlying the Web-usage-related lifestyle scale. Cluster analysis based on these six factors identified four shopper and four nonshopper segments among American online households.Shoppers were defined as household heads with home Internet access who had made an online purchase in the2 months preceding the study.Internet shoppers were divided into these segments:shopping lovers(11.1%),Internet explorers(8.9%),suspicious learners(9.6%),and business users(12.4%).Internet nonshoppers were similarly iden-tified as fearful browsers(10.7%),shopping avoiders (15.6%),technology muddlers(13.6%),and fun seekers (12.1%).Shopping lovers are competent computer users who frequently buy online and really enjoy doing so.Internet explorers believe Internet shopping is fun and could be considered opinion leaders for online buying.Suspicious learners are not very computer literate,but are open-minded for learning new things and are suspicious of giving their credit card number.Business users do not often make personal online purchases.They mainly use the Internet for business purposes and look at the Internet in terms of what it can do for their professional life.Fearful browsers are very computer literate and often practice‘‘Internet-window-shopping.’’They do not buy online for the moment since they distrust the security on the Internet,dislike shipping charges,and are reluctant to buying things without seeing them in person.Shopping avoiders will be difficult to turn into online shoppers since they do not want to wait for product delivery and want to see things in person before they buy.Technology muddlers do not spend much time online,are somewhat computer illiterate,and are not interested in increasing their computer knowledge—an uninteresting target for online retailers.Fun seekers value the entertainment of the Internet,but are afraid of buying online.Furthermore,they have a relatively lowM.Brengman et al./Journal of Business Research58(2005)79–88 80education and income level leaving them not much spend-ing power.4.Cross-cultural validation of the Web-usage-related lifestyle instrumentCollecting data in different cultures with the aim of obtaining comparative results requires that the measurement instrument possesses cross-cultural validity.The purpose of this study is to replicate the study of Smith and Swinyard (2001)in Belgium to determine the cross-cultural validity of Smith and Swinyard’s Internet shopper lifestyle scale.In doing so,both conceptual and methodological issues in cross-cultural research are considered.More specifically, sample equivalence,construct equivalence,and measure-ment equivalence are ensured or tested following the guide-lines offered by Steenkamp and Ter Hofstede(2002). Afterwards,the instrument is used to segment the American and the Belgian online audience.4.1.Sampling equivalenceSampling equivalence is not about obtaining a sample with the same sociodemographic characteristics cross-nationally,nor about drawing a sample representative for the population as a whole,but is about the representative-ness of the sample with respect to the relevant target population(Kumar,2000).The target population in this study is made up of household heads with home Internet and e-mail access.Due to the costs involved in obtaining a sufficiently large representative sample by mail,it was decided not to carry out a mail,but an online e-mail survey. Therefore,both in the United States and in Belgium an online sample was drawn in a more or less identical way. The online American sample was obtained through .E-mail requests to participate in the survey were sent out to20,000representative e-mail users in Spring2001.The response rate was about11%.In Belgium,a random sample of e-mail addresses was also compiled.According to a procedure outlined in Sheehan and Hoy(1999),e-mail addresses were randomly selected from user lists,generated by means of search engines(People Finders),provided by different Belgian Internet Service Providers.An invitation e-mail to visit the site on which the questionnaire was posted was sent out to11,500Belgian Internet users in November2001.A total of2188correctly completed questionnaires were returned(response rate of 19.0%).Since the purpose of the current study is not to provide a cross-cultural segmentation but rather to validate a scale and to compare which segments can be detected in the United States and in Belgium,the sample sizes of the United States and Belgium need not be proportionate to the respect-ive population sizes(Steenkamp and Ter Hofstede,2002). The characteristics of both the American and Belgian online samples are described in Table1.Despite the fact that efforts were taken to obtain repres-entative samples,both the American and Belgian samples turn out to be nonrepresentative of the online populations. Therefore,for subsequent analyses,U.S.data are weighed for age and gender according to the results of Harris Interactive,while the Belgian data is weighed according to the results of the InSites2001telephone survey.4.2.Construct equivalenceIn this study,construct equivalence mainly deals with the question of whether the Internet shopper lifestyle measure-ment instrument is equivalent across the United States and Belgium.Special attention should be devoted to items that are highly American in content and to dimensions that may not be included in the U.S.scale while they are