Communications of the ACM
Home/Magazine Archive/November 2004 (Vol. 47, No. 11)/What Americans Like About Being Online/Full Text
Communications of the ACM
What Americans Like About Being Online
Society is moving away from traditional mass-exposure media, in favor of the interactive interfaces of the Internet [3, 11]. It once was sufficient to advertise on the major television networks in order to market effectively to consumers; now, the Internet is a necessary component of most consumer marketing plans [9]. Even so, Internet-based e-commerce is still in its formative stages [2, 4], particularly in light of the dot-com shakeup, and it continues to be difficult to estimate the effect of Internet use on consumer behavior. Yet, the motivations that bring consumers to the Internet are critical to the success of online businesses [11, 12], since consumers must be online in order to participate in e-commerce [7, 10].
This article reports the results of a survey of online consumers. The survey was executed with the assistance of America Online (AOL) and HotWired.com, to determine the types of gratifications that drive consumer use of the Internet. The reasoning is that if we know what people enjoy about using the Internet, we will be better equipped to motivate them to happily use our part of the Internet in ways that fit with our e-commerce business models.
The survey utilized the results of a qualitative study of the HotWired site [11], in which 98 users were invited to help the site serve them better by providing words that described what they liked about Internet use. The questionnaire was active for one week, and a list of 45 terms resulted from application of a psychological word-association technique [11] designed to elicit trait terms that closely represented user attitudes. The list of terms generated in the qualitative study is displayed in Table 1.
Using the list of 45 descriptive terms as a potential range of gratifying online activities, a questionnaire was developed for a survey of AOL users. In the questionnaire, each descriptive term was linked to a seven-point semantic differential scale anchored by "very important" and "very unimportant." The questionnaire contained instructions for the respondents to use the importance scales to indicate how important or unimportant each characteristic or activity was to them, personally, as a reason for using the Internet.
Some 915 AOL users were recruited during a one-week period at the AOL Opinion Place site. Survey respondents were reasonably balanced across age and gender categories, with 16.6% of respondents between 1824 years of age, 21% between 2534, 19.8% between 3544, 20.4% between 4554, and 22.2% reporting they were 55 years of age or older. Of these, over 48% were men, and 51.9% were women.
Gratifications for Internet Use
Factor analysis was used to identify potential themes among the 45 descriptive terms. Principal components analysis was performed in SPSS 10.0, using the promax oblique rotation method in view of likely correlations between factors [5]. Eight factors emerged with eigenvalues greater than one, and a scree plot of eigenvalues demonstrated a leveling of explanatory power after the fourth factor, indicating a four-factor solution. In view of the oblique rotation method utilized, cross-loadings were expected. Hence, a variable was considered "salient" to a factor if it produced the strongest loading on that factor, despite cross-loadings [5]. In accordance with standards for "very significant" associations between variables and factors [6], loadings of 0.5 or greater were considered. Cross-loadings above 0.5 were subsequently examined as a counterpoint to the interpretation. Table
As an overview, sources of gratification for Internet use among AOL members range from searching and browsing activities, to communication and information retrieval, to using the Internet for socialization. People seem to like using the Web as a search tool; they also appear to appreciate the Internet as a source of information in support of learning and research. Users freely recognize the value of the Internet as a communications medium, and appear to derive personal gratification from using it as a socialization venue. A number of specific factors are involved in these results, including:
Searching, the strongest factor in the analysis, is characterized by terms that relate to using it to find things. The salient variables on this factor are suggestive of searching the Web for information related to computers and their use, given the emphasis on technology, software, resources, and Internet sites.
- Search engines
- Software
- Speed
- Surfing
- Technology
- Web sites
- Work
The loading of "work" on this factor is interesting, given the ubiquity of computers in the modern workplace and their increasing work-related use at home. The fact that "Web sites" loads here seems intuitive, since most Internet searches end with a site, and many begin with one, as well (for instance, Google), but the loading of the "speed" variable is difficult to assess unless it relates to an appreciation for prompt search returns for search efforts.