important for Belgium.With respect to the Internet shopper lifestyle scale,we believe construct equivalence problems are min-imal.Items such as‘‘I want to see things in person before I buy’’and‘‘I often go to the Internet for product reviews or recommendations’’are assumed to have the same meaning in the United States and Belgium.Only for the item‘‘I like it that no car is necessary when shopping on the Internet’’and for items referring to the price of an Internet purchase,the meaning may be different for Belgians and Americans.In a densely populated country such as Belgium,the distance to the nearest supermarket or shopping mall is on average very short,but parking troubles may make people hesitant of taking their car.In the United States,it is quite the opposite,parking usually is no problem but distances are on average much longer.Concerning price, we have to take into account that purchases on the Internet typically are relatively more expensive in Belgium than in the United States.The dimensions of the Internet shopper lifestyle scale are expected to cover the main interests and opinions of Belgian Table1Characteristics of the American and Belgian samplesUnited States(%)Belgium(%) GenderMale24.189.1 Female75.910.9Age<18 4.80.519–2922.628.630–4440.846.845–6430.124.2+65 1.70.9 EducationLower26.722.8 Higher73.378.2Hours online at home per week515.233.56–1018.029.711–2028.624.0>2039.212.8M.Brengman et al./Journal of Business Research58(2005)79–8881Internet users.A recent qualitative Belgian study in which eight focus groups were organized with regard to the‘‘ideal future store’’revealed that to Belgians,the dimensions that mattered included convenience of the Internet(not needing a car,time-saving),the costs involved,the delivery prob-lems,the physical experience,and the security threat of insecure payments.However,no aspects that are not cov-ered by the38item Web-usage-related lifestyle scale were found(Geuens et al.,2000).To conclude,the construct equivalence of the Web-usage-related lifestyle scale is assumed moderate to high.4.3.Measure equivalenceMeasure equivalence pertains to whether the variables and items used in the questionnaire are comparable across the United States and Belgium.For a distinction among calibration,translation,and score equivalence,see Steen-kamp and Ter Hofstede(2002).4.3.1.Calibration equivalenceCalibration equivalence deals with the question of whether monetary units,measures of weight,distance,volume,etc. are equivalent across the United States and Belgium.The U.S.questionnaire was composed of the Web-usage-related lifestyle scale,computer-usage items,themes of Internet usage,Internet shopping and Internet spending questions, computer literacy items,demographics,and psychographics.Some of the demographic questions or variables involved posed a problem since they are classified in a different way in both countries.Therefore,these variables were rescaled to broad categories.The education measure,for example,was revised to correspond to the norms of the two countries, where higher education for the United States means that people have a higher degree than high school while for Belgium it indicates a higher degree than secondary school. Similarly,income was classified to better represent the median income of the respective countries.Spending amounts in dollars were translated in euros.Type of housing (apartment,condominium,mobile/trailer,single-unit home) and zip code were omitted in the Belgian study since they were less relevant or not applicable.4.3.2.Translation equivalenceTranslation equivalence pertains to verbatim and mean-ing equivalence.The U.S.questionnaire was translated both in French and Dutch.A second person back-translated the questionnaire and adaptations were made where necessary. Afterwards,the questionnaire was pretested and no prob-lems were encountered.4.3.3.Score equivalenceScore equivalence can be defined as the equivalence of the observed scores on the ck of score equi-valence may be due to cross-national differences(1)in response styles or(2)in responses to specific items/ques-tions.Response styles refer to a tendency to respond in a systematic manner to questionnaire items on some other criterion than what the items were designed to measure (Paulhus,1991).Baumgartner and Steenkamp(2001)out-line five forms of stylistic responses(acquiescence and disacquiescence,extreme response style,midpoint respond-ing,and noncontingent responding),as well as how to measure them and how to correct for these response style biases by regressing the raw summated scale scores on response bias indices.In both samples,our regression analyses revealed signific-ant effects on each of the five response styles.On average,the Belgian respondents scored higher on the acquiescence response style[t(4118)=À4.493],while American respond-ents scored higher on the disacquiescence response style [t(4305)=7.541],extreme response style[t(4128)=8.332], midpoint response style[t(3895)=13.333],and response range[t(4119)=3.299].Therefore,in subsequent analyses,‘‘corrected’’instead of‘‘observed’’scores were used.But most importantly,lack of score equivalence may further arise from cross-national differences.We are espe-cially interested in the question of whether Belgians react differently to the scale than Americans.Multigroup con-firmatory factor analysis was used to test this notion,since