Five variables cross-loaded at high levels with another factor, and these should be considered in the context of both the main variables of the searching factor, and the likely relationships between searching and the second factor, information:
- Interests
- Resources
- Searching
- Updates
- Variety
Information was the second strongest factor in the analysis, and aside from the eight salient variables that loaded on it, there are five cross-loadings from searching to consider, as well as one cross-loading with a third factorcommunication. The information factor is primarily characterized by terms associated with the information content of the Internet, as evidenced by terms such as "education," "learning," "knowledge," and "ideas." This dimension of gratification seems to reflect user desires for informational content in the online experience:
- Education
- Ideas
- Information
- Knowledge
- Learning
- News
- Research
In addition to the five cross-loadings with the searching factor, the variable "current" cross-loads on the third factor, communication. There is also an anomalous variable loading on this factor: "progressive." Perhaps using the Internet to satisfy one's information needs is a progressive practice. User gratification from informational content online should be considered in the context of the relationships of information to other gratifications of Internet use, such as searching for information or communicating information with others.
Communication highlights terms that reference using the Internet as an interpersonal communications channel. There is a sense of "staying in touch," or connecting interpersonally with others, here:
- Access
- Answers
- Communication
Keeping in mind that "current" cross-loads from the information factor, there is one other variable that influences the likely interpretation of this factor. "Easy" loads cleanly on the factor, but its meaning cannot be fully discerned. AOL respondents had been instructed to indicate in the survey how important "easy" was to them as a reason for using the Internet, but in the response format provided, it was difficult to determine exactly how users perceived the term as related to their online experiences. We suspect ease of use to be an important aspect of AOL's appeal, but it's a speculative point.
Socialization, the fourth and final factor in the analysis, is associated with aspects of communication that further interpersonal relationships. With communication already identified as gratifying, it makes sense that people would also appreciate the social aspects of Internet use, since both characteristics are specific facets of gratification related to using the Internet for interpersonal communications [11, 12].
The socialization factor is similar in many ways to the communications factor, but what distinguishes it might be the more targeted reason for using the Internet as a communication channel. The variables that load on this factor are characteristic of friendly personal interactions:
- Chatting
- Friends
- Fun
- Interaction
- Newsgroups
- People
- Relaxing
It appears that using the Internet for social purposes is an enjoyable activity, in light of loadings from variables such as "relaxing" and "fun." "Newsgroups" is an interesting variable; it does not connote direct, live Internet interactions, although it does provide asynchronous connections between individuals sharing similar interests.
Heavy vs. Light Users
Internet companies like to know what separates the heavy users from the light users, since heavy users are potentially better targets for all sorts of online offerings. Thus, the sample was split to perform an analysis of variance between heavy and light users, based on respondent scores for an eight-point semantic differential questionnaire item that asked them to indicate how frequently or infrequently they used AOL online services. "Heavy users" were defined by usage frequency scores that fit into the top quartile of responses, and "light users" were defined as those who were in the bottom quartile for usage frequency. The quartile split of the sample identified 347 respondents as light users and 401 respondents as heavy users.
Analysis of variance was performed on the salient variables for each usage factor, and results show that heavy users score higher than light users on each of these variables. However, mean values are generally high on all variables across both usage levels. Most Internet uses examined here are considered important in some degree by all users, but some uses are prized even more by the heavier users, while several are considered relatively less important by light users.
As summarized in Table 3, significant differences are found between heavy and light users for all salient variables, and, with several exceptions, both groups demonstrate mean scores on the variables above the mid-point for the scales that were used. In other words, given a seven-point scale (with 4 indicating a neutral point), means were generally above the neutral rating.
One exception was "chatting," which loaded on the socialization factor. "Chatting" scored lowest (3.34) for the light usage group, and also generated the lowest score for the high usage group (4.04). Another exception was "shopping," which did not load on any of the factors, but was analyzed as a standalone variable in light of the rising popularity of e-commerce. "Shopping" produced the second lowest mean for the heavy use group (4.64), and the next-to-lowest mean for the light use group (3.72).
The practical implication of means generally trending above the neutral point for all variables is that AOL customers are not indifferent about what they liked in the Internet experience. These users are rather inclined toward the online experience.
Discussion
The implications of low scores for "chatting" and "shopping" among AOL users are worth considering. Putting the low score for "chatting" into the context of the seven-point scale used, the heavy use mean of 4.11 indicates a neutral opinion, at the midpoint between "very important" and "very unimportant." The light use mean of 3.34 for "chatting" is actually on the "unimportant" side of the neutral point on the scale. This could mean that AOL's Instant Messenger utility does not appeal to users as much as other aspects of the online service.
Since the "chatting" variable loads on the socialization factor along with "friends," "interaction," and "people," we can assume that users perceive "chatting" is an aspect of socializing online. So, why would "chatting" rate so poorly, relative to other social or communication usage gratifications, such as "friends," "interaction," and "communication," which all have strong means? Although there can be no way of knowing without specifically probing the issue in another study, there has been some concern expressed at AOL about how easy Instant Messenger is to set up and use, especially for newer users.
Perhaps some users just don't find AOL chat utilities as easy to set up and use as the AOL email service. This does not mean that Internet use is not socially motivated; far from it. There are important social aspects of the Internet [1]; if the low score for "chatting" leads us to question this assumption, one of the highest scoring variables reinforces the value of online social dynamics. In contrast to "chatting," "email" has the highest mean for the light users, and one of the highest for the heavy users.
It might be that Instant Messenger appeals more to the youth segment of the online marketplace. Research conducted by AOL [8] shows that a vast majority of teen AOL users utilize Instant Messenger, and since our survey was limited to adults of 18 or older (in view of AOL privacy policies) responses could be skewed toward uses that appeal to adults. Though we have no way to empirically assess the point for ages below 18, Instant Messenger does appear to have an attraction for the youth market.
As to shopping, the low importance score for this variable also requires some thought. Ask anyone; they'll tell you online shopping is an important part of Internet use. But conventional wisdom sometimes fails to provide clear guidance for business decisions, and neither light nor heavy AOL users appear to feel that shopping is an important part of their online experience, given the results here. Is it possible that AOL users are simply not serious shoppers? Not likely; a more realistic possibility might relate to do with usage characteristics of the AOL shopping interface, which guides shoppers through AOL-specific shopping areas, as compared to user-initiated navigation in standard Web browsers. Or perhaps the common AOL experience of pop-up sales ads that must be cleared during the log-on process has alienated users from the idea of online purchasing. Given the data on hand, we cannot explain with any degree of certainty why shopping is not more popular among the AOL users surveyed in this study. The fact simply remains that variables associated with searching, browsing, information, and communication ("chatting" aside) all display higher importance scores for online activity among a large sample of AOL users than shopping does.
Are Internet users Lone Browsers or Social Animals? You decide; but, keep in mind that all the factors described here have value.
Conclusion
Any Internet service provider is motivated to maximize enjoyment of the online experience for its heavy users, and to design its interfaces and experiences to entice more usage from light users in order to broaden the heavy usage base. This survey defined four broad dimensions of gratification related to the use of Internet services, and identified areas in which heavy users are different from light users. The strongest usage gratification, and the second strongest, both have much to do with personal enjoyment of Internet-provided utility and functionality, in view of the searching and information factors that were identified. The last two factors, accounting for about one quarter of the explained variance between them, involve the interactive qualities of the Internet in support of interpersonal goals related to communication.
In other words, the most powerful gratifications might be considered utilitarian, related to what the Internet provides in content and experience, while the more moderate gratifications are social. With the caution that our results are specific to the AOL market, a general prescription for increasing the base of heavy users at an ISP might lie in the seamless and trouble-free integration of browsing and information access capabilities into the user interface. To the extent that social dynamics are considered an important selling point for Internet service (and, of course, they are), their weight in promotion and planning should probably be balanced against the strength of the motivational factors related to the personal utility found in Internet content.