there is general agreement that this procedure is the most powerful approach to testing for cross-national measure-ment invariance.As a first step,exploratory factor analysis was carried out on the U.S.sample,followed by confirm-atory factor analysis.Afterwards,multigroup confirmatory factor analysis was employed on both the American and Belgian sample.5.ResultsA Varimax-rotated PCA extracted seven factors that explained56%of the variance.A six-factor solution was identified based on the eigenvalues,screeplot,the interpret-ability of the factors,and the coefficient alphas of the separate factors.Internet convenience,composed of the items:‘‘for me, shopping in stores is a hassle’’(reversed scaled),‘‘I think Internet shopping would avoid the hassle of local shopping,’’‘‘I would like not having to leave home when shopping,’’‘‘I like it that no car is necessary when shopping on the Internet,’’and‘‘I like having products delivered to me at home’’(alpha=.87),Perceived self-inefficacy,with items:‘‘I’d have a hard time searching the Internet to find what I need,’’‘‘I don’t think Internet stores carry things I want,’’‘‘I find the Internet ordering process is hard to understand and use,’’and‘‘I don’t know much about using the Internet’’(alpha=.79),Internet logistics,pointing to the items:‘‘I dislike the delivery problems and backorders of Internet buying,’’‘‘IM.Brengman et al./Journal of Business Research58(2005)79–88 82Fig.1.CFA result for the American and Belgian samples.M.Brengman et al./Journal of Business Research 58(2005)79–8883want to see things in person before I buy,’’‘‘I don’t like having to wait for products to arrive in the mail,’’‘‘it would be a real hassle to return merchandise bought online,’’and‘‘it’s hard to judge the quality of merchandise on the Internet’’(alpha=.75),Internet distrust,reflecting the items:‘‘I don’t want to give out my credit card number to a computer’’and‘‘I worry about my credit card number being stolen on the Internet’’(Pearson r=.65,P<.001),Internet offer,identified by the items:‘‘Local stores have better prices and promotions than Internet stores,’’‘‘I think Internet shopping offers better quality than local stores,’’and‘‘I think Internet shopping offers better selection than local stores’’(reverse-scored)(alpha=.79),andInternet window-shopping,composed of the items:‘‘I often go to the Internet to preview products’’and‘‘I often go to the Internet for product reviews or recommenda-tions’’(Pearson r=.68,P<.001).These dimensions resemble the ones distinguished by Smith and Swinyard(2001),and several of them correspond with the ones previously found by Vellido et al.(1999):our Internet convenience and Internet logistics factors boil down in their control and convenience dimension,our perceived self-efficacy resembles their ease of use factor,and our Internet distrust factor corresponds with their trust and security factor.Our Internet offer and Internet window-shopping factors,however,do not appear in the Vellido et al.(1999)study,while their affordability and effort/cus-tomer service cannot be revealed in our data.Only21of the38original Web-usage-related lifestyle items were retained.Confirmatory factor analysis using AMOS indicated an acceptable fit for the six-factor structure (G FI=.924,A F G I=.899,T L I=.897,C F I=.914, RMSEA=.066).If the same pattern of factor loadings can be found in the Belgian sample,the basic meaning and structure of the Web-usage-related lifestyle scale is said to be equivalent in the United States and Belgium.The fit of the multigroup confirmatory factor analysis model was satisfact-ory(GFI=.927,AGFI=.903,TLI=.887,CFI=.906, RMSEA=.046)indicating that the same factors underlie the Web-usage-related lifestyle scale in Belgium as in the United States.Factor loadings for both countries are shown in Fig.1.6.Online segments in the United States and BelgiumAs is common in segmentation studies,we clustered respondents on the factors found in the previous analysis,a practice also referred to as the tandem approach(Vellido et al., 2002;Steenkamp and Ter Hofstede,2002).The cluster analysis was done separately for Internet shoppers and Inter-net nonshoppers for each of the American group and Belgian group.Several segments resemble the ones detected in Smith and Swinyard’s previous study.Furthermore,the segment profiles are very similar for the United States and Belgium.Tables2and3show the profiles of the eight segments with respect to their scores on the Web-usage-related lifestyle dimensions,as well as on computer-usage themes,psycho-graphics,demographics,and attitude towards the Internet.6.1.Internet shopper segmentsAmong Internet users with online shopping experience in the past2months,four segments could be identified:‘‘tentative shoppers,’’‘‘suspicious learners,’’‘‘shopping lov-ers,’’and‘‘business users.’’Cluster1,the tentative shoppers,scores moderate to low on all Web-usage-related lifestyle factors.Although they have shopping experience and are not suspicious of giving their credit card number,they do not really enjoy online shopping.Moreover,they are not convinced of the fact that online shopping provides more convenience or a better offer. They are not excited about exploring new websites,and do not consider the Web as a real contribution to their life.Although they e-mail friends and family,they do not excel in social,informative,entertainment,or business