Are Internet users Lone Browsers or Social Animals? You decide; but, keep in mind that all the factors described here have value. The comparison between factors of gratification for Internet use is strictly relative, offered for purposes of suggesting optimization strategies for ISP operators. The dimensions of enjoyment for Internet use described here are broadly and generally descriptive of what users seek in the online experience. Any of the four dimensions could be used as a basis for designing, improving, or selling Internet access services. Of course, the obvious attractiveness of heavy users as a potentially profitable customer segment for products and services suggests that the motivational dimensions rated highly by heavy users should be particularly useful for improving online offerings so that they appeal to the more lucrative consumer segments.
Although "shopping" was not one of the highly ranked gratifications for Internet use among this group of AOL members, only the light users rate shopping as mildly unimportant; heavy users just don't rate it as very important, which is a subtle but important distinction. The general philosophy at AOL is that more time spent online inevitably means more opportunities to engage users with e-commerce offerings or fee-for-service promotions that generate revenue and profit for the ISP. The unspoken assumption is that shopping activity increases with Internet use, and more time spent logged on eventually translates into more sales opportunities.
The practical implication is that increased time spent online results in the conversion of more light users to heavy users over time, and a success strategy for online shopping at AOL might well be tied to simply promoting increased use of the service. This study identifies the uses and gratifications of Internet use most prized by heavy AOL users; it is a simple manner to concentrate managerial attention on maximizing the customer experience related to these variables to ensure more extensive use of the Internet service, in the expectation that shopping profits will rise with online experience, and over time with satisfied and repeated use of the interface.
References
1. Armstrong, A.G., and Hagel, J. The real value of online communities. Harvard Bus. Rev. 74, 3 (1996), 134141.
2. Bakos, Y. The emerging role of electronic marketplaces on the Internet. Commun. ACM 41, 8 (Aug. 1998), 3542.
3. Drëze, X., and Zufryden, F. Testing Web site design and promotional content. J. Advertising Research 37, 2 (1997), 7791.
4. Gallaugher, J. Challenging the new conventional wisdom of Net commerce strategies. Commun. ACM 42, 7 (July 1999), 2729.
5. Gorsuch, R.L. Exploratory factor analysis: Its role in item analysis. J. Personality Assess. 68, 3 (1997), 532560.
6. Hair, J.F., Anderson, R.E., and Tatham, R.L. Multivariate Data Analysis. Macmillan Publishing, New York, 1987.
7. Lohse, G.L., and Spiller, P. Electronic shopping. Commun, ACM 41, 7 (July 1998), 8187.
8. Pastore, M. Internet key to communication among youth. CyberAtlas, (2002); cyberatlas.internet.com/big_picture/demographics/article/0,, 5901_961881,00.html.
9. Peterson, R.A., Balasubramanian, S., and Bronnenberg, B.J. Exploring the implications of the Internet for consumer marketing. J. Acad. Marketing Sci. 25 (1997), 329346.
10. Schonberg, E., Cofino, T., Hoch, R., Podlaseck, M., and Spraragen, S.L. Measuring success. Commun. ACM 43, 8 (Aug. 2000), 5357.
11. Stafford, T.F., and Stafford, M.R. Uses and gratifications of the World Wide Web: A preliminary study. In Proceedings of the 1998 American Academy of Advertising Conference. D. Muehling, Ed. Washington State University, Pullman, WA.
12. Stafford, T.F., and Stafford, M.R. Identifying motivations for the use of commercial Web sites. Inf. Resources Manage. J. 14 (2001), 2230.
©2004 ACM 0001-0782/04/1100 $5.00
Permission to make digital or hard copies of all or part of this work for personal or classroom use is granted without fee provided that copies are not made or distributed for profit or commercial advantage and that copies bear this notice and the full citation on the first page. To copy otherwise, to republish, to post on servers or to redistribute to lists, requires prior specific permission and/or a fee.
The Digital Library is published by the Association for Computing Machinery. Copyright © 2004 ACM, Inc.
No entries found