activities on the net.Representing14%of the Internet shoppers in the United States and13.9%in Belgium, tentative shoppers form a rather large segment.Retailers should try to give them an added value they appreciate to increase their online shopping frequency.Cluster2,the suspicious learners,is the least computer-and Internet-literate.From all segments,they are the most fearful one and hold the most negative opinion on Internet logistics,the online shopping offer,and the convenience of the Internet.They are least likely to practice online window-shopping,make the least online purchases,and use the Internet the least(whether it is for fun,information,or business).Their negative attitude may be a result of their computer and Internet illiteracy.This tough segment repre-sents15.6%of the American online shoppers and8.9%of Belgian er-friendly procedures,training and guidance,and a solution for online security problems may turn this segment in convinced shoppers although patience probably is needed.Cluster3,the shopping lovers,is a dream for online retailers.They love Internet shopping and all its facets(the convenience,the offer,and the possibility of window-shopping).They frequently buy online,use the Internet for information,fun,and business,and are really excited about exploring millions of websites.About9.8%of the American Internet shoppers belong to this segment,while in Belgium it is as large as15.3%.Representing ideal opinion leaders,Internet retailers should try to encourage them even more to spread positive word-of-mouth.Cluster4,the business users,includes confident com-puter and Internet users who use the Internet mainly for competitive and business purposes.They love to explore websites and see the Internet as a contribution to their life. Business users represent12%of the American and10.5%of the online Belgian shoppers.M.Brengman et al./Journal of Business Research58(2005)79–88 84The shopping lovers and business users segments are identical to the ones described by Smith and Swinyard (2001).Suspicious learners in the present and Smith and Swinyard’s study share some characteristics,but an import-ant difference is that in the present study’s suspicious learners do not trust to give their credit card to the computer, making it a less interesting target segment for Internet retailers.Moreover,Smith and Swinyard’s adventurous explorers had to be replaced by the less Internet-enthusiastic segment,tentative shoppers.The resemblance of American and Belgian segments, especially the suspicious learners and shopping lovers,is striking.On every variable,the differences between the segments point in the same direction,the only exception being‘‘percent gifts bought online.’’Surprisingly,Belgian shopping lovers score extremely low on this.A possible explanation could be that Belgians are less inclined to use online shopping and express mail to send gifts(flowers, books,etc.),but to shop for gifts in traditional stores. Further research should shed more light on this.For tent-ative shoppers and business users,a few more differences were detected,although it is clear that the American and Belgian clusters are the same.6.2.Internet nonshopper segmentsAmong Internet users with no online shopping experience in the past2months,four segments could be discerned:‘‘fearful browsers,’’‘‘positive technology muddlers,’’‘‘nega-tive technology muddlers,’’and‘‘adventurous browsers.’’Table2Profiling the different Internet shopper segmentsP Tentative Shoppers Suspicious learners Shopping lovers Business usersUnited States Belgium UnitedStatesBelgium UnitedStatesBelgium UnitedStatesBelgium UnitedStatesBelgiumWeb-usage lifestyleF1:Internet convenience<.01<.01À+ÀÀÀÀ+++++ÀF2:Perceived self-inefficacy<.01<.01+++++ÀÀÀÀÀF3:Internet logistics<.01<.010À+++ÀÀÀÀ0+F4:Internet distrust<.01<.01ÀÀ++++ÀÀÀÀ++F5:Internet offer<.01<.01À+ÀÀÀ+++++ÀF6:Internet window-shopping<.01<.01ÀÀÀÀÀ++++++ Internet usageShopping1<.01<.01+0ÀÀÀÀ++++À0Fun2<.01<.010ÀÀÀÀ+++++ Information3<.01<.010ÀÀÀ+++0ÀBusiness4<.01<.01ÀÀÀÀÀÀÀ+++++++E-mail.77>.01ns0ns0nsÀns+ WebographicsOnline spending last2months<.01.070nsÀns+nsÀns Hours per week online.10<.01ns+ns+nsÀns0%gifts bought online<.01<.0100ÀÀ++++ÀÀ00%gifts bought in real stores<.01.21+ns+nsÀÀns+ns Computer literacy5<.01<.01+ÀÀÀ++++++ Internet literacy6<.01<.01+ÀÀÀ++++++ DemographicsAge.22<.01ns+ns0nsÀÀnsÀEducation<.01.37++nsÀns0ns+ns Personal income.05<.01++À++ÀÀÀ0 Attitude towards the InternetExcited to explore websites<.01<.01ÀÀÀÀÀÀÀÀ++++++ Love Internet shopping<.01<.0100ÀÀÀ++++00The Web contributes to my life<.01<.01ÀÀÀÀÀ++++++ Scale:ÀÀvery lowÀlow0moderate+high++very high.1Shopping:tickets and reservation,retail sites,auctions,and financial information(alpha=.62).2Fun:games,downloading photos and images,chat,and software(alpha=.63).3Information:news and magazines and hobby sites(r=.51,P<.001).4Business:use Internet to conduct business,get more done,and be more competitive(alpha=.65).5Computer literacy able to use word-processing programs,install computer software,configure computer drivers,fix system problems,and install an operating system(alpha=.88).6Internet literacy able to e-mail,browse the Internet,use a search engine,make an online purchase,find the best price on the Internet,use an Internet shopping bot,find Internet retailer ratings(alpha=.81).M.Brengman et al./Journal of Business Research58(2005)79–8885。
